A power station load intelligent distribution method and system under multi-unit cooperative constraint

Abstract

The present application relates to the technical field of smart grid, in particular to a power station load intelligent distribution method and system under multi-unit cooperative constraint, comprising the following steps: obtaining total load, rotating reserve, available state, grid-connected mode, water head, start-stop state, repair state and phase modulation state to determine feasible unit combination, traversing unit feasible output and eliminating vibration area, combining water consumption characteristics to form static load distribution table, extracting and screening selected scheme according to real-time constraint, calculating power generation water consumption flow and operation penalty flow to determine comprehensive water consumption flow and select distribution scheme, and recalculating when operation condition changes.In the present application, by incorporating unit combination constraint, vibration area avoidance, rotating reserve checking and water consumption evaluation into the same decision link, the load distribution takes into account both safety boundary and economic operation, reduces frequent start-stop and vibration area crossing, reduces dependence on experience correction, improves multi-unit joint operation stability, adapts to power station working condition changes in real time and improves water energy utilization efficiency.

Description

Technical Field

[0001] This invention relates to the field of smart grid technology, and in particular to a method and system for intelligent load allocation in power plants under multi-unit collaborative constraints. Background Technology

[0002] The field of smart grid technology involves the acquisition of power operation data, power allocation, and transmission of control commands between power plants, generator sets, grid-connected buses, dispatch control terminals, and monitoring servers. Among these, intelligent load allocation of power plants under multi-unit collaborative constraints refers to the collection of active power, reactive power, operating status, and load demand data of each generator set. The dispatch controller then calculates the output power of each generator set according to preset unit capacity, ramp-up constraints, and grid-connected operating parameters, and sends load allocation commands to the corresponding unit control units.

[0003] The existing load allocation within hydropower plants mainly relies on single-unit efficiency curves, empirical rules, or pre-set allocation strategies: operators select the combination of units to participate in operation based on the total load of the power plant, head conditions, and unit availability, and allocate the output of each unit according to empirical proportions or the principle of optimal local water consumption; when the units are under maintenance, shut down, undergo phase adjustment, have changed grid connection mode, or need to meet spinning standby requirements, operators usually make manual adjustments in conjunction with dispatching requirements.

[0004] The main technical problems are as follows:

[0005] First, the selection of unit combination and load allocation calculation are disconnected from each other. Operational constraints are often manually checked after allocation is completed. This can easily lead to situations where the generated allocation results do not meet the requirements of availability, maintenance, phase adjustment or grid connection mode, resulting in repeated adjustments to the scheme.

[0006] Second, the existing allocation process focuses on the efficiency of a single unit or the local water consumption level, and fails to incorporate the operational costs such as rotating standby, vibration zone avoidance, and start-up and shutdown changes into the evaluation in a unified manner. This may result in allocation results with low water consumption per instance but high overall operational risks and comprehensive costs.

[0007] Third, the vibration zone usually relies on operators to manually avoid it when adjusting the output. When the total load changes frequently or multiple units are adjusted in coordination, the number of times the unit crosses the vibration zone is easily increased, affecting the safety and stability of the unit operation.

[0008] Fourth, the impact of changes in start-up and shutdown status on load distribution lacks quantitative processing. Unnecessary unit start-up and shutdown switching may occur during load adjustment, increasing equipment wear and operational disturbances, which is not conducive to the long-term stable operation of the power plant.

[0009] Fifth, there is a lack of effective linkage between static load allocation data and real-time operating status. When the total load of the power plant, the availability of units, the maintenance status, or the phase adjustment status changes, it is difficult to obtain a better solution that meets multiple constraints in a timely manner by relying on manual re-judgment and correction.

[0010] Therefore, the existing plant load allocation method, which is mainly characterized by experience allocation, local efficiency judgment, and manual constraint verification, is difficult to simultaneously meet the requirements of safety constraints, economic water consumption, real-time adaptation, and stable regulation under the joint operation of multiple units. Summary of the Invention

[0011] The purpose of this invention is to provide a method and system for intelligent load allocation in power plants under multi-unit collaborative constraints. This method enables a hydropower station to determine the unit combination and load allocation scheme that meets the on-site operating conditions when multiple units are operating together, by combining operating constraints such as total load, spinning reserve, unit availability, grid connection mode, head, start-up and shutdown status, maintenance status, and phase adjustment status. Furthermore, it generates a plant load allocation result suitable for real-time operation, based on avoiding vibration zones and considering the water consumption characteristics of the units and the impact of start-up and shutdown changes.

[0012] To achieve the above objectives, the present invention adopts the following technical solution: a method for intelligent load allocation in power plants under multi-unit collaborative constraints, comprising the following steps:

[0013] Step S1: Obtain the total load of the power plant, spinning reserve demand, unit availability status, grid connection mode, head, start-up and shutdown status, maintenance status and unit phase adjustment status to form operating constraints, and determine feasible unit combinations based on the operating constraints.

[0014] Step S2: Based on the feasible unit combination, the feasible output of the units is traversed according to the head and the total load of the power station. The output range corresponding to the vibration zone is excluded within the feasible output range of each unit. Then, the static load allocation table is determined in combination with the water consumption characteristics of the units. The static load allocation table records the unit combination, unit output and power generation water consumption flow.

[0015] Step S3: Extract candidate load allocation schemes from the static load allocation table according to water head, total power plant load and feasible unit combination, and eliminate candidate load allocation schemes that do not meet the spinning reserve requirements, grid connection mode, maintenance status and unit phase adjustment status, and count the set of real-time available schemes.

[0016] Step S4: Calculate the power generation water consumption flow of each candidate load allocation scheme in the real-time available scheme set, determine the operation penalty flow based on start-stop changes and vibration zone crossing times, determine the comprehensive water consumption flow based on the power generation water consumption flow and the operation penalty flow, and determine the real-time load allocation scheme based on the comprehensive water consumption flow.

[0017] Step S5: When the total load of the power plant, the availability status of the generating units, the start-up and shutdown status, the maintenance status, or the phase adjustment status of the generating units changes, return to step S1 to redetermine the real-time load allocation scheme.

[0018] Specifically, in step S1, during the determination of feasible unit combinations:

[0019] The system obtains the total load of the power plant, spinning reserve requirement, unit availability status, grid connection mode, head, start-up / shutdown status, maintenance status, and unit phase adjustment status, and matches the unit availability status, start-up / shutdown status, and maintenance status with the unit number to generate the aforementioned operating constraints.

[0020] Using the aforementioned operational constraints, each unit is assessed to determine whether it meets the conditions for participating in power generation. When a unit is available, allowed to generate electricity in start-stop mode, and not under maintenance, the corresponding unit is included in the candidate unit set. The candidate unit set is then filtered based on the grid connection mode and the unit's phase adjustment status to obtain the feasible unit combination.

[0021] During the screening process, the first step is to determine whether each candidate unit can participate in the current grid-connected operation based on the grid connection method, and to eliminate candidate units that do not meet the current grid-connected operation requirements.

[0022] When a unit is in phase-shifting mode and cannot simultaneously generate power, the unit will be removed from the set of candidate units; when a unit is not in phase-shifting mode or is allowed to withdraw from phase-shifting mode to participate in power generation, the unit will retain its candidate status.

[0023] Specifically, in step S2, the process of determining the static load allocation table includes:

[0024] Obtain feasible unit combinations, head and total power plant load, and determine the feasible output range of each unit based on the head. Within the feasible output range, generate the corresponding unit output according to the preset output interval.

[0025] The unit outputs of each unit within the same feasible unit combination are combined, and the output combination whose sum matches the total load of the power station is selected to obtain the load-matching output combination.

[0026] The load matching output combinations are screened, and the unit outputs that fall into the output range corresponding to the vibration zone are eliminated. The power generation water consumption flow is determined in combination with the unit water consumption characteristics to obtain the static load allocation table.

[0027] Specifically, the process of determining the water consumption flow for power generation also includes:

[0028] Obtain the water consumption characteristics of each unit under the current head, and find the corresponding single unit water consumption flow rate according to the unit number and unit output combined with the unit water consumption characteristics;

[0029] When the unit output is located between two adjacent known output points, the single unit water consumption flow is determined according to the water consumption flow corresponding to the two adjacent known output points, and the single unit water consumption flow of each unit in the same load matching output combination is summarized to obtain the power generation water consumption flow.

[0030] Specifically, in step S3, during the statistical process of the set of real-time available solutions:

[0031] Using water head, total power plant load, and the feasible unit combination as search criteria, select candidate load allocation schemes with consistent unit combinations and matching load values ​​are extracted from the static load allocation table, and the unit output and power generation water consumption flow corresponding to each candidate load allocation scheme are read.

[0032] The candidate load allocation schemes are compared one by one with the spinning reserve requirement, grid connection mode, maintenance status and phase adjustment status of Unit 2. Candidate load allocation schemes with unmet constraints are deleted, and the remaining candidate load allocation schemes are sorted to obtain the set of real-time available schemes.

[0033] Specifically, in step S4, during the determination of the real-time load allocation scheme:

[0034] Read the unit output and power generation water consumption flow corresponding to each candidate load allocation scheme in the real-time available scheme set, and determine the number of start-stop changes based on the difference between the current start-stop status and the unit combination corresponding to the candidate load allocation scheme;

[0035] Based on whether the unit output crosses the output range corresponding to the vibration zone, the number of vibration zone crossings is recorded. The operating penalty flow is determined based on the number of start-stop changes and the number of vibration zone crossings. The power generation water consumption flow is added to the operating penalty flow to obtain the comprehensive water consumption flow.

[0036] Compare the total water consumption flow corresponding to each candidate load allocation scheme, and determine the candidate load allocation scheme with the minimum total water consumption flow as the real-time load allocation scheme.

[0037] Specifically, in the process of determining the penalty traffic:

[0038] Obtain the single start-stop penalty flow corresponding to the start-stop change and the single crossing penalty flow corresponding to the vibration zone crossing, and read the number of start-stop changes and the number of vibration zone crossings respectively;

[0039] The start-stop penalty flow is obtained by summing the single start-stop penalty flow with the corresponding number of start-stop changes, and the crossing penalty flow is obtained by summing the single crossing penalty flow with the corresponding number of crossings of the vibration zone. The start-stop penalty flow and the crossing penalty flow are then combined to obtain the operation penalty flow.

[0040] Specifically, the process for determining start / stop changes includes:

[0041] Obtain the current generating unit combination and the generating unit combination corresponding to the candidate load allocation scheme, and compare the participating status of the units in the two sets of generating unit combinations one by one according to the unit number;

[0042] When the same unit is in a state of participating in power generation in the current unit combination, but is in a state of not participating in power generation in the unit combination corresponding to the candidate load allocation scheme, it is recorded as a shutdown change;

[0043] When the same unit is not participating in power generation in the current unit combination, but is participating in power generation in the unit combination corresponding to the candidate load allocation scheme, it is recorded as a start-up change.

[0044] The total number of start-stop changes was obtained by combining the results.

[0045] Specifically, in the process of determining the number of vibration zone crossings:

[0046] The current output of each unit, the unit output corresponding to the selected load allocation scheme, and the output range corresponding to the vibration zone are obtained. The output change is judged according to the current output of the same unit to the direction of the selected output.

[0047] If the current output of the same unit is located on one side of the output range corresponding to the vibration zone and the selected output is located on the other side of the output range corresponding to the vibration zone, the corresponding unit is recorded as one vibration zone crossing. If the current output or the selected output is located within the output range corresponding to the vibration zone, the corresponding selected load allocation scheme is removed to obtain the number of vibration zone crossings.

[0048] A power plant load intelligent allocation system under multi-unit collaborative constraints includes a constraint module, a static table construction module, a scheme screening module, and an optimization control module.

[0049] The constraint module is used to obtain the total load of the power plant, spinning reserve demand, unit availability status, grid connection mode, head, start-up and shutdown status, maintenance status and unit phase adjustment status, form operating constraints, and determine feasible unit combinations based on the operating constraints.

[0050] The static table construction module is connected to the constraint module. It is used to traverse the feasible output of the units according to the feasible unit combination, the head and the total load of the power station, and exclude the output range corresponding to the vibration zone within the feasible output range of each unit. Then, it determines the static load allocation table in combination with the water consumption characteristics of the units. The static load allocation table records the unit combination, unit output and power generation water consumption flow.

[0051] The scheme selection module is connected to the static table construction module and the constraint module. It is used to extract candidate load allocation schemes from the static load allocation table according to the head, total load of the power station and feasible unit combinations, eliminate candidate load allocation schemes that do not meet the spinning reserve requirements, grid connection mode, maintenance status and unit phase adjustment status, and count the set of available schemes in real time.

[0052] The optimization control module is connected to the scheme selection module and is used to calculate the power generation water consumption flow of each candidate load allocation scheme in the real-time available scheme set. It determines the operation penalty flow based on start-up and shutdown changes and the number of vibration zone crossings. It determines the comprehensive water consumption flow based on the power generation water consumption flow and the operation penalty flow, and determines the real-time load allocation scheme based on the comprehensive water consumption flow. When the total load of the power plant, the unit availability status, start-up and shutdown status, maintenance status, or unit phase adjustment status changes, it returns to step S1 to redetermine the real-time load allocation scheme.

[0053] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0054] In this invention, a unified screening link is established around the total load of the hydropower station and the operating constraints of multiple units. The availability status, grid connection mode, head, start-up and shutdown status, maintenance status, and phase adjustment status are all incorporated into the unit combination judgment. When traversing feasible output, vibration zones are eliminated simultaneously and water consumption characteristics are associated to form a static load allocation basis that can be called in real time. Then, the candidate space is compressed by the constraints of spinning reserve and operating status. The power generation water consumption flow, start-up and shutdown changes, and vibration zone crossing times are converted into a comprehensive water consumption evaluation. This makes the load allocation shift from single-point efficiency comparison to a coordinated trade-off between safety constraints and operating costs, reducing ineffective screening and experience intervention, reducing frequent start-up and shutdown and vibration zone crossings, and improving the stability of load allocation within the plant and the efficiency of water energy utilization. Attached Figure Description

[0055] Figure 1 This is the main flowchart of power plant load allocation under multi-unit collaborative constraints according to the present invention;

[0056] Figure 2 This is a logic diagram for selecting feasible unit combinations according to the present invention;

[0057] Figure 3 A data flow graph is constructed for the static load allocation table of this invention;

[0058] Figure 4 This is a diagram showing the formation of a set of real-time available solutions and the calculation of comprehensive water consumption flow in this invention.

[0059] Figure 5 This is a data interaction diagram of the power plant load intelligent allocation system module of the present invention. Detailed Implementation

[0060] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0061] Please see Figures 1 to 4 This embodiment provides a method for intelligent load allocation in power plants under multi-unit collaborative constraints. The process unfolds as follows: First, the operating data is read in to form constraints for unit participation in power generation; then, a static load allocation table is generated based on the head and target load; during real-time operation, candidate schemes are retrieved from the static load allocation table, and unusable schemes are deleted according to the current constraints; finally, the power generation water consumption, start-up and shutdown changes, and vibration zone crossing are combined and calculated to select the scheme with the minimum comprehensive water consumption.

[0062] S1. Obtain operational data, formulate operational constraints, and determine feasible unit combinations.

[0063] The data entry points for S1 include the total power plant load, spinning reserve requirement, unit availability status, grid connection mode, head, start-up / shutdown status, maintenance status, and unit phase adjustment status. After this data is entered, it is first organized into unit status records by unit number. Each unit status record includes at least the unit number, unit availability status, start-up / shutdown status, maintenance status, phase adjustment status, and whether the unit meets the current grid connection mode.

[0064] S11. Establish unit status records.

[0065] During processing, the unit number is used as an index to write the unit's availability status, start-up / shutdown status, and maintenance status into the same record. If a unit lacks any of these statuses, it will not be included in the candidate unit set to avoid allocating power to units with incomplete statuses in subsequent processes. If duplicate status records exist for the same unit, the record corresponding to the current moment will be used as input; if the current status cannot be confirmed, the unit will not participate in this round of allocation.

[0066] S12. Select generating units that meet the conditions for participating in power generation.

[0067] Each generating unit is evaluated individually. If a unit is in an available state (available), allowed to generate electricity in a start-stop state, and not under maintenance, it is added to the candidate unit set. If any condition is not met, the unit is added to the rejection record, which retains the unit number and the reason for rejection.

[0068] S13. Select the set of candidate generating units according to the grid connection method.

[0069] Read the current grid connection mode and match each unit in the candidate unit set with the grid connection mode. Units that cannot participate in the current grid connection operation are removed from the candidate unit set; those that can participate are retained. This step outputs the grid connection-filtered candidate unit set, keeping the unit numbers unchanged.

[0070] S14. Continue screening according to the phase adjustment status of the unit.

[0071] Read the phase modulation status of the generating units. If a unit is in phase modulation status and cannot simultaneously generate power, it is removed from the candidate unit set; if a unit is not in phase modulation status, or is allowed to withdraw from phase modulation to participate in power generation, it is retained. After screening, the units in the candidate unit set are combined to form feasible unit combinations.

[0072] S1 outputs feasible unit combinations to S2 as combination boundaries for constructing the static load allocation table; it also sends them to S3 as combination conditions when retrieving the static load allocation table in real time. If no feasible unit combinations are formed after S1, the current round will not enter S2 to S4, and an abnormal state of "no feasible unit combinations" will be output. The round will then wait for changes in the total load of the power plant, unit status, maintenance status, or phase adjustment status before re-entering S1.

[0073] S2. Construct a static load allocation table based on feasible unit combinations, head, and total power plant load.

[0074] S2 receives the feasible unit combinations output by S1, and simultaneously reads the current head and total power plant load. The head is used to determine the feasible output range of each unit under the current conditions; the total power plant load is used to screen combinations in which the sum of the outputs of all units meets the target load; the output range corresponding to the vibration zone is used to eliminate unit outputs that are not allowed; the unit water consumption characteristics are used to calculate the water consumption flow of a single unit and the water consumption flow for power generation.

[0075] S21. Determine the feasible output range of each unit.

[0076] For each unit in a feasible unit combination, its feasible output range is determined by unit number and head. If the search is successful, the minimum and maximum feasible output of that unit are obtained. If no feasible output range corresponding to the current head is found, the combination containing that unit is not added to the static load allocation table. If any unit in a combination lacks a feasible output range, that combination will not participate in subsequent steps in this round.

[0077] S22, Generate candidate output points for the generating unit.

[0078] Within the feasible output range of each generating unit, candidate output points are generated according to a preset output interval. The preset output interval, used as the discrete traversal step size, is determined by the hydropower station's generating unit configuration. The generated candidate output points must not be lower than the unit's minimum feasible output, nor higher than the unit's maximum feasible output.

[0079] S23. Combine the output of each unit and match the total load of the power station.

[0080] Candidate unit output points within the same feasible unit combination are combined, and the sum of the unit outputs is calculated. If the sum of outputs matches the total load of the power plant, the output combination is written into the load matching output combination; otherwise, the output combination is discarded. Finally, the load matching output combination is output, and the data includes the unit combination, the unit number, and the corresponding unit output.

[0081] S24. Eliminate the output combinations within the vibration zone.

[0082] Read the output range corresponding to the vibration zone and check each load matching output combination. If the output of a certain unit falls within the output range corresponding to its vibration zone, the load matching output combination containing that unit's output is deleted. Load matching output combinations that do not fall within the output range corresponding to the vibration zone are included in the water consumption flow calculation.

[0083] S25. Calculate the water flow rate for power generation.

[0084] For the retained load matching output combinations, the unit's water consumption characteristics are searched by unit number, current head, and unit output. If the individual unit's water consumption flow can be directly found, the corresponding value is read; if the unit's output is between two adjacent known output points, the individual unit's water consumption flow is determined based on the water consumption flow corresponding to the two adjacent known output points. If a unit lacks water consumption characteristic data, or if the individual unit's water consumption flow cannot be determined based on adjacent known output points, then that load matching output combination is not written into the static load allocation table.

[0085] The water consumption of each unit in the same load-matching output combination is added together to obtain the power generation water consumption of that combination. The static load allocation table records the head, total power plant load, unit combination, output of each unit, and power generation water consumption. After the static load allocation table is generated, it is sent to S3 as a data source for real-time filtering.

[0086] If S2 does not generate any static load allocation table records, an abnormal status of "no static matching scheme" will be output. This round will not proceed to the comprehensive water consumption flow comparison; S1 and S2 will be re-executed after changes in head, total power plant load, or unit combination.

[0087] S3. Extract candidate load allocation schemes from the static load allocation table and form a set of real-time available schemes.

[0088] The inputs for S3 include head, total power plant load, feasible unit combinations, static load allocation table, spinning reserve demand, grid connection mode, maintenance status, and unit phase adjustment status. During processing, data is first retrieved according to the search criteria, and then schemes are deleted based on real-time constraints.

[0089] S31. Extract candidate load allocation schemes.

[0090] Using the current head, total power plant load, and feasible unit combinations as search criteria, records with consistent unit combinations and matching load values ​​are retrieved from the static load allocation table. Each retrieved result serves as a candidate load allocation scheme. Candidate load allocation schemes must include at least the unit combination, the output of each unit, and the water consumption for power generation.

[0091] If there is no matching record in the static load allocation table, S3 outputs an abnormal status of "no candidate load allocation scheme" and returns to S2 to rebuild or update the static load allocation table.

[0092] S32, Verify the rotational backup requirements.

[0093] Read each candidate load allocation scheme one by one and determine whether the scheme meets the current spinning reserve requirements. If it does not meet the requirements, delete the scheme; if it does meet the requirements, retain it. Deleted schemes will not be included in subsequent grid connection mode, maintenance status, and phase adjustment status checks.

[0094] S33, Verification of grid connection method.

[0095] Read the unit combinations from the remaining candidate load allocation schemes, and compare each unit in the combination with the current grid connection method. If any unit in the combination does not meet the current grid connection operation requirements, delete the scheme; if all units meet the requirements, retain the scheme.

[0096] S34, Check and maintain the maintenance status.

[0097] Read the unit combinations from the remaining candidate load allocation schemes, and compare each unit in the combination with its current maintenance status. If there are units in the combination that are under maintenance or not allowed to generate electricity, delete the scheme; if none of the units in the combination have triggered the maintenance removal conditions, retain the scheme.

[0098] S35. Check the phase adjustment status of the unit.

[0099] Read the phase shift status of the generating units. For the phase shift status of Unit 2, if the candidate load allocation scheme requires Unit 2 to undertake power generation, but Unit 2 is in phase shift status and cannot simultaneously undertake power generation, then the scheme is deleted; if Unit 2 is not in this restricted state, or is allowed to withdraw from phase shift to participate in power generation, the scheme is not deleted due to the phase shift status. The phase shift status of other units is handled according to the same constraint criteria.

[0100] S36. Compile a set of real-time available solutions.

[0101] Schemes that remain after steps S32 to S35 are written into the real-time available scheme set. Each record in the real-time available scheme set retains the unit combination, the output of each unit, the water consumption for power generation, and the status verified by real-time constraints. This set is sent to S4. If the real-time available scheme set is empty, an abnormal status of "no real-time available scheme" is output, no load allocation result is output, and the process returns to S1 to retrieve operating data again.

[0102] S4. Calculate the total water consumption flow and determine the real-time load distribution scheme.

[0103] S4 receives the set of available real-time solutions and simultaneously reads the current combination of generating units, their current start / stop status, the current output of each unit, the output range corresponding to the vibration zone, the single start / stop penalty flow, and the single cross-flow penalty flow. S4 calculates the comprehensive water consumption flow for each available real-time solution and then compares the values.

[0104] S41. Read the basic data of the scheme.

[0105] For each candidate load allocation scheme in the real-time available scheme set, the unit combination, the output of each unit, and the water consumption for power generation are read. The water consumption for power generation serves as the base value for the overall water consumption.

[0106] S42. Determine the number of start-stop changes.

[0107] Read the current generating unit combination and the generating unit combination corresponding to the candidate load allocation scheme. Compare the participation status of the two groups of units one by one according to their unit numbers. A unit currently participating in power generation but not participating in power generation under the candidate scheme is recorded as a shutdown change; a unit not currently participating in power generation but participating in power generation under the candidate scheme is recorded as an start-up change. Add the number of start-up changes to the number of shutdown changes to obtain the number of start-up / shutdown changes.

[0108] If the current participation status of a certain unit is missing, the candidate load allocation scheme including that unit will not be included in the comprehensive water consumption flow comparison, so as to avoid outputting a scheme when the number of start-up and shutdown changes is uncertain.

[0109] S43. Determine the number of times the vibration zone is crossed.

[0110] Read the current output of each unit, the unit output in the candidate load allocation scheme, and the output range corresponding to the vibration zone. For the same unit, determine the change path from the current output to the candidate output. When the current output is on one side of the output range corresponding to the vibration zone and the candidate output is on the other side, it is recorded as a vibration zone crossing. When the current output or the candidate output falls into the output range corresponding to the vibration zone, the candidate load allocation scheme is directly deleted, and the comprehensive water consumption flow is no longer calculated.

[0111] If a unit lacks current output data, it cannot be determined whether it has crossed the vibration zone. The candidate load allocation scheme will not be included in the comparison and will be recorded as missing current output.

[0112] S44. Calculate the running penalty flow.

[0113] Read the single start / stop penalty flow and the single crossing penalty flow. Add the single start / stop penalty flow according to the number of start / stop changes to obtain the start / stop penalty flow; add the single crossing penalty flow according to the number of times the vibration zone is crossed to obtain the crossing penalty flow; combine the start / stop penalty flow and the crossing penalty flow to obtain the running penalty flow.

[0114] S45. Calculate the total water consumption flow.

[0115] The power generation water consumption and the operation penalty flow of the candidate load allocation scheme are added together to obtain the comprehensive water consumption of the scheme. After the calculation is completed, the unit combination, the output of each unit, the power generation water consumption, the number of start-up and shutdown changes, the number of vibration zone crossings, the operation penalty flow, and the comprehensive water consumption are recorded in the scheme comparison record.

[0116] S46, Output real-time load distribution scheme.

[0117] All schemes are compared based on their total water consumption, and the scheme with the lowest total water consumption is selected as the real-time load allocation scheme. The output includes the combination of generating units, the output of each unit, and the corresponding total water consumption. If all schemes are deleted during vibration zone or data integrity checks, an abnormal status "No comparable schemes" is output, and the system returns to S1 to retrieve operating data again.

[0118] S5. After the operating status changes, it re-enters S1.

[0119] The system monitors the total load, unit availability, start-up / shutdown status, maintenance status, and unit phase adjustment status of the power plant. If any data changes, the current real-time load allocation scheme is no longer the sole basis for the current output. The changed data is returned to S1 as a new input, re-establishing operating constraints, re-determining feasible unit combinations, and continuing execution from S2 to S4. If the head changes, the feasible output range and water consumption characteristics of the units are re-evaluated, and the static load allocation table is updated if necessary.

[0120] Please see Figure 1 and Figure 5 This embodiment provides a power plant load intelligent allocation system under multi-unit collaborative constraints, including a constraint module, a static table construction module, a scheme screening module, and an optimization control module.

[0121] The constraint module receives the power plant's total load, spinning reserve demand, unit availability status, grid connection mode, head, start-up / shutdown status, maintenance status, and unit phasing status. The constraint module first generates unit status records by unit number, then checks the availability, start-up / shutdown, and maintenance status of each unit. Units meeting the conditions for participating in power generation are added to the candidate unit set; units not meeting the conditions are added to the rejection record. Subsequently, the constraint module continues to filter the candidate unit set according to grid connection mode and unit phasing status, outputting feasible unit combinations. These feasible unit combinations are sent to the static table construction module and the scheme screening module, respectively.

[0122] The static table construction module receives feasible unit combinations, head, and total power plant load. First, it searches for the feasible output range of a unit by head and unit number; if no matching range is found, it deletes the combination containing that unit. Then, it generates candidate unit output points within the feasible output range, combines the output of each unit, and filters the load-matching output combinations according to the total power plant load. The static table construction module reads the output interval corresponding to the vibration zone, deletes output combinations falling into the vibration zone, and then searches for the unit's water consumption characteristics by unit number, unit output, and head to obtain the water consumption flow of a single unit, summarizing them to form the power generation water consumption flow. Finally, the static table construction module outputs a static load allocation table and sends it to the scheme filtering module.

[0123] The scheme selection module receives the static load allocation table, current head, total power plant load, feasible unit combinations, spinning reserve requirements, grid connection mode, maintenance status, and unit phasing status. The module first reads candidate load allocation schemes from the static load allocation table based on head, total power plant load, and feasible unit combinations. If the read fails, it outputs a status indicating no candidate load allocation schemes and returns an update request to the static table construction module. Upon successful read, the module sequentially verifies the spinning reserve requirements, grid connection mode, maintenance status, and unit phasing status. Schemes that fail verification are deleted, and verified schemes are added to the real-time available scheme set. The real-time available scheme set is then sent to the optimization control module.

[0124] The optimization control module receives a set of real-time available solutions and simultaneously reads the current generating unit combination, current start / stop status, current output of each unit, output range corresponding to the vibration zone, single start / stop penalty flow, and single cross-vibration penalty flow. For each candidate load allocation scheme, the optimization control module reads the power generation water consumption flow and compares the current generating unit combination with the candidate scheme's unit combination to obtain the number of start / stop changes. Subsequently, the optimization control module compares the current output, the candidate output, and the output range corresponding to the vibration zone to obtain the number of vibration zone cross-vibration times; if the current output or the candidate output falls into the output range corresponding to the vibration zone, the scheme is deleted. Schemes that are not deleted are entered into the operation penalty flow calculation; the power generation water consumption flow and the operation penalty flow are added to form the comprehensive water consumption flow. The optimization control module compares the comprehensive water consumption flow of each scheme and outputs the real-time load allocation scheme.

[0125] When the total load of the power plant, the availability status of the units, their start-up / shutdown status, their maintenance status, or their phase adjustment status changes, the optimization control module returns a recalculated trigger status to the constraint module. The constraint module receives this trigger status and then rereads the operating data. If the scheme selection module outputs an empty set of real-time available schemes, or if the optimization control module deletes all schemes in the vibration zone judgment, current output reading, or start-up / shutdown status reading, the system will not output load allocation schemes that do not meet the conditions. Instead, it will return to the constraint module to reorganize feasible unit combinations or wait for updated operating data.

[0126] Example

[0127] For example, a hydropower station is equipped with four vertical turbine generator units, designated as Unit 1, Unit 2, Unit 3, and Unit 4, each with a rated capacity of 25MW, for a total installed capacity of 100MW. In this round of real-time calculations, the total load of the power station is 68MW, the spinning reserve requirement is 4MW, the current head is 128.5m, the power station operates via a 220kV I busbar, the preset output interval is 1MW, the single start-stop penalty flow is 0.45m³ / s, and the single vibration zone crossing penalty flow is 0.25m³ / s.

[0128] In step S1, the system reads the current operating data. The total load of the power station is 68MW, the spinning reserve requirement is 4MW, the head is 128.5m, and the current grid connection mode is 220kV I bus connection. Unit 1 is currently generating electricity and its current output is 25MW. Unit 2 is in phase adjustment mode and its current active power output is 0MW. Unit 3 is currently generating electricity and its current output is 23MW. Unit 4 is under maintenance and its current output is 0MW.

[0129] The unit status record generated in this round is as follows:

[0130] Unit number Unit availability status Start-stop status Maintenance status Phase modulation state Does it meet the requirements for 220kV I busbar grid connection? Current output Unit 1 Available Allowing power generation Uninspected Unadjusted phase satisfy 25MW Unit 2 Available Allowing power generation Uninspected During phasing, it is permissible to withdraw from phasing and participate in power generation. satisfy 0MW Unit 3 Available Allowing power generation Uninspected Unadjusted phase satisfy 23MW Unit 4 Unavailable Power generation is not allowed Planned maintenance Unadjusted phase satisfy 0MW

[0131] After screening for participation in power generation conditions, Units 1, 2, and 3 met the conditions for participating in power generation and were added to the candidate unit set; Unit 4 was added to the rejection record because "the unit is unavailable, power generation is not allowed in the start-up / shutdown state, and the maintenance status is planned maintenance". At this time, the candidate unit set is {Unit 1, Unit 2, Unit 3}.

[0132] Under the 220kV I busbar grid connection mode, Unit 1, Unit 2 and Unit 3 all meet the grid connection conditions of this busbar and will continue to be retained; the set of candidate units after grid connection screening is still {Unit 1, Unit 2, Unit 3}.

[0133] Unit 2 was in phase-shifting mode during this round, but dispatch permitted it to exit phase-shifting mode and participate in power generation; therefore, Unit 2 was not deleted due to its phase-shifting mode. Units 1 and 3 were not in phase-shifting mode and were retained. The resulting feasible unit combinations include {Unit 1}, {Unit 2}, {Unit 3}, {Unit 1, Unit 2}, {Unit 1, Unit 3}, {Unit 2, Unit 3}, and {Unit 1, Unit 2, Unit 3}.

[0134] The feasible unit combination has been formed in this round S1, so we continue to enter S2; among them, since the total load of the power plant is 68MW, the maximum feasible output of a single unit or any two units is insufficient to meet the target load of 68MW and the spinning reserve requirements. In the subsequent static table construction, the load matching is mainly carried out by {Unit 1, Unit 2, Unit 3}.

[0135] In step S2, under the current head of 128.5m, the feasible output range, vibration zone and typical water consumption characteristics of Unit 1, Unit 2 and Unit 3 are given by the power station test curves and operation database.

[0136] At a head of 128.5m, the minimum feasible output of Unit 1 is 14MW and the maximum feasible output is 26MW; the minimum feasible output of Unit 2 is 15MW and the maximum feasible output is 25MW; and the minimum feasible output of Unit 3 is 13MW and the maximum feasible output is 24MW. Unit 4 is not included in this round of feasible output range queries because it is under maintenance.

[0137] The preset output interval for this round is 1MW. Therefore, Unit 1 generates 13 candidate points ranging from 14MW, 15MW, 16MW to 26MW; Unit 2 generates 11 candidate points ranging from 15MW, 16MW, 17MW to 25MW; and Unit 3 generates 12 candidate points ranging from 13MW, 14MW, 15MW to 24MW. For the combination of {Unit 1, Unit 2, Unit 3}, the original number of discrete combinations is 13 × 11 × 12 = 1716.

[0138] With a total power plant load of 68MW as the matching target, 1716 original discrete combinations were summed and filtered to obtain 36 load matching output combinations that satisfy P1+P2+P3=68MW, such as: {Unit 1 21MW, Unit 2 23MW, Unit 3 24MW}, {Unit 1 22MW, Unit 2 22MW, Unit 3 24MW}, {Unit 1 23MW, Unit 2 21MW, Unit 3 24MW}, {Unit 1 24MW, Unit 2 21MW, Unit 3 23MW}, {Unit 1 26MW, Unit 2 20MW, Unit 3 22MW}.

[0139] At the current head of 128.5m, the vibration zone for Unit 1 is 18MW to 20MW, for Unit 2 it is 17MW to 19MW, and for Unit 3 it is 15MW to 17MW. Of the 36 load matching output combinations, those containing 18MW, 19MW, or 20MW from Unit 1, or 17MW, 18MW, or 19MW from Unit 2, or 15MW, 16MW, or 17MW from Unit 3, are deleted. After eliminating the vibration zones, 29 load matching output combinations are retained for water consumption flow calculation.

[0140] At a head of 128.5m, the single-unit water consumption and power generation water consumption of some retained combinations are as follows:

[0141] Static scheme number Unit 1 output Unit 2 output Unit 3 output Unit 1 water consumption Water consumption of Unit 2 Unit 3 water consumption Total water consumption for power generation A1 23MW 21MW 24MW 20.156 m³ / s 18.501 m³ / s 21.080 m³ / s 59.737 m³ / s A2 23MW 22MW 23MW 20.156 m³ / s 19.366 m³ / s 20.216 m³ / s 59.738 m³ / s A3 22MW 22MW 24MW 19.296 m³ / s 19.366 m³ / s 21.080 m³ / s 59.742 m³ / s A4 24MW 21MW 23MW 21.040 m³ / s 18.501 m³ / s 20.216 m³ / s 59.757 m³ / s A5 21MW 23MW 24MW 18.460 m³ / s 20.257 m³ / s 21.080 m³ / s 59.797 m³ / s A6 26MW 20MW 22MW 22.880 m³ / s 17.662 m³ / s 19.374 m³ / s 59.916 m³ / s

[0142] The above water consumption data matches the target load of 68MW, and the output of each unit does not fall into the corresponding vibration zone. For example, in Scheme A1, Unit 1 (23MW), Unit 2 (21MW), and Unit 3 (24MW) are all in the stable operating zone, and the sum of the water consumption of the three units is 59.737m³ / s. Therefore, Scheme A1 is included in the static load allocation table.

[0143] In this round S2, a total of 29 static load allocation table records were generated, so the abnormal state of "no static matching scheme" was not triggered, and the process continued into S3.

[0144] In step S3, the inputs for this round S3 are: head 128.5m, total power plant load 68MW, feasible unit combination includes {unit 1, unit 2, unit 3}, corresponding records in the static load allocation table are 29, spinning reserve requirement is 4MW, the current grid connection mode is 220kV I bus connection, unit 4 is under maintenance, unit 2 is phase-tuned but is allowed to withdraw from phase-tuning to participate in power generation.

[0145] This round of searching was performed according to "head 128.5m, total power plant load 68MW, unit combination {unit 1, unit 2, unit 3}", and 29 candidate load allocation schemes were retrieved. Therefore, the abnormal state of "no candidate load allocation scheme" was not triggered.

[0146] The maximum feasible output of Unit 1 in this round is 26MW, Unit 2 is 25MW, and Unit 3 is 24MW, for a combined maximum feasible output of 75MW. For any candidate scheme that meets the target load of 68MW, its spinning reserve capacity is 75MW-68MW=7MW, which is 4MW greater than the current spinning reserve requirement. Therefore, all 29 candidate load allocation schemes have passed the spinning reserve check.

[0147] All 29 candidate load allocation schemes consist of Unit 1, Unit 2, and Unit 3. All three units meet the grid connection requirements of the 220kV I busbar. Therefore, all 29 schemes have been verified through grid connection.

[0148] None of the 29 candidate load allocation schemes include Unit 4 which is under maintenance, and Units 1, 2 and 3 are all in an unmaintained state. Therefore, all 29 schemes have passed the maintenance status verification.

[0149] All 29 candidate load allocation schemes in this round require Unit 2 to generate 20MW to 25MW of power. Although Unit 2 is in phase-shifting mode, dispatch has permitted it to withdraw from phase-shifting mode and participate in power generation; therefore, the scheme will not be deleted due to Unit 2's phase-shifting status. Units 1 and 3 are not in phase-shifting mode and do not trigger the phase-shifting rejection conditions.

[0150] After verification of the spin-on standby, grid connection mode, maintenance status, and phase adjustment status, all 29 schemes were retained, forming a set of real-time available schemes. Each record in the set of real-time available schemes is marked as "Spin-on standby passed, grid connection verification passed, maintenance verification passed, phase adjustment verification passed".

[0151] In step S4, the current generating unit combination participating in power generation is {Unit 1, Unit 3}, where Unit 1 currently outputs 25MW and Unit 3 currently outputs 23MW; Unit 2 is currently in phase adjustment mode and has an active power output of 0MW. If it switches to the alternative power generation scheme, it will be counted as one start-up change in the start-up / shutdown change count. The penalty flow rate for a single start-up / shutdown is 0.45m³ / s, and the penalty flow rate for a single crossover is 0.25m³ / s.

[0152] For example, the data read for scheme A1 is: unit combination {unit 1, unit 2, unit 3}, unit 1 23MW, unit 2 21MW, unit 3 24MW, power generation water consumption flow rate 59.737m³ / s; the data read for scheme A2 is: unit 1 23MW, unit 2 22MW, unit 3 23MW, power generation water consumption flow rate 59.738m³ / s.

[0153] The generating units currently participating in power generation in this round are Unit 1 and Unit 3. All alternative schemes require Unit 1, Unit 2, and Unit 3 to participate in power generation. Therefore, the participation status of Unit 1 and Unit 3 remains unchanged. Unit 2 changes from not generating power to generating power, which is recorded as one start-up change. There are no shutdown changes, and the number of start-up / shutdown changes is 1.

[0154] The vibration zones for this round remain the same: Unit 1 from 18MW to 20MW, Unit 2 from 17MW to 19MW, and Unit 3 from 15MW to 17MW. Taking Scheme A1 as an example, Unit 1's adjustment from 25MW to 23MW follows a path that remains above Unit 1's vibration zone, so vibration zone crossing is not counted. Unit 2's adjustment from 0MW to 21MW follows a path that moves from below Unit 2's vibration zone to above it, crossing the 17MW to 19MW vibration zone, which is counted as one vibration zone crossing. Unit 3's adjustment from 23MW to 24MW follows a path that remains above Unit 3's vibration zone, so vibration zone crossing is not counted. Therefore, Scheme A1 has one vibration zone crossing. For the 29 schemes retained in this round, none of the candidate outputs fall into the vibration zone, and the current outputs of 25MW, 0MW, and 23MW also do not fall into their corresponding vibration zones. Therefore, schemes will not be deleted simply because the current or candidate outputs fall into the vibration zone.

[0155] The penalty flow rate for a single start-stop operation in this round is 0.45 m³ / s, and the penalty flow rate for a single crossing is 0.25 m³ / s. Taking Scheme A1 as an example, the number of start-stop changes is 1, with a start-stop penalty flow rate of 0.45 m³ / s; the number of crossings of the vibration zone is 1, with a crossing penalty flow rate of 0.25 m³ / s; the total operational penalty flow rate is 0.45 m³ / s + 0.25 m³ / s = 0.70 m³ / s.

[0156] The comparison of some of the proposed solutions is recorded below:

[0157] Scheme Number Unit Combination and Output Water consumption for power generation Number of start-stop changes Number of vibration zone crossings Run penalty traffic Comprehensive water consumption flow A1 Unit 1: 23MW, Unit 2: 21MW, Unit 3: 24MW 59.737 m³ / s 1 time 1 time 0.700m³ / s 60.437 m³ / s A2 Unit 1: 23MW, Unit 2: 22MW, Unit 3: 23MW 59.738 m³ / s 1 time 1 time 0.700m³ / s 60.438 m³ / s A3 Unit 1: 22MW, Unit 2: 22MW, Unit 3: 24MW 59.742 m³ / s 1 time 1 time 0.700m³ / s 60.442 m³ / s A4 Unit 1: 24MW, Unit 2: 21MW, Unit 3: 23MW 59.757 m³ / s 1 time 1 time 0.700m³ / s 60.457 m³ / s A5 Unit 1: 21MW, Unit 2: 23MW, Unit 3: 24MW 59.797 m³ / s 1 time 1 time 0.700m³ / s 60.497 m³ / s A6 Unit 1: 26MW, Unit 2: 20MW, Unit 3: 22MW 59.916 m³ / s 1 time 1 time 0.700m³ / s 60.616 m³ / s

[0158] Specifically, the comprehensive water consumption flow rate for Scheme A1 is 59.737 m³ / s + 0.700 m³ / s = 60.437 m³ / s; the comprehensive water consumption flow rate for Scheme A2 is 59.738 m³ / s + 0.700 m³ / s = 60.438 m³ / s. Since both schemes have the same number of start-stop cycles and vibration zone crossings, the difference in comprehensive water consumption flow rate is mainly determined by the water consumption flow rate for power generation.

[0159] All 29 schemes in this round have completed the comprehensive water consumption flow calculation, and no schemes were completely deleted. After comparison, Scheme A1 has the lowest comprehensive water consumption flow at 60.437 m³ / s. Therefore, the output real-time load allocation scheme is as follows: Participating generator unit combination {Unit 1, Unit 2, Unit 3}, with Unit 1 allocated 23MW, Unit 2 allocated 21MW, and Unit 3 allocated 24MW, for a total output of 68MW, corresponding to a comprehensive water consumption flow of 60.437 m³ / s, and a spinning reserve capacity of 7MW, meeting the 4MW spinning reserve requirement.

[0160] In step S5, when the total load of the power plant changes from 68MW to 72MW, or the head changes from 128.5m to 126.0m, or Unit 4 is restored to service after maintenance, or the phase modulation status of Unit 2 changes to "not allowed to withdraw from phase modulation to participate in power generation", the system will rewrite the current data into the data entry of S1 and re-select the unit combination. If the head changes to 126.0m, the feasible output range of Unit 1 can be updated to 13MW to 25MW, the feasible output range of Unit 2 can be updated to 14MW to 24MW, and the feasible output range of Unit 3 can be updated to 12MW to 23MW. The system will then reread the water consumption characteristic curve under this head and update the static load allocation table.

[0161] The above indicates that, under the operating scenario of a water head of 128.5m, a total power plant load of 68MW, and a spinning reserve requirement of 4MW, this scheme can select a real-time available scheme from the static load allocation table by combining the unit availability status, maintenance status, grid connection mode, phase adjustment status, vibration zone, start-up and shutdown changes, and water consumption characteristics. Furthermore, by comparing the comprehensive water consumption flow, scheme A1 is determined to be the optimal real-time load allocation scheme. This scheme enables units 1, 2, and 3 to handle 23MW, 21MW, and 24MW of output respectively, satisfying both the total load and spinning reserve requirements while achieving the minimum comprehensive water consumption flow of 60.437 m³ / s among the comparable schemes in this round.

[0162] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for intelligent load allocation in a power plant under multi-unit collaborative constraints, used for load allocation within the plant under the condition of joint operation of multiple generating units in a hydropower station, characterized in that, Includes the following steps: Step S1: Obtain the total load of the power plant, spinning reserve demand, unit availability status, grid connection mode, head, start-up and shutdown status, maintenance status and unit phase adjustment status to form operating constraints, and determine feasible unit combinations based on the operating constraints. Step S2: Based on the feasible unit combination, the feasible output of the units is traversed according to the head and the total load of the power station. The output range corresponding to the vibration zone is excluded within the feasible output range of each unit. Then, the static load allocation table is determined in combination with the water consumption characteristics of the units. The static load allocation table records the unit combination, unit output and power generation water consumption flow. Step S3: Extract candidate load allocation schemes from the static load allocation table according to water head, total power plant load and feasible unit combination, and eliminate candidate load allocation schemes that do not meet the spinning reserve requirements, grid connection mode, maintenance status and unit phase adjustment status, and count the set of real-time available schemes. Step S4: Calculate the power generation water consumption flow of each candidate load allocation scheme in the real-time available scheme set, determine the operation penalty flow based on start-stop changes and vibration zone crossing times, determine the comprehensive water consumption flow based on the power generation water consumption flow and the operation penalty flow, and determine the real-time load allocation scheme based on the comprehensive water consumption flow. Step S5: When the total load of the power plant, the availability status of the generating units, the start-up and shutdown status, the maintenance status, or the phase adjustment status of the generating units changes, return to step S1 to redetermine the real-time load allocation scheme.

2. The intelligent load allocation method for power plants under multi-unit collaborative constraints according to claim 1, characterized in that, In step S1, during the determination of feasible unit combinations: The system obtains the total load of the power plant, spinning reserve requirement, unit availability status, grid connection mode, head, start-up / shutdown status, maintenance status, and unit phase adjustment status, and matches the unit availability status, start-up / shutdown status, and maintenance status with the unit number to generate the aforementioned operating constraints. Using the aforementioned operational constraints, each unit is assessed to determine whether it meets the conditions for participating in power generation. When a unit is available, allowed to generate electricity in start-stop mode, and not under maintenance, the corresponding unit is included in the candidate unit set. The candidate unit set is then filtered based on the grid connection mode and the unit's phase adjustment status to obtain the feasible unit combination. During the screening process, the first step is to determine whether each candidate unit can participate in the current grid-connected operation based on the grid connection method, and to eliminate candidate units that do not meet the current grid-connected operation requirements. When a unit is in phase-shifting mode and cannot simultaneously generate power, the unit will be removed from the set of candidate units; when a unit is not in phase-shifting mode or is allowed to withdraw from phase-shifting mode to participate in power generation, the unit will retain its candidate status.

3. The intelligent power plant load allocation method under multi-unit collaborative constraints according to claim 1, characterized in that, In step S2, the process of determining the static load allocation table includes: Obtain feasible unit combinations, head and total power plant load, and determine the feasible output range of each unit based on the head. Within the feasible output range, generate the corresponding unit output according to the preset output interval. The unit outputs of each unit within the same feasible unit combination are combined, and the output combination whose sum matches the total load of the power station is selected to obtain the load-matching output combination. The load matching output combinations are screened, and the unit outputs that fall into the output range corresponding to the vibration zone are eliminated. The power generation water consumption flow is determined in combination with the unit water consumption characteristics to obtain the static load allocation table.

4. The intelligent load allocation method for power plants under multi-unit collaborative constraints according to claim 3, characterized in that, The process of determining the water consumption flow for power generation also includes: Obtain the water consumption characteristics of each unit under the current head, and find the corresponding single unit water consumption flow rate according to the unit number and unit output combined with the unit water consumption characteristics; When the unit output is located between two adjacent known output points, the single unit water consumption flow is determined according to the water consumption flow corresponding to the two adjacent known output points, and the single unit water consumption flow of each unit in the same load matching output combination is summarized to obtain the power generation water consumption flow.

5. The intelligent power plant load allocation method under multi-unit collaborative constraints according to claim 1, characterized in that, In step S3, during the statistical analysis of the set of available real-time solutions: Using water head, total power plant load, and the feasible unit combination as search criteria, select candidate load allocation schemes with consistent unit combinations and matching load values ​​are extracted from the static load allocation table, and the unit output and power generation water consumption flow corresponding to each candidate load allocation scheme are read. The candidate load allocation schemes are compared one by one with the spinning reserve requirement, grid connection mode, maintenance status and phase adjustment status of Unit 2. Candidate load allocation schemes with unmet constraints are deleted, and the remaining candidate load allocation schemes are sorted to obtain the set of real-time available schemes.

6. The intelligent load allocation method for power plants under multi-unit collaborative constraints according to claim 1, characterized in that, In step S4, during the determination of the real-time load allocation scheme: Read the unit output and power generation water consumption flow corresponding to each candidate load allocation scheme in the real-time available scheme set, and determine the number of start-stop changes based on the difference between the current start-stop status and the unit combination corresponding to the candidate load allocation scheme; Based on whether the unit output crosses the output range corresponding to the vibration zone, the number of vibration zone crossings is recorded. The operating penalty flow is determined based on the number of start-stop changes and the number of vibration zone crossings. The power generation water consumption flow is added to the operating penalty flow to obtain the comprehensive water consumption flow. Compare the total water consumption flow corresponding to each candidate load allocation scheme, and determine the candidate load allocation scheme with the minimum total water consumption flow as the real-time load allocation scheme.

7. The intelligent load allocation method for power plants under multi-unit collaborative constraints according to claim 6, characterized in that, During the process of determining the penalty traffic: Obtain the single start-stop penalty flow corresponding to the start-stop change and the single crossing penalty flow corresponding to the vibration zone crossing, and read the number of start-stop changes and the number of vibration zone crossings respectively; The start-stop penalty flow is obtained by summing the single start-stop penalty flow with the corresponding number of start-stop changes, and the crossing penalty flow is obtained by summing the single crossing penalty flow with the corresponding number of crossings of the vibration zone. The start-stop penalty flow and the crossing penalty flow are then combined to obtain the operation penalty flow.

8. The intelligent power plant load allocation method under multi-unit collaborative constraints according to claim 7, characterized in that, The process for determining start / stop changes includes: Obtain the current generating unit combination and the generating unit combination corresponding to the candidate load allocation scheme, and compare the participating status of the units in the two sets of generating unit combinations one by one according to the unit number; When the same unit is in a state of participating in power generation in the current unit combination, but is in a state of not participating in power generation in the unit combination corresponding to the candidate load allocation scheme, it is recorded as a shutdown change; When the same unit is not participating in power generation in the current unit combination, but is participating in power generation in the unit combination corresponding to the candidate load allocation scheme, it is recorded as a start-up change. The total number of start-stop changes was obtained by combining the results.

9. The intelligent power plant load allocation method under multi-unit collaborative constraints according to claim 7, characterized in that, In determining the number of vibration zone crossings: The current output of each unit, the unit output corresponding to the selected load allocation scheme, and the output range corresponding to the vibration zone are obtained. The output change is judged according to the current output of the same unit to the direction of the selected output. If the current output of the same unit is located on one side of the output range corresponding to the vibration zone and the selected output is located on the other side of the output range corresponding to the vibration zone, the corresponding unit is recorded as one vibration zone crossing. If the current output or the selected output is located within the output range corresponding to the vibration zone, the corresponding selected load allocation scheme is removed to obtain the number of vibration zone crossings.

10. A power plant load intelligent allocation system under multi-unit collaborative constraints, characterized in that, The system is used to implement the method according to any one of claims 1-9, comprising: The constraint module is used to obtain the total load of the power plant, spinning reserve demand, unit availability status, grid connection mode, head, start-up and shutdown status, maintenance status and unit phase adjustment status, form operating constraints, and determine feasible unit combinations based on the operating constraints. The static table construction module, connected to the constraint module, is used to traverse the feasible output of the units according to the head and the total load of the power station based on the feasible unit combination, and exclude the output range corresponding to the vibration zone within the feasible output range of each unit. Then, combined with the water consumption characteristics of the units, the static load allocation table is determined. The static load allocation table records the unit combination, unit output and power generation water consumption flow. The scheme selection module, connected to the static table construction module and the constraint module, is used to extract candidate load allocation schemes from the static load allocation table based on head, total power plant load and feasible unit combinations, eliminate candidate load allocation schemes that do not meet the spinning reserve requirements, grid connection mode, maintenance status and unit phase adjustment status, and compile a set of real-time available schemes. The optimization control module, connected to the scheme selection module, is used to calculate the power generation water consumption flow of each candidate load allocation scheme in the real-time available scheme set. It determines the operation penalty flow based on start-up and shutdown changes and the number of vibration zone crossings. It determines the comprehensive water consumption flow based on the power generation water consumption flow and the operation penalty flow, and determines the real-time load allocation scheme based on the comprehensive water consumption flow. When the total load of the power plant, the availability status of the units, the start-up and shutdown status, the maintenance status, or the phase adjustment status of the units changes, it returns to step S1 to redetermine the real-time load allocation scheme.

Images