Intelligent substation energy management method based on cloud computing

Abstract

The invention belongs to the field of power dispatching management, and discloses an intelligent substation energy management method based on cloud computing, which comprises the following steps: constructing a'prediction-dispatching 'game framework; a three-dimensional uncertainty cloud picture with space-time coupling characteristics is generated by fusing regional meteorological data and output correlation of adjacent substations, and an edge layer solves a Nash equilibrium point based on the three-dimensional uncertainty cloud picture to generate a scheduling scheme; performing dynamic electrical partitioning based on the feeder line voltage-power sensitivity matrix to obtain a partitioning result, implanting programmable impedance characteristics in a photovoltaic inverter and an energy storage power conversion system, and performing equipment autonomous response based on a scheduling scheme; the source load fluctuation event chain model carries out cross-time-scale linkage control based on a partitioning result; and updating a scheduling scheme based on an execution result, and adjusting an equipment operation state based on the updated scheduling scheme, thereby realizing intelligent substation energy management based on cloud computing.

Description

Technical Field

[0001] The present invention relates to the field of power dispatching management, and more specifically, to a cloud computing-based smart substation energy management method. Background Art

[0002] As the proportion of renewable energy sources (such as photovoltaics and wind power) in the power system continues to increase, their output volatility and random load demand pose significant challenges to power system scheduling and operation. Renewable energy output is affected by meteorological conditions such as sunlight and wind speed, exhibiting minute-by-minute power fluctuations. This high volatility leads to forecast errors (especially in short-term and ultra-short-term forecasts), which directly impacts the accuracy of scheduling plans. Specifically, if the actual output of renewable energy sources exceeds the forecast, it may cause overvoltage in the back-feeding power grid, threatening grid stability. If the actual output is lower than the forecast, backup resources (such as energy storage) must be urgently deployed to compensate for the power loss, significantly increasing operating costs.

[0003] At the same time, the random nature of load demand further exacerbates the operational challenges of the power system. Sudden increases or decreases in industrial load, coupled with unpredictable residential electricity consumption, often cause load curves to deviate from predicted values. These forecast deviations necessitate real-time balancing of unplanned power shortages or redundancies, placing extreme demands on the dispatching system's responsiveness, often requiring adjustments to be completed within seconds.

[0004] However, existing technologies mainly rely on complex algorithm models to improve the accuracy of renewable energy output forecasts, but these methods often lack economic incentive mechanisms for forecasting entities, making it difficult to fundamentally solve the problem of forecast errors. Especially when facing minute-level power fluctuations, the forecast accuracy is still limited and the scheduling efficiency is low, resulting in equipment response delays.

[0005] In view of this, a smart substation energy management method based on cloud computing is designed. Summary of the Invention

[0006] To overcome the above-mentioned shortcomings of the prior art and achieve the above-mentioned objectives, the present invention provides the following technical solutions: a cloud computing-based smart substation energy management method, comprising: constructing a "forecasting-dispatching" game framework at the source-load forecasting layer, whereby new energy operators and load aggregators submit forecast commitment intervals to the cloud platform, and implementing reward and penalty rules based on actual deviations; generating a three-dimensional uncertainty cloud map with spatiotemporal coupling characteristics by integrating regional meteorological data and the output correlation of adjacent substations, which serves as the decision-making basis for the dispatching and allocation layer; At the dispatch and allocation layer, substation resources are divided into four autonomous decision-making entities: new energy clusters, energy storage alliances, adjustable load groups, and grid interface agents. The edge layer solves Nash equilibrium points based on a three-dimensional uncertainty cloud map and generates a dispatch plan. Dynamic electrical partitioning is performed based on the feeder voltage-power sensitivity matrix, and the partitioning results are obtained. The dispatch plan and partitioning results serve as input to the execution control layer. At the execution control layer, programmable impedance characteristics are embedded in the photovoltaic inverter and energy storage power conversion system, enabling autonomous device response based on the scheduling plan. The source-load fluctuation event chain model performs cross-time-scale linkage control based on the partition results. The execution results are fed back to the integration layer. At the integration layer, the cloud platform conducts game arbitration and market settlement based on the execution results. The edge layer uses quantum-inspired algorithms to accelerate the search for equilibrium points and updates the scheduling plan based on the execution results. The terminal layer dynamically reconstructs the impedance characteristics of the equipment through the power electronic switch array and adjusts the equipment operating status based on the updated scheduling plan.

[0007] Preferably, the method for generating the three-dimensional uncertainty cloud map includes: Initialize the game participants, with new energy operators and load aggregators as the main participants in the game framework. Set the game rules, including obtaining dispatch priority and financial compensation if the actual value is within the commitment range, and paying deviation penalties if the actual value exceeds the commitment range. The penalties are injected into the backup resource pool as the incentive mechanism of the game framework. Participants submit the initial forecast commitment range as the input of the game framework, building a "forecast-dispatch" game framework. Based on a game theory framework, environmental data is hierarchically decomposed to extract disturbance characteristics at different time and spatial scales. Environmental data includes wind speed, light intensity, and load demand. Disturbance characteristics include minute-level mutation characteristics, hour-level trend characteristics, local mutation characteristics, and regional trend characteristics, which serve as the basis for reshaping the commitment interval. For disturbance features with different scales, minute-level mutation features have higher priority than hour-level trend features, and local mutation features have higher priority than regional trend features. For disturbance features with the same scale, priority is ranked according to the degree of impact of the disturbance features on prediction accuracy. Dynamically calibrate the looseness of the initial forecast commitment interval based on the priority sequence and the strength of the disturbance feature. That is, if a high-priority minute-level mutation feature is detected, the tolerance range of the commitment interval is significantly relaxed and the penalty is reduced. If only a medium-priority hourly trend feature is detected, the commitment interval is moderately relaxed. The adjusted forecast commitment interval is generated as the updated input of the game framework. Based on the adjusted forecast commitment range, new energy operators and load aggregators submit forecast commitments to the cloud platform. The platform then implements reward and penalty rules based on actual deviations to generate forecast deviation data, which serves as a basis for correcting the three-dimensional uncertainty cloud map. By integrating forecast deviation data, regional meteorological data, and the correlation between the outputs of adjacent substations, a three-dimensional uncertainty cloud map with spatiotemporal coupling characteristics is generated. The three-dimensional uncertainty cloud map includes the distribution of time, power, and occurrence probability.

[0008] Preferably, the method for solving the Nash equilibrium point based on the three-dimensional uncertainty cloud map and generating a scheduling plan includes: Substation resources are divided into four autonomous decision-making entities: new energy clusters, energy storage alliances, adjustable load groups, and grid interface agents. The goal of the new energy cluster is to maximize absorption capacity, the goal of the energy storage alliance is to minimize lifespan loss, the goal of the adjustable load group is to minimize electricity costs, and the goal of the grid interface agent is to meet superior dispatch instructions. This serves as the basis for policy conflict negotiation. Conflict features are extracted from the objective functions of each autonomous decision-making entity to generate a conflict feature set. These conflict features include target priority features and constraint boundary features. The target priority features include the priority of the new energy cluster's consumption capacity and the priority of the energy storage's lifespan loss. The constraint boundary features include the energy storage's SOC protection boundary and the adjustable load reduction capacity boundary, which serve as the basis for hierarchical negotiation. Decompose the conflict feature set into multiple conflict levels based on the severity and impact of the conflict. The conflict levels include the initial conflict level and the severe conflict level. The initial conflict level corresponds to conflicts where the SOC is close to exceeding the limit, while the severe conflict level corresponds to conflicts where the SOC actually exceeds the limit. These conflict levels serve as input for constraint convergence. Based on a three-dimensional uncertainty cloud map, a multi-stage negotiation mechanism is used to resolve strategic conflicts at different levels. In the initial conflict stage, the new energy cluster adjusts its output plan to meet the energy storage's SOC protection boundary. At the severe conflict stage, the energy storage alliance relaxes the SOC protection boundary. The negotiation process is iterative, with each iteration narrowing the feasible domain of the strategies. The resulting negotiated strategy set is then generated as input for solving the Nash equilibrium. Based on the negotiated strategy set and three-dimensional uncertainty cloud map, the edge layer solves the Nash equilibrium point and generates a scheduling plan, which includes the active / reactive power regulation range, charging and discharging strategy, load transfer plan, and power purchase and sales plan of each entity.

[0009] Preferably, the method for performing dynamic electrical partitioning based on the feeder voltage-power sensitivity matrix and obtaining partitioning results includes: Perform multi-dimensional decomposition of substation status data to extract change characteristics. Substation status data includes voltage, power, and power flow distribution. Change characteristics serve as the basis for disturbance perception and include instantaneous change characteristics, trend change characteristics, and spatial distribution characteristics. Based on the three-dimensional uncertainty cloud map, a comprehensive evaluation of the change characteristics is performed to generate a disturbance index for the substation status. The comprehensive evaluation is based on the intensity and duration of the change characteristics. If the instantaneous change characteristics are strong and short in duration, the disturbance index is high. If the trend change characteristics are weak and long in duration, the disturbance index is low. This disturbance index serves as input for frequency optimization. Dynamically adjust the update frequency of the feeder voltage-power sensitivity matrix based on the disturbance index. If the disturbance index is higher than a first threshold, the update frequency is increased to a first frequency value. If the disturbance index is lower than a second threshold, the update frequency is reduced to a second frequency value. The optimized update frequency is generated as the basis for dynamic electrical zoning. Based on the optimized update frequency and scheduling scheme, the feeder voltage-power sensitivity matrix is ​​calculated to generate dynamic electrical partitions. The dynamic electrical partitions include strong coupling areas and weak coupling areas. The strong coupling areas are subject to centralized optimization scheduling, while the weak coupling areas are subject to autonomous decision scheduling. The partition results are generated. When the partition changes, the scheduling rules are adjusted through the continuous optimization function to generate the adjusted partition results as the update input of the linkage control of the execution control layer.

[0010] Preferably, the method of implanting programmable impedance characteristics in a photovoltaic inverter and energy storage power conversion system and performing autonomous device response based on a scheduling scheme includes: Perform multi-level decomposition of equipment operating data to extract impact features. The operating data includes the operating status of the photovoltaic inverter and energy storage power conversion system. The impact features include geographic impact features, power impact features, and sensitivity impact features, which serve as the basis for equipment impact assessment. Based on the zoning results, the impact characteristics are prioritized and a comprehensive importance index is generated for each device. The evaluation is based on the contribution of the impact characteristics to system stability. If the device's geographical impact characteristics indicate that it is located in a strong coupling zone and its sensitivity impact characteristics indicate that it has a large impact on voltage, then the importance index is high and serves as the input for priority adjustment. According to the comprehensive importance index, the impedance parameter adjustment priority of the devices is dynamically sorted. Among them, the impedance parameters of devices with high importance index are adjusted first, and the adjustment of devices with low importance index is delayed. The device adjustment sequence is generated as the basis for the autonomous response of the devices; Based on the scheduling plan and the equipment adjustment sequence, programmable impedance characteristics are implanted in the photovoltaic inverter and energy storage power conversion system to execute autonomous equipment response, where the autonomous response includes the new energy equipment simulating the inertial response of the synchronous machine and the energy storage equipment providing short-circuit capacity support, and the equipment response results are generated as input for the execution control layer linkage control; When the system state changes, the comprehensive importance index is re-evaluated through the continuity update function, and the device adjustment sequence is adjusted to generate an updated device adjustment sequence as the update input for the linkage control of the execution control layer.

[0011] Preferably, the method for performing cross-time-scale linkage control based on the partition results of the source-load fluctuation event chain model includes: Decompose the indicators involved in the event chain and extract impact features. The indicators include PV output, load demand, and voltage level. The impact features include instantaneous impact features, trend impact features, and coupling impact features, which serve as the basis for correlation fusion. Based on the partitioning results, dynamic weights are assigned to the influencing features to generate a weight sequence. The weight sequence is based on the degree of influence of the influencing features on the substation stability. If the instantaneous influencing features have a large impact on the substation stability, a high weight is assigned. If the trend influencing features have a small impact, a low weight is assigned. The weight is dynamically adjusted according to the strong coupling area and weak coupling area of ​​the partitioning results, and serves as the input for condition optimization. The influence features and weight sequences are correlated and fused through the weight fusion function to generate composite trigger conditions and comprehensive influence indexes as the basis for linkage control; Based on device response results and compound trigger conditions, cross-timescale linkage control is executed. This linkage control includes millisecond-level energy storage compensation, second-level load shedding, and minute-level grid support. The linkage control results are generated as input to the execution results of the integration layer. The trigger threshold of the composite trigger condition is dynamically optimized according to the disturbance index of the substation status. If the disturbance index of the substation status is high, the trigger threshold is lowered; if the disturbance index of the substation status is low, the trigger threshold is increased. The optimized composite trigger condition is generated as the update input of the execution result of the integration layer.

[0012] Preferably, the method for the cloud platform to perform game arbitration and market liquidation based on the execution results includes: Perform multi-dimensional decomposition of substation operation scenarios and extract impact characteristics. Operation scenarios include voltage collapse risk, insufficient new energy consumption, and sudden load changes. Impact characteristics include disturbance impact characteristics, scope impact characteristics, and economic impact characteristics, which serve as the basis for scenario impact assessment. Based on the zoning results, the impact characteristics are prioritized and a comprehensive emergency index for the scenario is generated. The evaluation is based on the degree of impact of the impact characteristics on substation operation. If the disturbance impact characteristics indicate a high risk of voltage collapse and the range impact characteristics indicate an impact in a strongly coupled area, the emergency index is high, which serves as an input for capital optimization. Dynamically optimize the allocation ratio of backup resource pool funds based on the comprehensive emergency index. If the scenario's comprehensive emergency index is above the first threshold, funds are prioritized for purchasing energy storage services. If the emergency index is below the second threshold, funds are prioritized for subsidizing affected parties. This generates a fund allocation plan that serves as the basis for game arbitration and market settlement. Based on the equipment response results and linkage control results, verify the technical feasibility of the dispatch plan, where technical feasibility includes power flow safety and voltage stability, and generate arbitration results as input for market clearing; Based on the arbitration results and fund allocation plan, the reward and punishment funds and ancillary service fees are settled. Among them, the reward and punishment funds include forecast deviation penalties and economic compensation for new energy operators and load aggregators, and the ancillary service fees include the fees for energy storage services and load reduction services. The settlement results are generated as input for the edge layer's updated scheduling plan.

[0013] Preferably, the edge layer uses a quantum-inspired algorithm to accelerate the equilibrium point search, and the method for updating the scheduling scheme based on the execution result includes: Perform multi-dimensional decomposition on the execution results to extract deviation features, where the execution results include device response results and linkage control results. The deviation features include device response deviation features and linkage control deviation features, which serve as the basis for strategy optimization; Deviation characteristics are prioritized based on the dispatch plan to generate a deviation priority sequence. The evaluation is based on the degree to which the deviation characteristics affect the stability and economic efficiency of the substation. If the equipment response deviation characteristics indicate that power regulation exceeds the dispatch plan range, the priority is high. If the linkage control deviation characteristics indicate that voltage fluctuations exceed the safe range, the priority is also high. This sequence serves as input for the equilibrium point search. A quantum-inspired algorithm is used to accelerate equilibrium point search based on liquidation results. The liquidation results are used to adjust the reward and punishment rules of the game framework and generate an optimized strategy set as the basis for scheduling plan updates. Based on the optimized strategy set and deviation priority sequence, the dispatch plan is optimized. This optimization includes adjusting the active / reactive power regulation range, charging and discharging strategies, load transfer plans, and power purchase and sales plans of each entity. An updated dispatch plan is generated as input for adjusting the operating status of equipment at the terminal layer. When the system state changes, the deviation priority sequence is re-evaluated through the continuous optimization function, and the optimized strategy set is adjusted to generate an adjusted update scheduling scheme as the update input for the terminal layer to adjust the device operating state.

[0014] Preferably, the method for updating the scheduling scheme based on the execution result further includes: After adding the liquidation results to the execution results, multi-dimensional integration is performed to extract optimization features, including equipment response optimization features, linkage control optimization features, and economic optimization features, which serve as the basis for rolling optimization; Based on the scheduling plan, the optimization features are prioritized and an optimization priority sequence is generated. The evaluation is based on the degree to which the optimization features improve the stability and economic efficiency of the substation. If the equipment response optimization feature shows a reduction in power regulation deviation, the priority is high. If the economic optimization feature shows a reduction in cost, the priority is high. This sequence is used as the input for the rolling optimization. Based on the optimized priority sequence and the updated scheduling plan, a rolling optimization of the scheduling plan is performed. The rolling optimization includes refreshing the three-dimensional uncertainty cloud map and the scheduling plan at preset time intervals to generate an optimized scheduling plan as input for the terminal layer to adjust the equipment operating status; When an emergency is detected, the system immediately switches to emergency mode based on the linkage control results. In this mode, rapid response resources are prioritized, including energy storage devices and adjustable load resources, and an emergency management plan is generated as a temporary input for the terminal layer to adjust the operating status of the equipment. When the substation status changes, the optimization priority sequence is re-evaluated through the continuous optimization function, and the optimized scheduling plan is adjusted to generate the adjusted optimized scheduling plan as the update input for adjusting the equipment operating status at the terminal layer.

[0015] Preferably, the terminal layer dynamically reconstructs the device impedance characteristics through the power electronic switch array and adjusts the device operating state based on the updated scheduling plan, and the method includes: Perform multi-dimensional decomposition on the updated scheduling plan and extract adjustment features. The updated scheduling plan includes the updated scheduling plan and the optimized scheduling plan. The adjustment features include impedance adjustment features, power adjustment features, and time adjustment features, which serve as the basis for dynamic reconstruction. Based on the adjustment characteristics, the device impedance characteristics are dynamically reconstructed through the power electronic switch array. The impedance characteristics include the real and imaginary values ​​of the virtual impedance. The reconstruction includes the new energy equipment simulating the inertial response of the synchronous machine and the energy storage device providing short-circuit capacity support. The reconstructed impedance characteristics are generated as input for adjusting the device operating status. Based on the reconstructed impedance characteristics, the device operating state is adjusted. The operating state includes the actual output, state of charge, and impedance value. The adjusted device operating state is generated as input for the next round of prediction. Feedback the adjusted device operating status to the edge layer in real time to update the execution results, generate the feedback execution results, and serve as input for the next round of game arbitration and market liquidation in the integration layer; When the substation status changes, the adjustment characteristics are re-evaluated through the continuous optimization function, and the reconstructed impedance characteristics are adjusted to generate the updated equipment operation status as the updated input for the next round of prediction.

[0016] The technical effects and advantages of the cloud computing-based smart substation energy management method of the present invention are as follows: By constructing a "forecast-dispatch" game framework and leveraging economic incentives, this approach encourages new energy operators and load aggregators to improve forecast accuracy. The dynamic calibration logic of the forecast commitment interval adaptively adjusts the degree of slack based on the hierarchical characteristics of environmental disturbances, significantly reducing forecast errors. By transforming uncertainty into an economically driven problem, participants are incentivized to proactively optimize the forecast model, reducing reliance on algorithmic complexity and improving the robustness and fairness of forecasts.

[0017] Distributed autonomous decision-making is achieved by dividing substation resources into four autonomous decision-making entities: new energy clusters, energy storage alliances, adjustable load groups, and grid interface agents. A multi-stage negotiation mechanism and Nash equilibrium solution are employed to achieve this. Dynamic electrical zoning logic adaptively divides control granularity based on the feeder voltage-power sensitivity matrix, optimizing dispatch efficiency. This significantly improves the real-time and targeted nature of dispatch plan generation, making it particularly suitable for managing complex fluctuations in scenarios with a high proportion of new energy access.

[0018] By embedding programmable impedance characteristics in the photovoltaic inverter and energy storage power conversion system, the equipment can autonomously adjust its electrical characteristics according to the scheduling plan, enabling new energy equipment to simulate the inertial response of synchronous generators and energy storage equipment to provide short-circuit capacity support. Cross-timescale linkage control logic optimizes response speed and stability through complex trigger conditions. Through physical-level intelligent reconstruction of device characteristics, command latency is eliminated, significantly improving the ability to respond in seconds and enhancing the system's dynamic stability.

[0019] Through game arbitration and market clearing on the cloud platform, combined with quantum-inspired algorithms at the edge layer to accelerate equilibrium point search, real-time updates and rolling optimization of the scheduling plan are achieved. The terminal layer dynamically reconfigures device impedance characteristics through power electronic switch arrays, ensuring efficient execution of the scheduling plan. This forms a closed-loop management mechanism for prediction, scheduling, execution, and integration, significantly improving the system's economic consistency and physical feasibility, and optimizing resource utilization efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A schematic diagram of the steps of a cloud computing-based smart substation energy management method of the present invention; Figure 2 This is a structural diagram of a cloud computing-based smart substation energy management method of the present invention. DETAILED DESCRIPTION

[0021] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0022] Example 1

[0023] See also Figure 1 As shown, the cloud computing-based smart substation energy management method described in this embodiment includes: Traditional scheduling solutions often adopt a centralized optimization paradigm, where a single control center manages all resources. However, this centralized scheduling approach is computationally complex in scenarios with a high proportion of renewable energy access, making it difficult to meet the rapid response requirements of distributed energy systems. This is especially true when faced with real-time power shortages or redundancies, resulting in significantly insufficient scheduling efficiency.

[0024] The existing execution control layer primarily relies on traditional power command control, which ensures responses by issuing specific power adjustment commands to devices. However, this approach suffers from response delays, making it difficult to meet system stability requirements, especially in scenarios requiring sub-second responses. Furthermore, existing devices lack intelligent features and are unable to autonomously adjust their operating behavior based on system status.

[0025] Existing technologies lack closed-loop management mechanisms that are both economically self-consistent and physically enforceable at the system integration level. Traditional game arbitration and market clearing methods often fail to fully consider feedback from execution results, resulting in a lack of real-time and targeted scheduling optimization. Furthermore, the allocation method for the reserve resource pool is relatively simple, making it difficult to dynamically optimize based on the urgency of different scenarios, reducing the system's economic efficiency and stability.

[0026] Therefore, a cloud-based smart substation energy management approach is urgently needed. This approach, through innovative forecasting mechanisms, dispatching strategies, execution control, and system integration, can address the core challenges posed by source-load uncertainty and achieve a paradigm shift from "dispatching systems adapting to fluctuations" to "taming fluctuations as dispatchable resources." This approach effectively addresses the complex fluctuations encountered in scenarios with a high proportion of renewable energy access, providing an innovative dispatching path for smart substation operation that is both economically self-consistent and physically feasible.

[0027] The output of renewable energy sources (such as photovoltaics and wind power) is affected by meteorological conditions such as sunlight and wind speed, resulting in minute-by-minute power fluctuations. Forecast errors directly impact the accuracy of dispatch plans, potentially leading to overvoltage on the reverse power grid or the urgent deployment of backup resources, increasing operating costs. Sudden increases and decreases in load by industrial users and uncertain residential electricity consumption behavior can cause load curves to deviate from forecasts. Unplanned power shortfalls or redundancies require real-time balancing, placing extreme demands on the dispatch system's responsiveness. Existing forecasting methods lack economic incentives for forecasting entities, making it difficult to fundamentally address forecast errors. This is particularly true when faced with minute-by-minute power fluctuations, resulting in limited forecast accuracy.

[0028] This design constructs a "forecast-scheduling" game framework, introduces hierarchical perception of environmental disturbances and interval reshaping logic, dynamically adjusts the forecast commitment interval, and generates a three-dimensional uncertainty cloud map with spatiotemporal coupling characteristics, thereby improving forecast accuracy, quantifying uncertainty, and providing a reliable decision-making basis for the scheduling and allocation layer.

[0029] Specific content includes: At the source-load forecasting layer, a "forecast-dispatch" game framework is constructed. New energy operators and load aggregators submit forecast commitment intervals to the cloud platform, and reward and penalty rules are implemented based on actual deviations. By integrating regional meteorological data and the output correlation of adjacent substations, a three-dimensional uncertainty cloud map with spatiotemporal coupling characteristics is generated to serve as the decision-making basis for the dispatching and allocation layer. The method for generating the three-dimensional uncertainty cloud map includes: Initialize the game participants. The process of initializing the game participants includes registering the participants on the cloud platform, submitting basic information, and establishing a data interface. The steps are as follows: Step A1: New energy operators and load aggregators register accounts through the cloud platform's user management system and obtain unique participant identifiers (e.g., ID numbers). Step A2: Participants need to submit their basic information, including name, location, equipment capacity, historical operating data, etc. This information is submitted through the cloud platform's form or data upload interface and stored in the cloud database; Step A3: The cloud platform allocates a dedicated data interface for each participant to upload forecast data, receive scheduling instructions, and obtain reward and punishment results in real time. The data interface uses a standard communication protocol (such as MQTT or REST API) to ensure secure and real-time data transmission.

[0030] New energy operators [referring to entities responsible for the operation and management of new energy power generation equipment (such as photovoltaic power stations and wind farms), whose main task is to predict the output of new energy and participate in power dispatch. For example, a photovoltaic power station operating company A is responsible for managing a photovoltaic power station with an installed capacity of 50MW, predicting its output in the next 15 minutes, and submitting a forecast commitment interval to the cloud platform (such as "the probability of output ≥ 40MW is 90%)"] and load aggregators [referring to entities that manage and forecast load on behalf of multiple load users (such as industrial users and residential users), whose main task is to aggregate load demand, predict load curves and participate in power dispatch. For example, a load aggregation company B represents 10 factories in an industrial park, predicts their total load demand in the next hour, and submits a forecast commitment interval to the cloud platform (such as "the probability of load demand ≤ 100MW is 85%)"] as participants in the game framework; Game rules are set, including scheduling priority (referring to the participant's predicted output or load range and its corresponding probability) if the actual value is within the commitment range [referring to the participant's predicted output or load range and its corresponding probability. For example, new energy operator A predicts that its photovoltaic power station will output 40MW in the next 15 minutes and promises "a 90% probability of output ≥ 40MW," meaning the commitment range is [40MW, ∞) with a probability of 90%). For example, if A's actual output is within the commitment range (e.g., the actual output is 45MW), then A's photovoltaic power station will be given priority in the dispatch allocation layer, meaning its output will be included in the dispatch plan and will not be curtailed] and economic compensation (referring to the financial reward given to the participant with accurate predictions). For example, if A's actual output is within the commitment range, the cloud platform will pay compensation based on the output, such as 0.05 yuan per kWh. If A's actual output is 45MW, the compensation is 45,000kWh × 0.05 yuan / kWh = 2,250 yuan. If the actual output exceeds the committed range, a deviation penalty is paid. [Deviation penalties are penalties imposed on participants with inaccurate forecasts. For example, if A's actual output is 35MW, which is below the lower limit of the committed range of 40MW, a deviation penalty is required, such as 0.1 yuan per kWh of deviation. If the deviation is 40MW - 35MW = 5MW, the penalty is 5,000kWh × 0.1 yuan / kWh = 500 yuan.] This penalty is invested in the reserve resource pool [a fund pool used to smooth system fluctuations, composed of deviation penalties and other revenue]. For example, the 500 yuan penalty paid by A is injected into a reserve resource pool, which can be used to purchase energy storage services (e.g., pay energy storage operator C to provide 50kW of emergency compensation power) or subsidize affected parties (e.g., subsidize load aggregator B's increased electricity costs due to a sudden load surge). This serves as an incentive mechanism for the game framework. Participants submit initial forecast commitment intervals as input to the game framework, constructing a "forecast-dispatch" game framework. Specifically, taking a smart substation as an example, the participants in the game framework include new energy operator A (managing a 50MW photovoltaic power plant) and load aggregator B (representing an industrial park). During initialization, A and B register accounts on the cloud platform. A submits the installed capacity, historical output data, and geographic location of the photovoltaic power plant, while B submits historical load data and load characteristics for the industrial park. The cloud platform allocates data interfaces for A and B to upload subsequent forecast data and receive reward and penalty results.

[0031] Based on a game theory framework, environmental data is hierarchically decomposed to extract disturbance characteristics at different temporal and spatial scales. Environmental data (obtained through sensors, monitoring equipment, and external data interfaces) includes wind speed (collected in real time by wind speed sensors installed at wind farm meteorological stations, with data uploaded to the cloud platform at a frequency of seconds), light intensity (collected in real time by light sensors installed at photovoltaic power plants, with data uploaded to the cloud platform at a frequency of minutes), and load demand (real-time load data from industrial and residential users collected through smart meters and load monitoring equipment, with data uploaded to the cloud platform at a frequency of minutes). Disturbance characteristics include minute-level mutation characteristics (referring to drastic changes within a short period of time (minutes). For example, a sudden increase in wind speed from 5 m / s to 15 m / s within one minute represents a minute-level mutation characteristic), hourly trend characteristics (referring to stable changes over a longer period of time (hours). For example, a continuous decrease in light intensity from 800 W / m² to 600 W / m² within one hour represents an hour-level trend characteristic), and local mutation characteristics (referring to drastic changes in a specific area). For example, a 15% drop in light intensity in a photovoltaic panel area, while remaining stable in other areas, indicates a local mutation. Regional trend characteristics refer to stable changes across an entire region. For example, a drop in wind speed across the entire substation area from 10 m / s to 8 m / s within an hour indicates a regional trend. These characteristics serve as the basis for reshaping the commitment interval. Hierarchical decomposition refers to decomposing environmental data into multiple levels according to time and space scales in order to extract different features. Specific methods include: For time scale decomposition: Wavelet transform can be used to decompose environmental data into components at different time scales. For example, wind speed data can be decomposed into minute-level components (to capture sudden changes), hour-level components (to capture trends), and daily components (to capture long-term changes) by performing wavelet transform.

[0032] Corresponding spatial scale decomposition: Spatial interpolation methods (such as Kriging interpolation) can be used to decompose environmental data into components at different spatial scales. For example, spatial interpolation of light intensity data can be performed to decompose it into a local component (capturing changes in a photovoltaic panel area) and a regional component (capturing changes in the entire substation area).

[0033] The extraction of disturbance features is achieved through statistical analysis and feature detection methods. The specific steps are as follows: Statistical analysis: Calculate statistical indicators of environmental data, such as mean, variance, and rate of change. For example, calculate the minute-level rate of change of wind speed. If the rate of change exceeds 10%, it is extracted as a minute-level mutation feature.

[0034] Feature detection: Use threshold detection or pattern recognition methods to identify abnormal patterns in environmental data. For example, if the light intensity in a certain local area drops by 15%, it is extracted as a local mutation feature.

[0035] Of course, the above decomposition method can also be adopted by other means, as long as the purpose of this technology is achieved.

[0036] Disturbance characteristics quantify environmental uncertainty and guide the dynamic adjustment of commitment intervals. For example, if minute-level mutation characteristics (such as a sudden increase in wind speed) are detected, indicating severe environmental fluctuations and high forecast difficulty, the commitment interval should be relaxed (for example, from ±10% to ±20%) to avoid unfair penalties. If hour-level trend characteristics (such as a steady decrease in light intensity) are detected, indicating minor environmental fluctuations and low forecast difficulty, the commitment interval should be tightened (for example, from ±10% to ±5%) to improve forecast accuracy requirements.

[0037] For disturbance features with different scales, minute-level mutation features have higher priority than hour-level trend features, and local mutation features have higher priority than regional trend features. For disturbance features with the same scale, priority is ranked according to the degree of impact of the disturbance features on prediction accuracy. The degree of impact can be quantified by the following indicators: Fluctuation amplitude: The greater the amplitude of the disturbance characteristic, the greater the impact on forecast accuracy. For example, a minute-level sudden change in wind speed of 10 m / s has a greater impact than an hour-level trend characteristic of a 2 m / s drop in wind speed.

[0038] Duration: The shorter the duration of a disturbance feature, the greater its impact on prediction accuracy. For example, minute-level mutation features have a short duration and are difficult to predict, so they have a higher priority than hour-level trend features.

[0039] Spatial scope: The smaller the impact range of a disturbance feature, the greater its impact on prediction accuracy. For example, a local sudden change feature (such as a 15% drop in sunlight in a photovoltaic panel area) is more difficult to predict than a regional trend feature (such as a 5% drop in sunlight across the entire region), and therefore has a higher priority.

[0040] A weighted scoring method can be used to calculate the impact score by combining the fluctuation amplitude, duration, and spatial range. For example, the score of the minute-level mutation feature = fluctuation amplitude weight (0.5) × amplitude score (10) + duration weight (0.3) × time score (8) + spatial range weight (0.2) × range score (6) = 8.3; the score of the hourly trend feature = 4.5. The higher the score, the higher the priority.

[0041] Based on the priority sequence and the strength of the disturbance characteristics, the looseness of the initial forecast commitment interval is dynamically calibrated. That is, if a high-priority minute-level mutation characteristic is detected, the tolerance range of the commitment interval is significantly relaxed and the penalty intensity is reduced. [The penalty amount can be reduced by adjusting the penalty calculation coefficient. For example, the initial penalty coefficient is 0.1 yuan / kWh. If a minute-level mutation characteristic is detected, it is significantly reduced to 0.05 yuan / kWh; if an hour-level trend characteristic is detected, it is moderately reduced to 0.08 yuan / kWh.] If only a medium-priority hour-level trend characteristic is detected, the commitment interval is moderately relaxed, and the adjusted forecast commitment interval is generated as the update input of the game framework. Based on the adjusted forecast commitment interval [referring to the final forecast results submitted by the participants to the cloud platform based on the adjusted forecast commitment interval. For example, based on the adjusted commitment interval "90% probability of output ≥ 35MW", new energy operator A submits a forecast commitment of "40MW output in the next 15 minutes, 90% probability".], the new energy operator and load aggregator submit forecast commitments to the cloud platform, and based on the actual deviation [referring to the difference between the predicted value and the actual value. For example, if A's forecast commitment is "output of 40MW" and the actual output is 45MW, the actual deviation is 45MW - 40MW = 5MW (positive deviation); if the actual output is 30MW, the actual deviation is 30MW - 40MW = -10MW (negative deviation).], reward and punishment rules [referring to the economic incentives or penalties implemented based on the actual deviation]. For example, if A's actual output is 45MW, within the committed range, it will receive dispatch priority (output priority) and financial compensation (2,250 yuan). If the actual output is 30MW, outside the committed range, it will pay a deviation penalty (deviation of 10MW × 0.05 yuan / kWh = 500 yuan). This generates forecast deviation data, which serves as a basis for correcting the three-dimensional uncertainty cloud map. [By quantifying the forecast error, the probability distribution of the three-dimensional uncertainty cloud map is corrected. For example, if A's forecast deviation is -10MW, indicating that the forecast value is too high, the power distribution probability is adjusted toward lower power when generating the cloud map (for example, "40MW probability is 90%" is corrected to "35MW probability is 90%"), thereby improving the accuracy of the cloud map.] The system integrates forecast deviation data, regional meteorological data (referring to meteorological information covering the substation area, including wind speed, light intensity, temperature, etc.). For example, weather radar grid data can be used to obtain the wind speed distribution (e.g., wind speed in one area is 10m / s, while in adjacent areas it is 8m / s) and light intensity distribution (e.g., 800W / m² in one area, while in adjacent areas it is 600W / m²) in a substation area). It also integrates output correlations between adjacent substations (referring to the mutual influence between the outputs of adjacent substations). For example, if the photovoltaic output of substation A decreases by 10%, historical data analysis indicates that the wind power output of adjacent substation B may decrease by 8%, with a correlation coefficient of 0.8. This correlation is calculated using a correlation coefficient matrix. This generates a model with spatiotemporal coupling features (referring to the interrelationship between the time and space dimensions in a three-dimensional uncertainty cloud map). For example, the output of a photovoltaic power station is characterized by a 10% decrease in the temporal dimension and a localized decrease in spatial dimension. The spatiotemporal coupling characteristic is expressed as an 80% probability of a 10% decrease in the localized output in the next 15 minutes. [The following text appears to be unrelated and should likely be omitted:] A three-dimensional uncertainty cloud map is created, which includes the distribution of time, power, and probability of occurrence.

[0042] Through the reward and penalty rules of a game-themed framework, the forecast accuracy issue is transformed into an economic incentive issue, incentivizing new energy operators and load aggregators to proactively improve forecast accuracy. By leveraging hierarchical perception of environmental disturbances and interval reshaping logic, forecast commitment intervals are dynamically adjusted based on the priority and intensity of disturbance characteristics, improving forecast fairness and adaptability. By generating a three-dimensional uncertainty cloud map with spatiotemporal coupling features, it replaces traditional point forecasts and provides information on the distribution of time, power, and probability of occurrence, providing a more comprehensive basis for decision-making at the dispatch and allocation level. A feedback correction mechanism for forecast deviation data forms a closed-loop optimization of forecasting, scheduling, and execution, improving the overall forecast accuracy and stability of the system.

[0043] Due to the volatility of renewable energy output and the randomness of load demand, various resources within a substation (such as renewable energy, energy storage, loads, and grid interfaces) face conflicting objectives during scheduling. For example, maximizing renewable energy consumption can lead to over-discharge of energy storage equipment, impacting its lifespan; minimizing costs for adjustable loads can conflict with grid stability requirements. Traditional centralized optimization scheduling methods are computationally complex and struggle to respond quickly to real-time fluctuations. However, Claim 3 resolves this conflicting strategy through distributed autonomous decision-making and a multi-stage negotiation mechanism, improving scheduling efficiency and system stability.

[0044] At the dispatch and allocation layer, substation resources are divided into new energy clusters [referring to the collection of all new energy generation equipment within a substation, such as photovoltaic power plants and wind farms. These devices are connected to the grid through inverters and can adjust active and reactive power output. For example, a substation may contain 10 photovoltaic power generation units and 5 wind turbines, which together form a new energy cluster], energy storage alliances [referring to the collection of all energy storage equipment within a substation, such as lithium-ion battery packs and supercapacitors. These devices are connected to the grid through a power conversion system (PCS) and can perform charging and discharging operations. For example, a substation may contain 3 large lithium-ion battery packs and 2 supercapacitors, which together form an energy storage alliance], and adjustable load groups [referring to the collection of all adjustable loads within a substation, such as interruptible loads for industrial users and smart home appliances for residential users. These loads can adjust their power consumption through demand response mechanisms. For example, a substation may contain 5 interruptible production lines for industrial users and 1,000 smart air conditioners for residential users, which together form an adjustable load group]. 】 and grid interface agent [refers to the interface equipment or virtual agent between the substation and the upper power grid, which is responsible for executing the dispatch instructions of the upper power grid, such as power purchase or sales plan. The goal of the grid interface agent is to ensure that the power exchange of the substation meets the requirements of the upper power grid. For example, a substation may be connected to the upper power grid through a transformer, and the grid interface agent is responsible for managing the power flow between the substation and the upper power grid] four types of autonomous decision-making entities [refers to the above four types of resource collections that have independent decision-making capabilities in the dispatch process. Each entity independently formulates a strategy based on its own goals and constraints, and negotiates with other entities to achieve the global optimum. For example, the new energy cluster independently determines the output plan, and the energy storage alliance independently determines the charging and discharging strategy. ], the edge layer solves the Nash equilibrium point based on the three-dimensional uncertainty cloud map and generates a dispatch plan; performs dynamic electrical partitioning based on the feeder voltage-power sensitivity matrix, obtains the partitioning results, and the dispatch plan and partitioning results are used as inputs to the execution control layer; The method of solving the Nash equilibrium point based on the three-dimensional uncertainty cloud map and generating a scheduling plan includes: Substation resources are divided into four autonomous decision-making entities: new energy clusters, energy storage alliances, adjustable load groups, and grid interface agents. The goal of the new energy cluster is to maximize power consumption [this means that the new energy cluster outputs as much power as possible to reduce the curtailment of wind and solar power. For example, the goal of a photovoltaic power station is to generate as much power as possible when there is sufficient sunlight, with output power reaching over 90% of its installed capacity.], the goal of the energy storage alliance is to minimize lifespan loss, the goal of the adjustable load group is to minimize electricity costs, and the goal of the grid interface agent is to meet the superior dispatch instructions, which serves as the basis for policy conflict negotiation. Conflict features are extracted from the objective functions of each autonomous decision-making subject to generate a conflict feature set, where the conflict features include target priority features [referring to the importance of each subject's goals, which are used to measure the relative importance of each other. For example, the priority of the new energy cluster's absorption capacity may be higher than the priority of the life loss of energy storage, because new energy absorption is the core goal of system operation.] and constraint boundary features [referring to the restrictions that each subject must meet when achieving its goals. For example, the SOC protection boundary of energy storage means that the SOC must not be lower than 20% or higher than 80% to protect the battery life; the reduction capacity boundary of the adjustable load means that the load reduction amount must not exceed its adjustable range, such as industrial users can reduce electricity consumption by up to 30%.], the target priority features include the absorption capacity priority of the new energy cluster and the life loss priority of energy storage, and the constraint boundary features include the SOC protection boundary of energy storage and the reduction capacity boundary of the adjustable load, which serve as the basis for hierarchical negotiation; The objective function is a mathematical expression that describes the optimization goal of each autonomous decision-making entity and is used to quantify its decision-making objectives. Conflict feature extraction involves analyzing the contradictions between the objective functions and extracting the key features that lead to the conflict. For example, by comparing the output plan of the new energy cluster with the charging and discharging plan of the energy storage alliance, the conflict feature "increased output of new energy leads to decreased SOC of energy storage" can be extracted.

[0045] Decompose the conflict feature set into multiple conflict levels based on the severity and impact of the conflict. The conflict levels include the initial conflict level and the severe conflict level. The initial conflict level corresponds to conflicts where the SOC is close to exceeding the limit, while the severe conflict level corresponds to conflicts where the SOC actually exceeds the limit. These conflict levels serve as input for constraint convergence. The decomposition is based on the severity and scope of the conflict. Severity refers to the threat the conflict poses to system stability. For example, a conflict approaching SOC exceeding the limit poses a lower threat, while a conflict actually exceeding the limit poses a higher threat. Scope refers to the number of entities or regions involved in the conflict. For example, a conflict involving only new energy and energy storage has a smaller scope, while a conflict involving all entities has a larger scope.

[0046] Conflict level: The conflict level includes the initial conflict level and the serious conflict level.

[0047] Initial conflict level: corresponds to the conflict where the SOC is close to exceeding the limit. For example, the SOC of the energy storage drops to 25%, close to the protection limit of 20%, but has not yet exceeded it.

[0048] Severe conflict level: corresponds to conflicts where the SOC actually exceeds the limit. For example, the SOC of the energy storage drops to 18%, exceeding the protection limit of 20%, which may cause battery damage.

[0049] As input to constraint convergence: The conflict level serves as an input to constraint convergence, guiding the subsequent negotiation mechanism. For example, initial conflict levels can be resolved with minor adjustments to the policy, while severe conflict levels require major adjustments to the policy or even relaxation of constraint boundaries.

[0050] Based on a three-dimensional uncertainty cloud map, a multi-stage negotiation mechanism is used to resolve strategic conflicts at different levels. In the initial conflict stage, the new energy cluster adjusts its output plan to meet the energy storage's SOC protection boundary. At the severe conflict stage, the energy storage alliance relaxes the SOC protection boundary. The negotiation process is iterative, with each iteration narrowing the feasible domain of the strategies. The resulting negotiated strategy set is then generated as input for solving the Nash equilibrium. The edge layer solves the Nash equilibrium point based on the negotiated policy set (e.g., "renewable energy output 185kW, energy storage discharge 45kW, adjustable load reduction 8kW, and grid power purchase 35kW") and the three-dimensional uncertainty cloud map, and generates a dispatch plan. The dispatch plan includes each entity's active / reactive power regulation range, charging and discharging strategies, load transfer plan, and power purchase and sales plan.

[0051] For example, suppose a new energy cluster within a substation plans to output 200 kW of power within the next 15 minutes. The energy storage alliance plans to discharge 50 kW to support new energy consumption. The adjustable load group plans to reduce load by 10 kW to reduce costs. The grid interface agent plans to purchase 30 kW of power to meet superior instructions. Conflict feature extraction revealed conflict features such as "the 200 kW output of new energy causes the energy storage SOC to drop to 25%, approaching the 20% protection limit" and "the 10 kW reduction in adjustable load causes the grid to purchase 40 kW of power, exceeding the planned limit." These features constitute the conflict feature set. In the conflict hierarchy decomposition, "the SOC drop to 25%" is classified as an early conflict level because it has not yet breached the protection limit; "the grid purchase increase to 40 kW" is classified as a severe conflict level because it exceeds the planned limit. These conflict hierarchies serve as inputs for constraint convergence and guide subsequent negotiations.

[0052] Based on the three-dimensional uncertainty cloud map, the new energy cluster discovered that while there was an 80% probability of outputting 200kW, this would cause the energy storage SOC to drop to 25%, representing an initial conflict level. Therefore, in the first iteration, the new energy cluster reduced its output to 190kW, restoring the energy storage SOC to 22%, still close to the protection boundary. In the second iteration, the new energy cluster further reduced its output to 185kW, restoring the energy storage SOC to 20%, meeting the protection boundary. Meanwhile, the adjustable load group discovered that a 10kW reduction would result in an increase in grid purchases to 40kW, representing a severe conflict level. In the first iteration, the adjustable load group reduced its reduction to 8kW, reducing grid purchases to 38kW. In the second iteration, the energy storage alliance relaxed the SOC protection boundary to 15%, supporting the new energy output of 185kW. Ultimately, grid purchases dropped to 35kW, meeting the planned range. After multiple rounds of negotiation, the negotiated policy set was generated: "new energy output 185kW, energy storage discharge 45kW, adjustable load reduction 8kW, grid purchase 35kW," which served as input for solving the Nash equilibrium.

[0053] Methods for performing dynamic electrical partitioning based on the feeder voltage-power sensitivity matrix and obtaining partitioning results include: Perform multi-dimensional decomposition of substation status data to extract change characteristics. Substation status data includes voltage, power, and power flow distribution. Change characteristics serve as the basis for disturbance perception, including instantaneous change characteristics (referring to the sudden change characteristics of data in a short period of time), trend change characteristics (referring to the trend of data change over a longer period of time), and spatial distribution characteristics (referring to the spatial distribution differences of data). Among them, multi-dimensional decomposition refers to decomposing substation status data in three dimensions: time, space and physical quantity to extract change characteristics at different scales.

[0054] Time dimension decomposition: Use time series analysis methods (such as wavelet transform and Fourier transform) to decompose data into components of different time scales, such as minutes and hours.

[0055] Spatial dimension decomposition: Decompose the data into components of different spatial scales, such as a local area or the entire substation area, through spatial interpolation methods (such as Kriging interpolation) or network topology analysis.

[0056] Physical quantity dimension decomposition: Decompose data into components of different physical quantities, such as voltage, power, and current, through statistical analysis methods (such as principal component analysis).

[0057] Power flow distribution refers to the distribution of power flow within each feeder and node within a substation, including the direction and magnitude of active and reactive power. For example, a substation has three feeders. Feeder 1 has an active power flow of 50 MW and a reactive power flow of 20 Mvar; feeder 2 has an active power flow of 30 MW and a reactive power flow of 10 Mvar; and feeder 3 has an active power flow of -10 MW (in the reverse direction) and a reactive power flow of 5 Mvar.

[0058] Based on the three-dimensional uncertainty cloud map, a comprehensive evaluation of the change characteristics is performed to generate a disturbance index for the substation status. The comprehensive evaluation is based on the intensity and duration of the change characteristics. If the instantaneous change characteristics are strong and short in duration, the disturbance index is high. If the trend change characteristics are weak and long in duration, the disturbance index is low. This disturbance index serves as input for frequency optimization. Comprehensive assessment is to integrate the intensity and duration of change characteristics into a unified disturbance index through a weighted fusion method to quantify the degree of disturbance of the system state. Specific methods include: Step B1: Calculate the intensity values ​​for each of the instantaneous change feature, trend change feature, and spatial distribution feature. For example, the intensity of the instantaneous change feature can be calculated by the absolute value of the rate of change (e.g., a voltage change rate of -2% / minute has an intensity of 2%); the intensity of the trend change feature can be calculated by the total amount of change (e.g., a power change of +50kW / hour has an intensity of 50kW); and the intensity of the spatial distribution feature can be calculated by the variance (e.g., a power distribution variance of 1000kW² has an intensity of 1000).

[0059] Step B2: Normalize the duration of each feature. For example, if the duration of an instantaneous change feature is 1 minute, it is normalized to 0.1 (based on 10 minutes); if the duration of a trend change feature is 1 hour, it is normalized to 1.

[0060] Step B3: Determine the weight of each feature based on the probability distribution provided by the three-dimensional uncertainty cloud map. For example, if the cloud map shows an 80% probability of power fluctuation in the next 15 minutes, then the instantaneous change feature has a high weight (e.g., 0.6), the trend change feature has a low weight (e.g., 0.3), and the spatial distribution feature has a moderate weight (e.g., 0.1). The perturbation index is calculated as follows: Perturbation Index = Weight 1 × Intensity 1 × (1 - Duration 1) + Weight 2 × Intensity 2 × Duration 2 + Weight 3 × Intensity 3, where Weight 1, Weight 2, and Weight 3 correspond to the instantaneous, trend, and spatial distribution features, respectively.

[0061] Suppose that the voltage on feeder A at a substation suddenly drops from 10 kV to 9.8 kV within one minute (transient variation characteristics: 2% intensity, 0.1 duration). Feeder B's power increases from 500 kW to 550 kW within one hour (trend variation characteristics: 50 kW intensity, 1 duration). The power distribution variance of feeders A and B increases from 500 kW² to 1000 kW² (spatial distribution characteristics: intensity 1000, 0.5 duration). Based on the three-dimensional uncertainty cloud map, the probability of power fluctuation in the next 15 minutes is 80%. The weights are determined as 0.6 for transient, 0.3 for trend, and 0.1 for spatial. The disturbance index is calculated as: 0.6 × 2 × (1 - 0.1) + 0.3 × 50 × 1 + 0.1 × 1000 × 0.5 = 1.08 + 15 + 50 = 66.08. This disturbance index of 66.08 is used as input for frequency optimization.

[0062] Dynamically adjust the update frequency of the feeder voltage-power sensitivity matrix based on the disturbance index. If the disturbance index is above a first threshold, the update frequency is increased to the first value. If the disturbance index is below a second threshold, the update frequency is reduced to the second value. This optimized update frequency is used as the basis for dynamic electrical zoning. The first threshold indicates a high disturbance state, while the second threshold indicates a low disturbance state. If the disturbance index is between the first and second thresholds, the current frequency (e.g., updating every 5 minutes) is maintained.

[0063] Assume the disturbance index is 66.08, the first threshold is 80, and the second threshold is 30. Since 66.08 is between 30 and 80, the system status is determined to be medium disturbance, and the current update frequency of every 5 minutes is maintained. If the voltage sag subsequently intensifies and the disturbance index rises to 85 (above the first threshold of 80), the system status is determined to be high disturbance, and the update frequency is increased to every 1 minute. If the voltage stabilizes and the disturbance index drops to 25 (below the second threshold of 30), the system status is determined to be low disturbance, and the update frequency is reduced to every 10 minutes. The optimized update frequency (for example, every 1 minute) serves as the basis for dynamic electrical zoning.

[0064] Based on the optimized update frequency and scheduling scheme, the feeder voltage-power sensitivity matrix is ​​calculated to generate dynamic electrical partitions. The dynamic electrical partitions include strong coupling areas and weak coupling areas. The strong coupling areas are subject to centralized optimization scheduling, while the weak coupling areas are subject to autonomous decision scheduling. The partition results are generated. The feeder voltage-power sensitivity matrix is ​​used to quantify the impact of each feeder power change on the node voltage. The specific calculation method includes: Step C1: Based on the power flow equation (P = V²G - VYcosθ, Q = -V²B + VYsinθ, where P is active power, Q is reactive power, V is voltage, G is conductance, B is susceptance, Y is admittance, and θ is phase angle), calculate the sensitivity matrix using a linearization method. For example, the sensitivity of voltage to power can be expressed as and , which is usually calculated by taking the inverse of the Jacobian matrix.

[0065] Step C2: For a substation with n nodes and m feeders, the sensitivity matrix is ​​an n×m matrix, where the matrix elements represent the change in voltage at each node when the power of a feeder changes by 1 unit. For example, a substation has three nodes (N1, N2, N3) and two feeders (F1, F2). The calculated sensitivity matrix is: ; Assume the calculation result is: ; This means that when the power of feeder F1 increases by 1 MW, the voltage at node N1 increases by 0.05 kV, the voltage at node N2 increases by 0.02 kV, and the voltage at node N3 increases by 0.01 kV.

[0066] When the partition changes, the scheduling rules are adjusted through the continuous optimization function to generate the adjusted partition results as the update input of the linkage control of the execution control layer.

[0067] By dividing substation resources into four autonomous decision-making entities—new energy clusters, energy storage alliances, adjustable load groups, and grid interface agents—the system breaks away from the traditional centralized optimization paradigm, enabling distributed autonomous decision-making and reducing computational complexity. By extracting conflict features and decomposing conflict hierarchies, complex policy conflicts are addressed in layers, improving the targeted and efficient nature of conflict resolution. Through a multi-stage negotiation mechanism and iterative convergence approach, the strategies of each entity are dynamically adjusted to ensure the global optimality of the scheduling solution. The probability distribution information from the three-dimensional uncertainty cloud map guides policy conflict negotiation and Nash equilibrium solution, improving the accuracy and adaptability of the scheduling solution.

[0068] Through multi-dimensional decomposition and comprehensive evaluation, the system accurately perceives the degree of disturbance in the state, avoiding the blindness of fixed-frequency updates. By dynamically adjusting the update frequency, the efficiency of computing resources is optimized. Through continuous optimization functions, scheduling rule adjustments when partitions change are smoothed, enhancing system stability.

[0069] At the execution control layer, programmable impedance characteristics are embedded in the photovoltaic inverter and energy storage power conversion system, enabling autonomous device response based on the scheduling plan. The source-load fluctuation event chain model performs cross-time-scale linkage control based on the partition results. The execution results are fed back to the integration layer. Methods for embedding programmable impedance characteristics in photovoltaic inverters and energy storage power conversion systems to enable autonomous device responses based on scheduling schemes include: Perform multi-level decomposition of equipment operating data to extract impact features. The operating data includes the operating status of the photovoltaic inverter and energy storage power conversion system. The impact features include geographic impact features, power impact features, and sensitivity impact features, which serve as the basis for equipment impact assessment. Based on the zoning results, the impact characteristics are prioritized and a comprehensive importance index is generated for each device. The evaluation is based on the contribution of the impact characteristics to system stability. If the device's geographical impact characteristics indicate that it is located in a strong coupling zone and its sensitivity impact characteristics indicate that it has a large impact on voltage, then the importance index is high and serves as the input for priority adjustment. According to the comprehensive importance index, the impedance parameter adjustment priority of the devices is dynamically sorted. Among them, the impedance parameters of devices with high importance index are adjusted first, and the adjustment of devices with low importance index is delayed. The device adjustment sequence is generated as the basis for the autonomous response of the devices; Based on the scheduling plan and the equipment adjustment sequence, programmable impedance characteristics are implanted in the photovoltaic inverter and energy storage power conversion system to execute autonomous equipment response, where the autonomous response includes the new energy equipment simulating the inertial response of the synchronous machine and the energy storage equipment providing short-circuit capacity support, and the equipment response results are generated as input for the execution control layer linkage control; When the system state changes, the comprehensive importance index is re-evaluated through the continuity update function, and the device adjustment sequence is adjusted to generate an updated device adjustment sequence as the update input for the linkage control of the execution control layer.

[0070] For example, a substation contains 10 PV inverters (numbered PV1-PV10) and 5 energy storage power conversion systems (numbered ESS1-ESS5). Operational data includes the actual output of the PV inverters (kW), the state of charge (SOC, unit: %), and the output power (kW) of the energy storage power conversion systems.

[0071] Extract impact features through multi-level decomposition method: Geographical impact characteristics: Record the distance between each device and the substation load center. For example, PV1 is 100 meters away from the load center, and ESS1 is 50 meters away from the load center.

[0072] Power impact characteristics: Calculate the rated power and actual output of each device. For example, PV1 has a rated power of 500kW and an actual output of 450kW; ESS1 has a rated power of 300kW and an actual output of 200kW.

[0073] Sensitivity Impact Characteristics: Calculates the sensitivity of each device to the substation bus voltage (unit: V / kW). For example, the sensitivity of PV1 to the bus voltage is 0.05 V / kW, and that of ESS1 is 0.08 V / kW.

[0074] Prioritization based on zoning results: Assuming the dispatching and allocation layer has generated zoning results, the substation is divided into a strong coupling zone (including busbar 1 and feeders 1-3) and a weak coupling zone (including busbar 2 and feeders 4-6). PV1, PV2, and ESS1 are located in the strong coupling zone, while the remaining equipment is located in the weak coupling zone. Based on the zoning results, the impact characteristics are prioritized: Evaluation basis: If the geographical impact characteristics show that the device is located in a strong coupling area, the contribution is high; if the sensitivity impact characteristics show a large impact on voltage, the contribution is high; if the power impact characteristics show a large rated power, the contribution is high.

[0075] Comprehensive importance index calculation: A weighted summation method is used, with weights of 0.4 for geographic impact, 0.4 for sensitivity impact, and 0.2 for power impact. For example, the comprehensive importance index of ESS1 is 0.4 × (1 / 50) + 0.4 × 0.08 + 0.2 × (300 / 500) = 0.16; the index of PV1 is 0.4 × (1 / 100) + 0.4 × 0.05 + 0.2 × (500 / 500) = 0.11.

[0076] Device impedance parameter adjustment priority is dynamically ranked based on the comprehensive importance index: ESS1 (0.16) > PV1 (0.11) > PV2 (0.10) > ... > PV10 (0.05). Devices with a high importance index (such as ESS1) have their impedance parameters adjusted first, while devices with a low importance index (such as PV10) have their impedance adjusted later.

[0077] Based on the dispatch plan, the equipment performs autonomous response. Assume that the dispatch plan requires the PV output to increase by 100kW and the energy storage to discharge by 50kW. According to the equipment adjustment sequence, the impedance parameters of ESS1 are adjusted first: Programmable impedance feature: Through the power electronic switch array, the virtual impedance of ESS1 is adjusted to 0.1+j0.05 ohms, providing short-circuit capacity support and increasing the discharge power by 50kW.

[0078] New energy equipment response: Adjust the virtual impedance of PV1 to 0.15+j0.08 ohms to simulate the inertia response of the synchronous machine, and the output will increase by 60kW; adjust the impedance of PV2, and the output will increase by 40kW.

[0079] Device response results: Generate device response results, including ESS1 discharging 50kW, PV1 output increasing 60kW, and PV2 output increasing 40kW, as inputs for linkage control.

[0080] When the system state changes, the sequence is updated. For example, assume that the load center migrates to bus 2, transforming the weak coupling zone into a strong coupling zone. The comprehensive importance index is reassessed using a continuity update function (e.g., linear interpolation). For example, if the weight of PV10's geographic influence characteristic increases, its index rises from 0.05 to 0.12, and the adjustment sequence is updated to ESS1 > PV10 > PV1 > ... . This updated device adjustment sequence is generated and serves as the update input for the linkage control.

[0081] The method of cross-time-scale linkage control based on partition results of the source-load fluctuation event chain model includes: Decompose the indicators involved in the event chain and extract impact features. The indicators include PV output, load demand, and voltage level. The impact features include instantaneous impact features, trend impact features, and coupling impact features, which serve as the basis for correlation fusion. Based on the partitioning results, dynamic weights are assigned to the influencing features to generate a weight sequence. The weight sequence is based on the degree of influence of the influencing features on the substation stability. If the instantaneous influencing features have a large impact on the substation stability, a high weight is assigned. If the trend influencing features have a small impact, a low weight is assigned. The weight is dynamically adjusted according to the strong coupling area and weak coupling area of ​​the partitioning results, and serves as the input for condition optimization. The influence features and weight sequences are correlated and fused through the weight fusion function to generate composite trigger conditions and comprehensive influence indexes as the basis for linkage control; Based on device response results and compound trigger conditions, cross-timescale linkage control is executed. This linkage control includes millisecond-level energy storage compensation, second-level load shedding, and minute-level grid support. The linkage control results are generated as input to the execution results of the integration layer. The trigger threshold of the composite trigger condition is dynamically optimized according to the disturbance index of the substation status. If the disturbance index of the substation status is high, the trigger threshold is lowered; if the disturbance index of the substation status is low, the trigger threshold is increased. The optimized composite trigger condition is generated as the update input of the execution result of the integration layer.

[0082] At the integration layer, the cloud platform conducts game arbitration and market settlement based on the execution results. The edge layer uses quantum-inspired algorithms to accelerate the search for equilibrium points and updates the scheduling plan based on the execution results. The terminal layer dynamically reconstructs the impedance characteristics of the equipment through the power electronic switch array and adjusts the equipment operating status based on the updated scheduling plan.

[0083] Assume that the event chain involves a drop in PV output, a sudden increase in load demand, and a voltage drop. Extract impact features: Transient impact characteristics: PV output drops by 10% within 1 minute (for example, from 500kW to 450kW).

[0084] Trend impact characteristics: Load demand continues to increase by 5% within 1 hour (for example, from 1000kW to 1050kW).

[0085] Coupling impact characteristics: The correlation coefficient between photovoltaic output drop and voltage drop is 0.8.

[0086] Dynamic weight allocation based on partition results Assume that the partitioning results show that the voltage fluctuation in the strong coupling area is large. Based on the partitioning results, the weights are dynamically assigned: Allocation basis: The instantaneous impact feature has a greater impact on system stability and has a weight of 0.6; the trend impact feature has a smaller impact and has a weight of 0.3; the coupling impact feature has a weight of 0.1.

[0087] Strong coupling region adjustment: The weight of the strong coupling region is increased by 10%. For example, the weight of the instantaneous impact feature is adjusted to 0.66.

[0088] Generate a weight sequence: The weight sequence is [0.66, 0.27, 0.07], which is used as the input for conditional optimization.

[0089] Associated fusion generates compound trigger conditions Generate a comprehensive impact index through a weighted fusion function (such as weighted summation): Calculation formula: Comprehensive impact index = 0.66×(PV output decrease ratio) + 0.27×(load demand increase ratio) + 0.07×(correlation coefficient).

[0090] Example: PV output decreases by 10%, load demand increases by 5%, the correlation coefficient is 0.8, and the comprehensive impact index = 0.66×0.1+0.27×0.05+0.07×0.8=0.125.

[0091] Compound trigger condition: If the comprehensive impact index > 0.1, linkage control will be triggered.

[0092] Execute linkage control based on the device response results. Based on the device response results (ESS1 discharges 50kW, PV1 output increases 60kW) and the compound trigger condition (index 0.125>0.1), execute linkage control: Millisecond-level energy storage compensation: ESS1 discharges 50kW within 10 milliseconds to compensate for the drop in PV output.

[0093] Second-level load shedding: 10% of non-critical loads (e.g. 100kW) in the strong coupling zone are shelved within 5 seconds.

[0094] Minute-level grid support: Apply for 50kW of support from the upper-level grid within 1 minute.

[0095] Linkage control results: Generate linkage control results, including 50kW of energy storage compensation, 100kW of load shedding, and 50kW of grid support, as input for the execution results of the integration layer.

[0096] The trigger threshold is optimized based on the disturbance index. For example, if the substation's disturbance index is 0.8 (higher than the threshold of 0.6), the trigger threshold is lowered to 0.08. If the disturbance index drops to 0.4 (lower than the threshold of 0.5), the trigger threshold is raised to 0.12. This optimized composite trigger condition is generated and used as an input to update the execution results of the integration layer.

[0097] The cloud platform's methods for game arbitration and market clearing based on execution results include: Perform multi-dimensional decomposition of substation operation scenarios and extract impact characteristics. Operation scenarios include voltage collapse risk, insufficient new energy consumption, and sudden load changes. Impact characteristics include disturbance impact characteristics, scope impact characteristics, and economic impact characteristics, which serve as the basis for scenario impact assessment. Based on the zoning results, the impact characteristics are prioritized and a comprehensive emergency index for the scenario is generated. The evaluation is based on the degree of impact of the impact characteristics on substation operation. If the disturbance impact characteristics indicate a high risk of voltage collapse and the range impact characteristics indicate an impact in a strongly coupled area, the emergency index is high, which serves as an input for capital optimization. Dynamically optimize the allocation ratio of backup resource pool funds based on the comprehensive emergency index. If the scenario's comprehensive emergency index is above the first threshold, funds are prioritized for purchasing energy storage services. If the emergency index is below the second threshold, funds are prioritized for subsidizing affected parties. This generates a fund allocation plan that serves as the basis for game arbitration and market settlement. Based on the equipment response results and linkage control results, verify the technical feasibility of the dispatch plan, where technical feasibility includes power flow safety and voltage stability, and generate arbitration results as input for market clearing; Based on the arbitration results and fund allocation plan, the reward and punishment funds and ancillary service fees are settled. Among them, the reward and punishment funds include forecast deviation penalties and economic compensation for new energy operators and load aggregators, and the ancillary service fees include the fees for energy storage services and load reduction services. The settlement results are generated as input for the edge layer's updated scheduling plan.

[0098] Assume the operating scenario is voltage collapse risk. Extract impact features: Disturbance impact characteristics: voltage drops by 5% (for example, from 110kV to 104.5kV).

[0099] Range impact characteristics: affects three feeders in the strong coupling area.

[0100] Economic impact characteristics: Potential economic losses are 100,000 yuan.

[0101] Priority assessment based on zoning results. Based on zoning results (large voltage fluctuations in strong coupling areas), evaluate impact characteristics: Assessment basis: The disturbance impact characteristics show that the voltage drop is serious, with a weight of 0.5; the range impact characteristics show that the strong coupling area is affected, with a weight of 0.3; the economic impact characteristics show that the loss is large, with a weight of 0.2.

[0102] Calculation of comprehensive emergency index: Comprehensive emergency index = 0.5×(5 / 10)+0.3×(3 / 6)+0.2×(10 / 20)=0.5.

[0103] Fund allocation is optimized based on the emergency index. For example, if the comprehensive emergency index is 0.5 and exceeds the first threshold of 0.4, priority is given to purchasing energy storage services from the backup resource pool (for example, 70% of the funds, or 7,000 yuan). If the emergency index drops to 0.3 and is less than the second threshold of 0.35, priority is given to subsidizing affected parties (for example, 60% of the funds, or 6,000 yuan). This generates a fund allocation plan that serves as the basis for game arbitration and market settlement.

[0104] The technical feasibility was verified based on the execution results. The dispatch plan was also verified based on the equipment response results (ESS1 discharge 50kW, PV1 output increase 60kW) and the linkage control results (energy storage compensation 50kW, load shedding 100kW, grid support 50kW): Power flow safety: Power flow calculation shows that the power of feeder 1 does not exceed the limit.

[0105] Voltage stability: Bus 1 voltage recovered to 105kV, meeting the safety range.

[0106] Arbitration result: The scheduling plan is technically feasible and can serve as input for market clearing.

[0107] Market liquidation based on arbitration results and fund allocation plan Based on the arbitration result (technical feasibility) and the fund allocation plan (procurement of energy storage services for RMB 7,000), settlement: Reward and Penalty Fund: New energy operators pay a penalty of 500 yuan for forecast deviations, and load aggregators pay 300 yuan for deviations.

[0108] Auxiliary service fees: energy storage service fee 4,000 yuan, load reduction service fee 2,000 yuan.

[0109] Liquidation result: Generates the liquidation result, including a penalty of 800 yuan and an auxiliary service fee of 6,000 yuan, as input for the edge layer update scheduling plan.

[0110] The edge layer uses quantum-inspired algorithms to accelerate equilibrium point search and update the scheduling plan based on the execution results. The methods include: Perform multi-dimensional decomposition on the execution results to extract deviation features, where the execution results include device response results and linkage control results. The deviation features include device response deviation features and linkage control deviation features, which serve as the basis for strategy optimization; Deviation characteristics are prioritized based on the dispatch plan to generate a deviation priority sequence. The evaluation is based on the degree to which the deviation characteristics affect the stability and economic efficiency of the substation. If the equipment response deviation characteristics indicate that power regulation exceeds the dispatch plan range, the priority is high. If the linkage control deviation characteristics indicate that voltage fluctuations exceed the safe range, the priority is also high. This sequence serves as input for the equilibrium point search. A quantum-inspired algorithm is used to accelerate equilibrium point search based on liquidation results. The liquidation results are used to adjust the reward and punishment rules of the game framework and generate an optimized strategy set as the basis for scheduling plan updates. Based on the optimized strategy set and deviation priority sequence, the dispatch plan is optimized. This optimization includes adjusting the active / reactive power regulation range, charging and discharging strategies, load transfer plans, and power purchase and sales plans of each entity. An updated dispatch plan is generated as input for adjusting the operating status of equipment at the terminal layer. When the system state changes, the deviation priority sequence is re-evaluated through the continuous optimization function, and the optimized strategy set is adjusted to generate an adjusted update scheduling scheme as the update input for the terminal layer to adjust the device operating state.

[0111] Assume that the integration layer has generated execution results, including device response results (for example, the actual output of photovoltaic inverter PV1 increases by 60kW, and the energy storage power conversion system ESS1 discharges 50kW) and linkage control results (for example, energy storage compensation 50kW, load shedding 100kW, and grid support 50kW). Deviation features are extracted through multi-dimensional decomposition methods: Equipment response deviation characteristics: Compare the difference between the actual equipment response and the dispatch plan. For example, the dispatch plan requires PV1 to increase its output by 50kW, but the actual increase is 60kW, with a deviation of +10kW; ESS1 requires 50kW discharge, but the actual discharge is 50kW, with a deviation of 0kW.

[0112] Linkage Control Deviation: This feature compares the actual linkage control effect with the target. For example, if the target is to restore bus 1 voltage to 105kV, but the actual voltage is restored to 104kV, the deviation is -1kV.

[0113] Priority evaluation based on the dispatch plan: Based on the dispatch plan (for example, PV output increased by 50kW, energy storage discharged by 50kW, load shedding by 100kW), the priority of the deviation characteristics is evaluated: Evaluation basis: If the equipment response deviation characteristics show that the power regulation exceeds the scheduling plan range (for example, PV1 deviation +10kW), the priority is high; if the linkage control deviation characteristics show that the voltage fluctuation exceeds the safety range (for example, voltage deviation -1kV, safety range ±0.5kV), the priority is high.

[0114] Deviation priority sequence: Generates a sequence, such as voltage deviation (-1 kV, priority 0.8) > PV1 power deviation (+10 kW, priority 0.6) > ESS1 power deviation (0 kW, priority 0.1), as input for equilibrium point search.

[0115] Based on the liquidation results, a quantum-inspired algorithm is used to accelerate the equilibrium search. Assume that the integration layer liquidation results show that the new energy operator pays a forecast deviation penalty of 500 yuan and a storage service fee of 4,000 yuan. Based on the liquidation results, the reward and punishment rules of the game framework are adjusted, for example, increasing the penalty ratio by 10% (from 0.1 yuan per kW deviation to 0.11 yuan). A quantum-inspired algorithm (such as the quantum annealing algorithm) is used to accelerate the equilibrium search: Search process: Using the deviation priority sequence as a constraint, optimize the strategies of each entity (e.g., new energy clusters reduce output deviations, energy storage alliances adjust discharge strategies).

[0116] Optimized strategy set: Generate an optimized strategy set, such as controlling the output deviation of the new energy cluster within ±5kW and the discharge deviation of the energy storage alliance within ±2kW, as the basis for updating the scheduling plan.

[0117] Tune the scheduling scheme based on the optimized policy set. Tune the scheduling scheme based on the optimized policy set and deviation priority sequence: Optimization content: Adjust the active power regulation range of the new energy cluster to ±5kW, the charging and discharging strategy of the energy storage alliance to a discharge deviation of ±2kW, the load transfer plan to a removal deviation of ±10kW, and the power purchase and sales plan to a support deviation of ±5kW.

[0118] Updated dispatch plan: Generate an updated dispatch plan, such as PV1 output increase of 45kW, ESS1 discharge of 48kW, load shedding of 90kW, and grid support of 45kW, as input for the terminal layer to adjust the equipment operating status.

[0119] Adjust the policy set when the system state changes. Assuming a system state change (for example, a sudden 10% increase in load demand), the deviation priority sequence is re-evaluated using a continuous optimization function (such as an exponential decay function). For example, if a sudden load increase causes the voltage deviation priority to increase from 0.8 to 0.9, the optimized policy set is adjusted (for example, the load transfer plan deviation is relaxed to ±15kW). This generates an adjusted updated scheduling plan, which serves as the updated input for adjusting the operating state of the terminal layer.

[0120] The method for updating the scheduling plan based on the execution result also includes: After adding the liquidation results to the execution results, multi-dimensional integration is performed to extract optimization features, including equipment response optimization features, linkage control optimization features, and economic optimization features, which serve as the basis for rolling optimization; Based on the scheduling plan, the optimization features are prioritized and an optimization priority sequence is generated. The evaluation is based on the degree to which the optimization features improve the stability and economic efficiency of the substation. If the equipment response optimization feature shows a reduction in power regulation deviation, the priority is high. If the economic optimization feature shows a reduction in cost, the priority is high. This sequence is used as the input for the rolling optimization. Based on the optimized priority sequence and the updated scheduling plan, a rolling optimization of the scheduling plan is performed. The rolling optimization includes refreshing the three-dimensional uncertainty cloud map and the scheduling plan at preset time intervals to generate an optimized scheduling plan as input for the terminal layer to adjust the equipment operating status; When an emergency is detected, the system immediately switches to emergency mode based on the linkage control results. In this mode, rapid response resources are prioritized, including energy storage devices and adjustable load resources, and an emergency management plan is generated as a temporary input for the terminal layer to adjust the operating status of the equipment. When the substation status changes, the optimization priority sequence is re-evaluated through the continuous optimization function, and the optimized scheduling plan is adjusted to generate the adjusted optimized scheduling plan as the update input for adjusting the equipment operating status at the terminal layer.

[0121] Add the liquidation results (e.g., a penalty of 800 yuan and an auxiliary service fee of 6,000 yuan) to the execution results, perform multi-dimensional integration, and extract optimization features: Equipment response optimization feature: PV1 power regulation deviation is reduced from +10kW to +5kW, with an optimization range of 50%.

[0122] Linkage control optimization feature: voltage deviation is reduced from -1kV to -0.5kV, with an optimization range of 50%.

[0123] Economic optimization features: operating costs were reduced from 100,000 yuan to 90,000 yuan, with an optimization range of 10%.

[0124] Based on the updated dispatch plan (for example, PV1 output increases by 45kW and ESS1 discharges by 48kW), evaluate the priority of the optimization features: Evaluation basis: If the equipment response optimization feature shows a reduction in power regulation deviation (for example, a PV1 optimization range of 50%), then the priority is high; if the economic optimization feature shows a cost reduction (for example, a 10% optimization range), then the priority is high.

[0125] Optimization priority sequence: Generate a sequence, such as PV1 power optimization (priority 0.7) > voltage optimization (priority 0.6) > cost optimization (priority 0.4), as input for rolling optimization.

[0126] Based on the optimization priority sequence and the updated scheduling plan, perform rolling optimization: Rolling optimization process: The 3D uncertainty cloud map (e.g., updating the PV output probability distribution) and the dispatch plan are refreshed every 5 minutes. For example, the PV1 output plan is adjusted to increase by 40kW, and the ESS1 discharge plan is adjusted to 46kW.

[0127] Optimized dispatch plan: Generate an optimized dispatch plan, such as PV1 output increase of 40kW, ESS1 discharge of 46kW, load shedding of 85kW, and grid support of 40kW, as input for the terminal layer to adjust the equipment operating status.

[0128] Assume that an emergency event is detected (for example, a 20% drop in PV output), and the system immediately switches to emergency mode based on the linkage control result (for example, 50kW of energy storage compensation): Emergency mode: Prioritize fast-response resources, such as increasing ESS1 discharge to 60kW and adjustable load shedding to 120kW.

[0129] Emergency management plan: Generate an emergency management plan, such as ESS1 discharging 60kW and load shedding 120kW, as temporary input for the terminal layer to adjust the equipment operating status.

[0130] Assuming the system state changes (for example, load demand returns to normal), the optimization priority sequence is re-evaluated using a continuous optimization function (such as a linear interpolation function). For example, if the cost optimization priority increases from 0.4 to 0.6, the optimized dispatch plan is adjusted (for example, the load shedding plan is reduced to 80kW). The adjusted optimized dispatch plan is generated and used as the updated input for adjusting the operating status of the equipment at the terminal layer.

[0131] The terminal layer dynamically reconstructs the device impedance characteristics through the power electronic switch array and adjusts the device operating status based on the updated scheduling plan. The methods include: Perform multi-dimensional decomposition on the updated scheduling plan and extract adjustment features. The updated scheduling plan includes the updated scheduling plan and the optimized scheduling plan. The adjustment features include impedance adjustment features, power adjustment features, and time adjustment features, which serve as the basis for dynamic reconstruction. Based on the adjustment characteristics, the device impedance characteristics are dynamically reconstructed through the power electronic switch array. The impedance characteristics include the real and imaginary values ​​of the virtual impedance. The reconstruction includes the new energy equipment simulating the inertial response of the synchronous machine and the energy storage device providing short-circuit capacity support. The reconstructed impedance characteristics are generated as input for adjusting the device operating status. Based on the reconstructed impedance characteristics, the device operating state is adjusted. The operating state includes the actual output, state of charge, and impedance value. The adjusted device operating state is generated as input for the next round of prediction. Feedback the adjusted device operating status to the edge layer in real time to update the execution results, generate the feedback execution results, and serve as input for the next round of game arbitration and market liquidation in the integration layer; When the substation status changes, the adjustment characteristics are re-evaluated through the continuous optimization function, and the reconstructed impedance characteristics are adjusted to generate the updated equipment operation status as the updated input for the next round of prediction.

[0132] Perform multi-dimensional decomposition on the updated scheduling plan (including the updated scheduling plan and the optimized scheduling plan, such as PV1 output increased by 40kW and ESS1 discharge 46kW) to extract adjustment features: Impedance adjustment feature: PV1 needs to adjust the virtual impedance to 0.12 + j0.06 ohms, and ESS1 needs to adjust it to 0.09 + j0.04 ohms.

[0133] Power adjustment characteristics: PV1 output is adjusted to 490kW, and ESS1 discharge is adjusted to 46kW.

[0134] Time adjustment feature: The adjustment time is completed within 5 seconds.

[0135] Based on the adjustment characteristics, the device impedance characteristics are dynamically reconstructed through the power electronic switch array: PV1 reconstruction: Adjust the virtual impedance to 0.12+j0.06 ohms to simulate the inertial response of the synchronous machine, and increase the output to 490 kW.

[0136] ESS1 reconstruction: Adjust the virtual impedance to 0.09+j0.04 ohms to provide short-circuit capacity support and discharge 46kW.

[0137] Reconstructed impedance characteristics: Generates reconstructed impedance characteristics, such as PV1 impedance 0.12 + j0.06 ohms and ESS1 impedance 0.09 + j0.04 ohms, as input for adjusting the device operating status.

[0138] Adjust the device operating status based on the reconstructed impedance characteristics: PV1 operating status: actual output 490kW, impedance 0.12+j0.06 ohms.

[0139] ESS1 operating status: actual discharge 46kW, state of charge SOC dropped from 80% to 78%, impedance 0.09+j0.04 ohms.

[0140] Adjusted equipment operating status: Generates the adjusted equipment operating status, such as PV1 output 490kW and ESS1 state of charge 78%, as input for the next round of prediction.

[0141] The adjusted equipment operating status is fed back to the edge layer in real time, such as PV1 output 490kW and ESS1 state of charge 78%, to update the execution results. The generated execution results, such as equipment response deviation reduced to +2kW and voltage deviation reduced to -0.3kV, serve as input for the next round of game arbitration and market settlement at the integration layer.

[0142] Assuming a system state change (for example, a sudden 10% increase in load demand), the adjustment characteristics are re-evaluated using a continuous optimization function (such as an exponential decay function). For example, the impedance adjustment characteristic of PV1 needs to be adjusted to 0.10 + j0.05 ohms. After adjusting the reconstructed impedance characteristics, an updated device operating state is generated, such as PV1 output 495 kW and ESS1 state of charge 76%, which serves as the updated input for the next round of prediction.

[0143] Example 2

[0144] This embodiment discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the operation mode of the above-mentioned cloud computing-based smart substation energy management method is implemented.

[0145] Since the electronic device introduced in this embodiment is an electronic device used to implement a cloud computing-based smart substation energy management method in the embodiment of this application, based on the cloud computing-based smart substation energy management method introduced in the embodiment of this application, those skilled in the art can understand the specific implementation of the electronic device of this embodiment and its various variations, so how the electronic device implements the method in the embodiment of this application will not be described in detail here. As long as those skilled in the art implement the electronic device used in the cloud computing-based smart substation energy management method in the embodiment of this application, it falls within the scope of protection of this application.

[0146] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters and thresholds in the formulas are set by technicians in this field according to actual conditions.

[0147] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiment. All technical solutions based on the principles of the present invention are within the scope of protection of the present invention. It should be noted that for users of ordinary skill in the art, various improvements and modifications that do not depart from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A cloud computing-based smart substation energy management method, characterized in that: include: At the source-load forecasting layer, a "forecast-dispatch" game framework is constructed. New energy operators and load aggregators submit forecast commitment intervals to the cloud platform, and reward and penalty rules are implemented based on actual deviations. By integrating regional meteorological data and the output correlation of adjacent substations, a three-dimensional uncertainty cloud map with spatiotemporal coupling characteristics is generated to serve as the decision-making basis for the dispatching and allocation layer. At the dispatching and allocation layer, substation resources are divided into four autonomous decision-making entities: new energy clusters, energy storage alliances, adjustable load groups, and grid interface agents. The edge layer solves the Nash equilibrium point based on the three-dimensional uncertainty cloud map and generates a dispatch plan. Perform dynamic electrical partitioning based on the feeder voltage-power sensitivity matrix, obtain the partitioning results, and use the scheduling plan and partitioning results as input to the execution control layer; At the execution control layer, programmable impedance characteristics are embedded in the photovoltaic inverter and energy storage power conversion system, enabling autonomous device responses based on the scheduling scheme. The source-load fluctuation event chain model performs cross-time-scale linkage control based on partition results; the execution results are fed back to the integration layer At the integration layer, the cloud platform conducts game arbitration and market settlement based on the execution results. The edge layer uses quantum-inspired algorithms to accelerate the search for equilibrium points and updates the scheduling plan based on the execution results. The terminal layer dynamically reconstructs the impedance characteristics of the equipment through the power electronic switch array and adjusts the equipment operating status based on the updated scheduling plan.

2. The cloud computing-based smart substation energy management method according to claim 1, characterized in that: The method for generating the three-dimensional uncertainty cloud map includes: Initialize the game participants, with new energy operators and load aggregators as the main participants in the game framework. Set the game rules, including obtaining dispatch priority and financial compensation if the actual value is within the commitment range, and paying deviation penalties if the actual value exceeds the commitment range. The penalties are injected into the backup resource pool as the incentive mechanism of the game framework. Participants submit the initial forecast commitment range as the input of the game framework, thus establishing the "forecast-dispatch" game framework. Based on a game theory framework, environmental data is hierarchically decomposed to extract disturbance characteristics at different time and spatial scales. Environmental data includes wind speed, light intensity, and load demand. Disturbance characteristics include minute-level mutation characteristics, hour-level trend characteristics, local mutation characteristics, and regional trend characteristics, which serve as the basis for reshaping the commitment interval. For disturbance features with different scales, minute-level mutation features have higher priority than hour-level trend features, and local mutation features have higher priority than regional trend features. For disturbance features with the same scale, priority is ranked according to the degree of impact of the disturbance features on prediction accuracy. Dynamically calibrate the looseness of the initial forecast commitment interval based on the priority sequence and the strength of the disturbance feature. That is, if a high-priority minute-level mutation feature is detected, the tolerance range of the commitment interval is significantly relaxed and the penalty is reduced. If only a medium-priority hourly trend feature is detected, the commitment interval is moderately relaxed. The adjusted forecast commitment interval is generated as the updated input of the game framework. Based on the adjusted forecast commitment range, new energy operators and load aggregators submit forecast commitments to the cloud platform. The platform then implements reward and penalty rules based on actual deviations to generate forecast deviation data, which serves as a basis for correcting the three-dimensional uncertainty cloud map. By integrating forecast deviation data, regional meteorological data, and the correlation between the outputs of adjacent substations, a three-dimensional uncertainty cloud map with spatiotemporal coupling characteristics is generated. The three-dimensional uncertainty cloud map includes the distribution of time, power, and occurrence probability.

3. The cloud computing-based smart substation energy management method according to claim 2, characterized in that: The method for solving the Nash equilibrium point based on the three-dimensional uncertainty cloud map and generating a scheduling plan includes: Substation resources are divided into four autonomous decision-making entities: new energy clusters, energy storage alliances, adjustable load groups, and grid interface agents. The goal of the new energy cluster is to maximize absorption capacity, the goal of the energy storage alliance is to minimize lifespan loss, the goal of the adjustable load group is to minimize electricity costs, and the goal of the grid interface agent is to meet superior dispatch instructions. This serves as the basis for policy conflict negotiation. Conflict features are extracted from the objective functions of each autonomous decision-making entity to generate a conflict feature set. These conflict features include target priority features and constraint boundary features. The target priority features include the priority of the new energy cluster's consumption capacity and the priority of the energy storage's lifespan loss. The constraint boundary features include the energy storage's SOC protection boundary and the adjustable load reduction capacity boundary, which serve as the basis for hierarchical negotiation. Decompose the conflict feature set into multiple conflict levels based on the severity and impact of the conflict. The conflict levels include the initial conflict level and the severe conflict level. The initial conflict level corresponds to conflicts where the SOC is close to exceeding the limit, while the severe conflict level corresponds to conflicts where the SOC actually exceeds the limit. These conflict levels serve as input for constraint convergence. Based on a three-dimensional uncertainty cloud map, a multi-stage negotiation mechanism is used to resolve strategic conflicts at different levels. In the initial conflict stage, the new energy cluster adjusts its output plan to meet the energy storage's SOC protection boundary. At the severe conflict stage, the energy storage alliance relaxes the SOC protection boundary. The negotiation process is iterative, with each iteration narrowing the feasible domain of the strategies. The resulting negotiated strategy set is then generated as input for solving the Nash equilibrium. Based on the negotiated strategy set and three-dimensional uncertainty cloud map, the edge layer solves the Nash equilibrium point and generates a scheduling plan, which includes the active / reactive power regulation range, charging and discharging strategy, load transfer plan, and power purchase and sales plan of each entity.

4. The cloud computing-based smart substation energy management method according to claim 3, characterized in that: The method for performing dynamic electrical partitioning based on the feeder voltage-power sensitivity matrix and obtaining partitioning results includes: Perform multi-dimensional decomposition of substation status data to extract change characteristics. Substation status data includes voltage, power, and power flow distribution. Change characteristics serve as the basis for disturbance perception and include instantaneous change characteristics, trend change characteristics, and spatial distribution characteristics. Based on the three-dimensional uncertainty cloud map, a comprehensive evaluation of the change characteristics is performed to generate a disturbance index for the substation status. The comprehensive evaluation is based on the intensity and duration of the change characteristics. If the instantaneous change characteristics are strong and short in duration, the disturbance index is high. If the trend change characteristics are weak and long in duration, the disturbance index is low. This disturbance index serves as input for frequency optimization. Dynamically adjust the update frequency of the feeder voltage-power sensitivity matrix based on the disturbance index. If the disturbance index is higher than a first threshold, the update frequency is increased to a first frequency value. If the disturbance index is lower than a second threshold, the update frequency is reduced to a second frequency value. The optimized update frequency is generated as the basis for dynamic electrical zoning. Based on the optimized update frequency and scheduling scheme, the feeder voltage-power sensitivity matrix is ​​calculated to generate dynamic electrical partitions. The dynamic electrical partitions include strong coupling areas and weak coupling areas. The strong coupling areas are subject to centralized optimization scheduling, while the weak coupling areas are subject to autonomous decision scheduling. The partition results are generated. When the partition changes, the scheduling rules are adjusted through the continuous optimization function to generate the adjusted partition results as the update input of the linkage control of the execution control layer.

5. The cloud computing-based smart substation energy management method according to claim 4, characterized in that: The method for implanting programmable impedance characteristics in a photovoltaic inverter and energy storage power conversion system and enabling autonomous device response based on a scheduling scheme includes: Perform multi-level decomposition of equipment operating data to extract impact features. The operating data includes the operating status of the photovoltaic inverter and energy storage power conversion system. The impact features include geographic impact features, power impact features, and sensitivity impact features, which serve as the basis for equipment impact assessment. Based on the zoning results, the impact characteristics are prioritized and a comprehensive importance index is generated for each device. The evaluation is based on the contribution of the impact characteristics to system stability. If the device's geographical impact characteristics indicate that it is located in a strong coupling zone and its sensitivity impact characteristics indicate that it has a large impact on voltage, then the importance index is high and serves as the input for priority adjustment. According to the comprehensive importance index, the impedance parameter adjustment priority of the devices is dynamically sorted. Among them, the impedance parameters of devices with high importance index are adjusted first, and the adjustment of devices with low importance index is delayed. The device adjustment sequence is generated as the basis for the autonomous response of the devices; Based on the scheduling plan and the equipment adjustment sequence, programmable impedance characteristics are implanted in the photovoltaic inverter and energy storage power conversion system to execute autonomous equipment response, where the autonomous response includes the new energy equipment simulating the inertial response of the synchronous machine and the energy storage equipment providing short-circuit capacity support, and the equipment response results are generated as input for the execution control layer linkage control; When the system state changes, the comprehensive importance index is re-evaluated through the continuity update function, and the device adjustment sequence is adjusted to generate an updated device adjustment sequence as the update input for the linkage control of the execution control layer.

6. The cloud computing-based smart substation energy management method according to claim 5, characterized in that: The method for performing cross-time-scale linkage control based on partition results in the source-load fluctuation event chain model includes: Decompose the indicators involved in the event chain and extract impact features. The indicators include PV output, load demand, and voltage level. The impact features include instantaneous impact features, trend impact features, and coupling impact features, which serve as the basis for correlation fusion. Based on the partitioning results, dynamic weights are assigned to the influencing features to generate a weight sequence. The weight sequence is based on the degree of influence of the influencing features on the substation stability. If the instantaneous influencing features have a large impact on the substation stability, a high weight is assigned. If the trend influencing features have a small impact, a low weight is assigned. The weight is dynamically adjusted according to the strong coupling area and weak coupling area of ​​the partitioning results, and serves as the input for condition optimization. The influence features and weight sequences are correlated and fused through the weight fusion function to generate composite trigger conditions and comprehensive influence indexes as the basis for linkage control; Based on device response results and compound trigger conditions, cross-timescale linkage control is executed. This linkage control includes millisecond-level energy storage compensation, second-level load shedding, and minute-level grid support. The linkage control results are generated as input to the execution results of the integration layer. The trigger threshold of the composite trigger condition is dynamically optimized according to the disturbance index of the substation status. If the disturbance index of the substation status is high, the trigger threshold is lowered; if the disturbance index of the substation status is low, the trigger threshold is increased. The optimized composite trigger condition is generated as the update input of the execution result of the integration layer.

7. The cloud computing-based smart substation energy management method according to claim 6, characterized in that: The method for the cloud platform to perform game arbitration and market liquidation based on the execution results includes: Perform multi-dimensional decomposition of substation operation scenarios and extract impact characteristics. Operation scenarios include voltage collapse risk, insufficient new energy consumption, and sudden load changes. Impact characteristics include disturbance impact characteristics, scope impact characteristics, and economic impact characteristics, which serve as the basis for scenario impact assessment. Based on the zoning results, the impact characteristics are prioritized and a comprehensive emergency index for the scenario is generated. The evaluation is based on the degree of impact of the impact characteristics on substation operation. If the disturbance impact characteristics indicate a high risk of voltage collapse and the range impact characteristics indicate an impact in a strongly coupled area, the emergency index is high, which serves as an input for capital optimization. Dynamically optimize the allocation ratio of backup resource pool funds based on the comprehensive emergency index. If the scenario's comprehensive emergency index is above the first threshold, funds are prioritized for purchasing energy storage services. If the emergency index is below the second threshold, funds are prioritized for subsidizing affected parties. This generates a fund allocation plan that serves as the basis for game arbitration and market settlement. Based on the equipment response results and linkage control results, verify the technical feasibility of the dispatch plan, where technical feasibility includes power flow safety and voltage stability, and generate arbitration results as input for market clearing; Based on the arbitration results and fund allocation plan, the reward and punishment funds and ancillary service fees are settled. Among them, the reward and punishment funds include forecast deviation penalties and economic compensation for new energy operators and load aggregators, and the ancillary service fees include the fees for energy storage services and load reduction services. The settlement results are generated as input for the edge layer's updated scheduling plan.

8. The cloud computing-based smart substation energy management method according to claim 7, characterized in that: The method of using a quantum-inspired algorithm to accelerate the search for equilibrium points at the edge layer and updating the scheduling scheme based on the execution results includes: Perform multi-dimensional decomposition on the execution results to extract deviation features, where the execution results include device response results and linkage control results. The deviation features include device response deviation features and linkage control deviation features, which serve as the basis for strategy optimization; Deviation characteristics are prioritized based on the dispatch plan to generate a deviation priority sequence. The evaluation is based on the degree to which the deviation characteristics affect the stability and economic efficiency of the substation. If the equipment response deviation characteristics indicate that power regulation exceeds the dispatch plan range, the priority is high. If the linkage control deviation characteristics indicate that voltage fluctuations exceed the safe range, the priority is also high. This sequence serves as input for the equilibrium point search. A quantum-inspired algorithm is used to accelerate equilibrium point search based on liquidation results. The liquidation results are used to adjust the reward and punishment rules of the game framework and generate an optimized strategy set as the basis for scheduling plan updates. Based on the optimized strategy set and deviation priority sequence, the dispatch plan is optimized. This optimization includes adjusting the active / reactive power regulation range, charging and discharging strategies, load transfer plans, and power purchase and sales plans of each entity. An updated dispatch plan is generated as input for adjusting the operating status of equipment at the terminal layer. When the system state changes, the deviation priority sequence is re-evaluated through the continuous optimization function, and the optimized strategy set is adjusted to generate an adjusted update scheduling scheme as the update input for the terminal layer to adjust the device operating state.

9. The cloud computing-based smart substation energy management method according to claim 8, characterized in that: The method for updating the scheduling scheme based on the execution result further includes: After adding the liquidation results to the execution results, multi-dimensional integration is performed to extract optimization features, including equipment response optimization features, linkage control optimization features, and economic optimization features, which serve as the basis for rolling optimization; Based on the scheduling plan, the optimization features are prioritized and an optimization priority sequence is generated. The evaluation is based on the degree to which the optimization features improve the stability and economic efficiency of the substation. If the equipment response optimization feature shows a reduction in power regulation deviation, the priority is high. If the economic optimization feature shows a reduction in cost, the priority is high. This sequence is used as the input for the rolling optimization. Based on the optimized priority sequence and the updated scheduling plan, a rolling optimization of the scheduling plan is performed. The rolling optimization includes refreshing the three-dimensional uncertainty cloud map and the scheduling plan at preset time intervals to generate an optimized scheduling plan as input for the terminal layer to adjust the equipment operating status; When an emergency is detected, the system immediately switches to emergency mode based on the linkage control results. In this mode, rapid response resources are prioritized, including energy storage devices and adjustable load resources, and an emergency management plan is generated as a temporary input for the terminal layer to adjust the operating status of the equipment. When the substation status changes, the optimization priority sequence is re-evaluated through the continuous optimization function, and the optimized scheduling plan is adjusted to generate the adjusted optimized scheduling plan as the update input for adjusting the equipment operating status at the terminal layer.

10. The cloud computing-based smart substation energy management method according to claim 9, characterized in that: The terminal layer dynamically reconstructs the device impedance characteristics through the power electronic switch array and adjusts the device operating status based on the updated scheduling plan. The method includes: Perform multi-dimensional decomposition on the updated scheduling plan and extract adjustment features. The updated scheduling plan includes the updated scheduling plan and the optimized scheduling plan. The adjustment features include impedance adjustment features, power adjustment features, and time adjustment features, which serve as the basis for dynamic reconstruction. Based on the adjustment characteristics, the device impedance characteristics are dynamically reconstructed through the power electronic switch array. The impedance characteristics include the real and imaginary values ​​of the virtual impedance. The reconstruction includes the new energy equipment simulating the inertial response of the synchronous machine and the energy storage device providing short-circuit capacity support. The reconstructed impedance characteristics are generated as input for adjusting the device operating status. Based on the reconstructed impedance characteristics, the device operating state is adjusted. The operating state includes the actual output, state of charge, and impedance value. The adjusted device operating state is generated as input for the next round of prediction. Feedback the adjusted device operating status to the edge layer in real time to update the execution results, generate the feedback execution results, and serve as input for the next round of game arbitration and market liquidation in the integration layer; When the substation status changes, the adjustment characteristics are re-evaluated through the continuous optimization function, and the reconstructed impedance characteristics are adjusted to generate the updated equipment operation status as the updated input for the next round of prediction.

Images