A Method and System for Determining Collaborative Construction Schemes for Multiple Hydropower Stations Based on Privacy-Preserving Computation

By performing local data processing at the hydropower station through privacy computing, a common operating condition center and representative scenarios are formed. Combined with physical verification, the problems of data exposure and inaccurate evaluation of construction plans in the collaborative operation of multiple hydropower stations are solved, and higher consistency and reliability of collaboration and construction plan determination are achieved.

CN123122018APending Publication Date: 2026-11-17NORTHWEST ENGINEERING CORPORATION LIMITED
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611627582.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-21
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In the current collaborative operation of multiple hydropower stations, the centralized uploading of data increases the scope of operational information exposure. Fixed models are difficult to adapt to the collaborative processing of different projects, and the evaluation of construction schemes cannot reflect the actual technical contributions of different alliance combinations and construction capacities, resulting in inconsistencies between the calculation boundaries and actual operating conditions.

Method used

By using a privacy-based computation method, local privacy computation is performed at the station to form protected aggregate quantities, identify common operating condition centers and representative scenarios, determine scenario boundaries by combining the current prediction status, coordinate local results within the feasible domain of hydraulic and power tasks, perform physical verification using a complete local model, and determine the construction plan and construction responsibility allocation for the target shared facilities.

Benefits of technology

It improves the consistency between the collaborative results of multiple hydropower stations and actual physical constraints, enhances the reliability of the construction plan, reduces the risk of raw data exposure, and ensures that the evaluation results are in line with current operating conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123122018A_ABST
    Figure CN123122018A_ABST
Patent Text Reader

Abstract

This disclosure provides a method and system for determining collaborative construction schemes for multiple hydropower stations based on privacy-preserving computation, relating to the field of data processing technology. The method includes: generating task identifiers; establishing a mapping between participating entities, station nodes, and callable assets, and binding control authorization; each station node performing local privacy computation when the original data does not leave the station, determining the common operating condition center, representative scenario, scenario weight, and scenario boundary when both thresholds are met; forming evaluation scenarios for alliance entity combinations and candidate shared facility construction schemes, closing uncallable assets and reconstructing the task feasibility domain, coordinating local station results and verifying them through a complete local model, determining and distributing target candidate operating results; conducting technical evaluation based on the verification results, determining the target shared facility construction scheme, construction capacity, and construction responsibility allocation ratio; and invalidating and recalculating scenarios when binding relationships, candidate schemes, construction capacity, verification results, or execution feedback change.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This disclosure relates to the field of data processing technology, and more specifically, to a method and system for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing. Background Technology

[0002] Conventional hydropower stations, pumped storage power stations, and their associated reservoirs, water conveyance channels, transmission lines, and communication facilities are interconnected by hydraulic and electrical systems. When multiple hydropower stations operate collaboratively, the inflow, reservoir capacity, unit status, and transmission capacity of each station influence each other. Current methods for joint operation of multiple hydropower stations typically employ centralized processing, where each station uploads its hydrological, equipment, and operational data to a central computing platform for unified calculation. However, different hydropower projects differ in data fields, sampling periods, equipment codes, topological relationships, and control authorization scope. Fixed models are insufficient to adapt to the collaborative processing of different projects, and centralized uploading of raw data increases the exposure scope of station-level operational information.

[0003] Meanwhile, existing joint operation models are typically built upon fixed participating entities, equipment, capacity, and topology relationships. When participating entities, control authorization, communication status, facility capacity, or topology relationships change, the original model may continue to utilize currently unavailable units, reservoir capacity, water transmission channels, outgoing lines, or control capabilities, leading to inconsistencies between computational boundaries and actual operating conditions. Although privacy-preserving computation can reduce raw data exchange, local statistical results obtained at the central side may still reflect station-level operational information, and the numerical convergence results at the central side cannot replace the verification of operational results by the complete physical model at the station level.

[0004] Furthermore, using fixed proportions or static capacities for evaluation during the determination of shared facility construction plans makes it difficult to reflect the actual technical contributions of each participating entity under different alliance combinations and construction capacities. When relevant task conditions change, continuing to use the original evaluation results may also affect the accuracy of the construction plan determination. Therefore, improvements are still needed to the evaluation process for the collaborative operation and construction plans of multiple hydropower stations. Summary of the Invention

[0005] The purpose of this disclosure is to provide a method, system, electronic device, and computer-readable storage medium for determining collaborative construction schemes for multiple hydropower stations based on privacy computing, which can perform cross-station privacy computing without the original data leaving the station. It can dynamically reconstruct the feasible domain of the task for different alliance entity combinations and candidate shared facility construction schemes, and combine station-side physical verification, operation feedback, and recalculation of affected scenarios for collaborative processing, thereby improving the consistency between the collaborative results of multiple hydropower stations and the actual physical constraints, as well as the reliability of the construction scheme determination.

[0006] Other features and advantages of this disclosure will become apparent from the following detailed description, or may be learned in part by practice of this disclosure.

[0007] According to a first aspect of the present disclosure, a method for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing is provided, comprising: generating a task identifier; establishing a mapping relationship between participating entities, station-end nodes corresponding to multiple hydropower stations, and callable assets associated with multiple hydropower stations; binding the mapping relationship and corresponding control authorization to the task identifier; each station-end node performs local privacy computing based on the task identifier under the condition that the original data of the corresponding hydropower station does not leave the station, forming a protected aggregate quantity; when the number of valid station-end nodes submitting the protected aggregate quantity is not less than the node threshold and the number of valid samples is not less than the sample threshold, determining the common operating condition center, representative scenario, and scenario weight based on the protected aggregate quantity, and determining the scenario boundary in combination with the current prediction state; determining multiple alliance entity combinations and multiple candidate shared facility construction schemes; forming an evaluation scenario corresponding to the combination of each alliance entity combination and each candidate shared facility construction scheme based on the representative scenario and scenario weight; and closing uncallable assets and reconstructing the water corresponding to the evaluation scenario according to the binding relationship of the task identifier and the scenario boundary. The feasible domain for hydraulic and electric tasks is defined as follows: Within this domain, local results from various stations are coordinated to obtain candidate operational results. Physical verification results from each station, based on a complete local model, are received. Target candidate operational results for actual operation are determined from the physically verified candidate operational results. After control authorization and security verification, these results are sent to the corresponding stations, and execution feedback is received. Based on the physically verified evaluation scenario results, a technical evaluation is conducted on the combination of each alliance entity and each candidate shared facility construction scheme. Technical evaluation results are obtained, and the target shared facility construction scheme, its corresponding construction capacity, and the construction responsibility allocation ratio for each participating entity are determined based on these results. When any of the following changes occur: the binding relationship of the task identifier, the candidate shared facility construction scheme, its corresponding construction capacity, the physical verification results, or the execution feedback, the existing results affected by the corresponding change are invalidated, and the scenario affected by the corresponding change is recalculated.

[0008] In an example embodiment of this disclosure, based on the aforementioned scheme, the step of performing local privacy calculations based on the task identifier to form a protected aggregate quantity, and determining the common working condition center, representative scenario, and scenario weight according to the protected aggregate quantity when the number of valid stations submitting the protected aggregate quantity is not less than the node threshold and the number of valid samples is not less than the sample threshold, includes: each station assigning its local running window to different working condition categories, and performing pruning, fixed-point encoding, and masking processing on the working condition count and the sum of working condition feature vectors corresponding to each working condition category to form the protected aggregate quantity; when the number of valid stations submitting the protected aggregate quantity is not less than the node threshold and the number of valid samples is not less than the sample threshold, restoring the cross-station combined metering corresponding to each working condition category, and determining the common working condition center, representative scenario, and scenario weight according to the working condition feature vector. The common operating condition center is determined by the ratio of the total cross-station combined metering to the total cross-station operating condition count. The corresponding scenario weight is determined based on the proportion of operating condition counts for each operating condition category to all valid operating condition counts. A representative scenario is generated for operating condition categories with cross-station operating condition counts greater than zero. No representative scenario is generated for any operating condition category with a cross-station operating condition count of zero. The scenario boundary corresponding to the representative scenario is determined based on the common operating condition center and the current prediction state, where the current prediction state includes the current predicted water inflow, water level, equipment status, and node injection. The cross-station combined metering and the scenario weight are not released to the central ordinary calculation process if either the number of valid stations or the number of valid samples fails to reach the corresponding threshold.

[0009] In an example embodiment of this disclosure, based on the aforementioned scheme, the method for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing further includes: each station generating a temporary key for the task identifier and task session, and generating paired masks with opposite signs between each station, superimposing the paired masks onto the sum of the operating condition count and the operating condition feature vector, so that the paired masks corresponding to the online station cancel each other out during the cross-station summation process; wherein, the temporary key and the paired mask are not reused across task identifiers; for a station that goes offline after submitting the protected aggregate quantity, when the offline proof is passed, the recovery share is not less than the recovery threshold, and the recovery share verification is passed, the uncancelled mask corresponding to the offline station is restored, and the mask corresponding to the still online station is not restored; when any of the following occurs: the offline proof fails, the recovery share is less than the recovery threshold, or the recovery share verification fails, the uncancelled mask corresponding to the offline station is not restored.

[0010] In an example embodiment of this disclosure, based on the aforementioned scheme, the method for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing further includes: using either a trusted execution environment or a security comparison protocol to compare the number of effective samples with the sample threshold, wherein the central ordinary computing process only obtains the threshold passing flag and the cross-station combined metering after the threshold is passed; and for multiple tasks formed based on the same data window version and whose corresponding data windows overlap, the query frequency of the multiple tasks is limited.

[0011] In an example embodiment of this disclosure, based on the aforementioned scheme, the step of generating a task identifier, establishing a mapping relationship between participating entities, station-end nodes corresponding to multiple hydropower stations, and callable assets associated with the multiple hydropower stations, and binding the mapping relationship and corresponding control authorization to the task identifier, includes: identifying the participating entity with a subject identifier, identifying the station-end node with a node identifier, and identifying the callable asset with an asset identifier, wherein the callable asset includes at least one of generating units, reservoir capacity, water conveyance channels, transmission lines, and control capabilities; determining the attribution relationship between the subject identifier and the node identifier, and the relationship between the node identifier and the... The callable relationship between asset identifiers is used to form the mapping relationship based on the ownership relationship and the callable relationship, and the corresponding control authorization certificate is recorded; the participating entity snapshot, station node snapshot, callable asset, mapping relationship, field whitelist, model version, authorization version, output permission version and topology version are determined as the task boundary corresponding to the task identifier, and the task boundary is kept unchanged under the same task identifier; when any item constituting the task boundary changes, a new task identifier is generated; when the task boundary does not change but an execution deviation occurs, a new candidate running result version is generated under the original task identifier.

[0012] In an example embodiment of this disclosure, based on the aforementioned scheme, closing uncallable assets includes: configuring at least one necessary node group for each of the callable assets; for any callable asset, when the participation status, control authorization status, communication status, and task execution verification status of all station nodes within at least one of the necessary node groups are all valid, and the callable asset is located within the current valid topology connectivity component, the asset activation status of the callable asset is determined to be activated; otherwise, the asset activation status of the callable asset is determined to be deactivated; based on the callable assets in the activated state... The available capacity under the current task is determined by the common upper limit of the asset's physical capacity, authorized capacity, task registration capacity, and topology connectivity. The positive capacity indicator is determined based on whether the available capacity is greater than zero. The sum of the generator's power generation status flag and pumping status flag is constrained to be no greater than the product of the asset's activation status and the positive capacity indicator. The generator's power generation flow rate is limited to no greater than the product of the power generation status flag and the maximum allowable flow rate corresponding to the power generation status. The generator's pumping flow rate is also limited to no greater than the product of the pumping status flag and the maximum allowable flow rate corresponding to the pumping status.

[0013] In an example embodiment of this disclosure, based on the aforementioned scheme, the reconstructing of the feasible domain of the hydraulic and power tasks corresponding to the evaluation scenario includes: closing the variables and capacities corresponding to callable assets that do not belong to the current alliance entity combination; regenerating water balance constraints, reservoir capacity continuity constraints, unit operating condition constraints, node power balance constraints, line capacity constraints, and shared boundary consistency constraints according to the scenario boundary to form the feasible domain of the hydraulic and power tasks corresponding to the current evaluation scenario; when any of the following occurs: physical waterway disconnection, lack of reference nodes for necessary electrical connectivity components, or the shared boundary projection set being empty, the corresponding evaluation scenario is determined as a physically infeasible scenario, and no candidate operating results corresponding to the physically infeasible scenario are generated.

[0014] In an example embodiment of this disclosure, based on the aforementioned scheme, the method for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing further includes: updating the authorization version when the control authorization changes; updating the topology version when the connectivity of either the water conveyance channel or the transmission line changes; generating a new task identifier when either the authorization version or the topology version changes; and switching the main transmission line to the registered backup line when the main transmission line is disconnected and a registered backup line exists, updating the topology version, and reconstructing the corresponding hydraulic and power task feasible domains based on the new task identifier according to the capacity of the registered backup line.

[0015] In an example embodiment of this disclosure, based on the aforementioned scheme, the step of coordinating the local results of each station within the feasible domain of the hydraulic and power tasks to obtain candidate operating results includes: performing station-level local updates, central shared boundary updates, and multiplier updates based on the same boundary mapping, wherein the same boundary mapping is used to establish the correspondence between local variables of each station and the central shared boundary; determining local consistency residuals, global physical conservation residuals, adjacent iteration shared boundary change indices, and local feasibility flags corresponding to each station based on the station-level local updates, the central shared boundary updates, and the multiplier updates; generating fixed candidate operating results and corresponding candidate operating result versions when the local consistency residuals, the global physical conservation residuals, and the adjacent iteration shared boundary change indices are all not greater than the corresponding thresholds, and all local feasibility flags corresponding to all stations indicate local feasibility; and not publishing the candidate operating results when the maximum number of iterations is reached and the conditions that the local consistency residuals, the global physical conservation residuals, and the adjacent iteration shared boundary change indices are all not greater than the corresponding thresholds, and all local feasibility flags corresponding to all stations indicate local feasibility are not simultaneously met.

[0016] In an example embodiment of this disclosure, based on the aforementioned scheme, the process of updating the central shared boundary includes: determining the target conservation set according to the task topology template; wherein the target conservation set includes at least one of the cross-station water conservation set and the power conservation set; determining the intersection of the shared capacity set and the target conservation set; when the intersection is not empty, performing weighted least squares projection on the boundary mapping results corresponding to the local variables at each station within the intersection, and updating the central shared boundary according to the projection results; when the intersection is empty, not generating the candidate running results, and reconstructing the feasible domain of the hydraulic and electric tasks.

[0017] In an example embodiment of this disclosure, based on the aforementioned scheme, receiving the physical verification results returned by each station based on the complete local model for the candidate operation results includes: each station, for a fixed candidate operation result, performs item-by-item checks on water balance, unit operating conditions, water level and reservoir capacity boundaries, line injection boundaries, and station safety constraints based on the complete water level and reservoir capacity curve, complete unit characteristics, safe operating range, and station line constraints, and determines the normalized violation amount of each constraint according to the corresponding physical verification scale; when all the normalized violations are not greater than the corresponding threshold, a physical verification pass result is returned; when at least one of the normalized violations is greater than the corresponding threshold, a physical verification fail result is returned, and only the constraint category, the maximum normalized violation amount, and the allowable adjustment boundary are returned, without returning the complete original data used to determine the normalized violation amount; for evaluation scenarios where the physical verification fails, the corresponding evaluation scenario results are not used for the technical evaluation.

[0018] In an example embodiment of this disclosure, based on the aforementioned scheme, the step of sending the target candidate operation result to the corresponding station after the control authorization verification and security verification, and receiving execution feedback, includes: setting a candidate operation result version for the candidate operation result, and generating a control plan version according to the target candidate operation result, wherein the control plan version is bound to the candidate operation result version corresponding to the target candidate operation result; wherein, the control plan version includes at least one of unit power generation, pumping power, flow rate, water level boundary, and line injection limit; when both the control authorization verification and the security verification pass, providing the control reference value corresponding to the control plan version to the existing control interface at the station, and ensuring that the control reference value does not cover the existing protection and interlocking at the station; normalizing the difference between the observed value and the control reference value according to the feedback scale to obtain the feedback deviation, and binding the execution feedback including the feedback deviation to the control plan version and the candidate operation result version corresponding to the target candidate operation result.

[0019] In an example embodiment of this disclosure, based on the aforementioned scheme, the technical evaluation of the combination of each alliance entity combination and each candidate shared facility construction scheme based on the evaluation scenario results that have passed physical verification, to obtain the technical evaluation results, includes: obtaining at least one hydraulic technical indicator and at least one of power safety indicators and reliability indicators corresponding to the evaluation scenario results that have passed physical verification; wherein, the hydraulic technical indicator includes at least one of water wastage and operating water loss, and the power safety indicators and reliability indicators include at least one of power supply deficit, line congestion, reserve shortage, and constraint violation; dividing each of the indicators by the corresponding pre-fixed positive scale to obtain each normalized indicator; weighting each of the normalized indicators according to the preset technical loss weight to obtain the scenario technical loss of the corresponding evaluation scenario, and using the scenario technical loss as a component of the technical evaluation results; wherein, when any indicator used to determine the scenario technical loss is missing and cannot be supplemented based on audited data under the same task identifier, the corresponding evaluation scenario is not used for technical evaluation.

[0020] In an example embodiment of this disclosure, based on the aforementioned scheme, the technical evaluation result includes technical value and marginal technical contribution. The process for determining the technical value and the marginal technical contribution includes: for any combination of alliance entities and any candidate shared facility construction scheme, obtaining the reference technical loss of the combination of alliance entities under a fixed reference scenario, and the candidate technical loss of the combination of alliance entities after adopting the candidate shared facility construction scheme and completing coordination and physical verification; determining the technical value of the combination of alliance entities under the candidate shared facility construction scheme based on the difference between the reference technical loss and the candidate technical loss; for any participating entity, under the same candidate shared facility construction scheme, determining a first combination of alliance entities including the participating entity and a combination of the first... The alliance entity combination removes the participating entities to form a second alliance entity combination. The marginal technical contribution of each participating entity is determined based on the difference between the technical value of the first alliance entity combination under the candidate shared facility construction scheme and the technical value of the second alliance entity combination under the same scheme. The fixed reference scenario corresponding to the alliance entity combination and the evaluation scenario formed by the alliance entity combination using the candidate shared facility construction scheme employ the same technical indicator caliber, positive scale, technical loss weight, model version, and physical verification rules. The reference technical loss is determined based on the original technical loss indicator, the positive scale, and the technical loss weight corresponding to the fixed reference scenario, and a station-side physical verification is performed on the fixed reference scenario.

[0021] In an example embodiment of this disclosure, based on the aforementioned scheme, the method for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing further includes: for any participating entity, under the same candidate shared facility construction scheme, weighting the marginal technical contribution corresponding to the different joining order of the participating entity in different alliance entity combinations to obtain the comprehensive marginal technical contribution of the participating entity; using the comprehensive marginal technical contribution as one of the bases for determining the construction responsibility allocation ratio, and combining at least two of the physically verified shared facility utilization contribution, new capacity demand, and reliability contribution to determine the construction responsibility allocation ratio.

[0022] In an example embodiment of this disclosure, based on the aforementioned scheme, the step of weighting the marginal technical contribution of any participating entity under the same candidate shared facility construction scheme according to the marginal technical contribution corresponding to the different joining order of the participating entity in different alliance entity combinations to obtain the comprehensive marginal technical contribution of the participating entity includes: when the number of participating entities is not greater than the precise enumeration upper limit, performing hydraulic and power task feasible domain reconstruction, local result coordination, and physical verification on all alliance entity combinations respectively; determining the marginal technical contribution of each participating entity under different alliance entity combinations according to the technical value corresponding to each alliance entity combination; and determining the marginal technical contribution based on the marginal technical contribution. The comprehensive marginal technical contribution of each participating entity is calculated. When the number of participating entities exceeds the upper limit of the precise enumeration, the comprehensive marginal technical contribution of each participating entity is estimated by random sampling, and the half-width of the confidence interval for the estimated value corresponding to each participating entity is determined. Sampling is stopped when the half-width of the confidence interval for the estimated value corresponding to all participating entities is not greater than the error boundary, and the estimation result of the comprehensive marginal technical contribution is used to determine the construction responsibility allocation ratio. When the maximum number of samplings is reached, if the half-width of the confidence interval for the estimated value corresponding to at least one participating entity is still greater than the error boundary, the estimation result of the comprehensive marginal technical contribution is marked as not reaching the target accuracy.

[0023] In an example embodiment of this disclosure, based on the aforementioned scheme, the process of determining the construction responsibility allocation ratio includes: obtaining the non-negative portion of the comprehensive marginal technical contribution, shared facility utilization contribution, new capacity demand, and reliability contribution corresponding to each participating entity; for each contribution category, normalizing the contribution category according to the contribution amount corresponding to that contribution category of each participating entity; determining the category weight corresponding to each available contribution category according to the construction scheme rule version; weighting the normalized contribution corresponding to each participating entity according to the category weight to obtain the comprehensive contribution amount corresponding to each participating entity, and normalizing it according to the comprehensive contribution amount corresponding to each participating entity to obtain the construction responsibility allocation ratio; when the normalization denominator of any contribution category is zero, the contribution category is determined as an unavailable contribution category, and the category weights corresponding to the remaining available contribution categories are re-normalized according to the pre-registration rules; when all contribution categories are unavailable contribution categories, the construction responsibility allocation ratio is not generated.

[0024] In an example embodiment of this disclosure, based on the aforementioned scheme, the method for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing further includes: when the independent operation scheme corresponding to the alliance subject combination is determined to be physically feasible, the independent operation scheme is determined as a fixed reference scenario of the alliance subject combination; when the independent operation scheme corresponding to the alliance subject combination is determined to be physically infeasible, the external selection scheme registered before the establishment of the construction scheme evaluation task is determined as the fixed reference scenario of the alliance subject combination; the fixed reference scenario is bound to the participating subject snapshot, the binding relationship of the task identifier, the model version, the topology version, the capacity and service level, the representative scenario, the scenario weight, the original technical loss index, the positive scale, the technical loss weight, the station-side physical verification result and audit evidence, and the fixed reference scenario remains unchanged during the task.

[0025] In an example embodiment of this disclosure, based on the aforementioned scheme, the method for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing further includes: generating corresponding scene identifiers for each combination of alliance entities and each candidate shared facility construction scheme, and forming evaluation scenarios corresponding to each scene identifier based on the representative scenario and the scene weight; for each evaluation scenario corresponding to a scene identifier, closing callable assets that do not belong to the alliance entity combination corresponding to the scene identifier and callable assets that have not obtained effective control authorization, and re-determining the asset activation status and callable capacity according to the binding relationship of the task identifier; reconstructing the hydraulic and power task feasible domain corresponding to the scene identifier based on the re-determined asset activation status, callable capacity, and scene boundary, and based on the reconstructed water The feasible domain of the power and electricity tasks performs local result coordination and physical verification; based on the binding relationship of the task identifier, the shared facility utilization, new capacity demand, and reliability improvement of the station node are aggregated to the corresponding participating entities, respectively obtaining the shared facility utilization contribution, the new capacity demand, and the reliability contribution; when a station node belongs to multiple participating entities, the shared facility utilization, new capacity demand, and reliability improvement of the station node are allocated to the corresponding multiple participating entities according to the control responsibility coefficient pre-registered in the task identifier and whose sum is one, and the control responsibility coefficient remains unchanged during the corresponding task; wherein, only the evaluation scenario results that pass the physical verification are used to determine the shared facility utilization contribution, the new capacity demand, and the reliability contribution.

[0026] In an example embodiment of this disclosure, based on the aforementioned scheme, the method for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing further includes: obtaining multiple candidate shared facility construction schemes with different candidate construction capacities; screening each candidate shared facility construction scheme according to the technical loss threshold and reserve threshold pre-registered in the construction scheme rule version; for each candidate shared facility construction scheme that passes the screening, determining the technical value of the unit construction cost corresponding to each candidate shared facility construction scheme based on the technical value and construction cost under the candidate construction capacity corresponding to each candidate shared facility construction scheme, and comparing the candidate shared facility construction schemes that pass the screening based on the technical value of the unit construction cost to determine the target shared facility construction scheme and the construction capacity corresponding to the target shared facility construction scheme; performing sensitivity analysis on the technical loss weight and the cost basis corresponding to each candidate shared facility construction scheme respectively; and re-screening and comparing the candidate shared facility construction schemes when any one of the technical loss weight and the cost basis corresponding to each candidate shared facility construction scheme changes.

[0027] In an example embodiment of this disclosure, based on the foregoing scheme, invalidating existing results affected by the corresponding change includes: determining at least one target version among candidate runtime result versions, control plan versions, and construction scheme versions affected by the change; changing the target version from a referenceable state to an invalid state through an indivisible state update; not publishing alternative results corresponding to the target version before the state update is successful, and maintaining the state of refusing to reference the target version when the state update write fails.

[0028] In an example embodiment of this disclosure, based on the aforementioned scheme, the recalculation of scenarios affected by the corresponding changes includes: when any one of the participation qualifications of the station node and the field whitelist changes, a new task identifier is generated, and based on the new task identifier, the protected aggregate quantity submitted by each station is re-received, the common working condition center, the representative scenario, the scenario weight, and the scenario boundary are re-determined, the corresponding hydraulic and power task feasible domain is reconstructed, and local result coordination and physical verification are re-performed; when any one of the callable assets, the control authorization, and the topology connectivity changes, a new task identifier is generated, and based on the new task identifier, the uncallable assets are closed again, the corresponding hydraulic and power task feasible domain is reconstructed, and local result coordination and physical verification are re-performed; when the task boundary does not change and any one of the execution deviation and coordination parameter adjustment occurs, a new candidate running result version is generated under the original task identifier, and local result coordination and physical verification are re-performed; wherein, only scenarios affected by the corresponding changes are recalculated, and the results corresponding to unaffected scenarios remain unchanged.

[0029] In an example embodiment of this disclosure, based on the aforementioned scheme, the method for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing further includes: when the physical verification result is a failure, not generating a control plan version for the candidate operation result corresponding to the physical verification result; when the physical verification result on which the issued control plan version is based is revoked, invalidating the issued control plan version and the candidate operation result version bound to the issued control plan version; when the feedback deviation determined according to the execution feedback exceeds the corresponding feedback threshold, invalidating the corresponding control plan version and the candidate operation result version bound to the control plan version; when any one of the construction capacity corresponding to the target shared facility construction scheme, the cost basis version corresponding to the candidate shared facility construction scheme, the construction scheme rule version, and the construction scheme evidence bound to the construction scheme version changes, invalidating the corresponding construction scheme version; when the execution feedback causes any one of the feasibility and technical indicators of the corresponding scenario to change, invalidating the construction scheme version associated with the corresponding scenario.

[0030] In an example embodiment of this disclosure, based on the aforementioned scheme, the method for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing further includes: setting candidate operation result versions for each of the candidate operation results; generating a control plan version corresponding to the target candidate operation result based on the target candidate operation result; and generating a construction scheme version corresponding to the target shared facility construction scheme; generating scene identifiers corresponding to the combination of each alliance entity combination and each candidate shared facility construction scheme, reference scene identifiers corresponding to each fixed reference scene, and representative scene identifiers corresponding to each representative scene; generating a participant snapshot based on the participating entities; determining the data window version, model version, authorization version, output permission version, and topology version corresponding to the task identifier; determining the cost basis version corresponding to each candidate shared facility construction scheme; and determining the construction scheme rule version corresponding to the task identifier; The construction scheme version is bound to the task identifier, each of the scenario identifiers, each of the reference scenario identifiers, each of the representative scenario identifiers, each candidate shared facility construction scheme associated with the construction scheme version, the participating entity snapshot, the common operating condition center, the scenario weight, the data window version, each of the candidate operation result versions, the control plan version corresponding to the target candidate operation result, the binding relationship of the task identifier, the model version, the authorization version, the output permission version, the topology version, the cost basis version corresponding to each candidate shared facility construction scheme, the construction scheme rule version, the original technical loss index, the positive scale, the technical loss weight, the feasible domain of the hydraulic and electric tasks, the physical verification result, the technical value, the marginal technical contribution, the construction responsibility allocation ratio, and the corresponding signature, proof, and audit evidence citation.

[0031] According to a second aspect of the present disclosure, a system for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing is provided. This system includes a central alliance computing platform and multiple station-side privacy computing nodes. The central alliance computing platform is used to generate task identifiers, establish mapping relationships between participating entities, station-side nodes corresponding to multiple hydropower stations, and callable assets associated with the multiple hydropower stations, and bind the mapping relationships and corresponding control authorizations to the task identifiers. Each station-side privacy computing node is used to perform local privacy computing based on the task identifiers, forming protected aggregate quantities, without requiring the original data of the corresponding hydropower station to leave the station. The central alliance computing platform is used to determine the common operating condition center, representative scenario, and scenario weight based on the protected aggregation quantity when the number of valid stations submitting the protected aggregation quantity is not less than the node threshold and the number of valid samples is not less than the sample threshold, and to determine the scenario boundary in combination with the current prediction state. The central alliance computing platform is used to determine multiple alliance entity combinations and multiple candidate shared facility construction schemes, and based on the representative scenario and the scenario weight, to form evaluation scenarios corresponding to the combinations of each alliance entity combination and each candidate shared facility construction scheme, and to close uncallable assets and reconstruct the evaluation scenario according to the binding relationship of the task identifier and the scenario boundary. The corresponding feasible domain for hydraulic and electric power tasks; the central consortium computing platform is used to coordinate the local results of each station-end privacy computing node within the feasible domain for hydraulic and electric power tasks, obtain candidate operation results, and receive physical verification results returned by each station-end privacy computing node based on a complete local model for the candidate operation results; determine the target candidate operation results for actual operation from the candidate operation results that have passed the physical verification, and after the control authorization verification and security verification, send the target candidate operation results to the corresponding station-end privacy computing node and receive execution feedback; the central consortium computing platform is used to evaluate the scenario results that have passed the physical verification, A technical evaluation is conducted on the combination of each alliance entity and each candidate shared facility construction scheme to obtain the technical evaluation results. Based on the technical evaluation results, the target shared facility construction scheme, the construction capacity corresponding to the target shared facility construction scheme, and the construction responsibility allocation ratio corresponding to each participating entity are determined. The central alliance computing platform is used to invalidate the existing results affected by the corresponding change and recalculate the scenario affected by the corresponding change when any one of the following changes occurs: the binding relationship of the task identifier, the candidate shared facility construction scheme, the construction capacity corresponding to the candidate shared facility construction scheme, the physical verification result, and the execution feedback.

[0032] According to a third aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory storing computer-readable instructions, which, when executed by the processor, implement the method for determining a multi-hydropower station collaborative construction scheme based on privacy computing as described in the first aspect.

[0033] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method for determining a multi-hydropower station collaborative construction scheme based on privacy computing as described in the first aspect.

[0034] The technical solutions provided in this disclosure may have the following beneficial effects: According to the method for determining the collaborative and construction scheme of multi-hydropower stations based on privacy computing in this example embodiment, on the one hand, by generating task identifiers, the mapping relationship and corresponding control authorization between participating entities, station nodes, and callable assets are bound, and each station performs local privacy computing under the condition that the original data does not leave the station. Only when the number of effective station nodes and the number of effective samples simultaneously meet the threshold, a common operating condition center, representative scenario, and scenario weight are formed based on the protected aggregation quantity. In this way, while avoiding the centralized uploading of original hydrological, equipment, and operation data, different station nodes can participate in cross-station collaboration under a unified task boundary, and reduce the possibility of direct exposure of the original operation information of a single station.

[0035] On the other hand, by combining the results of common operating conditions with the current predicted state to determine the scenario boundary, and forming evaluation scenarios for different alliance subject combinations and candidate shared facility construction schemes, and closing uncallable assets and reconstructing the corresponding task feasible domain according to the binding relationship of task identifiers, when the conditions such as participating subjects, control authorization, callable assets, and construction capacity differ, the corresponding collaborative calculation can be based on the current actual callable resources and the current operating boundary, reducing the situation where unavailable units, reservoir capacity, water transmission channels, transmission lines or control capabilities are still included in the calculation due to the use of fixed subjects, fixed capacity or fixed asset relationships.

[0036] On the other hand, by coordinating the local results of each station within the reconstructed feasible domain, and having each station use a complete local model to physically verify the candidate operation results, only the candidate operation results that pass the physical verification are further used in actual operation. This means that the numerical results obtained by the central side coordination need to be re-verified by the actual hydraulic, electrical and safety constraints at the station, avoiding the substitution of the station's complete physical constraint judgment with only the convergence results calculated by the central side.

[0037] On the other hand, by evaluating the results of the physical verification process alone, technical evaluations are conducted on different combinations of alliance entities and different candidate shared facility construction schemes. Furthermore, the construction scheme, construction capacity, and the proportion of construction responsibility allocation for each participating entity are determined. This allows the evaluation of the construction scheme to be based on different combinations of entities, different construction capacities, and their actual collaborative operation results, reducing the bias caused by using only fixed proportions or static capacity to determine construction responsibility.

[0038] Furthermore, by invalidating existing results and recalculating the affected scenarios when changes occur in task binding relationships, candidate construction schemes, construction capacity, physical verification results, or execution feedback, results that no longer meet current operating conditions can be prevented from participating in subsequent collaboration and construction scheme determination. Thus, the process of determining collaborative construction schemes for multiple hydropower stations can be constrained and updated at multiple levels, including data boundaries, task boundaries, physical feasibility, operational feedback, and scheme evaluation, thereby improving the consistency between the collaborative results of multiple hydropower stations and actual physical constraints, as well as the reliability of the determined construction schemes.

[0039] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0040] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure. It is obvious that the drawings described below are merely some embodiments of this disclosure, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0041] Figure 1 The illustration shows a flowchart of a method for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing, according to some embodiments of the present disclosure.

[0042] Figure 2 The illustration shows a schematic diagram of the heterogeneous station access and task establishment process according to some embodiments of the present disclosure.

[0043] Figure 3 The illustration shows common operating conditions and joint coordination processes according to some embodiments of the present disclosure.

[0044] Figure 4 The illustration shows a schematic diagram of the candidate run result review and failure process according to some embodiments of the present disclosure.

[0045] Figure 5 The illustration schematically depicts a relationship between tasks and results according to some embodiments of this disclosure.

[0046] Figure 6 The diagram illustrates a block diagram of a system for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing, according to some embodiments of the present disclosure.

[0047] Figure 7 The diagram illustrates the interaction relationships of a system for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing, according to some embodiments of the present disclosure.

[0048] Figure 8 The diagram illustrates a site-side privacy computing node structure according to some embodiments of the present disclosure.

[0049] Figure 9 A schematic block diagram of a central consortium computing platform according to some embodiments of the present disclosure is shown.

[0050] Figure 10 The illustration shows a schematic diagram of the process for determining construction schemes and allocating responsibilities according to some embodiments of this disclosure.

[0051] Figure 11 The schematic diagram illustrates the structural schematic of a computer system of an electronic device according to some embodiments of the present disclosure.

[0052] In the accompanying drawings, the same or corresponding reference numerals indicate the same or corresponding parts. Detailed Implementation

[0053] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this specification. Rather, they are merely examples of systems and methods consistent with some aspects of this specification as detailed in the appended claims.

[0054] The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of this specification. The singular forms “a,” “the,” and “the” as used in this specification and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0055] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be more thorough and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art.

[0056] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a thorough understanding of embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced without one or more of the specific details, or other methods, components, systems, steps, etc., can be employed. In other instances, well-known methods, systems, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of this disclosure.

[0057] Furthermore, the accompanying drawings are for illustrative purposes only and are not necessarily drawn to scale. The block diagrams shown in the drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor systems and / or microcontroller systems.

[0058] In the exemplary embodiments disclosed herein, a method for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing is first provided. This method can be implemented by a system for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing. Figure 1 The illustration schematically depicts a flowchart of a method for determining a collaborative construction scheme for multiple hydropower stations based on privacy computing, according to some embodiments of this disclosure. (Reference) Figure 1 As shown, the method for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing may include the following steps: Step S110: Generate task identifier, establish mapping relationship between participating entities, station-end nodes corresponding to multiple hydropower stations, and callable assets associated with multiple hydropower stations, and bind the mapping relationship and corresponding control authorization to task identifier; Step S120: Under the condition that the original data of the corresponding hydropower station does not leave the station, each station node performs local privacy calculation based on the task identifier to form a protected aggregate. When the number of valid stations submitting the protected aggregate is not less than the node threshold and the number of valid samples is not less than the sample threshold, the common working condition center, representative scenario and scenario weight are determined according to the protected aggregate, and the scenario boundary is determined in combination with the current prediction status. Step S130: Determine multiple alliance entity combinations and multiple candidate shared facility construction schemes. Based on representative scenarios and scenario weights, form evaluation scenarios corresponding to the combinations of each alliance entity combination and each candidate shared facility construction scheme. According to the binding relationship of task identifiers and scenario boundaries, close uncallable assets and reconstruct the feasible domain of hydraulic and electric tasks corresponding to the evaluation scenario. Step S140: Coordinate the local results of each station within the feasible domain of the hydraulic and electric tasks to obtain candidate operation results, and receive the physical verification results returned by each station based on the complete local model for the candidate operation results; Step S150: Determine the target candidate operation result for actual operation from the candidate operation results that have passed the physical review, and after control authorization verification and security verification, send the target candidate operation result to the corresponding station and receive execution feedback; Step S160: Based on the evaluation scenario results that have passed the physical verification, conduct a technical evaluation on the combination of each alliance entity combination and each candidate shared facility construction scheme to obtain the technical evaluation results, and determine the target shared facility construction scheme, the construction capacity corresponding to the target shared facility construction scheme, and the construction responsibility allocation ratio corresponding to each participating entity based on the technical evaluation results. Step S170: When any one of the following changes, such as the binding relationship of the task identifier, the candidate shared facility construction plan, the construction capacity corresponding to the candidate shared facility construction plan, the physical verification result, and the execution feedback, the existing results affected by the corresponding change are invalidated, and the scenario affected by the corresponding change is recalculated.

[0059] According to the method for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing in this example embodiment, on the one hand, by generating task identifiers, the mapping relationship and corresponding control authorization between participating entities, station nodes, and callable assets are bound together. Each station performs local privacy computing without leaving the station, and a common operating condition center, representative scenario, and scenario weight are formed only when the number of valid stations and the number of valid samples simultaneously meet the threshold. This reduces the centralized uploading of original hydrological, equipment, and operational data while lowering the possibility of direct exposure of single-station operational information. On the other hand, by combining the common operating condition results with the current prediction status to determine the scenario boundary, evaluation scenarios are formed for different alliance entity combinations and candidate shared facility construction schemes. Based on the binding relationship of the task identifiers, uncallable assets are closed and the corresponding task feasible domain is reconstructed, enabling collaborative computing to be based on the current actual callable resources and operational boundaries, reducing the situation where unavailable units, reservoir capacity, water transmission channels, transmission lines, or control capabilities continue to participate in the calculation. On the other hand, by coordinating the local results of each station within the reconstructed feasible domain, and having each station physically verify the candidate operational results using a complete local model, only the physically verified candidate operational results are used in actual operation. This further verifies the central calculation results against the actual hydraulic, electrical, and safety constraints of the stations, avoiding the substitution of numerical convergence results for complete physical constraint judgments. Furthermore, by evaluating different alliance entity combinations and candidate shared facility construction schemes based solely on physically verified evaluation scenario results, and determining the target shared facility construction scheme, construction capacity, and the allocation ratio of construction responsibilities among participating entities, the evaluation of construction schemes can be based on different entity combinations, different construction capacities, and their collaborative operation results, reducing biases caused by fixed-ratio or static capacity evaluations. Moreover, by invalidating the affected existing results and recalculating the affected scenarios when changes occur in task binding relationships, candidate construction schemes, construction capacities, physical verification results, or execution feedback, invalidated results are prevented from continuing to participate in subsequent collaboration and construction scheme determination, thereby improving the consistency between multi-hydropower station collaborative results and actual physical constraints, as well as the reliability of construction scheme determination.

[0060] The following will further explain the method for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing in this example embodiment.

[0061] In step S110, a task identifier is generated, a mapping relationship is established between participating entities, station-end nodes corresponding to multiple hydropower stations, and callable assets associated with multiple hydropower stations, and the mapping relationship and corresponding control authorization are bound to the task identifier.

[0062] The task identifier identifies a multi-hydropower station collaborative construction scheme evaluation task, linking participating entities, station nodes, callable assets, and corresponding control permissions. Participating entities represent the entities involved in the current collaborative task or construction scheme evaluation; for example, they could be entities corresponding to hydropower stations A, B, and C. Station nodes represent data processing nodes located at the corresponding hydropower station and participating in the current task; for example, they could be privacy computing nodes deployed at hydropower stations A, B, and C. Callable assets represent physical resources or control capabilities related to the operation of the corresponding hydropower station and capable of participating in collaborative computing in the current task; for example, they could include generating units, reservoir capacity, water conveyance channels, transmission lines, or control capabilities. The mapping relationship represents the attribution and calling correspondence between participating entities, station nodes, and callable assets. Control authorization represents the scope of permissions allowed for corresponding participating entities or station nodes to call corresponding assets or perform related control operations.

[0063] For example, a unique task identifier can be generated for a single collaborative task involving multiple hydropower stations. A mapping can be established based on the attribution relationships between participating entities and station nodes, as well as the relationships between station nodes and callable assets. This ensures that all entities, nodes, assets, and permissions involved in the task are associated with the same task identifier. For instance, the station node corresponding to hydropower station A can establish a calling relationship with the generating units, reservoir capacity, or transmission lines on hydropower station A that can participate in the current task. Hydropower stations B and C can establish corresponding relationships in the same way. Subsequent privacy calculations, scenario construction, joint operation, and result updates can all determine the task boundaries based on this task identifier, thereby reducing the possibility of mixing entity, asset, or permission information between different tasks.

[0064] In step S120, each station node performs local privacy calculations based on the task identifier under the condition that the original data of the corresponding hydropower station does not leave the station, forming a protected aggregate quantity. When the number of valid stations submitting the protected aggregate quantity is not less than the node threshold and the number of valid samples is not less than the sample threshold, the common working condition center, representative scenario and scenario weight are determined according to the protected aggregate quantity, and the scenario boundary is determined in combination with the current prediction state.

[0065] Here, raw data can represent hydrological data, equipment data, and operational data generated and stored at the hydropower station during its operation. For example, it may include water level, unit operating status, power output, and time-series data related to the hydropower station's operation. Local privacy computation can represent the processing of raw data locally by station nodes to generate computation results that can participate in cross-station computation without providing complete raw data to the central side. Protected aggregation quantity can represent the results generated through local privacy computation and allowed for cross-station aggregation within the current task scope. Valid station number can represent the number of station nodes that have completed valid data submission in the current task; valid sample number can represent the number of valid samples that can participate in this cross-station aggregation computation; node threshold and sample threshold can represent the minimum number of station nodes and minimum number of samples required to allow cross-station aggregation, respectively.

[0066] A common operating condition center can represent aggregated features obtained from the protected aggregation results of multiple station nodes, used to characterize the common operational characteristics of multiple hydropower stations. A representative scenario can represent a typical operational scenario formed based on the corresponding common operating condition center; for example, it can represent the typical operational state of multiple hydropower stations under a certain type of common operational characteristic. Scenario weights can represent the relative proportion of the corresponding representative scenario in the current evaluation process. The current predicted state can represent the predicted operational state corresponding to the current collaborative task. Scenario boundaries can represent the boundary conditions formed by combining the common operating condition center and the current predicted state, used to limit the operational range of the corresponding representative scenario. The representative scenario boundaries are jointly determined by the common operating condition center and the current predicted state, rather than simply using historical operational features to replace the current state.

[0067] Specifically, each station node can utilize the local raw data of its corresponding hydropower station to form a protected aggregate quantity, and associate this protected aggregate quantity with the current task identifier. When both the number of valid stations and the number of valid samples reach the corresponding thresholds, shared operational features across stations can be extracted based on the protected aggregate results corresponding to multiple station nodes, forming a common operating condition center, and further forming representative scenarios and corresponding scenario weights. Subsequently, the operational patterns reflected by the common operating condition center are combined with the current prediction state to determine the scenario boundaries of the representative scenarios. This reduces the centralized transmission range of raw operational data while enabling subsequent scenarios to take into account both shared operational patterns across stations and current operational conditions.

[0068] In step S130, multiple alliance entity combinations and multiple candidate shared facility construction schemes are determined. Based on representative scenarios and scenario weights, evaluation scenarios corresponding to the combinations of each alliance entity combination and each candidate shared facility construction scheme are formed. According to the binding relationship of task identifiers and scenario boundaries, uncallable assets are closed and the feasible domain of hydraulic and electric tasks corresponding to the evaluation scenario is reconstructed.

[0069] The term "alliance entity combination" refers to a combination of entities formed from the participating entities involved in the current task for collaborative evaluation. For example, if the participating entities include A, B, and C, the alliance entity combination could include combinations such as AB, AC, BC, or ABC. "Candidate shared facility construction schemes" can represent the construction configurations to be evaluated for shared facilities among multiple hydropower stations. For example, it could refer to different construction capacity schemes for multiple hydropower stations sharing a transmission line. "Evaluation scenario" can represent a scenario object jointly determined by the alliance entity combination, candidate shared facility construction schemes, and representative scenarios, used to characterize the collaborative operation of a specific entity combination under specific candidate construction schemes and operating conditions. For example, an alliance entity combination and a candidate shared facility construction scheme can form a corresponding evaluation scenario under a representative scenario.

[0070] Uncallable assets can refer to callable assets that cannot participate in the current evaluation scenario calculation based on the binding relationship of the current task identifier and the current scenario conditions. For example, for a participating entity that is not included in the current alliance entity combination, the related assets corresponding to that participating entity and that cannot be called by the current alliance can be treated as uncallable assets.

[0071] The feasible domain of hydraulic and electric tasks can be represented as a set of feasible operating states under the current evaluation scenario, which is jointly constrained by hydraulic operating conditions, electric operating conditions, and cross-station collaborative conditions. It is a task feasible domain that is simultaneously constrained by hydraulic and electric conditions, rather than two independent feasible domains.

[0072] Specifically, each alliance entity combination and each candidate shared facility construction scheme can be combined separately, and corresponding evaluation scenarios can be formed by combining representative scenarios and their scenario weights. For each evaluation scenario, the assets that can currently participate in collaboration can be determined based on the entities, nodes, assets, and permission relationships bound to the task identifier. Assets that do not meet the current calling conditions can be shut down, and the allowable hydraulic and power operation ranges can be redefined based on the scenario boundaries corresponding to the evaluation scenario, so as to form a task feasible domain matching the evaluation scenario. For example, for an evaluation scenario formed by alliance entity combination AB and a certain shared transmission line candidate construction scheme, only the assets that the alliance entity combination can call under the current task can be included in the corresponding task feasible domain. In this way, different alliance entity combinations and different candidate construction schemes adopt feasible operation ranges that match their current participating objects and physical conditions, reducing the situation where the fixed model continues to call currently unavailable assets.

[0073] In step S140, the local results of each station are coordinated within the feasible domain of the hydraulic and electric tasks to obtain candidate operation results, and the physical verification results returned by each station based on the complete local model are received.

[0074] The local results can represent the local operational calculation results generated by each station node based on the local operating conditions and current task requirements of the corresponding hydropower station. For example, it can be the local operational status calculation result generated by the corresponding station in the current evaluation scenario. The candidate operational results can represent the joint operational results generated after cross-station coordination of the local results of multiple station nodes within the task feasible domain corresponding to the current evaluation scenario. The complete local model can represent the local model stored at the corresponding station that reflects the actual physical operational relationships and safety constraints of the corresponding hydropower station. The physical verification result can represent the verification result generated by the station node after verifying the candidate operational results using the complete local model, which is used to characterize whether the candidate operational results meet the actual physical operational constraints of the corresponding hydropower station.

[0075] Specifically, each station node can generate corresponding local results under the constraints of the current task's feasible domain. Through cross-station coordination, the local results from different stations can meet the shared operational requirements of the current evaluation scenario, thus forming candidate operational results. After the candidate operational results are formed, the result portion corresponding to each station is sent to the respective station. The station then performs physical verification based on the complete model stored locally and returns the physical verification results, without needing to upload the complete local model to the central side. For example, for candidate operational results jointly generated by multiple stations, each station can perform local physical verification on the candidate result portion related to its own hydropower station. Thus, while keeping the local detailed model from leaving the station, station-level physical constraint verification can be performed on the candidate operational results generated through cross-station coordination, reducing the physical deviation caused by determining the operational scheme solely based on the central side's calculation results.

[0076] In step S150, the target candidate operation result for actual operation is determined from the candidate operation results that have passed the physical review. After control authorization verification and security verification, the target candidate operation result is sent to the corresponding station and execution feedback is received.

[0077] Specifically, the target candidate execution result can represent the candidate execution result determined from the candidate execution results that have passed physical review and used for the current actual execution. Control authorization verification can represent the verification of whether the control objects and control scope involved in the target candidate execution result conform to the control permissions corresponding to the current task. Security verification can represent the verification of whether the target candidate execution result meets the security conditions that the corresponding station can control before it is issued. Execution feedback can represent the feedback information formed and returned by the corresponding station based on the actual execution status after the target candidate execution result is issued; for example, it can represent the execution status feedback or execution deviation information formed by the corresponding station after actually running according to the target candidate execution result.

[0078] Specifically, target candidate operation results are selected for actual operation only from those that have passed physical verification at the station. Before distribution, control authorization and security verifications are performed on the target candidate operation results. Only after the corresponding verifications meet the requirements are the target candidate operation results provided to the corresponding station for actual operation control at that station. After the corresponding operation is performed at the station, execution feedback is obtained to characterize the operation response between the actual execution status and the current target candidate operation results. This avoids candidate operation results that have not undergone physical verification or do not meet the current control permissions and security conditions from directly entering actual operation, and allows the effectiveness of the distributed operation results to be reflected by the actual execution status.

[0079] In step S160, based on the evaluation scenario results that have passed the physical verification, a technical evaluation is conducted on the combination of each alliance entity combination and each candidate shared facility construction scheme to obtain the technical evaluation results. Based on the technical evaluation results, the target shared facility construction scheme, the construction capacity corresponding to the target shared facility construction scheme, and the construction responsibility allocation ratio corresponding to each participating entity are determined.

[0080] The technical evaluation results can represent the evaluation results based on the physical verification passed evaluation scenario, comparing the technical performance of different alliance entity combinations and different candidate shared facility construction schemes in the coordinated operation of multiple hydropower stations. The target shared facility construction scheme can represent the shared facility construction scheme determined from the candidate shared facility construction schemes for the current task based on the technical evaluation results; for example, it can be the target construction scheme determined from multiple candidate construction schemes for shared transmission lines. The construction capacity can represent the facility configuration capacity corresponding to the target shared facility construction scheme; for example, in the case of shared transmission lines, it can represent the new capacity of the line corresponding to the target construction scheme. The construction responsibility allocation ratio can represent the construction responsibility ratio of each participating entity in the target shared facility construction scheme, determined based on the technical evaluation results; for example, the construction responsibility allocation ratios for participating entity A, participating entity B, and participating entity C can be determined separately.

[0081] Specifically, the scenario results for each evaluation scenario can be obtained after task feasibility domain reconstruction, station-side local coordination, and complete local model verification. Only the evaluation scenario results that pass physical verification are used for subsequent technical evaluation, and the combinations of each alliance entity combination and each candidate shared facility construction scheme are compared. Based on the technical evaluation results obtained from the comparison, the target shared facility construction scheme and corresponding construction capacity that match the current multi-hydropower station collaborative needs can be determined, and the construction responsibility allocation ratio of each participating entity can be further determined. For example, the evaluation scenario results formed by different alliance entity combinations under different shared facility construction capacities can be compared separately, and then the target scheme can be determined from multiple candidate shared facility construction schemes. In this way, the determination of construction schemes and construction responsibilities is based on the actual feasible collaborative operation results of different alliance entity combinations and candidate construction schemes, reducing the deviation caused by determining solely based on static entity relationships or static capacity.

[0082] In step S170, when any one of the following changes, such as the binding relationship of the task identifier, the candidate shared facility construction plan, the construction capacity corresponding to the candidate shared facility construction plan, the physical verification result, and the execution feedback, the existing results affected by the corresponding change are invalidated, and the scenario affected by the corresponding change is recalculated.

[0083] Existing results can refer to operational results or construction scheme evaluation results that were formed based on the current task, evaluation scenario, or candidate construction scheme before the corresponding change occurred and can be referenced by subsequent steps. Scenarios affected by the corresponding change can refer to evaluation scenarios where the scenario's effectiveness, physical feasibility, or technical evaluation results depend on the changed task binding relationship, candidate shared facility construction scheme, construction capacity, physical verification results, or execution feedback. For example, when the construction capacity corresponding to a candidate shared facility construction scheme changes, the evaluation scenario related to that candidate shared facility construction scheme can be considered a scenario affected by the corresponding change; when a physical verification result changes, the evaluation scenario using that physical verification result can be considered an affected scenario.

[0084] Specifically, the status of the current task can be continuously updated. When the object relationship bound to the task identifier changes, the scenario matching the new task boundary can be redefined; when the construction scheme or construction capacity of the candidate shared facility changes, the evaluation scenario affected by the corresponding construction configuration can be redefined; when the physical verification result or execution feedback changes, it can be determined whether the original operation result or evaluation result is still valid based on the changed content. For existing results confirmed to be affected by the corresponding changes, they are invalidated, and the affected scenarios are recalculated; for scenarios not affected by the corresponding changes, their existing results can be maintained. For example, when only the construction capacity of a candidate shared facility construction scheme changes, the scenario related to that candidate shared facility construction scheme can be recalculated without repeating the calculation of other unaffected evaluation scenarios. Thus, the participation of historical results that no longer meet the current task conditions in the multi-hydropower station collaboration and construction scheme determination can be reduced, while avoiding indiscriminate duplication of calculations for all scenarios.

[0085] The following is a detailed explanation of the steps described above.

[0086] In some embodiments, establishing a mapping relationship between participating entities, station-end nodes corresponding to multiple hydropower stations, and callable assets associated with multiple hydropower stations, and binding the mapping relationship and corresponding control authorization with task identifiers, can be achieved through the following technical steps: The first step is to identify participating entities with entity identifiers, station nodes with node identifiers, and callable assets with asset identifiers. Callable assets include at least one of the following: generating units, reservoir capacity, water transmission channels, transmission lines, and control capabilities.

[0087] The task identifier represents identification information used to distinguish between a single multi-hydropower station collaborative and construction scheme evaluation task. The participating entity represents the main object participating in the current collaborative task or construction scheme evaluation; for example, it could be a main object corresponding to a conventional hydropower station or pumped storage power station participating in the current task. The station-side node represents a data processing node set up on the corresponding hydropower station side and participating in the current task. The callable asset represents physical resources or control capabilities related to the operation of the corresponding hydropower station and capable of participating in collaborative calculations in the current task; for example, it can include at least one of generating units, reservoir capacity, water conveyance channels, transmission lines, and control capabilities. Control authorization represents the scope of permissions allowed for the station-side node to call upon the corresponding asset or generate control reference values ​​under the current task.

[0088] The entity identifier can represent identification information used to distinguish different participating entities; the node identifier can represent identification information used to distinguish different station-end nodes; and the asset identifier can represent identification information used to distinguish different callable assets. For example, a single hydropower station can have one or more station-end nodes, and each station-end node can correspond to one or more assets such as generating units, reservoir capacity, water conveyance channels, transmission lines, or control capabilities. By setting entity identifiers, node identifiers, and asset identifiers respectively, subsequent scenario construction and callable asset determination can be traced back to the specific participating objects in the current task.

[0089] The second step is to determine the attribution relationship between the subject identifier and the node identifier, as well as the callability relationship between the node identifier and the asset identifier. Based on the attribution and callability relationships, a mapping relationship is formed, and the corresponding control authorization credentials are recorded.

[0090] Specifically, the attribution relationship between a subject and a node represents the participating subject corresponding to a certain station node in the current task, and the callability relationship between a node and an asset represents the assets that a certain station node can call within the current control authorization scope. Mapping values ​​can be represented in two states: valid and invalid. When an attribution relationship permitted by the current task exists between a subject and a node, the corresponding subject-node mapping is valid; when a callability relationship permitted by the current task exists between a node and an asset, the corresponding node-asset mapping is valid. Control authorization credentials and the corresponding callability relationships are recorded together to confirm whether control permissions remain valid during subsequent asset activation judgments, control plan issuance, and version updates.

[0091] The third step is to determine the task boundaries corresponding to the task identifier by identifying the participating entity snapshot, site node snapshot, callable assets, mapping relationships, field whitelist, model version, authorization version, output permission version, and topology version, and keep the task boundaries unchanged under the same task identifier.

[0092] Among them, the participant snapshot represents the set of participant states when the task identifier is generated; the station node snapshot represents the set of station node states when the task identifier is generated; the field whitelist represents the range of fields allowed to participate in calculations or outputs in the current task; the model version represents the version information corresponding to the station-side model and the central coordination model used in the current task; the authorization version represents the version information corresponding to the current control authorization status and authorized capacity; the output permission version represents the version information corresponding to the data types and ranges allowed to be output to the central side in the current task; and the topology version represents the version information corresponding to the current hydraulic and power connection relationships. Once the task boundary is bound to the task identifier, it remains unchanged during the duration of the same task identifier's validity, ensuring that subsequent protected aggregation, scenario construction, feasible domain reconstruction, candidate operation results, and construction scheme evaluation all correspond to the same task boundary.

[0093] The fourth step is to generate a new task identifier when any of the elements constituting the task boundary changes; and to generate a new candidate execution result version under the original task identifier when the task boundary remains unchanged but an execution deviation occurs.

[0094] Specifically, when any of the following changes occur, resulting in a change in the task boundary: participating entity, station node, callable asset, mapping relationship, field whitelist, model version, authorization version, output permission version, or topology version, the original task identifier is not used for new scenario calculations. Instead, a new task identifier is generated and re-associated with the corresponding task boundary. When there is only a runtime deviation that does not change the task boundary, a new candidate runtime result version can be formed under the original task identifier, allowing different candidate results under the same task boundary to be distinguished from each other. This reduces the likelihood of results corresponding to different entities, assets, permissions, or topology conditions being mixed and referenced within the same task.

[0095] Figure 2 This illustration schematically depicts a heterogeneous station access and task establishment process according to some embodiments of the present disclosure. (Reference) Figure 2 As shown, the data source for heterogeneous hydropower projects can be obtained first, and the data at each station can be accessed and its quality checked to determine whether the data at each station meets the data access requirements of the current task. After completing the access and quality check, it can be determined whether the corresponding station data meets the requirements. If the corresponding station data does not meet the requirements, the corresponding field mapping can be corrected, or the station that does not meet the requirements can be excluded; if the corresponding station data meets the requirements, candidate station nodes can be further determined and tasks can be orchestrated to determine the candidate station nodes participating in the current task and their corresponding task relationships.

[0096] After determining candidate site nodes and orchestrating tasks, task verification can be performed to confirm whether the current task configuration meets the task creation conditions. Upon successful verification, a task identifier can be generated and bound to the task boundary, establishing a correspondence between the participating entities, site nodes, callable assets, and corresponding version and permission information involved in the current task and the task identifier. After generating the task identifier and binding the task boundary, it can be further determined whether the node participation eligibility meets the conditions. If the node participation eligibility does not meet the conditions, node reorganization or termination of the current task can be performed; if the node participation eligibility meets the conditions, a task session can be established and local privacy computation can be performed, allowing site nodes that meet the participation conditions to enter the subsequent privacy computation process under the current task identifier and task boundary.

[0097] In some embodiments, performing local privacy computation based on task identifiers and determining common operating condition centers, representative scenarios, and scenario weights can be achieved through the following technical steps: The first step is for each station to allocate its local operating window to different operating condition categories, and to perform pruning, fixed-point encoding, and masking on the operating condition counts and the sum of operating condition feature vectors corresponding to each operating condition category to form protected aggregate quantities.

[0098] The local operating window represents a segment of operating data formed from the station's local operating data according to the current data window version. The protected aggregate quantity represents the result that can participate in cross-station aggregation, formed after the station performs pruning, fixed-point encoding, and masking on the sum of operating condition counts and operating condition feature vectors. The common operating condition center represents the shared operating characteristics formed by cross-station combined metering based on the protected aggregate quantities from multiple stations. The representative scenario represents a typical operating scenario corresponding to the corresponding common operating condition center. The scenario weight represents the proportion of the operating condition count corresponding to the representative scenario to the total valid operating condition count. The current predicted state represents the predicted operating state formed by the current predicted inflow, water level, equipment status, and node injection. The scenario boundary represents the representative scenario operating boundary determined jointly by the common operating condition center and the current predicted state.

[0099] The work condition category represents the category used to distinguish different operational characteristics under the current model version; the work condition count represents the number of local operation windows assigned to a certain station node to the corresponding work condition category; the sum of work condition feature vectors represents the result of summing the work condition feature vectors of each local operation window within the corresponding work condition category. Pruning can be used to limit the work condition count or the sum of work condition feature vectors to the allowed range of the current task registration; fixed-point encoding can be used to convert the values ​​to be aggregated into a numerical representation that matches the current protected aggregation process; masking can be used to protect the submission volume of a single station before cross-station aggregation. After these processes, the station submits protected aggregated data to the central side, rather than complete original data or unprotected single-station statistical results.

[0100] The second step is to restore the cross-station combined metering corresponding to each working condition category when the number of valid stations submitting the protected aggregate quantity is not less than the node threshold and the number of valid samples is not less than the sample threshold. The common working condition center is determined according to the ratio of the cross-station combined metering of the sum of working condition feature vectors to the cross-station total quantity of working condition counts. The corresponding scenario weight is determined according to the proportion of the working condition count corresponding to each working condition category to the total valid working condition count.

[0101] Among them, the number of valid stations can represent the number of station nodes that have completed valid protected aggregation submissions and meet the validity requirements of the current task; the number of valid samples can represent the number of valid running windows participating in this cross-site aggregation; the node threshold and the sample threshold represent the minimum number of stations and the minimum number of samples required to allow the recovery of cross-site combined metering, respectively. Cross-site combined metering can include cross-site combined metering of operating condition counts corresponding to each operating condition category and cross-site combined metering of the sum of operating condition feature vectors.

[0102] The relationship between the common operating condition center and the scene weight can be expressed as: ; ; in, Indicates the site-side node index; Indicates the working condition category index; This indicates the common operating condition center calculation round index; Indicates the index of the scheduling period; Indicates the scheduling period The set of station-end nodes; This represents the set of working condition categories included in the statistics; Represents the set of working condition categories The summation index in the table; Indicates the station node In the In the wheel calculation, it belongs to the working condition category. The number of running windows; Indicates the station node In the In the wheel calculation, it belongs to the working condition category. The sum of the characteristic vectors of the working conditions; Indicates the station node In the In the wheel calculation, it belongs to the working condition category. The number of running windows; Indicates according to the first Operating condition categories obtained from cross-station combined metering updates Common operating condition center; Indicates the first Wheel calculation working condition categories The corresponding scene weights.

[0103] Specifically, in terms of operating condition categories If the cross-station total metering of the corresponding operating condition count is greater than zero, the ratio of the cross-station total metering of the sum of the operating condition feature vectors corresponding to that operating condition category to the cross-station total metering of the operating condition count is determined as the updated common operating condition center for that operating condition category. ; Classify working conditions The ratio of the cross-site combined metering of the corresponding operating condition count to the sum of the cross-site combined metering of the operating condition counts for all operating condition categories is determined as the scenario weight corresponding to that operating condition category. The calculation of scenario weights is based on the condition that the sum of cross-site aggregated counts for all operating condition categories is greater than zero. Therefore, common operating condition information for subsequent scenario construction can be determined based on cross-site aggregation results without directly aggregating the raw data from each station.

[0104] The third step is to generate a representative scenario for each work condition category whose cross-station work condition count is greater than zero. When the cross-station work condition count for any work condition category is zero, no representative scenario is generated for that work condition category.

[0105] Specifically, a cross-station operating condition count greater than zero indicates that there is at least one valid operating window that can characterize the corresponding operating condition category, and a representative scenario can be formed based on the common operating condition center of this operating condition category; when the cross-station operating condition count is zero, there is no valid operating condition sample that can be used to form the representative scenario, so no corresponding representative scenario is generated, and this category is not included in the subsequent evaluation scenario calculation.

[0106] The fourth step is to determine the scene boundary corresponding to the representative scene based on the common working condition center and the current prediction state. The current prediction state includes the current predicted water inflow, water level, equipment status, and node injection.

[0107] Specifically, the common operating condition center is used to characterize the shared operational characteristics formed by multiple stations under corresponding operating condition categories, while the current predicted inflow, water level, equipment status, and node injection are used to characterize the real-time prediction conditions corresponding to the current task. Using both together to determine the scenario boundary allows subsequent feasible domain reconstruction to both adopt cross-station shared operational characteristics and be constrained by the current prediction state, avoiding the formation of an evaluation boundary inconsistent with the current physical state based solely on historical operating condition centers.

[0108] Furthermore, if either the number of valid stations or the number of valid samples fails to reach the corresponding threshold, cross-station combined metering and scenario weights will not be released to the central ordinary computing process.

[0109] If the number of valid stations is below the node threshold, cross-station aggregated measurement and scenario weights will not be released even if the number of valid samples reaches the sample threshold; similarly, if the number of valid stations is below the sample threshold, cross-station aggregated measurement and scenario weights will not be released even if the number of valid stations reaches the node threshold. Only when both thresholds are met simultaneously can the cross-station aggregated result enter the common operating condition center, the process of determining representative scenarios and scenario weights, thereby reducing the risk of single-station information exposure caused by releasing aggregated results when the participation scope is too small or the number of samples is insufficient.

[0110] In some embodiments, setting a temporary key and a pairwise mask corresponding to the task identifier and task session for the protected aggregate quantity can be performed through the following technical steps: The first step is that each station generates a temporary key for the task identifier and task session, and generates paired masks with opposite signs between each station. The paired masks are then superimposed on the sum of the condition count and the condition feature vector, so that the paired masks corresponding to the online stations cancel each other out during the cross-station summation process. The temporary key and the paired mask are not reused across task identifiers.

[0111] Here, a task session can represent the scope of a session under a task identifier used to complete a protected aggregate message interaction. A temporary key can represent key material used only for the current task identifier and the current task session. A pairwise mask can represent mask quantities formed jointly by two station nodes with opposite signs. For two station nodes that remain online and have completed the submission, their corresponding pairwise masks cancel each other out during cross-station summation, allowing the aggregated result to be recovered. The central ordinary computing process does not need to obtain the unmasked condition counts or the sum of condition feature vectors from a single station. Temporary keys and pairwise masks are not reused across task identifiers, ensuring that different tasks do not share the same set of mask relationships.

[0112] The second step is to restore the unoffset mask corresponding to the offline site when the offline proof is passed, the number of restored shares is not less than the restoration threshold, and the restoration share verification is passed. However, the mask corresponding to the still online site is not restored.

[0113] The offline proof represents the verification result confirming that the corresponding site node did not continue participating in the current session after completing the protected aggregation submission; the recovery share represents the share information used to recover the unoffset mask of the offline site; and the recovery threshold represents the minimum number of valid recovery shares required to allow unoffset mask recovery. The unoffset mask corresponding to the offline site is only recovered when all three conditions—offset proof, recovery share quantity, and recovery share verification—are met. Masks still on the online site remain unrecovered, thus limiting the offline recovery process to the mask range necessary to complete the aggregation.

[0114] Furthermore, if any of the following occurs: the disconnection proof fails, the recovery share is less than the recovery threshold, or the recovery share verification fails, the unoffset mask corresponding to the disconnected station will not be restored. Specifically, the three recovery conditions are simultaneously satisfied; if any one condition is not met, the unoffset mask of the currently disconnected station will not enter the recovery process, and the corresponding protected aggregation process can remain incomplete or not release the result. This avoids restoring the mask when the station is disconnected or the validity of the recovery share cannot be confirmed.

[0115] In some embodiments, a protected comparison between the number of valid samples and the sample threshold, as well as limiting the query frequency of overlapping tasks, can be performed through the following technical steps: The first step involves using either a trusted execution environment or a secure comparison protocol to compare the number of valid samples with the sample threshold. The central ordinary calculation process only obtains the threshold passing flag and the cross-site combined metering after the threshold is passed.

[0116] The threshold passing flag indicates whether the number of valid samples has reached the sample threshold. When using a trusted execution environment, the comparison between the number of valid samples and the sample threshold can be performed within the protected environment; when using a secure comparison protocol, the comparison result can be obtained without directly disclosing the number of valid samples to the central ordinary computing process. Before the threshold passes, the central ordinary computing process only obtains the pass / fail status required for threshold judgment, and does not obtain intermediate values ​​used to infer the sample size of a single station; after the threshold passes, it then receives the allowed cross-station combined metering.

[0117] The second step is to limit the query frequency of multiple tasks that are based on the same data window version and whose corresponding data windows overlap.

[0118] The data window version represents the data time window and its version information used to generate the current common operating condition center and representative scenario. Overlapping data windows indicate that different tasks use data windows that contain common data segments. For such tasks, identical or highly overlapping aggregation queries are not initiated continuously or without restriction; instead, they are processed according to the query frequency limit registered for the current task. This restriction, together with the aforementioned node threshold and sample threshold, reduces the possibility of repeatedly inferring single-station operation information through differences in results between multiple overlapping tasks.

[0119] In some embodiments, disabling uncallable assets can be achieved through the following technical steps: The first step is to configure at least one necessary node group for each callable asset.

[0120] Here, a necessary node group can represent a set of station-side nodes that can support the same callable asset to remain callable in the current task. A callable asset can be configured with one necessary node group, or multiple alternative necessary node groups can be configured; when multiple necessary node groups exist, as long as at least one of the necessary node groups meets the asset activation conditions, it can continue to determine whether the asset is within a valid topology connectivity component.

[0121] The second step is to determine the asset activation status of any callable asset as activated when the participation status, control authorization status, communication status, and task execution verification status of all station nodes in at least one necessary node group are valid, and the callable asset is located within the current valid topology connectivity component; otherwise, the asset activation status of the callable asset is determined as deactivated.

[0122] Among these, participation status indicates whether the corresponding station node belongs to the node range allowed to participate in the current task; control authorization status indicates whether the corresponding station node's control authority over the target asset is valid; communication status indicates whether the corresponding station node meets the communication participation conditions of the current task; task execution verification status indicates the verification status of the station node for the current task code, task environment, or task session; currently valid topology connectivity components indicate the hydraulic or power connection parts that are determined according to the current topology version and remain connected under the current task. Asset activation status can be represented by 1 and 0, where 1 indicates enabled and 0 indicates disabled.

[0123] The third step is to determine the callable capacity under the current task based on the common upper limit of the physical capacity, authorized capacity, task registration capacity and topology connectivity of the callable assets in the enabled state, and to determine the positive value indicator of the capacity based on whether the callable capacity is greater than zero.

[0124] Physical capacity represents the maximum capacity that the asset itself can provide under current physical conditions; authorized capacity represents the maximum capacity allowed by control authorization; task registration capacity represents the maximum capacity registered in the current task; and topology connectivity represents the maximum capacity allowed for transfer or provision by the asset under the current valid topology. The common upper limit can be determined by the minimum value among these capacity limits. Positive capacity indicators can be represented by 1 and 0, taking 1 when the available capacity under the current task is greater than zero, and 0 otherwise.

[0125] The relationship between asset activation status, available capacity under the current task, and positive capacity indicator can be expressed as: ; ; ; in, Indicates a callable asset index; Indicates the index of the scheduling period; Assets During the scheduling period The asset activation status; Assets During the scheduling period Is it located within the currently valid topological connected components? This indicates a conditional indicator function; it takes the value 1 if the condition within the parentheses is true, and 0 otherwise. This indicates that there exists at least one necessary group of nodes that satisfies the conditions within the parentheses; Assets The set of alternative necessary node groups; Indicates the necessary node group index; Assets The A necessary group of nodes; Indicates the site-side node index; Indicates the station node During the scheduling period The participation status; Indicates the station node During the scheduling period Control authorization status; Indicates the station node During the scheduling period The communication status; Indicates the station node During the scheduling period Task execution verification status; Assets Physical capacity; Indicates authorized capacity; Indicates the task registration capacity; Indicates topological connectivity; Indicates the available callable capacity for the current task; This indicates a positive capacity value.

[0126] The fourth step is to restrict the sum of the generator's power generation status indicator and pumping status indicator to no greater than the product of the asset activation status and the positive capacity indication, and to limit the generator's power generation flow rate to no greater than the product of the power generation status indicator and the maximum allowable flow rate corresponding to the power generation status, and to limit the generator's pumping flow rate to no greater than the product of the pumping status indicator and the maximum allowable flow rate corresponding to the pumping status.

[0127] The power generation and pumping status flags represent the state variables indicating whether the unit is in power generation or pumping mode, respectively. By subjecting the sum of these two status flags to both asset activation status and positive capacity indication, the unit can be prevented from entering power generation or pumping mode when the asset is off or the available capacity is zero. Furthermore, by limiting the power generation and pumping flow rates according to the corresponding status flags, the flow rate can be synchronously set to zero when the status flag is zero. Therefore, whether a callable asset participates in the subsequent feasible domain calculation is jointly determined by node status, authorization status, communication status, task verification status, topology status, and capacity.

[0128] In some embodiments, reconstructing the feasible domain of hydraulic and electric power tasks corresponding to the evaluation scenario can be achieved through the following technical steps: The first step is to disable the variables and capacities corresponding to callable assets that do not belong to the current alliance entity combination.

[0129] Here, "alliance entity combination" can represent a combination of one or more participating entities in the current task. "Candidate shared facility construction scheme" can represent the construction configuration to be evaluated for shared facilities among multiple hydropower stations. "Evaluation scenario" can represent the technical evaluation object formed by a specific alliance entity combination, a specific candidate shared facility construction scheme, and representative scenario conditions. "Hydraulic and power task feasible domain" can represent the set of feasible operating states under the current evaluation scenario, jointly defined by hydraulic operation constraints, power operation constraints, and cross-station sharing boundary constraints.

[0130] Specifically, based on the mapping relationship between the participating entities, station nodes, and callable assets bound to the current task identifier, the station nodes and callable assets that the current alliance entity combination can use are determined. For participating entities that do not belong to the current alliance entity combination, their corresponding units, reservoir capacity, water transmission channels, transmission lines, or control capabilities that cannot be called by the current alliance are not included in the current evaluation scenario, and related variables and capacities are treated as closed. This processing, together with the aforementioned asset activation status, determines the actual range of resources that can be used in the current evaluation scenario.

[0131] The second step involves regenerating water balance constraints, reservoir capacity continuity constraints, unit operating condition constraints, node power balance constraints, line capacity constraints, and shared boundary consistency constraints based on the scenario boundaries, thus forming the feasible domain for the hydraulic and power tasks corresponding to the current evaluation scenario.

[0132] Specifically, the scenario boundary provides the operational boundary jointly determined by the current representative scenario and the current prediction state, while the asset activation status and available capacity provide the range of assets and capacity that the current alliance entity combination can use under the current candidate shared facility construction scheme. Based on this, station-side local variables, the central shared boundary, and cross-station water or power conservation relationships jointly form the task feasible region of the current evaluation scenario. The task feasible region can be represented as: ; in, Indicates the scheduling period The corresponding feasible domains for hydraulic and electric power tasks; Represents the set of all local variables at the station level; Indicates that the center shares a boundary; Indicates the site-side node index; Indicates the scheduling period The set of station-end nodes; Indicates the station node Local variables; This represents the set of asset activation statuses corresponding to the current evaluation scenario; This represents the set of available task capacity corresponding to the current evaluation scenario; This represents the locally feasible set of stations, which is jointly defined by the current asset activation status, the current available capacity of the task, water volume relationships, unit operating conditions, and station-side safety constraints. This represents the use of station nodes in the unified boundary mapping. The mapping of local variables to the space corresponding to the shared boundary; This represents the function used in unified boundary mapping to map the central shared boundary to station nodes. Mapping to the corresponding boundary space; Indicates the scheduling period The corresponding shared boundary set; Indicates the station node The corresponding cross-station water volume or power conservation mapping; Indicates the scheduling period The corresponding external physical quantity. Only , Only when both the cross-site conservation relationship and the condition are met can the corresponding operating state belong to the feasible domain of the current task.

[0133] For water balance and reservoir capacity continuity constraints, the relationship between reservoir capacity changes during adjacent scheduling periods can be represented. For example, the reservoir capacity continuity relationship can be expressed as: ; in, Represents a reservoir index; Indicates the index of the scheduling period; Reservoir During the scheduling period Storage capacity; Reservoir Storage capacity in the next scheduling period; Indicates the length of the scheduling period; Reservoir During the scheduling period External inflow; Reservoir During the scheduling period Outflow rate; Indicates the relationship with the reservoir A collection of generating units or water transmission channels where water exchange occurs; Indicates the index of the generating unit or water transmission channel; Indicates the generating unit or water supply channel During the scheduling period Traffic; This indicates the corresponding flow rate relative to the reservoir. The directional relationship. Through this reservoir capacity continuity relationship, the inflow, outflow, and unit or channel flow under the current scenario boundary can be reflected in the reservoir capacity continuity conditions of adjacent time periods.

[0134] For unit operating condition constraints, node power balance constraints, and line capacity constraints, the unit extraction and conversion, operating condition gating, and grid node and line boundaries can be represented as follows: ; ; ; ; ; ; in, Indicates the unit index; Indicates the index of the scheduling period; Indicates the unit During the scheduling period The power generation capacity; Indicates the unit During the scheduling period Pumping power; Indicates the density of water; Represents gravitational acceleration; Indicates the unit During the scheduling period The net head or net lift; Indicates the overall efficiency under power generation conditions; Indicates the overall efficiency of the pumping operation; Indicates power generation flow rate; Indicates the pumping flow rate; Indicates power generation operating condition; Indicates pumping operation status; Indicates the unit The maximum power generation flow allowed through the unit when it is in power generation operation; Indicates the unit The maximum pumping flow rate allowed through the unit when it is in pumping operation.

[0135] Represents the index of the power grid node; Indicates mapping to power grid nodes The collection of station terminals; Indicates the site-side node index; Indicates the station node Mapped to power grid nodes During the scheduling period Net injection power; Indicates the scheduling period Internally, in addition to mapping to power grid nodes In addition to participating hydropower stations, other power sources supply grid nodes Injected active power; Indicates the scheduling period Internally, connected to the power grid node The active power consumed by the electrical load.

[0136] Indicates the line index; Represents a power grid node With the line The direction of the relationship between them; Indicates the line During the scheduling period Line power; Indicates the line Line coefficients in the current DC power relationship; Indicates the line During the scheduling period, both ends of the node The phase angle difference; Indicates the line The asset activation status; Indicates the line Available capacity under the current task. By using power generation and pumping status indicators, asset activation status, positive capacity indicators, and line available capacity, assets that are shut down or have zero available capacity can be prevented from entering effective operating conditions.

[0137] The third step is to determine the corresponding evaluation scenario as a physically infeasible scenario when any of the following occurs: the physical waterway is disconnected, the necessary electrical connection component lacks a reference node, or the shared boundary projection set is in the air. Instead, candidate running results for the physically infeasible scenario will not be generated.

[0138] Specifically, when the physical waterway is disconnected, the hydraulic relationship between the current alliance entity combination and the candidate shared facility construction scheme cannot form an effective water transfer path that satisfies the scenario boundary; when the necessary electrical connectivity component lacks reference nodes, the current electrical connectivity part cannot satisfy the reference conditions required for the power relationship of the corresponding node; when the shared boundary projection set is empty, there is no feasible intersection between the station-end local boundary and the central shared boundary that can simultaneously satisfy the shared capacity and conservation relationship. In any of these cases, the corresponding evaluation scenario is determined to be physically infeasible before the candidate operation results are generated, thereby avoiding continued joint coordination for scenarios without physically feasible solutions.

[0139] In some embodiments, when control authorization or topology connectivity changes, version updates and feasible domain reconstruction can be performed through the following technical steps: The first step is to update the authorization version when control authorization changes. Specifically, when the effective scope or capacity of control authorization changes, the corresponding authorization information is no longer completely consistent with the original authorization version under the current task identifier. Therefore, an updated authorization version is generated to represent the changed control permission status.

[0140] The second step involves updating the topology version whenever the connectivity of either the water supply channel or the power transmission line changes. The topology version can reflect both the hydraulic and electrical connections used in the current task. When the connection status of the water supply channel or the power transmission line changes, the corresponding hydraulic or electrical connectivity changes, thus requiring an update to the topology version. This ensures that subsequent asset activation status and task feasible regions are determined based on the changed topology.

[0141] The third step involves generating a new task identifier when either the authorized version or the topology version changes. Both the authorized version and the topology version belong to the task boundary corresponding to the current task identifier; therefore, a change in either one indicates a change in the original task boundary. At this point, a new task identifier is generated, and subsequent asset activation, scene construction, task feasible domain reconstruction, and candidate execution result calculation are transferred to the new task identifier, instead of continuing to use the original task boundary.

[0142] The fourth step is to switch the main transmission line to the registered backup line when the main transmission line is disconnected and a registered backup line exists, update the topology version, and reconstruct the corresponding hydraulic and power task feasible domains based on the new task identifier according to the capacity of the registered backup line.

[0143] Specifically, the backup line must already exist in the current task's registration information. If a registered backup line exists after the primary outgoing line is disconnected, the effective outgoing path will be switched to that backup line, and the registered capacity of the backup line will be used as the corresponding line capacity boundary. Since the connectivity of the outgoing lines changes, the topology version is updated first, and a new task identifier is generated. Then, asset activation judgment and task feasible domain construction are re-performed based on the new topology. This avoids using the original line capacity and topology for joint computation even after the primary outgoing line has been disconnected.

[0144] In some embodiments, coordinating local results at various stations within the feasible domain of hydraulic and power tasks to obtain candidate operational results can be achieved through the following technical steps: The first step is to perform station-level local updates, central shared boundary updates, and multiplier updates based on the same boundary mapping. The same boundary mapping is used to establish the correspondence between local variables at each station and the central shared boundary.

[0145] First, local results can represent the local operational calculation results formed by each station node within the local feasible set corresponding to the current evaluation scenario. Candidate operational results can represent the joint operational results formed after coordinating multiple station local results based on the same boundary mapping. The complete local model can represent the local physical model stored at the corresponding station, containing complete water level and reservoir capacity curves, complete unit characteristics, safe operating range, and station line constraints.

[0146] Specifically, the station-side local update, the central shared boundary update, and the multiplier update all adopt the unified boundary mapping used when constructing the feasible domain of the current hydraulic and electric power tasks. Instead of using different boundary variable definitions in the three update processes, local station updates are performed within the corresponding local feasible set; central shared boundary updates are performed within the allowable range of the current shared boundary; and multiplier updates are performed based on the consistency deviation between the local station boundary and the central shared boundary. The relationship between the three types of updates can be expressed as: ; ; ; ; in, Indicates the iteration round of joint coordination; Indicates the station node In the The local variables obtained by adding one round; Indicates the station node The station-side objective adopted under the current task; Indicates the first The consistency multiplier of the wheel; Indicates a positive augmentation coefficient; This represents the use of station nodes in the unified boundary mapping. The mapping of local variables to the space corresponding to the shared boundary; This represents the function used in unified boundary mapping to map the central shared boundary to station nodes. Mapping to the corresponding boundary space; Indicates the station node The corresponding dimensionless positive definite scaling matrix; Indicates the first The centers of the wheels share the same boundary; Indicates the first The centers of the wheels share the same boundary; Indicates the scheduling period The intersection of the set of internal shared capacities and the set of global conservation constraints; This represents the cross-site conservation mapping within the feasible region of the current hydraulic and power tasks; This represents the external physical quantity within the feasible domain of the current hydraulic and power tasks; This represents the shared boundary conservation mapping registered in the task topology template, which corresponds to the station-side cross-station conservation mapping on the feasible boundary. Indicates that the center shares a boundary; Indicates transpose; Represents the L2 norm; This represents the variable that minimizes the objective function within the corresponding feasible set.

[0147] The second step involves determining the local consistency residual, global physical conservation residual, adjacent iteration shared boundary change index, and local feasibility flag for each station based on the station-side local update, central shared boundary update, and multiplier update.

[0148] Among them, the local consistency residual can represent the consistency deviation between the local boundary mapping results of each station and the central shared boundary; the global physical conservation residual can represent the deviation of the combination of local results from multiple stations relative to the cross-station water or power conservation relationship; the adjacent iteration shared boundary change index can represent the degree of change between the central shared boundary in two consecutive rounds; and the local feasibility flag can represent whether the local update result of the corresponding station meets the current local feasible set of that station. The three types of residuals and the local feasibility flag are used together to determine whether the current joint coordination can form a fixed candidate operating result.

[0149] The three types of stopping indicators and the combined stopping conditions can be expressed as follows: ; ; ; ; in, Indicates the first Local consistency residuals of the wheel; Indicates the first Global physical conservation residuals; Indicates the first Adjacent iterations share boundary change indices; This represents the scaling matrix used for global physical conservation residual normalization; This represents the scaling matrix used for normalizing the shared boundary change index; Indicates the station node In the The local feasibility flag for the wheel is set to 1, indicating that it is locally feasible. This represents the stopping threshold corresponding to the locally consistent residual. This represents the stopping threshold corresponding to the global physical conservation residual; This indicates the stopping threshold corresponding to the change index of the shared boundary between adjacent iterations; Indicates the first The indication result of whether the wheel meets the joint stopping conditions. This represents a conditional indicator function; it takes the value 1 if the condition within the parentheses is true, and 0 otherwise. The joint stopping condition is only satisfied when all three types of residuals are no greater than their corresponding thresholds and all local feasibility flags at all stations indicate local feasibility. Indicates the station node The corresponding dimensionless positive definite scaling matrix; This represents the cross-site conservation mapping within the feasible region of the current hydraulic and power tasks; Indicates the station node In the Local variables of the wheel; d This represents the external physical quantity corresponding to the current scheduling period; Indicates the first The center of the wheel shares the boundary.

[0150] The third step is to generate fixed candidate running results and corresponding candidate running result versions when the local consistency residual, global physical conservation residual, and adjacent iteration shared boundary change index are all no greater than the corresponding thresholds, and all local feasibility flags corresponding to all stations indicate local feasibility.

[0151] Specifically, after the joint stopping conditions are met, the corresponding station-side local results, central shared boundaries, and related scenario information for this round are fixed as candidate running results, and candidate running result versions are set for these results. The candidate running result versions correspond to the current task identifier and evaluation scenario, and are used for subsequent station-side physical verification and control plan version binding. Numerical stopping only indicates that the joint coordination process has met the currently registered stopping conditions and does not replace the subsequent physical verification of the complete local model.

[0152] Fourth, if the maximum number of iterations is reached but the conditions that the local consistency residual, global physical conservation residual, and adjacent iteration shared boundary change index are all not greater than the corresponding thresholds, and the local feasibility flags corresponding to all stations indicate local feasibility are not met, then candidate running results will not be published.

[0153] When the maximum number of iterations is reached, if any of the three types of residuals is still greater than the corresponding threshold, or if at least one station's local feasibility flag indicates local infeasibility, no published candidate running results will be generated for subsequent physical verification and actual operation. This avoids using intermediate iteration results that do not meet the unified boundary and global conservation stopping conditions as candidate running results.

[0154] In some embodiments, central shared boundary updates can be performed through the following technical steps: The first step is to determine the target conservation set based on the task topology template, wherein the target conservation set includes at least one of the cross-station water conservation set and the power conservation set.

[0155] The task topology template can represent a template corresponding to the current task topology version, used to register shared boundary variables and their water or power conservation relationships. The target conservation set is determined based on the shared hydraulic and shared electrical relationships actually involved in the current evaluation scenario; when the evaluation scenario involves only one type of shared relationship, a corresponding conservation set can be used; when both cross-station water and power relationships are involved, both types of conservation relationships can be included simultaneously.

[0156] The second step is to determine the intersection of the shared capacity set and the objective conservation set.

[0157] The shared capacity set represents the allowable range of the central shared boundary, jointly defined by the current shared facility capacity, the asset callable capacity, and the scenario boundary. Intersecting the shared capacity set with the objective conservation set yields a feasible set of central shared boundaries that satisfies both the current shared capacity boundary and the corresponding cross-site conservation relationship.

[0158] The third step is to perform weighted least squares projection on the boundary mapping results corresponding to the local variables of each station within the intersection, and update the central shared boundary based on the projection results.

[0159] Specifically, after each station completes its local update, the shared boundary mapping result for each station is obtained through a unified boundary mapping. The central side does not directly take values ​​within arbitrary boundary ranges, but performs a weighted least squares projection within the intersection of the shared capacity set and the target conservation set. This ensures that the updated central shared boundary is simultaneously constrained by the shared capacity and cross-station conservation conditions, and is used for the next round of station local updates and multiplier updates.

[0160] Fourth, when the intersection is empty, no candidate running results are generated, and the feasible domain of the hydraulic and electric tasks is reconstructed.

[0161] An empty intersection indicates that there is no central shared boundary that can be mutually satisfied between the current shared capacity range and the target conservation condition. In this case, continued joint coordination cannot generate candidate operation results that conform to the current task boundary. Therefore, no candidate operation results are generated, and the feasible domain of the hydraulic and power tasks is reconstructed based on the current asset status, capacity, scenario boundary, and topology.

[0162] In some embodiments, each station performs a physical review of the candidate running results based on a complete local model, which can be done through the following technical steps: The first step involves each station checking the water balance, unit operating conditions, water level and reservoir capacity boundaries, line injection boundaries, and station safety constraints for fixed candidate operating results, based on the complete water level and reservoir capacity curve, complete unit characteristics, safe operating range, and station line constraints. The normalized violation amount of each constraint is then determined according to the corresponding physical verification scale.

[0163] The physical verification scale can represent the tolerance scale or nominal scale corresponding to the corresponding physical constraint, used to convert the degree of constraint violation of different dimensions into a normalized quantity that can be compared with its respective threshold. The physical verification uses the complete local model retained at the station, and does not replace the complete water level and reservoir capacity curve, complete unit characteristics, safe operating range, or station line constraints with the simplified boundary relationships used in the central coordination process. Each constraint can be converted into a form not greater than zero, and then a normalized violation quantity is formed based on the positive part of the constraint violation and the corresponding physical verification scale.

[0164] The second step is to return the physical verification result when all normalization violations are not greater than the corresponding threshold.

[0165] Specifically, each station checks the corresponding constraints involved in the current candidate operation result. Only when all normalization violations that the current station needs to check do not exceed the corresponding threshold will the station return a physical verification pass result; multiple stations complete the checks according to their own complete local models, so that the candidate results used in subsequent actual operation and technical evaluation have a verification status corresponding to the station's physical constraints.

[0166] The third step is to return a physical review failure result when at least one normalization violation is greater than the corresponding threshold. Only the constraint category, the maximum normalization violation, and the allowable adjustment boundary are returned, without returning the complete original data used to determine the normalization violation.

[0167] Among these, the constraint category can represent the type of constraint that exceeded the limit; the maximum normalization violation can represent the maximum normalization violation among all the exceeding items in the current review; and the allowable adjustment boundary can represent the boundary adjustment range that the station allows the central coordination process to use in subsequent re-coordination. By returning only this review summary information, the central side can identify the reasons why candidate running results failed the review and the adjustable range, while the complete original data and the complete local model are still retained at the station.

[0168] Furthermore, for evaluation scenarios that fail the physical verification, the corresponding evaluation scenario results will not be used for technical evaluation. Therefore, the numerical stopping results from the joint coordination phase can only proceed to subsequent actual operation processing or technical evaluation after the complete local model verification at the station has passed. Evaluation scenarios that fail the physical verification will not be included in the subsequent determination of technical losses, technical value, marginal technical contributions, and the proportion of construction responsibility allocation.

[0169] Figure 3 A schematic diagram illustrating common operating conditions and joint coordination processes according to some embodiments of this disclosure is shown. (Reference) Figure 3 As shown, the local operating window and operating condition characteristics at the station, as well as the current prediction status, can be obtained first. Then, dual-threshold protection aggregation is performed on the local operating window and operating condition characteristics. After completing the dual-threshold protection aggregation, common operating condition centers, representative scenarios, and scenario weights can be determined based on the aggregation results, and scenario boundaries can be further determined in conjunction with the current prediction status. Based on the determined representative scenarios, scenario weights, and scenario boundaries, corresponding evaluation scenarios can be formed, allowing subsequent processing to be performed on the current evaluation scenario. After forming the evaluation scenarios, the feasible domain of hydraulic and electric power tasks can be reconstructed according to the task conditions corresponding to the current evaluation scenario. Within the reconstructed feasible domain of hydraulic and electric power tasks, local station results are coordinated to form candidate operating results.

[0170] After candidate execution results are generated, each station can perform a physical review based on the complete local model and determine whether the physical review passes. If the physical review fails, local result coordination or task feasibility domain reconstruction can be performed again, and candidate execution results and station-level complete local model physical review can be performed again based on the adjusted results. If the physical review passes, control authorization verification and security verification can be further performed. After the control authorization verification and security verification pass, control reference values ​​can be issued and execution feedback can be collected, so that the actual execution feedback can correspond to the candidate execution results that have passed the physical review, and can be used for subsequent result status judgment and related scenario processing.

[0171] In some embodiments, sending the target candidate execution results to the corresponding station and receiving execution feedback can be done through the following technical steps: The first step is to set a candidate operation result version for the candidate operation result and generate a control plan version based on the target candidate operation result. The control plan version is bound to the candidate operation result version corresponding to the target candidate operation result. The control plan version includes at least one of the following: unit power generation, pumping power, flow rate, water level boundary and line injection limit.

[0172] Here, the target candidate operating result can represent the candidate operating result determined from the candidate operating results that have passed physical verification and used for actual operation. Execution feedback can represent the actual observation information collected by the station after the control reference value corresponding to the target candidate operating result is provided to the existing control interface at the station, and the deviation information between it and the control reference value.

[0173] A control plan version can represent a version of control reference information formed based on target candidate operational results that have passed physical verification and been determined for actual operation. A control plan version is bound to its source candidate operational result version, ensuring that subsequent authorization verification, security verification, and execution feedback are all mapped to the candidate operational results that formed the control plan. The specific control reference information included in the control plan is determined by the actual control objects involved in the target candidate operational results.

[0174] The second step is to provide the control reference value corresponding to the control plan version to the existing control interface at the station when both the control authorization verification and security verification pass, and to ensure that the control reference value does not cover the existing protection and interlocking at the station.

[0175] Specifically, the control authorization verification confirms that the objects, scope, and capacity involved in the current control plan are still within the current authorization range; the security verification confirms that the control plan meets the security conditions allowed by the corresponding station. Only when both verifications pass will the control reference value be provided to the existing control interface at the station. The control reference value serves as a reference input corresponding to the current collaborative operation result and does not change the operational boundaries of the existing protection and interlocking at the station. If either verification fails, the control reference value corresponding to the current control plan version will not be used in the actual operation at the station.

[0176] The third step is to normalize the difference between the observed value and the control reference value according to the feedback scale to obtain the feedback deviation, and then bind the execution feedback and control plan version, including the feedback deviation, as well as the candidate operation result version corresponding to the target candidate operation result.

[0177] Here, the observed value can represent the corresponding actual operating quantity collected after the control reference value is provided to the existing control interface at the station; the feedback scale can represent the scale used to normalize the difference between the observed value and the control reference value. The execution feedback is simultaneously bound to the control plan version and the candidate operating result version, so that subsequent result failures and recalculations of affected scenarios can be mapped to specific control plans and candidate operating results based on the feedback.

[0178] The normalized violation and the feedback deviation after actual operation can be expressed as: ; ; in, Indicates the station node The corresponding normalized physical violation quantity; Indicates the physical constraint index at the station end; Indicates the station node The A physical constraint function is converted into a constraint form that is not greater than zero; Indicates the first Violation of the constraint; This indicates the physical verification scale corresponding to the respective constraints; Indicates the station node The corresponding feedback deviation; Indicates the index of the control reference item; This represents the observed value of the corresponding control reference item; Indicates the corresponding control reference value; This indicates the feedback scale for the corresponding control reference item. By employing both the physical verification scale and the feedback scale, the degree of violation and feedback deviation corresponding to different physical quantities can be compared under their respective registered scales.

[0179] Figure 4 A schematic diagram illustrating the candidate execution result verification and failure process according to some embodiments of this disclosure is shown. (Reference) Figure 4 As shown, candidate execution results and versions can be obtained first, and a complete local model physical verification can be performed on the candidate execution results. After completing the complete local model physical verification, control authorization verification and security verification can be further performed. After the corresponding verifications pass, control reference values ​​are issued, and then execution feedback is collected. After collecting execution feedback, it can be determined whether the physical verification result has been revoked or whether the feedback deviation exceeds the threshold. When the physical verification result has not been revoked and the feedback deviation has not exceeded the threshold, the current version can be kept valid, so that the current candidate execution result and corresponding control processing remain valid.

[0180] When a physical review result is revoked or a feedback deviation exceeds a threshold, the control plan version and the execution result version can be invalidated. After the corresponding version becomes invalid, subsequent processing can be performed based on the type of change that caused the version invalidation. Specifically, when the task boundary changes, appropriate processing can be performed according to the changed task boundary; when execution deviations or coordination parameter adjustments occur, the corresponding execution results can be reprocessed; when callable assets or control authorizations change, appropriate processing can be performed based on the changed asset status or control authorization status.

[0181] In some embodiments, the technical evaluation of the evaluation scenario results that have passed the physical verification can be carried out through the following technical steps: The first step is to obtain at least one hydraulic technical indicator and at least one power safety indicator and reliability indicator corresponding to the evaluation scenario results that have passed the physical verification; wherein, the hydraulic technical indicator includes at least one of water wastage and operating water loss, and the power safety indicator and reliability indicator includes at least one of power supply deficit, line congestion, reserve insufficiency and constraint violation.

[0182] The technical evaluation results can represent the evaluation results based on the physically verified evaluation scenarios. These results are generated after processing different combinations of alliance entities and different candidate shared facility construction schemes according to a unified technical indicator standard. Specifically, after the evaluation scenario undergoes task feasibility domain reconstruction, joint coordination, and physical verification using a complete local model, original technical indicators corresponding to that scenario can be generated. To ensure that the technical evaluation simultaneously reflects hydraulic operation and power safety or reliability, at least one hydraulic technical indicator and at least one power safety or reliability indicator are selected. Different evaluation scenarios are compared using the same indicator standard.

[0183] The second step is to divide each indicator by its corresponding pre-fixed positive scale to obtain the normalized indicators.

[0184] Among them, the positive scale can represent a normalized scale that is fixed in advance and is greater than zero for the corresponding technical indicators. When water wastage, power supply deficit, line congestion, reserve shortage, constraint violation, or operational water loss have different dimensions, they are normalized using their respective corresponding positive scales, and the unnormalized indicators with different dimensions are not directly added together.

[0185] The third step is to weight each normalized indicator according to the preset technical loss weight to obtain the scene technical loss of the corresponding evaluation scenario, and to include the scene technical loss as a component of the technical evaluation result.

[0186] Here, the technical loss weight can represent the non-negative weight registered for each technical loss indicator in the construction scheme rule version, and the sum of all effective technical loss weights can be 1. The determination relationship of scenario technical loss can be expressed as: ; in, Indicates the combination of the main entities in the alliance; Indicates candidate shared facility construction plans; Indicates the combination of alliance entities Adopt candidate shared facility construction scheme The corresponding evaluation scenario's technical loss; This represents the set of technical indicators used in the current technical loss calculation; Indicates a technical indicator index; Indicates the first evaluation scenario. Non-negative original technical loss indicators; Indicates the first The fixed positive scale corresponding to each technical loss indicator, and Greater than zero; Indicates the first The technical loss weights corresponding to each technical loss indicator The sum of all non-negative and effective technical loss weights is 1. By normalizing first and then weighting, technical indicators with different dimensions can participate in the evaluation of a unified dimensionless technical loss.

[0187] Fourth, if any indicator used to determine the technical loss of a scenario is missing and cannot be supplemented based on audited data under the same task identifier, the corresponding evaluation scenario will not be used for technical evaluation.

[0188] Specifically, when an indicator is missing, the system first checks if there is audited data under the same task identifier that corresponds to the current evaluation scenario and can fill the missing indicator. If the missing data can be filled, the audited data is used to calculate the corresponding indicator; if the missing data cannot be filled, a manually assigned single value is not used as a substitute, and the evaluation scenario is not used for calculations related to scenario-based technical losses, technical value, and construction responsibility. This ensures the verifiability of the technical indicators used in different evaluation scenarios.

[0189] In some embodiments, before determining the reference technology loss using a fixed reference scenario, the reference scenario for the corresponding alliance subject combination can be determined and fixed through the following technical steps: The first step is to determine the independent operation scheme corresponding to the alliance entity combination as physically feasible, and then define the independent operation scheme as the fixed reference scenario for the alliance entity combination.

[0190] The independent operation scheme can represent a reference scheme formed by the corresponding alliance entity combination according to the pre-registered independent operation boundary, without adopting the current candidate shared facility construction scheme. The independent operation scheme needs to undergo task feasibility domain judgment and station-side physical verification corresponding to the candidate evaluation scenario; only when it is determined to be physically feasible will it be used as a fixed reference scenario.

[0191] The second step is to determine that when the independent operation plan corresponding to the alliance's main body combination is determined to be physically infeasible, the external alternative plan registered before the establishment of the construction plan evaluation task will be determined as the fixed reference scenario for the alliance's main body combination.

[0192] Here, an external alternative can refer to an external alternative that has been registered before the current construction scheme evaluation task is established, and is intended to form a reference technical loss when the independent operating scheme is physically infeasible. The external alternative is not constructed temporarily after the technical evaluation results of the candidate construction schemes are obtained, but is registered before the construction scheme evaluation task is established.

[0193] The third step involves binding the participating entity snapshots, task identifiers, model versions, topology versions, capacity and service levels, representative scenarios, scenario weights, original technical loss indicators, positive scale, technical loss weights, site-side physical verification results, and audit evidence to a fixed reference scenario, and keeping the fixed reference scenario unchanged during the task.

[0194] Specifically, the fixed reference scenario not only stores a reference technical loss value, but also the scenario boundaries, version, original technical indicators, normalization scale, technical loss weight, and site-side physical verification results required to form that reference technical loss. During the task, the fixed reference scenario is not temporarily changed based on the evaluation results of candidate construction schemes, thus ensuring that the reference technical loss and candidate technical loss in the same task adopt comparable fixed reference conditions.

[0195] In some embodiments, the determination of technological value and marginal technological contribution can be carried out through the following technical steps: The first step is to obtain the reference technology loss of any alliance entity combination and any candidate shared facility construction scheme under a fixed reference scenario, as well as the candidate technology loss of the alliance entity combination after adopting the candidate shared facility construction scheme and completing coordination and physical verification. Based on the difference between the reference technology loss and the candidate technology loss, the technical value of the alliance entity combination under the candidate shared facility construction scheme is determined.

[0196] The reference technical loss can be represented as the technical loss obtained by calculating the fixed reference scenario using the same technical indicators, positive scale, and technical loss weights as the candidate evaluation scenario, and then undergoing physical verification at the station. The candidate technical loss can be represented as the scenario technical loss obtained by adopting the corresponding candidate shared facility construction scheme for the current alliance entity combination, after task feasibility domain reconstruction, local result coordination, and station-level physical verification. The reference technical loss minus the candidate technical loss forms the technical value, which is used to characterize the change in technical loss brought about by the candidate shared facility construction scheme relative to the fixed reference scenario.

[0197] The second step involves identifying, for any participating entity, a first alliance entity combination containing the participating entity and a second alliance entity combination formed by removing participating entities from the first alliance entity combination under the same candidate shared facility construction scheme. Based on the difference between the technical value of the first alliance entity combination and the technical value of the second alliance entity combination under the candidate shared facility construction scheme, the marginal technical contribution of the participating entity is determined. The fixed reference scenario corresponding to the alliance entity combination and the evaluation scenario formed by the alliance entity combination using the candidate shared facility construction scheme adopt the same technical indicator caliber, positive scale, technical loss weight, model version, and physical verification rules.

[0198] Specifically, the second alliance entity combination can be denoted as one that does not include the target participating entities. Alliance main body combination The first alliance's main body combination is denoted as The two alliance entities combined in the same candidate shared facility construction plan. Each entity then completes the reconstruction of its own feasible domain, joint coordination, and physical verification, corresponding to its own scope, and then determines its technological value based on the same technical indicator standards. The difference in technological value between the two entities is used as the basis for determining the participating entities. Compared to the main combination of the alliance Marginal technological contribution.

[0199] The third step is to determine the reference technical loss based on the original technical loss index, positive scale, and technical loss weight corresponding to the fixed reference scenario, and to conduct a site-side physical verification of the fixed reference scenario.

[0200] Technological value and marginal technological contribution can be expressed as: ; ; in, This indicates that the target participants are not included. The main body of the alliance; Indicates candidate shared facility construction plans; Indicates the combination of alliance entities Reference technology loss in a fixed reference scenario; Indicates the combination of alliance entities Adopt candidate shared facility construction scheme And complete the joint coordination and station-side physical verification of candidate technology losses; Indicates the combination of alliance entities In candidate shared facility construction plans The technological value of the following; Indicates the target participants; Indicates the combination of the main body of the alliance Add participating entities The alliance's main components were subsequently formed; Indicates the participating entities Compared to the main combination of the alliance In candidate shared facility construction plans Marginal technological contribution.

[0201] In some embodiments, the determination of the types of contributions required for the allocation of comprehensive marginal technical contributions and construction responsibilities can be carried out through the following technical steps: The first step is to weight the marginal technical contribution of any participating entity under the same candidate shared facility construction scheme according to the marginal technical contribution corresponding to the different joining order of the participating entity in different alliance entity combinations, so as to obtain the comprehensive marginal technical contribution of the participating entity.

[0202] Specifically, the order in which the same participating entity joins different alliance entity combinations varies, and the alliance entity combinations it previously belonged to also differ. Therefore, the corresponding marginal technical contributions may differ. The marginal technical contributions of the comprehensive system are weighted among these marginal technical contributions under these different alliance entity combinations, so that the comprehensive result is formed by multiple scenario results that have completed the reconstruction of the feasible domain, coordination of local results, and physical verification, rather than using only the marginal technical contributions of a single alliance entity combination.

[0203] The second step involves using the comprehensive marginal technology contribution as one of the bases for determining the construction responsibility allocation ratio, and combining it with at least two of the following: the shared facility utilization contribution after physical verification, the new capacity demand, and the reliability contribution to determine the construction responsibility allocation ratio.

[0204] Among them, the contribution of shared facility utilization can represent the contribution of the corresponding participating entity to the actual utilization level of the shared facility in the evaluation scenario that has passed the physical verification; the new capacity demand can represent the new capacity demand of the corresponding participating entity under the candidate shared facility construction plan; and the reliability contribution can represent the contribution formed by the sum of the reliability improvements of the corresponding participating entities. The comprehensive marginal technical contribution is not used as the sole basis for the allocation ratio of construction responsibility, but is used in combination with the contribution generated by these physically verified scenarios.

[0205] In some embodiments, the calculation of the comprehensive marginal technological contribution based on the number of participating entities can be performed through the following technical steps: The first step is to reconstruct the feasible domain of hydraulic and electric tasks, coordinate local results, and conduct physical verification for all alliance entity combinations when the number of participating entities is no greater than the upper limit of precise enumeration. Based on the technical value corresponding to each alliance entity combination, the marginal technical contribution of each participating entity under different alliance entity combinations is determined, and the comprehensive marginal technical contribution of each participating entity is formed based on the marginal technical contribution.

[0206] The precise enumeration upper limit represents the maximum number of participating entities allowed when calculating each combination of alliance entities in the current construction scheme evaluation task. When the number of participating entities is not greater than this upper limit, alliance entity combinations are not omitted; each alliance entity combination is re-determined according to the same task process, the callable assets are reconstructed, the feasible domain of the task is restructured, local results are coordinated, and station-side physical verification is performed before determining the corresponding marginal technical contribution.

[0207] The second step is to estimate the comprehensive marginal technical contribution of each participant by random sampling when the number of participants exceeds the upper limit of precise enumeration, and to determine the half-width of the confidence interval for the estimated value of each participant.

[0208] Random permutation sampling can represent the random arrangement of the joining order of participating entities, and the determination of the corresponding marginal technical contribution sample based on the set of participating entities before the target participating entity joins in each random permutation. During the sampling process, the candidate shared facility construction scheme remains unchanged, and the corresponding alliance entity combination still generates technical value according to the current task's feasible domain reconstruction, joint coordination, and physical verification rules.

[0209] The third step is to stop sampling when the half-width of the confidence interval for the estimated values ​​of all participating entities is no greater than the error limit, and to use the estimated results of the comprehensive marginal technical contribution to determine the proportion of construction responsibility allocation.

[0210] The fourth step is to mark the estimation result of the combined marginal technical contribution as not reaching the target accuracy when there is still a confidence interval half-width greater than the error bound after reaching the maximum number of samplings.

[0211] The comprehensive marginal technical contribution estimate formed by random sampling and the stopping condition can be expressed as: ; ; in, Indicates the process Participating entities obtained from random permutation sampling The estimated comprehensive marginal technology contribution; Indicates the current sampling number; This indicates a randomized arrangement of sample indices; Indicates the first A random arrangement; Indicates the first In a random arrangement, the participating entities are located in The previous set of participating entities; Indicating the candidate shared facility construction plan The following will be the participating entities join in The technological value corresponding to the subsequent combination of alliance entities; Indicates joining the participating entities The technological value corresponding to the previous alliance's main components; Indicates confidence level The corresponding quantile coefficients; Indicates the significance level and satisfies ; Indicates the participating entities The sample standard deviation of the marginal technical contribution; Indicates the allowable error bound.

[0212] In some embodiments, the shared facility utilization contribution, new capacity demand, and reliability contribution are respectively formed through the following technical steps: The first step is to generate corresponding scenario identifiers for each combination of alliance entities and each candidate shared facility construction plan, and to form evaluation scenarios corresponding to each scenario identifier based on representative scenarios and scenario weights.

[0213] The scenario identifier can represent identification information used to distinguish specific alliance entity combinations, specific candidate shared facility construction schemes, and corresponding representative scenario conditions. Under the same task identifier, different alliance entity combinations or different candidate shared facility construction schemes will form different scenario identifiers, so that subsequent feasible domains, candidate operation results, physical verification results, and technical indicators can be mapped to specific evaluation scenarios.

[0214] The second step is to shut down callable assets that do not belong to the alliance entity combination corresponding to the scene identifier and callable assets that have not obtained effective control authorization for each scene identifier, and to re-determine the asset activation status and callable capacity according to the binding relationship of the task identifier.

[0215] Specifically, for each evaluation scenario, the available assets are re-determined based on the current alliance entity combination, control authorization status, site node status, and topology status, rather than directly using the asset status from other scenarios. Assets not belonging to the current alliance entity combination or assets that have not obtained valid control authorization are not included in this evaluation scenario. Assets in the enabled state have their available capacity for the current task determined according to the rules for determining the available capacity for the current task.

[0216] The third step involves reconstructing the feasible domains for hydraulic and power tasks corresponding to the scenario identifiers based on the redefined asset activation status, available capacity, and scenario boundaries. Local result coordination and physical verification are then performed based on these reconstructed feasible domains. This step ensures that each alliance entity combination and candidate shared facility construction scheme has a physically feasible domain corresponding to its participating entities, construction capacity, and scenario boundaries. Furthermore, a unified boundary coordination and complete local model verification process is followed to generate scenario results usable for contribution calculation.

[0217] The fourth step involves summarizing the shared facility utilization, new capacity requirements, and reliability improvement amounts at the station node level according to the task identifier binding relationship, and obtaining the shared facility utilization contribution, new capacity requirements, and reliability contribution respectively.

[0218] Specifically, the subject and station node affiliation relationship bound to the current task identifier are used to map station node-level technical quantities to participating subjects. Shared facility utilization, new capacity demand, and reliability improvement quantities are first formed in the corresponding station node or evaluation scenario, and then aggregated to participating subjects according to the fixed affiliation relationship of the current task identifier, so that the corresponding contribution and the physical verification that generated the technical quantity remain consistent through the scenario.

[0219] Fifth, when a station node belongs to multiple participating entities, the shared facility utilization, new capacity demand, and reliability improvement amount corresponding to the station node are allocated to the corresponding multiple participating entities according to the control responsibility coefficients that are pre-registered in the task identifier and sum to one, and the control responsibility coefficients remain unchanged during the corresponding task. Among them, only the evaluation scenario results that pass the physical verification are used to determine the shared facility utilization contribution, new capacity demand, and reliability contribution.

[0220] The control responsibility coefficient represents the pre-registered proportion used when allocating the node-level technical quantity of a shared station node among multiple participating entities. The sum of the control responsibility coefficients for multiple participating entities is one, and it is bound to the current task identifier, remaining unchanged during the task's execution based on subsequent construction responsibility results. This allows the technical quantity corresponding to the shared node to be included in the entity-level contribution calculation according to the pre-determined attribution rules of the current task.

[0221] In some embodiments, determining the target shared facility construction plan and corresponding construction capacity based on the technical evaluation results can be achieved through the following technical steps: The first step is to obtain multiple candidate shared facility construction plans with different candidate construction capacities.

[0222] Specifically, the same shared facility can form multiple candidate schemes with different candidate construction capacities. Different candidate schemes form evaluation scenarios according to their corresponding capacities, access relationships and scenario boundaries, and complete the reconstruction of the task feasibility domain, joint coordination, physical verification and technical value calculation.

[0223] The second step is to screen the candidate shared facility construction plans based on the technical loss threshold and backup threshold pre-registered in the construction plan rules version.

[0224] The construction scheme rule version indicates the version of the rules used in the current construction scheme evaluation task, including the technical loss threshold, reserve threshold, technical loss weight, and contribution category weight. The technical loss threshold determines whether the technical losses of candidate shared facility construction schemes meet the current task's admission criteria, while the reserve threshold determines whether the corresponding reserve indicators meet the current task's admission criteria. Only candidate schemes that meet the pre-registration screening criteria proceed to the subsequent comparison of the technical value of unit construction costs.

[0225] The third step involves determining the technical value of the unit construction cost of each candidate shared facility construction scheme based on its technical value and construction cost under the corresponding candidate construction capacity. Then, the candidate shared facility construction schemes are compared based on their technical value and unit construction cost to determine the target shared facility construction scheme and its corresponding construction capacity.

[0226] Construction cost can be represented as the amount of construction cost corresponding to the cost basis of the corresponding candidate shared facility construction scheme. The technical value of unit construction cost can be determined by the ratio between the technical value of the corresponding candidate scheme and the construction cost. Comparison is only conducted between candidate schemes that have passed the technical loss threshold and reserve threshold screening, ensuring that candidate schemes that do not meet the technical access conditions are not included in the technical value comparison of unit construction cost. After obtaining the target shared facility construction scheme through comparison, the candidate construction capacity registered for the corresponding candidate scheme is taken as the construction capacity corresponding to the target shared facility construction scheme.

[0227] The fourth step is to conduct sensitivity analysis on the weight of technical loss and the cost basis corresponding to each candidate shared facility construction scheme.

[0228] Specifically, while keeping other task boundaries and evaluation rules unchanged, the weight of technical loss can be changed or other registered cost bases for the corresponding candidate shared facility construction schemes can be adopted to re-compare the technical loss, technical value, screening, and unit construction cost technical value to observe whether the target shared facility construction scheme and construction capacity change with the corresponding inputs. This process does not introduce new technical evaluation indicators and still uses the technical loss indicators, positive scale, technical loss weights, and physical verification rules already registered for the current task.

[0229] The fifth step is to re-screen and compare the candidate shared facility construction schemes if any of the following changes: the weight of technical loss or the cost basis corresponding to each candidate shared facility construction scheme.

[0230] Changes in the weighting of technological losses will alter the technological losses in a given scenario and the resulting technological value; changes in the basis for cost estimation will change the construction costs and the technological value per unit of construction cost. Therefore, if any of these factors change, the original target shared facility construction plan and capacity will not be directly adopted; instead, the screening and comparison will be re-done based on the changed data.

[0231] In some embodiments, after determining the target shared facility construction plan and construction capacity, the construction responsibility allocation ratio of each participating entity can be determined based on the comprehensive marginal technical contribution, shared facility utilization contribution, new capacity demand, and reliability contribution corresponding to the plan. This can be done through the following technical steps: The first step is to obtain the non-negative portion of the comprehensive marginal technical contribution, the shared facility utilization contribution, the new capacity demand, and the reliability contribution for each participating entity.

[0232] The marginal technical contribution may have different signs, and its non-negative part is used when calculating the construction responsibility allocation ratio. The contribution of shared facility utilization, new capacity demand, and reliability are all obtained by summarizing the results of the evaluation scenarios that have passed the aforementioned physical verification according to the task identifier binding relationship.

[0233] The second step is to normalize each contribution category based on the contribution amount of each participating entity in that category.

[0234] Specifically, for the same contribution category, the sum of the contributions of all participating entities in that category is used as the normalization denominator. When the denominator is greater than zero, the normalized contribution of each participating entity is formed separately. Different contribution categories are normalized separately, and the original contribution amounts with different dimensions are not directly added together.

[0235] The third step is to determine the category weights corresponding to each available contribution category based on the version of the construction plan rules.

[0236] Category weights represent the non-negative weights registered in the construction scheme rule version for comprehensive marginal technical contributions, shared facility utilization contributions, new capacity demand, and reliability contributions. For currently available contribution categories, their category weights are determined according to the current construction scheme rule version; when unavailable contribution categories exist, the weights of the remaining available categories are renormalized according to subsequent steps.

[0237] The fourth step is to weight the normalized contributions of each participating entity according to the category weights to obtain the comprehensive contribution of each participating entity, and then normalize the comprehensive contribution of each participating entity to obtain the construction responsibility allocation ratio.

[0238] In the example where all four types of contributions are available, the relationship between the combined marginal technical contribution and the construction responsibility allocation ratio can be expressed as: ; ; ; ; in, Indicates the participating entities In candidate shared facility construction plans The comprehensive marginal technological contribution; Represents the current set of participating entities; Indicates the number of participating entities; Indicates the participating entity index; j Represents the current set of participating entities The summation index in the table; This indicates that it does not include participating entities. The main body of the alliance; This represents the factorial operation; Indicates the combination of alliance entities The number of participating entities; Indicates the participating entities Compared to the main combination of the alliance In candidate shared facility construction plans Marginal technological contribution below; Indicates the participating entities The comprehensive marginal technological contribution; Indicates the participating entities The normalized value of the non-negative part of the comprehensive marginal technical contribution; Indicates the participating entities Corresponding contribution to the utilization of shared facilities; Indicates the participating entities The corresponding increase in capacity requirements; Indicates the participating entities The corresponding reliability contribution; Indicates the participating entities The corresponding normalized value of shared facility utilization contribution; Indicates the participating entities The corresponding normalized value of the new capacity demand; Indicates the participating entities The corresponding normalized value of reliability contribution. This indicates the category weight corresponding to the overall marginal technical contribution; This indicates the category weight corresponding to the contribution to the use of shared facilities; This indicates the category weight corresponding to the new capacity demand; This represents the category weight corresponding to the reliability contribution, and the category weights of each category are non-negative and their sum after normalization is 1; Indicates the participating entities The corresponding construction responsibility allocation ratio; This indicates the current candidate construction plans for shared facilities. This represents the total construction cost corresponding to the target shared facility construction plan; When it is necessary to further specify the construction responsibility costs corresponding to the participating entities, it will be based on the construction responsibility allocation ratio. The liability costs arising from the total construction cost. The determination of the construction liability allocation ratio is based on... As a result, The representation does not change the normalization, weighting, and category availability determination process for each contribution category.

[0239] The fifth step is to determine the contribution category as an unavailable contribution category when the normalized denominator of any contribution category is zero, and to renormalize the category weights corresponding to the remaining available contribution categories according to the pre-registration rules.

[0240] For example, when the sum of contributions from all participating entities under a certain contribution category is zero, that contribution category cannot generate a valid normalization result. Therefore, the original category weight of that category is not included in the overall contribution. Instead, the category weights of the remaining available contribution categories are re-normalized according to the rules pre-registered in the construction plan rules version. This process only changes the weight normalization result between available categories, and does not change the original contribution of each entity.

[0241] Furthermore, when all contribution categories are unavailable, no construction responsibility allocation ratio is generated. When the non-negative portion of the comprehensive marginal technical contribution, shared facility utilization contribution, new capacity demand, and reliability contribution cannot form a valid normalization result, no fixed ratio or manual ratio is used as a substitute, and the current construction plan does not generate a construction responsibility allocation ratio. Therefore, the construction responsibility allocation ratio is jointly formed by the physically verified scenario technical contributions and the rules of the current construction plan.

[0242] Figure 5 A schematic diagram illustrating the relationship between tasks and results according to some embodiments of this disclosure is shown. (Reference) Figure 5 As shown, multiple evaluation scenarios can be formed based on task identifiers, including evaluation scenario one, evaluation scenario two, and other evaluation scenarios. Different evaluation scenarios correspond to the corresponding alliance subject combinations, candidate shared facility construction schemes, and scenario conditions.

[0243] For each evaluation scenario, the feasible domain of the task can be reconstructed and the local results at the station can be coordinated to form candidate operating results. Each candidate operating result corresponding to an evaluation scenario undergoes further physical verification using a complete local model at the station, ensuring that subsequent control processing and construction scheme evaluation adopt results verified by station-side physical constraints. After completing the physical verification of the complete local model at the station, for candidate operating results used in actual operation and meeting the corresponding conditions, a control plan version can be generated; for evaluation scenario results that pass the physical verification for construction scheme evaluation, a technical evaluation can be conducted, and a construction scheme version can be generated. This ensures that both the control plan version and the construction scheme version can be mapped to the corresponding task identifier, evaluation scenario, candidate operating result, and physical verification result.

[0244] Furthermore, when relevant conditions change, existing results can be invalidated and the affected scenarios can be recalculated to generate a new version. Specifically, the affected scope can be determined based on the changed conditions and their correspondence with the evaluation scenarios, candidate operation results, control plan versions, and construction scheme versions. The affected existing results can be invalidated, and the corresponding calculations can be performed for the affected scenarios to form a new version corresponding to the changed task conditions.

[0245] In some embodiments, invalidating existing results affected by the corresponding change can be achieved through the following technical steps: The first step is to identify at least one target version among the candidate operational outcome version, control plan version, and construction scheme version that are affected by the changes.

[0246] Specifically, existing results can include candidate operational result versions, control plan versions, and construction scheme versions. Scenarios affected by the corresponding changes can represent evaluation scenarios where the participating entities, available assets, scenario boundaries, physical feasibility, technical indicators, or construction scheme evaluation results depend on the changed information. The scope of impact is determined based on the binding relationship between the changed object and each version. If the change only affects a single candidate operational result, the corresponding candidate operational result version can be determined as the target version; if the change simultaneously affects existing control plans or construction scheme evaluation results, the corresponding control plan version or construction scheme version can also be determined as the target version.

[0247] The second step is to change the target version from a referenceable state to an invalid state through an indivisible state update.

[0248] The referrable state indicates that the current version is allowed to be referenced by subsequent control issuance, technical evaluation, or construction plan result queries; the invalid state indicates that the current version is no longer allowed to be referenced as a valid result. An indivisible state update means that when the target version changes from a referrable state to an invalid state, it is processed as a complete state write, avoiding the simultaneous existence of both referrable and invalid states in the same version.

[0249] The third step is to not publish the alternative result corresponding to the target version before the state update is successful, and to maintain the state of refusing to reference the target version when the state update write fails.

[0250] Specifically, the failure status of the old target version is processed first, and then the corresponding recalculated replacement result is allowed to be published. If the failure status write fails, the normal reference to the old target version is not restored, but the reference is rejected. This reduces the chance that the old version and the replacement version are referenced by downstream processes simultaneously during version switching.

[0251] In some embodiments, recalculating the scenarios affected by the corresponding changes can be performed through the following technical steps: The first step is to generate a new task identifier when any of the participation qualifications or field whitelist of the station nodes changes, and to re-receive the protected aggregate quantity submitted by each station based on the new task identifier. Then, the common working condition center, representative scenario, scenario weight and scenario boundary are redefined, the corresponding hydraulic and electric task feasible domains are reconstructed, and local result coordination and physical verification are carried out again.

[0252] Changes in the eligibility or field whitelist of participating nodes at the station end will alter the data sources or data field ranges that the current task can participate in protected aggregations, thus constituting a change in task boundaries. After the new task identifier is generated, the processing related to the change will be re-executed starting from the reception of protected aggregation quantities, so that the newly formed common working condition center, representative scenario, scenario weight, scenario boundary, and subsequent task feasible domain all correspond to the new participation scope and field range.

[0253] The second step involves generating a new task identifier when any of the callable assets, control authorization, or topology connectivity changes. Based on the new task identifier, the uncallable assets are then closed again, the corresponding hydraulic and power task feasible domains are reconstructed, and local result coordination and physical verification are performed again.

[0254] Changes in callable assets, control authorization, or topology connectivity can directly alter the asset activation status, the current task's callable capacity, or physical connectivity. Therefore, after generating a new task identifier, recalculation can begin from the stage of determining uncallable assets and reconstructing the task's feasible domain, without unconditionally repeating all historical processing when these changes do not affect the established common working condition center and representative scenario.

[0255] The third step is to generate a new candidate execution result version under the original task identifier when the task boundary has not changed and either the execution deviation or the coordination parameter adjustment occurs, and to re-perform local result coordination and physical verification.

[0256] When the execution deviation or coordination parameter adjustment does not change the participating entities, nodes, assets, fields, models, authorizations, output permissions, or topology boundaries, a new task identifier is not generated. Instead, a new candidate execution result version is formed under the original task identifier. The new candidate execution result undergoes local result coordination and complete local model physical verification to distinguish it from the candidate execution result before the deviation or parameter adjustment.

[0257] The fourth step is to recalculate only the scenarios affected by the corresponding changes, while keeping the results for the unaffected scenarios unchanged.

[0258] Specifically, the scope of impact is determined based on the binding relationships between the changed object and scenario identifier, candidate shared facility construction plans, asset status, and version. Scenarios that do not depend on the changed object retain their original results and version relationships; scenarios that depend on the changed object are recalculated according to the corresponding branch. This process avoids indiscriminate duplication of calculations for all evaluation scenarios after partial changes to the task.

[0259] In some embodiments, when the physical verification results, execution feedback, or construction plan-related information change, version invalidation can be handled through the following technical steps: The first step is to not generate a control plan version for the candidate running results corresponding to the physical review result when the physical review result is not passed.

[0260] Candidate run results can only be used to form a control plan version for actual operation after the physical verification of the complete local model has passed. If the physical verification fails, the candidate run result can retain its verification failure information, but will not form a control plan version that can be sent to the existing control interface at the station.

[0261] The second step is to invalidate the issued control plan version and the candidate execution result version bound to it when the physical verification result on which the issued control plan version is based is revoked.

[0262] The control plan version is bound to the candidate run result version corresponding to the target candidate run result. Therefore, after the physical verification result on which it is based is revoked, neither the control plan nor the source candidate run result will meet the original valid conditions, and both versions will be converted to an invalid state.

[0263] The third step is to invalidate the corresponding control plan version and the candidate execution result version bound to the control plan version when the feedback deviation determined by the execution feedback exceeds the corresponding feedback threshold.

[0264] Feedback deviation is obtained based on observed values, control reference values, and feedback scale. When the feedback deviation exceeds the corresponding feedback threshold registered in the current station-end rule, it indicates that the deviation between the actual operating state and the control reference value has exceeded the allowable range. Therefore, the relevant control plan version and the source candidate operating result version become invalid, and the process enters the corresponding task feasible domain reconstruction or local result re-coordination process according to the cause of the deviation.

[0265] The fourth step is to invalidate the corresponding construction plan version when any of the following changes: the construction capacity corresponding to the target shared facility construction plan, the cost basis version corresponding to the candidate shared facility construction plan, the construction plan rule version, or the construction plan evidence bound to the construction plan version.

[0266] Changes in construction capacity will alter the shared capacity boundaries of the corresponding evaluation scenarios; changes in the version of the cost basis will change the construction costs of candidate shared facility construction schemes; changes in the version of the construction scheme rules will change the weight of technical loss, threshold, or contribution category weight; changes in the evidence of the construction scheme will change the evidence content corresponding to the construction scheme version. Any of these changes may affect the target shared facility construction scheme, construction capacity, or the proportion of construction responsibility allocation; therefore, the original construction scheme version will no longer be used.

[0267] The fifth step is to invalidate the construction plan version associated with the corresponding scenario when the execution feedback causes any change in the feasibility or technical indicators of the corresponding scenario.

[0268] Specifically, the execution feedback can not only affect the version of the control plan, but when the changes in the operational status it reflects further change whether the corresponding evaluation scenario is physically feasible, or change the technical indicators used to calculate the technical loss of the scenario, the evaluation results of the construction scheme associated with the scenario will no longer be consistent with the current actual state. Therefore, the corresponding construction scheme version will become invalid and be recalculated according to the scope of impact.

[0269] In some embodiments, establishing version and evidence binding for candidate runtime results, control plans, and construction schemes can be performed through the following technical steps: The first step is to set a candidate operation result version for each candidate operation result, generate a control plan version corresponding to the target candidate operation result based on the target candidate operation result, and generate a construction plan version corresponding to the target shared facility construction plan.

[0270] Among them, the candidate operation result version is used to distinguish the candidate results formed under different coordination rounds or operation deviations in the same evaluation scenario; the control plan version is used to identify the control reference information formed by the target candidate operation results that have passed physical review and entered actual operation; and the construction plan version is used to identify the version status of the current target shared facility construction plan, construction capacity, construction responsibility allocation ratio and corresponding evidence information.

[0271] The second step is to generate scene identifiers corresponding to the combination of each alliance entity and each candidate shared facility construction plan, reference scene identifiers corresponding to each fixed reference scene, and representative scene identifiers corresponding to each representative scene.

[0272] Scenario identifiers are used to distinguish evaluation scenarios formed by combinations of alliance entities and candidate shared facility construction schemes; reference scenario identifiers are used to distinguish fixed reference scenarios corresponding to independent operation schemes or externally selected schemes; representative scenario identifiers are used to distinguish representative scenarios formed by common operating condition centers and current prediction states. Different identifiers are respectively bound to corresponding versions and technical indicators, enabling the calculation of technical value to correspond to specific candidate scenarios and fixed reference scenarios.

[0273] The third step is to generate a snapshot of the participating entities based on the participating entities, determine the data window version, model version, authorization version, output permission version and topology version corresponding to the task identifier, determine the cost basis version corresponding to each candidate shared facility construction scheme, and determine the construction scheme rule version corresponding to the task identifier.

[0274] Among them, the data window version can represent the local running data window version used to generate the common working condition center, representative scenario, and scenario weight; the cost basis version can represent the cost basis version used to determine the construction cost of the corresponding candidate shared facility construction scheme; and the construction scheme rule version can represent the version corresponding to the construction scheme evaluation rules such as the technical loss threshold, reserve threshold, technical loss weight, and contribution category weight. Each version has a clear correspondence with the current task identifier or candidate shared facility construction scheme.

[0275] Furthermore, the construction scheme version is bound to the task identifier, scenario identifiers, reference scenario identifiers, representative scenario identifiers, candidate shared facility construction schemes associated with the construction scheme version, participant snapshots, common operating condition centers, scenario weights, data window versions, candidate operation result versions, control plan versions corresponding to target candidate operation results, task identifier binding relationships, model versions, authorization versions, output permission versions, topology versions, cost basis versions corresponding to candidate shared facility construction schemes, construction scheme rule versions, original technical loss indicators, positive scale, technical loss weights, feasible domains for hydraulic and electric tasks, physical verification results, technical value, marginal technical contribution, construction responsibility allocation ratios, and corresponding signatures, certifications, and audit evidence citations.

[0276] Specifically, the construction plan version is associated not only with the final target shared facility construction plan and the proportion of construction responsibility allocation, but also with the task boundaries, representative scenarios, fixed reference scenarios, candidate operational results, control plans, original technical indicators, physical verification results, and related version information used to generate the result. Signatures, certifications, and audit evidence references are used to point to the corresponding message sources, task processing, and verification evidence. Therefore, when any bound object subsequently changes, the affected scenarios and results can be determined based on the bound content of the construction plan version, and corresponding failure and recalculation processes can be initiated.

[0277] In some embodiments, the binding content between task identifier, scenario identifier, candidate run result version, control plan version, and construction scheme version, as well as the triggering conditions for new versions, can be as shown in Table 1 below.

[0278] Table 1 Furthermore, node thresholds, sample thresholds, recovery thresholds, fixed-point coding parameters, joint coordination stopping thresholds, physical verification scales, feedback thresholds, positive scales for technical losses, and evaluation parameters for construction schemes can be pre-registered according to the current task and remain consistent with the corresponding version during the task. The main implementation parameters and setting principles are shown in Table 2.

[0279] Table 2 Furthermore, in this embodiment of the disclosure, a site-side privacy computing node is also provided, which includes a local data connector, a field mapper and quality checker, a local feature and model generator, a privacy task executor, a local physical checker, a control interface and feedback collector, and a local audit memory.

[0280] The local data connector and field mapping and quality checker are used to read hydrological data, equipment data, and operational data from the station, and to complete field semantic mapping, unit conversion, time alignment, missing and anomaly marking, and equipment encoding mapping. The local feature and model generator is used to generate local operational window features, operational condition category allocation, complete local model, and shared boundary mapping. The privacy task executor is used to perform local privacy computation based on task identifiers, and to perform pruning, fixed-point encoding, and masking on the operational condition counts and operational condition feature vector sums corresponding to the local operational window to form protected aggregate quantities. Under the shared boundary mapping, it performs station-side local updates and multiplier updates, and outputs station-side local results and station-side local feasibility flags. Hydrological data, equipment data, operational data, and the complete local model at the station do not leave the station; the local physical verifier is used to physically verify the candidate operational results corresponding to the task identifier based on the complete local model, and output the physical verification results; the control interface and feedback collector are used to receive the control plan version, provide control reference values ​​to the existing control interface at the station after the control authorization verification and security verification are passed, and collect the execution feedback bound to the control plan version; the local audit storage is used to store local evidence corresponding to the task identifier, candidate operational result version, control plan version, and construction scheme version. The hydrological data, equipment data, operational data, and the complete local model at the station are kept in the privacy computing node at the station.

[0281] In some embodiments, the privacy task executor is further configured to generate a temporary key for the task identifier and task session, and generate paired masks with opposite signs between stations, and superimpose the paired masks onto the sum of the condition count and the condition feature vector, so that the paired masks corresponding to the online station cancel each other out during the cross-station summation process, and the temporary key and paired masks are not reused across task identifiers; the privacy task executor is further configured to provide a recovery share when other station nodes go offline after submitting the protected aggregation amount, provided that the offline proof is passed, the recovery share is not less than the recovery threshold and the recovery share verification is passed, in order to restore the uncancelled mask corresponding to the offline station node, and not to provide a share for restoring the mask corresponding to the still online station node.

[0282] In some embodiments, the local physical checker checks the water balance, unit operating conditions, water level and reservoir capacity boundaries, line injection boundaries, and station-end safety constraints of candidate operating results item by item based on the complete water level and reservoir capacity curve, complete unit characteristics, safe operating range, and station-end line constraints. It then determines the normalized violation of each constraint according to the corresponding physical check scale. When all normalized violations are not greater than the corresponding threshold, the local physical checker returns a physical check pass result. When at least one normalized violation is greater than the corresponding threshold, the local physical checker returns a physical check fail result, and only returns the constraint category, the maximum normalized violation, and the allowable adjustment boundary, without returning the complete original data used to determine the normalized violation. The complete water level and reservoir capacity curve, complete unit characteristics, and complete local model are retained within the station-end privacy computing node.

[0283] In some embodiments, the control reference values ​​provided by the control interface and feedback collector include at least one of the unit's generating power, pumping power, flow rate, water level boundary, and line injection limit; the control interface and feedback collector ensure that the control reference values ​​do not cover existing protection and interlocking at the station end; the control interface and feedback collector normalize the difference between the observed value and the control reference value according to the feedback scale to obtain the feedback deviation, and bind the execution feedback and control plan version, including the feedback deviation, as well as the candidate operation result version corresponding to the target candidate operation result.

[0284] In some embodiments, the local evidence stored in the local audit memory is bound to the candidate running result version, control plan version, and construction scheme version. When the control plan version becomes invalid, the control interface and feedback collector stop providing the control reference value corresponding to the control plan version to the existing control interface at the station and maintain the state of refusing to reference the invalid control plan version until a new control plan version is received.

[0285] Furthermore, in the exemplary embodiments disclosed herein, a system for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing is also provided. (Refer to...) Figure 6 As shown, the privacy-based multi-hydropower station collaboration and construction scheme determination system 600 includes a central alliance computing platform 610 and multiple station-side privacy computing nodes. The multiple station-side privacy computing nodes may include a first station-side privacy computing node 621, a second station-side privacy computing node 622, and so on up to the nth station-side privacy computing node 623.

[0286] The central alliance computing platform is used to generate task identifiers, establish mapping relationships between participating entities, station-end nodes corresponding to multiple hydropower stations, and callable assets associated with multiple hydropower stations, and bind the mapping relationships and corresponding control authorizations to the task identifiers. Each station-end privacy computing node performs local privacy calculations based on the task identifiers to form protected aggregate quantities, provided that the original data of the corresponding hydropower station does not leave the station. The central alliance computing platform, when the number of valid station-end nodes submitting protected aggregate quantities is not less than the node threshold and the number of valid samples is not less than the sample threshold, determines the common operating condition center, representative scenario, and scenario weight based on the protected aggregate quantities, and determines the scenario boundaries in conjunction with the current prediction status. The central alliance computing platform is used to determine multiple alliance entity combinations and multiple candidate shared facility construction schemes. Based on the representative scenario and scenario weights, it forms evaluation scenarios corresponding to the combinations of each alliance entity combination and each candidate shared facility construction scheme. Based on the binding relationship of the task identifiers and the scenario boundaries, it closes uncallable assets and reconstructs the feasible domain of the hydraulic and power tasks corresponding to the evaluation scenario. The central consortium computing platform coordinates the local results of privacy computing nodes at various stations within the feasible domain of hydraulic and power tasks, obtains candidate operational results, and receives physical verification results from each station's privacy computing nodes based on a complete local model. From the physically verified candidate operational results, it determines the target candidate operational results for actual operation, and after control authorization and security verification, distributes the target candidate operational results to the corresponding station's privacy computing nodes and receives execution feedback. Based on the physically verified evaluation scenario results, the central consortium computing platform performs a technical evaluation of the combination of each consortium entity and each candidate shared facility construction scheme, obtaining technical evaluation results. Based on the technical evaluation results, it determines the target shared facility construction scheme, the corresponding construction capacity, and the construction responsibility allocation ratio for each participating entity. When any of the following changes occur: the task identifier binding relationship, the candidate shared facility construction scheme, the corresponding construction capacity, the physical verification results, or the execution feedback, the central consortium computing platform invalidates the existing results affected by the corresponding change and recalculates the scenario affected by the change.

[0287] Below, for reference Figure 7 The interaction relationship of the multi-hydropower station collaboration and construction scheme determination system based on privacy computing in this disclosure is explained.

[0288] refer to Figure 7As shown, the system for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing can include a central consortium computing platform and multiple station-side privacy computing nodes. These station-side privacy computing nodes can include a first station-side privacy computing node, a second station-side privacy computing node, and so on up to the nth station-side privacy computing node. Each station-side privacy computing node interacts with the central consortium computing platform for task-related data exchange, participating in the collaborative processing of multiple hydropower stations while retaining the original data at the station level.

[0289] The central consortium computing platform can include a trusted execution environment and a general computing environment. The trusted execution environment can perform sample number threshold comparisons and, when the corresponding threshold conditions are met, provide the threshold-passing results to the general computing environment to control the entry of the corresponding cross-site aggregation results into the subsequent processing. The general computing environment can include a task orchestration unit, a privacy-collaboration pattern extraction unit, a feasible domain generation unit, a joint decision-making unit, a construction scheme determination unit, and a version management unit. Specifically, the task orchestration unit generates task identifiers and binds them to relevant tasks; the privacy-collaboration pattern extraction unit determines common operating condition information based on relevant results that meet threshold conditions; the feasible domain generation unit reconstructs the feasible domains of hydraulic and power tasks corresponding to the evaluation scenarios; the joint decision-making unit coordinates local station-side results within the feasible domains of hydraulic and power tasks; the construction scheme determination unit conducts technical evaluations based on the physically verified evaluation scenario results and determines the construction scheme, construction capacity, and construction responsibility allocation ratio for the target shared facilities; and the version management unit manages the valid status of existing results and the recalculation of affected scenarios.

[0290] Furthermore, the task identifiers and binding relationships formed by the task orchestration unit can be used to associate the site-side privacy computing node, task processing process, and subsequent results corresponding to the current task, so that different processing stages are all associated with the corresponding tasks. After the construction scheme determination unit forms the construction scheme determination result, it can further form a construction scheme version and corresponding evidence to ensure that the target shared facility construction scheme, construction capacity, construction responsibility allocation ratio, and corresponding evaluation and review results remain consistent.

[0291] In addition, the structure of the site-side privacy computing node can be referenced. Figure 8 As shown. Figure 8 As shown, the station-side privacy computing node can connect to station-side hydrological, equipment, and operational data sources, and may include a local data connector, field mapping and quality checker, local feature and model generator, privacy task executor, local physical checker, control interface and feedback collector, and local audit storage.

[0292] Specifically, the system first accesses station-level hydrological, equipment, and operational data via a local data connector, and then performs field mapping and quality checks using a field mapping and quality checker. After processing, a local feature and model generator generates the local processing results required for subsequent privacy computation and physical verification, and provides them to the privacy task executor. The privacy task executor can interact with the central consortium computing platform and perform station-level local privacy computation based on the information corresponding to the current task. The results generated by the privacy task executor can be further provided to a local physical verifier, which performs physical verification based on the complete local model at the station. After physical verification, the control interface and feedback collector perform corresponding control processing and execute feedback collection, and interact with the station's existing scheduling, interlocking, and control systems. A local audit memory can store relevant local evidence generated during the station-level processing.

[0293] Further, refer to Figure 9 As shown, the central alliance computing platform 610 may include a task orchestration unit 611, a privacy collaboration pattern extraction unit 612, a feasible domain generation unit 613, a joint decision-making unit 614, a construction scheme determination unit 615, and a version management unit 616.

[0294] The task orchestration unit 611 is used to generate task identifiers and bind mapping relationships and control authorizations, and to generate new task identifiers when any of the mapping relationships, field whitelists, model versions, authorization versions, output permission versions and topology versions change.

[0295] The privacy collaboration pattern extraction unit 612 is used to determine the common working condition center, representative scenario and scenario weight when the number of valid stations is not less than the node threshold and the number of valid samples is not less than the sample threshold, and does not release cross-station combined measurement and scenario weight when either the number of valid stations or the number of valid samples fails to reach the corresponding threshold.

[0296] The feasible domain generation unit 613 is used to determine the asset activation status and available capacity of the callable assets, reconstruct the feasible domain of the hydraulic and power tasks corresponding to the evaluation scenario, and not output candidate running results when the corresponding evaluation scenario is determined to be a physically infeasible scenario.

[0297] The joint decision-making unit 614 is used to coordinate local results and generate candidate operation results within the feasible domain of hydraulic and electric tasks. It does not publish candidate operation results when the stopping conditions are not met, and does not generate a control plan version when the physical verification result is that the physical verification fails.

[0298] The construction scheme determination unit 615 is used to generate technical evaluation results and determine the construction scheme, construction capacity and construction responsibility allocation ratio of the target shared facility. It does not generate the construction responsibility allocation ratio when all contribution categories are unavailable contribution categories. Figure 10 The illustration schematically depicts a process for determining construction plans and allocating responsibilities according to some embodiments of this disclosure. (Reference) Figure 10 The construction plan determination unit can receive task identifiers and binding relationships, evaluation scenario results that have passed physical verification, and candidate shared facility construction plans, and then perform construction plan determination and construction responsibility allocation processing based on the above information.

[0299] Specifically, evaluation scenarios can be first established, allowing different combinations of alliance entities, candidate shared facility construction schemes, and corresponding scenario conditions to form scenario objects suitable for subsequent evaluation. Then, a technical evaluation is conducted on the established evaluation scenarios, and only the results of evaluation scenarios that pass physical verification are used to determine the corresponding technical evaluation results. After completing the technical evaluation, different candidate shared facility construction schemes and their construction capacities can be compared based on the technical evaluation results corresponding to each candidate shared facility construction scheme to determine the target shared facility construction scheme and construction capacity. Furthermore, based on available contribution categories such as comprehensive marginal technical contribution, shared facility utilization contribution, new capacity demand, and reliability contribution, the construction responsibility allocation ratio corresponding to each participating entity can be determined; when all contribution categories are unavailable, no construction responsibility allocation ratio is generated. After completing the above processing, the target shared facility construction scheme, construction capacity, and construction responsibility allocation ratio can be output. Simultaneously, a construction scheme version and corresponding evidence corresponding to the current construction scheme determination results can be generated, ensuring that the target shared facility construction scheme, construction capacity, construction responsibility allocation ratio, and the evaluation scenarios and physical verification results on which these results are based remain consistent.

[0300] Version management unit 616 is used to invalidate existing results and recalculate scenarios affected by corresponding changes, and to maintain the state of rejecting references to existing results before invalidation when the invalidation state update write fails.

[0301] Furthermore, in some embodiments, the central consortium computing platform includes a general computing environment and a trusted execution environment. The general computing environment is used to generate and bind task identifiers, reconstruct the feasible domain of hydraulic and power tasks, coordinate candidate operation results, receive physical verification results, conduct technical evaluations, and determine the construction responsibility allocation ratio. The trusted execution environment is used to compare the number of valid samples with the sample threshold and output a threshold pass flag to the general computing environment. The general computing environment does not receive cross-station combined metering and scenario weights when it does not obtain the threshold pass flag, and the protected aggregate quantity, station-side local results, station-side local feasibility flags, physical verification results, and execution feedback are received by the general computing environment. The complete local model is retained in the station-side privacy computing node.

[0302] It should be noted that although several modules or units of the multi-hydropower station collaboration and construction scheme determination system based on privacy computing have been mentioned in the detailed description above, this division is not mandatory. In fact, according to the embodiments of this disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0303] Furthermore, in an exemplary embodiment of this disclosure, an electronic device is also provided that can implement the above-described method for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing.

[0304] Those skilled in the art will understand that various aspects of this disclosure can be implemented as a system, method, or program product. Therefore, various aspects of this disclosure can be embodied in the following forms: a completely hardware embodiment, a completely software embodiment (including firmware, microcode, etc.), or an embodiment combining hardware and software aspects, collectively referred to herein as a "circuit," "module," or "system."

[0305] The following reference Figure 11 To describe an electronic device 1100 according to such an embodiment of the present disclosure. Figure 11 The electronic device 1100 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments disclosed herein.

[0306] like Figure 11 As shown, the electronic device 1100 is presented in the form of a general-purpose computing device. The components of the electronic device 1100 may include, but are not limited to: at least one processing unit 1110, at least one storage unit 1120, a bus 1130 connecting different system components (including storage unit 1120 and processing unit 1110), and a display unit 1140.

[0307] The storage unit stores program code that can be executed by the processing unit 1110, causing the processing unit 1110 to perform the steps described in the "Detailed Description" section of this specification according to various exemplary embodiments of this disclosure. The storage unit 1120 may include a readable medium in the form of a volatile storage unit, such as a random access memory (RAM) unit 1121 and / or a cache memory unit 1122, and may further include a read-only memory unit (ROM) unit 1123.

[0308] Storage unit 1120 may also include a program / utility 1124 having a set (at least one) program module 1125, such program module 1125 including but not limited to: operating system, one or more application programs, other program modules and program data, each or some combination of these examples may include an implementation of a network environment.

[0309] Bus 1130 can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the various bus structures.

[0310] Electronic device 1100 can also communicate with one or more external devices 1170 (e.g., keyboard, pointing device, Bluetooth device, etc.), one or more devices that enable a user to interact with electronic device 1100, and / or any device that enables electronic device 1100 to communicate with one or more other computing devices (e.g., router, modem, etc.). This communication can be performed via input / output (I / O) interface 1150. Furthermore, electronic device 1100 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 1160. As shown, network adapter 1160 communicates with other modules of electronic device 1100 via bus 1130. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 1100, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0311] Through the description of the above embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, terminal system, or network device, etc.) to execute the methods according to the embodiments of this disclosure.

[0312] In exemplary embodiments of this disclosure, a computer-readable storage medium is also provided, on which a program product capable of implementing the methods described above is stored. In some possible embodiments, various aspects of this disclosure may also be implemented as a program product including program code, which, when the program product is run on a terminal device, causes the terminal device to perform the steps according to various exemplary embodiments of this disclosure described in the "Detailed Description" section above.

[0313] The program product for implementing the above-described method for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing, according to embodiments of this disclosure, can be a portable compact disk read-only memory (CD-ROM) and include program code, and can run on a terminal device, such as a personal computer. However, the program product of this disclosure is not limited thereto. In this disclosure, the readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0314] The program product may employ any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0315] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium, capable of sending, propagating, or transmitting programs for use by or in conjunction with an instruction execution system, apparatus, or device.

[0316] The program code contained on the readable medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0317] Program code for performing the operations of this disclosure can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0318] Furthermore, the above figures are merely illustrative of the processes included in the method according to exemplary embodiments of this disclosure and are not intended to be limiting. It is readily understood that the processes shown in the above figures do not indicate or limit the temporal order of these processes. Additionally, it is readily understood that these processes may be executed synchronously or asynchronously, for example, in multiple modules.

[0319] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the claims.

[0320] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A multi-hydropower station collaboration and construction scheme determination method based on privacy computing, characterized in that, include: Generate a task identifier, establish a mapping relationship between participating entities, station-end nodes corresponding to multiple hydropower stations, and callable assets associated with the multiple hydropower stations, and bind the mapping relationship and the corresponding control authorization to the task identifier; Under the condition that the original data of the corresponding hydropower station does not leave the station, each station node performs local privacy calculation based on the task identifier to form a protected aggregate. When the number of valid stations submitting the protected aggregate is not less than the node threshold and the number of valid samples is not less than the sample threshold, the common working condition center, representative scenario and scenario weight are determined according to the protected aggregate, and the scenario boundary is determined in combination with the current prediction state. Multiple alliance entity combinations and multiple candidate shared facility construction schemes are determined. Based on the representative scenario and the scenario weight, an evaluation scenario corresponding to each combination of alliance entity combinations and each candidate shared facility construction scheme is formed. According to the binding relationship of the task identifier and the scenario boundary, uncallable assets are closed and the feasible domain of hydraulic and electric tasks corresponding to the evaluation scenario is reconstructed. Within the feasible domain of the hydraulic and electric tasks, coordinate the local results of each station to obtain candidate operation results, and receive the physical verification results returned by each station based on the complete local model for the candidate operation results; From the candidate operation results that have passed the physical review, the target candidate operation results for actual operation are determined. After the control authorization verification and security verification, the target candidate operation results are sent to the corresponding station and execution feedback is received. Based on the evaluation scenario results that have passed the physical verification, a technical evaluation is conducted on the combination of each alliance entity combination and each candidate shared facility construction scheme to obtain the technical evaluation results. Based on the technical evaluation results, the target shared facility construction scheme, the construction capacity corresponding to the target shared facility construction scheme, and the construction responsibility allocation ratio corresponding to each participating entity are determined. If any one of the following changes, such as the binding relationship of the task identifier, the candidate shared facility construction plan, the construction capacity corresponding to the candidate shared facility construction plan, the physical verification result, or the execution feedback, the existing results affected by the corresponding change shall be invalidated, and the scenario affected by the corresponding change shall be recalculated.

2. The method for multi-hydropower station coordination and construction scheme determination based on privacy computing according to claim 1, characterized in that, The process of performing local privacy calculations based on the task identifier to form a protected aggregate quantity, wherein the number of valid stations submitting the protected aggregate quantity is not less than the node threshold and the number of valid samples is not less than the sample threshold, determines the common working condition center, representative scenario, and scenario weight based on the protected aggregate quantity, including: Each station assigns its local operating window to different operating condition categories, and performs pruning, fixed-point encoding, and masking on the operating condition counts and the sum of operating condition feature vectors corresponding to each operating condition category to form the protected aggregate quantity; When the number of valid stations submitting the protected aggregate quantity is not less than the node threshold and the number of valid samples is not less than the sample threshold, the cross-station combined metering corresponding to each working condition category is restored. The common working condition center is determined according to the ratio of the cross-station combined metering of the sum of the working condition feature vectors to the cross-station total quantity of the working condition count. The corresponding scene weight is determined according to the proportion of the working condition count corresponding to each working condition category to the total valid working condition count. For work condition categories with cross-station work condition counts greater than zero, generate corresponding representative scenarios. When the cross-station work condition count for any work condition category is zero, do not generate representative scenarios for that work condition category. The scene boundary corresponding to the representative scene is determined based on the common working condition center and the current prediction state. The current prediction state includes the current predicted water inflow, water level, equipment status, and node injection. If either the number of valid stations or the number of valid samples fails to reach the corresponding threshold, the cross-station combined metering and the scenario weight will not be released to the central ordinary computing process.

3. The method for multi-hydropower station coordination and construction scheme determination based on privacy computing according to claim 2, characterized in that, Also includes: Each station generates a temporary key for the task identifier and task session, and generates paired masks with opposite signs between each station. The paired masks are then superimposed on the sum of the working condition count and the working condition feature vector, so that the paired masks corresponding to the online station cancel each other out during the cross-station summation process. The temporary key and the paired mask are not reused across task identifiers. For a station that goes offline after submitting the protected aggregate amount, when the offline proof is passed, the recovery share is not less than the recovery threshold, and the recovery share verification is passed, the unoffset mask corresponding to the offline station is restored, and the mask corresponding to the still online station is not restored. If any of the following occurs: the disconnection proof fails, the recovery share is less than the recovery threshold, or the recovery share verification fails, the unoffset mask corresponding to the disconnected station will not be restored.

4. The method for multi-hydropower station coordination and construction scheme determination based on privacy computing according to claim 1, characterized in that, The process of generating a task identifier establishes a mapping relationship between participating entities, station-end nodes corresponding to multiple hydropower stations, and callable assets associated with the multiple hydropower stations. This mapping relationship and corresponding control authorization are then bound to the task identifier, including: The participating entities are identified by a subject identifier, the station nodes are identified by a node identifier, and the callable assets are identified by an asset identifier. The callable assets include at least one of the following: generating units, reservoir capacity, water transmission channels, transmission lines, and control capabilities. Determine the attribution relationship between the subject identifier and the node identifier, and the callability relationship between the node identifier and the asset identifier; form the mapping relationship based on the attribution relationship and the callability relationship; and record the corresponding control authorization certificate. The snapshots of participating entities, site nodes, callable assets, the mapping relationship, field whitelist, model version, authorization version, output permission version, and topology version are determined as the task boundary corresponding to the task identifier, and the task boundary remains unchanged under the same task identifier; When any of the components constituting the task boundary changes, a new task identifier is generated; when the task boundary remains unchanged but an execution deviation occurs, a new candidate execution result version is generated under the original task identifier.

5. The method for multi-hydropower station coordination and construction scheme determination based on privacy computing according to claim 4, characterized in that, The closure of inaccessible assets includes: Configure at least one necessary node group for each of the aforementioned callable assets; For any callable asset, when the participation status, control authorization status, communication status, and task execution verification status of all station nodes in at least one of the necessary node groups are all valid, and the callable asset is located within the current valid topology connectivity component, the asset activation status of the callable asset is determined to be activated; otherwise, the asset activation status of the callable asset is determined to be deactivated. The callable capacity under the current task is determined based on the common upper limit of the physical capacity, authorized capacity, task registration capacity, and topology connectivity of the callable assets in the enabled state, and the positive capacity indication is determined based on whether the callable capacity is greater than zero; wherein, the sum of the generator operation status flag and the pumping operation status flag of the constrained unit is not greater than the product of the asset's enabled state and the positive capacity indication. The power generation flow rate of the unit is limited to no more than the product of the power generation condition indicator and the maximum allowable flow rate corresponding to the power generation condition, and the pumping flow rate of the unit is limited to no more than the product of the pumping condition indicator and the maximum allowable flow rate corresponding to the pumping condition.

6. The method for multi-hydropower station coordination and construction scheme determination based on privacy computing according to claim 1, characterized in that, The coordination of local results at each station within the feasible domain of the hydraulic and power tasks yields candidate operational results, including: Station-end local updates, central shared boundary updates, and multiplier updates are performed based on the same boundary mapping, whereby the same boundary mapping is used to establish the correspondence between local variables at each station and the central shared boundary. Based on the station-end local update, the central shared boundary update, and the multiplier update, determine the local consistency residual, the global physical conservation residual, the adjacent iteration shared boundary change index, and the local feasibility flag corresponding to each station end; When the local consistency residual, the global physical conservation residual, and the adjacent iteration shared boundary change index are all no greater than the corresponding thresholds, and all local feasibility flags corresponding to all stations indicate local feasibility, a fixed candidate running result and the corresponding candidate running result version are generated. If the maximum number of iterations is reached but the conditions that the local consistency residual, the global physical conservation residual, and the adjacent iteration shared boundary change index are all not greater than the corresponding thresholds, and the local feasibility flags corresponding to all stations indicate local feasibility are not met simultaneously, the candidate running results will not be published.

7. The method for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing according to claim 6, characterized in that, The process of updating the central shared boundary includes: The target conservation set is determined based on the task topology template; wherein, the target conservation set includes at least one of the cross-station water conservation set and the power conservation set; Determine the intersection of the shared capacity set and the target conservation set; When the intersection is not empty, a weighted least squares projection is performed on the boundary mapping results corresponding to the local variables of each station within the intersection, and the central shared boundary is updated according to the projection results. When the intersection is empty, the candidate running results are not generated, and the feasible domain of the hydraulic and electric tasks is reconstructed.

8. The method for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing according to claim 1, characterized in that, The receipt of physical verification results from each station based on the complete local model for the candidate running results includes: For the fixed candidate operating results, each station checks the water balance, unit operating conditions, water level and reservoir capacity boundaries, line injection boundaries, and station safety constraints item by item based on the complete water level and reservoir capacity curve, complete unit characteristics, safe operating range, and station line constraints, and determines the normalized violation amount of each constraint according to the corresponding physical verification scale. When all the normalization violations are not greater than the corresponding threshold, the physical review result is returned as passed. When at least one of the normalization violations is greater than the corresponding threshold, a physical review failure result is returned, and only the constraint category, the maximum normalization violation, and the allowable adjustment boundary are returned, without returning the complete original data used to determine the normalization violation. For evaluation scenarios where physical verification fails, the results of the corresponding evaluation scenarios will not be used in the technical evaluation.

9. The method for determining the collaborative construction scheme of multiple hydropower stations based on privacy computing according to claim 1, characterized in that, The technology evaluation results include technological value and marginal technological contribution, and the process for determining the technological value and the marginal technological contribution includes: For any combination of alliance entities and any candidate shared facility construction scheme, obtain the reference technology loss of the combination of alliance entities under a fixed reference scenario, and the candidate technology loss of the combination of alliance entities after adopting the candidate shared facility construction scheme and completing coordination and physical verification. Based on the difference between the reference technology loss and the candidate technology loss, determine the technical value of the combination of alliance entities under the candidate shared facility construction scheme. For any participating entity, under the same candidate shared facility construction scheme, a first alliance entity combination including the participating entity and a second alliance entity combination formed by removing the participating entity from the first alliance entity combination are determined. The marginal technical contribution corresponding to the participating entity is determined based on the difference between the technical value of the first alliance entity combination under the candidate shared facility construction scheme and the technical value of the second alliance entity combination under the candidate shared facility construction scheme. The fixed reference scenario corresponding to the alliance entity combination and the evaluation scenario formed by the alliance entity combination using the candidate shared facility construction scheme employ the same technical indicator caliber, positive scale, technical loss weight, model version, and physical verification rules. The reference technical loss is determined based on the original technical loss index corresponding to the fixed reference scenario, the positive scale, and the technical loss weight, and the fixed reference scenario is physically verified at the station end.

10. A system for determining collaborative construction schemes for multiple hydropower stations based on privacy-preserving computation, used to implement the method for determining collaborative construction schemes for multiple hydropower stations based on privacy-preserving computation as described in any one of claims 1 to 9, characterized in that, The system includes a central consortium computing platform and multiple station-side privacy computing nodes; The central alliance computing platform is used to generate task identifiers, establish mapping relationships between participating entities, station-end nodes corresponding to multiple hydropower stations, and callable assets associated with the multiple hydropower stations, and bind the mapping relationships and corresponding control authorizations to the task identifiers; Each of the aforementioned station-side privacy computing nodes is used to perform local privacy computing based on the task identifier, forming a protected aggregate, under the condition that the original data of the corresponding hydropower station does not leave the station. The central alliance computing platform is used to determine the common working condition center, representative scenario and scenario weight based on the protected aggregation quantity when the number of valid stations submitting the protected aggregation quantity is not less than the node threshold and the number of valid samples is not less than the sample threshold, and to determine the scenario boundary in combination with the current prediction state. The central alliance computing platform is used to determine multiple alliance entity combinations and multiple candidate shared facility construction schemes. Based on the representative scenario and the scenario weight, it forms an evaluation scenario corresponding to the combination of each alliance entity combination and each candidate shared facility construction scheme. According to the binding relationship of the task identifier and the scenario boundary, it closes the inaccessible assets and reconstructs the feasible domain of the hydraulic and electric tasks corresponding to the evaluation scenario. The central consortium computing platform is used to coordinate the local results of each station-end privacy computing node within the feasible domain of the hydraulic and power tasks, obtain candidate running results, and receive the physical verification results returned by each station-end privacy computing node based on the complete local model for the candidate running results; The target candidate execution result for actual execution is determined from the candidate execution results that have passed the physical review. After the control authorization verification and security verification, the target candidate execution result is sent to the corresponding site-side privacy computing node and execution feedback is received. The central alliance computing platform is used to conduct technical evaluations of the combination of each alliance entity and each candidate shared facility construction scheme based on the evaluation scenario results that have passed physical verification, to obtain technical evaluation results, and to determine the target shared facility construction scheme, the construction capacity corresponding to the target shared facility construction scheme, and the construction responsibility allocation ratio of each participating entity based on the technical evaluation results. The central alliance computing platform is used to invalidate existing results affected by any change in the binding relationship of the task identifier, the candidate shared facility construction plan, the construction capacity corresponding to the candidate shared facility construction plan, the physical verification result, and the execution feedback, and to recalculate the scenario affected by the corresponding change.