Water storage combined agc phased response method and system

Abstract

The application discloses a water storage combined AGC phased response method and system, relates to the technical field of power system automation, and aims at the technical problems that the existing water storage combined AGC control strategy is not designed in stages according to the response process time scale, the state of charge of energy storage lacks an active recovery mechanism, and the vibration area of a unit lacks systematic avoidance means. A NSGA-II multi-objective genetic algorithm is used to off-line optimize a standardized phased control parameter set, three-phase collaborative response including fast response, energy maintenance and accurate adjustment is realized, the power division optimization of water and energy storage in different time scales is realized, vibration area constraint correction and energy storage state of charge active recovery operation are synchronously coupled, and the technical effects of AGC comprehensive regulation performance improvement, energy storage continuous availability enhancement and unit vibration area residence time reduction are realized.

Description

Technical Field

[0001] This application relates to the field of power system automation technology, and in particular to a phased response method and system for a water-storage combined AGC. Background Technology

[0002] With the large-scale integration of new energy sources into the grid, the power output fluctuations of intermittent power sources such as wind and solar power are causing increasingly frequent grid frequency disturbances, significantly increasing the regulation demands on automatic generation control (AGC). Hydropower units, due to their rapid start-up and shutdown and wide regulation range, have become one of the main power sources for AGC frequency regulation. To compensate for the poor response performance, frequent crossing of vibration zones, and severe mechanical wear of hydropower units, energy storage batteries are typically configured to form a combined hydropower-storage AGC system. Existing combined hydropower-storage AGC control strategies mainly adopt a fixed-ratio allocation method, that is, allocating AGC power commands to hydropower units and energy storage batteries according to a preset ratio; or a hydropower priority deficit compensation method, that is, hydropower units take priority in undertaking regulation tasks, and energy storage batteries only compensate for the deficit between hydropower output and commands.

[0003] However, the existing strategies described above have the following technical problems: From receiving AGC commands to actual output adjustment, the hydropower unit needs to go through multiple series inertial stages, including the governor control loop, guide vane actuator, and water flow inertia, resulting in a response lag of several seconds. Furthermore, neither the fixed-ratio allocation nor the deficit compensation strategy addresses the phased design of power division between hydropower and energy storage based on the dynamic characteristics of different time scales in the response process. This prevents the rapid response capability of the energy storage battery and the continuous adjustment capability of the hydropower unit from being fully utilized at different stages, thus limiting the overall AGC adjustment performance. Simultaneously, the existing solutions do not adequately address the energy storage's charge status... The current system only sets upper and lower limit protections and lacks an active recovery mechanism. After frequent charging and discharging during continuous AGC events, the state of charge of the energy storage battery is prone to deviate from the median range for a long time. When it approaches the limit, it is forced to operate with limited power or even exit the response, which reduces the continuous frequency regulation capability of the joint system. In addition, there is a vibration zone in the specific output range of the hydropower unit. The existing solution does not incorporate the vibration zone constraint into the AGC power allocation logic and lacks systematic processing methods such as safety margin quantification, output redistribution, trajectory amplitude limiting and safety side locking. The target output path of the unit is prone to fall into or stay in the vibration zone for a long time, which aggravates the fatigue damage of the equipment. Summary of the Invention

[0004] The main objective of this application is to provide a phased response method and system for water-storage combined AGC, in order to solve the problems of limited comprehensive regulation performance, weak continuous response capability of energy storage SOC, and easy entry into or prolonged stay in the vibration zone in the existing water-storage combined AGC control strategy, which exacerbates equipment fatigue damage.

[0005] To achieve the above objectives, this application provides the following technical solution:

[0006] A phased response method for combined water storage and AGC, the method comprising:

[0007] Step S1: Obtain real-time operation data of the water-storage combined system and historical operation data of the unit, and input them into the water-power-storage coupling model for parameter identification. Input the parameter-identified water-power-storage coupling model into the NSGA-II multi-objective genetic algorithm for solution to obtain a standardized staged control parameter set.

[0008] Step S2: Perform instruction recognition on the standardized phased control parameter set and perform threshold determination calculation on the real-time AGC power instruction to obtain a complete and valid AGC event feature sequence.

[0009] Step S3: Perform a three-stage coordinated response on the effective AGC event feature sequence and the standardized phased control parameter set, dividing the response into a rapid response stage, an energy maintenance stage, and a precise adjustment stage according to the time sequence, to obtain the hydropower storage stratified power response benchmark value.

[0010] Step S4: Perform vibration zone constraint correction on the reference value of the stratified power response of the hydropower storage and the vibration zone constraint parameters in the standardized phased control parameter set to complete the unit output safety side locking and the crossing trajectory amplitude limit calculation, and obtain the target output of the hydropower unit after constraint correction.

[0011] Step S5: Perform energy storage power matching calculation on the target output of the hydropower unit after constraint correction and the energy storage recovery parameters in the standardized phased control parameter set, and simultaneously couple the energy storage state of charge boundary constraint and the active recovery calculation of the state of charge to obtain the real-time energy storage compensation power command.

[0012] Step S6: The real-time energy storage compensation power command is superimposed with the target output of the constrained hydropower unit to generate a joint output command for the grid connection point, and a power closed-loop tracking calculation is performed to obtain the grid connection regulation execution signal for the entire station.

[0013] Beneficial effects of steps S1 to S6:

[0014] This method addresses the technical problems of existing hydropower-storage combined AGC control strategies, such as the lack of phased design according to the response process time scale, the lack of active recovery mechanism for energy storage state of charge, and the lack of systematic avoidance measures for unit vibration zones. It adopts the NSGA-II multi-objective genetic algorithm to optimize the standardized phased control parameter set offline. Through the coordinated allocation of three stages—rapid response, energy maintenance, and precise regulation—it achieves power synergy optimization of hydropower and energy storage at different time scales. Simultaneously, it couples vibration zone constraint correction and active recovery calculation of energy storage state of charge, thereby achieving the technical effects of improving the overall regulation performance of AGC, enhancing the continuous availability of energy storage, and reducing the dwell time in the unit vibration zone.

[0015] In step S1, the parameters of the hydropower-storage coupling model are identified and solved using the NSGA-II multi-objective genetic algorithm. This outputs a standardized set of phased control parameters that takes into account AGC response performance, state of charge deviation, and residence time in the vibration zone, thus achieving systematic tuning of the control parameters. Step S2 identifies valid AGC events by performing threshold judgment calculations on real-time AGC power commands, eliminating short-term fluctuation interference and improving the accuracy of subsequent response triggering. Step S3 divides the time sequence into three stages: fast response, energy maintenance, and precise adjustment. This allows the energy storage battery to prioritize compensating for the dynamic lag of hydropower during the fast response stage and retain capacity during the energy maintenance stage. In the precise adjustment stage, residual errors are eliminated to fully leverage the complementary advantages of water and storage. Step S4, through vibration zone constraint correction and safety side locking, achieves controlled crossing and trajectory limiting when the unit's target output path crosses the vibration zone, reducing the risk of equipment fatigue damage. Step S5, through energy storage power matching calculation, synchronously couples state of charge boundary constraints and active recovery calculation, and performs priority arbitration between AGC response and SOC recovery, enabling the energy storage state of charge to return to the reference value and improving the availability of continuous frequency regulation. Step S6, through power closed-loop tracking calculation, performs closed-loop correction and constraint limiting on the joint output command, improving the power tracking accuracy at the grid connection point.

[0016] As a further improvement to this application, in step S1, real-time operating data of the combined hydropower and energy storage system and historical operating data of the units are obtained and input into the hydropower-energy storage coupling model for parameter identification. The parameter-identified hydropower-energy storage coupling model is then input into the NSGA-II multi-objective genetic algorithm for solution, resulting in a standardized set of staged control parameters, including:

[0017] Step S1.1: Collect the plant-wide AGC commands of the water storage combined system, the actual output of each operating hydropower unit, the guide vane opening, the unit's operating status, the energy storage battery power, and the energy storage battery's state of charge, and integrate them to obtain real-time operating data;

[0018] Step S1.2: Collect the equivalent time constant of the hydropower unit speed governor, the equivalent time constant of the guide vane actuator, the equivalent time constant of the water flow inertia, the boundary of the unit vibration zone, the rated power of the energy storage battery, and the rated energy of the energy storage battery, and integrate them to obtain the historical operation dataset of the unit;

[0019] Step S1.3: Input the real-time operation data and the historical operation dataset of the unit into the hydropower storage coupling model, and use the series inertial link to equivalently describe the governor, guide vane actuator and water flow inertial dynamic process of the hydropower unit to obtain the model to be identified;

[0020] Step S1.4: Use hydropower station test data or historical operation data to identify the parameters of the model to be identified and verify the accuracy of the model to be identified, so as to obtain the hydropower storage coupling model after parameter identification.

[0021] Step S1.5: Input the hydropower storage coupling model into the NSGA-II multi-objective genetic algorithm, and use the safe guide vane action rate, the exit threshold of the rapid response stage, the re-injection ratio threshold of the precise adjustment stage, and the compensation coefficient of the precise adjustment stage as optimization variables to obtain the first set of optimization variables;

[0022] Step S1.6: Using the vibration zone safety side lock-in duration, energy storage state of charge recovery gain, and energy storage state of charge recovery power maximum amplitude as optimization variables, the second set of optimization variables is obtained;

[0023] Step S1.7: Using the maximization of AGC comprehensive regulation performance index, the minimization of energy storage state of charge deviation, the minimization of the cumulative residence time in the vibration zone of the unit, and the minimization of the controlled crossing number of the vibration zone as objective functions, a multi-objective optimization model is constructed for the first set of optimization variables and the second set of optimization variables.

[0024] Step S1.8: Apply energy storage power constraints, energy storage state of charge constraints, hydropower output constraints, guide vane safety opening and closing constraints, and vibration zone crossing time constraints to the multi-objective optimization model, and solve and output the Pareto solution set.

[0025] Step S1.9: Select representative parameter groups from the Pareto solution set that preferentially meet the AGC response performance and whose energy storage state of charge deviation is less than a preset deviation threshold, and format them into a standardized phased control parameter set.

[0026] Beneficial effects of steps S1.1 to S1.9:

[0027] This series of steps addresses the technical problem of relying on manual experience to tune control parameters in combined hydropower and energy storage AGC systems, making it difficult to balance multiple performance indicators. By collecting real-time operating data and historical operating datasets of the units, a coupled hydropower and energy storage model is constructed. The NSGA-II multi-objective genetic algorithm is then used to perform multi-objective optimization with AGC comprehensive regulation performance, energy storage state of charge deviation, cumulative residence time in the vibration zone, and controlled crossing times as objective functions. This achieves systematic tuning and global optimal configuration of control parameters, enabling a balance between response performance, energy storage availability, and unit safety in the phased response strategy.

[0028] Specifically, steps S1.1 and S1.2 collect real-time operating data and historical operating datasets of the unit, respectively, to provide a complete data foundation for model construction and parameter identification; step S1.3 uses a series inertial link to equivalently describe the multi-stage dynamic process of the hydropower unit, establishing a model reflecting the coupling characteristics of hydropower and energy storage; step S1.4 uses experimental data or historical operating data to identify parameters and verify the accuracy of the model to be identified, improving the model's approximation of actual operating conditions; steps S1.5 and S1.6 divide the optimization variables into a first optimization variable group directly related to the stage response and a group related to vibration zone locking and S... The second set of optimization variables related to OC recovery is used to achieve structured management of optimization variables; Step S1.7 constructs a multi-objective optimization model with four performance indicators as objective functions, incorporating AGC response performance, energy storage availability and unit safety into a unified optimization framework; Step S1.8 ensures the engineering feasibility of the optimization results by applying constraints on energy storage power, state of charge, hydropower output, guide vane safe opening and closing and vibration zone crossing time; Step S1.9 selects representative parameter sets from the Pareto solution set through preset deviation thresholds and formats them to output a standardized phased control parameter set that can be directly used for online control.

[0029] As a further improvement of this application, step S2 involves performing instruction recognition on the standardized phased control parameter set and threshold determination calculation on the real-time AGC power instruction to obtain a fully labeled valid AGC event feature sequence, including:

[0030] Step S2.1: Extract the effective AGC instruction change threshold and minimum duration from the standardized phased control parameter set;

[0031] Step S2.2: Obtain the real-time AGC power command at the current sampling time, and calculate the change in the real-time AGC power command relative to the previous sampling time;

[0032] Step S2.3: Determine whether the change amount is greater than the change dead zone of the valid AGC instruction. If it is greater, mark it as a candidate AGC event.

[0033] Step S2.4: Determine whether the duration of the candidate AGC event is not less than the minimum duration. If the minimum duration is met, the event is identified as a valid AGC event.

[0034] Step S2.5: Record the start time, adjustment direction, command change amplitude, initial output of the hydropower unit, and initial state of charge of the energy storage battery of the effective AGC event to obtain a complete and labeled feature sequence of effective AGC events.

[0035] Beneficial effects of steps S2.1 to S2.5:

[0036] This series of steps addresses the technical problem of frequent short-term power fluctuations triggering responses in hydropower storage integrated AGC systems, leading to frequent start-stop adjustments of hydropower storage. By using a dual criterion of effective AGC command change threshold and minimum duration, real-time AGC power commands are filtered to identify genuine power adjustment events and record complete feature information. This achieves accurate identification of effective AGC events and effective filtering of short-term fluctuations, providing a reliable event triggering basis for subsequent three-stage coordinated allocation.

[0037] Specifically, step S2.1 extracts the effective AGC command change threshold and minimum duration from the standardized phased control parameter set to provide a quantitative basis for event identification; step S2.2 quantifies the amplitude characteristics of the current adjustment demand by calculating the change in real-time AGC power command relative to the previous sampling time; step S2.3 filters candidate AGC events by comparing the change with the effective AGC command change threshold, eliminating minor fluctuations with insufficient amplitude; step S2.4 further filters short-term pulse interference by determining the minimum duration, ensuring that only continuous power changes are identified as effective AGC events; and step S2.5 provides complete initial state information for the three-stage coordinated allocation by recording the event start time, adjustment direction, command change amplitude, initial output of the hydropower unit, and initial state of charge of the energy storage battery.

[0038] As a further improvement to this application, step S3 involves a three-stage collaborative allocation of the effective AGC event feature sequence and the standardized phased control parameter set, dividing the sequence into a rapid response stage, an energy maintenance stage, and a precise adjustment stage, to obtain a tiered power response benchmark value for hydropower storage, including:

[0039] Step S3.1: Extract the exit threshold of the rapid response stage, the re-investment ratio threshold of the precise adjustment stage, and the compensation coefficient of the precise adjustment stage from the standardized phased control parameter set.

[0040] Step S3.2: Determine the event start time and adjustment direction based on the effective AGC event characteristic sequence and enter the rapid response stage. Use the energy storage battery to compensate for the grid connection point power error according to the rated power and use the hydro-generator unit to adjust the power output to the target according to the allowable guide vane action rate.

[0041] Step S3.3: Determine whether the output change of each operating hydropower unit reaches the exit threshold of the rapid response stage. If it does, exit the rapid response stage and obtain the rapid response stage end signal.

[0042] Step S3.4: According to the signal indicating the end of the rapid response phase, the energy maintenance phase begins, in which the hydropower unit undertakes the main power regulation task and the energy storage battery stops the conventional power compensation.

[0043] Step S3.5: Calculate the output completion ratio of the hydropower unit and determine whether the output completion ratio of the hydropower unit is greater than the re-start ratio threshold of the precise adjustment stage. If it is greater, exit the energy maintenance stage and obtain the energy maintenance stage end signal.

[0044] Step S3.6: According to the end signal of the energy maintenance stage, enter the precise adjustment stage, and eliminate the residual power error at the grid connection point by means of the energy storage battery according to the compensation coefficient of the precise adjustment stage;

[0045] Step S3.7: Integrate the target output sequence of hydropower units and the power sequence of energy storage batteries in the rapid response stage, energy maintenance stage, and precise regulation stage to obtain the stratified power response benchmark value of hydropower energy storage.

[0046] Beneficial effects of steps S3.1 to S3.7:

[0047] This series of steps addresses the technical problems of existing hydropower-storage combined AGC strategies not being designed in stages according to the response process time scale and the fixed power division between hydropower and storage leading to the inability to fully leverage complementary advantages. By dividing the time sequence into three stages—prioritizing energy storage compensation in the rapid response stage, reserving capacity of hydropower main regulation and energy storage in the energy maintenance stage, and eliminating residual errors by proportion of energy storage in the precise regulation stage—this achieves optimized power division between hydropower and energy storage in different response stages, improves the overall regulation performance of AGC, and rationally allocates energy storage capacity consumption.

[0048] The process involves several steps: Step S3.1 extracts three-stage control parameters from a standardized set of phased control parameters, providing a quantitative basis for phase division and switching; Step S3.2, in the rapid response phase, prioritizes compensation for grid connection point power errors using energy storage batteries and adjusts the hydropower units according to the allowable guide vane movement rate, utilizing the rapid response capability of energy storage to compensate for hydropower dynamic lag; Step S3.3, by determining whether the hydropower unit output change reaches the rapid response phase exit threshold, timely exit from the rapid response phase and phase switching are achieved; Step S3.4, in the energy maintenance phase, hydropower units undertake the main power regulation task, and energy storage batteries cease routine power compensation, reserving energy storage capacity for subsequent precise regulation; Step S3.5, by comparing the hydropower unit output completion ratio with the re-injection ratio threshold, accurate switching from the energy maintenance phase to the precise regulation phase is achieved; Step S3.6, in the precise regulation phase, energy storage batteries eliminate residual grid connection point power errors according to the compensation coefficient, improving AGC power tracking accuracy; and Step S3.7, by integrating the three-stage hydropower unit target output sequence and the energy storage battery power sequence, a complete layered power response benchmark value is output for subsequent vibration zone constraint correction.

[0049] As a further improvement to this application, step S4 involves performing vibration zone constraint correction on the stratified power response benchmark value of the hydropower storage and the vibration zone constraint parameters in the standardized phased control parameter set, in order to complete the unit output safety side locking and the crossing trajectory amplitude limiting calculation, and obtain the target output of the hydropower unit after constraint correction, including:

[0050] Step S4.1: Extract the vibration zone boundary, vibration zone safety margin, and vibration zone safety side locking duration from the standardized phased control parameter set;

[0051] Step S4.2: Calculate the lower safety boundary and upper safety boundary of the vibration zone for each operational hydropower unit based on the vibration zone boundary and vibration zone safety margin.

[0052] Step S4.3: Compare the reference value of the stratified power response of the hydropower storage with the lower safe side boundary and the upper safe side boundary of the vibration zone of each operational hydropower unit, and determine whether the target output of the hydropower unit falls into the vibration zone.

[0053] Step S4.4: If the target output of the hydropower unit does not fall into the vibration zone, then execute according to the original target output to obtain the first constraint correction result;

[0054] Step S4.5: If the target output of the hydropower unit falls into the vibration zone, the target output of the unit is redistributed between the lower safety side and the upper safety side of the vibration zone, and the scheme with the smallest deviation from the average distribution and fewer safety side switching times is selected to obtain the second constraint correction result.

[0055] Step S4.6: If the hydropower unit needs to cross from one side of the vibration zone to the other safe side, a constrained climbing trajectory is generated according to the safe operating rate of the guide vanes. An amplitude limiting calculation is performed on the climbing trajectory to obtain the third constraint correction result.

[0056] Step S4.7: After the unit completes the crossing of the vibration zone, it remains on the current safe side for the duration of the vibration zone safety side lock. If a new AGC command does not require crossing to another safe side within the lock duration, the unit is prohibited from re-entering the vibration zone.

[0057] Step S4.8: Integrate the first constraint correction result, the second constraint correction result, and the third constraint correction result to obtain the target output of the hydropower unit after constraint correction.

[0058] Beneficial effects of steps S4.1 to S4.8:

[0059] This series of steps addresses the technical problem that the target output path of hydropower units is prone to falling into the vibration zone and that existing solutions lack a systematic avoidance and controlled crossing mechanism. By calculating the safety side boundary through the vibration zone boundary and safety margin, when the unit's target output falls into or crosses the vibration zone, the output is redistributed, the climbing trajectory is limited, and the safety side is locked. This enables the unit to actively avoid and control the crossing of the vibration zone, reduce the risk of equipment fatigue damage, and ensure the safe operation of the unit.

[0060] Specifically, step S4.1 extracts the vibration zone boundary, safety margin, and safety-side locking duration from the standardized phased control parameter set to provide quantitative parameters for constraint correction; step S4.2 calculates the upper and lower safety-side boundaries of each operating unit based on the vibration zone boundary and safety margin to clarify the safe output range allowed for unit operation; step S4.3 compares the layered power response benchmark value with the safety-side boundary to determine whether the unit's target output falls within the vibration zone, providing a basis for subsequent branch processing; step S4.4 executes the original target if the target output does not fall within the vibration zone to avoid unnecessary output adjustments; step S4.5 executes the target output... When the unit enters the vibration zone, the target output of the unit is redistributed between the upper and lower safety sides, and the scheme with the smallest deviation from the average distribution and fewer switching times is selected, taking into account both power balance and safety side switching frequency; Step S4.6 When the unit needs to cross the vibration zone, a constrained climbing trajectory is generated according to the guide vane safe operating rate and amplitude limiting calculation is performed to control the safety and stability of the crossing process; Step S4.7 After the crossing is completed, the safety side is locked for a set duration to prevent the unit from repeatedly entering the vibration zone during the lock period; Step S4.8 By integrating the three types of constraint correction results, the target output of the hydropower unit after vibration zone correction is output for subsequent energy storage power matching.

[0061] As a further improvement to this application, step S5 involves performing energy storage power matching calculations on the target output of the hydropower unit after constraint correction and the energy storage recovery parameters in the standardized phased control parameter set, and simultaneously coupling the energy storage state-of-charge boundary constraints and active state-of-charge recovery calculations to obtain real-time energy storage compensation power commands, including:

[0062] Step S5.1: Extract the upper limit of energy storage state of charge, the lower limit of energy storage state of charge, the reference value of energy storage state of charge, the recovery gain of energy storage state of charge, and the maximum value of the recovery power of energy storage state of charge from the standardized phased control parameter set.

[0063] Step S5.2: Calculate the grid connection point power error based on the target output of the hydropower unit after constraint correction and the AGC command of the whole plant, and use the grid connection point power error as the AGC response energy storage power command;

[0064] Step S5.3: Determine whether there is a valid AGC event or a controlled crossing of the unit's vibration zone. If so, use the AGC response energy storage power command as a candidate value for real-time energy storage compensation power.

[0065] Step S5.4: If there is no valid AGC event and no controlled crossing of the unit vibration zone, calculate the deviation between the current state of charge of the energy storage battery and the reference value of the energy storage state of charge.

[0066] Step S5.5: Determine whether the deviation is greater than the recovery dead zone. If it is, generate a SOC recovery power plan based on the energy storage state of charge recovery gain and the maximum value of the energy storage state of charge recovery power.

[0067] Step S5.6: Adjust the total target output of the hydropower unit according to the SOC recovery power plan, so that the state of charge of the energy storage battery returns to the reference value of the energy storage state of charge, and obtain the SOC recovery energy storage power command.

[0068] Step S5.7: Arbitrate the priority of the SOC recovery energy storage power command and the AGC response energy storage power command. When the priority of the AGC response is increased, reduce or suspend the SOC recovery power to obtain the candidate value of real-time energy storage compensation power.

[0069] Step S5.8: Apply the energy storage rated power constraint, the energy storage state of charge upper limit constraint, and the energy storage state of charge lower limit constraint to the energy storage real-time compensation power candidate value to obtain the energy storage real-time compensation power command.

[0070] Beneficial effects of steps S5.1 to S5.8:

[0071] This series of steps addresses the technical problems of the lack of an active recovery mechanism for the state of charge (SOC) of energy storage batteries and the tendency for the SOC to deviate from the median region for a long time after continuous AGC events, leading to a decrease in availability. By performing energy storage power matching during the AGC response, generating a recovery power plan based on the SOC deviation during the static steady period, and adjusting the hydropower output to bring the SOC back to the reference value, and by simultaneously coupling the state of charge boundary constraints and priority arbitration, coordinated control of AGC response and SOC recovery is achieved, thereby improving the availability of continuous frequency regulation of energy storage.

[0072] Specifically, step S5.1 extracts the upper and lower limits of the energy storage state of charge (SOC), reference value, recovery gain, and maximum recovery power from the standardized phased control parameter set, providing parameter basis for power matching and SOC recovery; step S5.2 calculates the grid connection point power error based on the target output of the hydropower unit after constraint correction and the AGC command of the entire plant, and uses the power error as the AGC response energy storage power command to achieve real-time determination of the energy storage compensation amount; step S5.3 uses the AGC response energy storage power command as a candidate value when there is a valid AGC event or controlled crossing of the vibration zone, ensuring that the adjustment priority is higher than SOC recovery; step S5.4 calculates the deviation between the current SOC of the energy storage battery and the reference value during the static stability period, quantifying the SOC deviation. Step S5.5: Determine whether SOC recovery is triggered by comparing the deviation with the recovery dead zone, and generate an SOC recovery power plan based on the recovery gain and the maximum recovery power when triggered; Step S5.6: Adjust the total target output of the hydropower unit according to the SOC recovery power plan to make the state of charge of the energy storage battery return to the reference value, thereby achieving active maintenance of energy storage availability; Step S5.7: Arbitrate the priority between the SOC recovery energy storage power command and the AGC response energy storage power command, and reduce or suspend the SOC recovery power when the priority of the AGC response is increased to ensure the priority execution of the regulation task; Step S5.8: Apply rated power and upper and lower limit constraints of state of charge to the candidate values ​​of real-time compensation power for energy storage to ensure the safe operation of energy storage.

[0073] As a further improvement to this application, step S6 involves superimposing the real-time energy storage compensation power command with the constrained corrected target output of the hydropower unit to generate a joint output command for the grid connection point and performing power closed-loop tracking calculation to obtain the grid-connected regulation execution signal for the entire station, including:

[0074] Step S6.1: The real-time energy storage compensation power command is superimposed with the target output of the constrained hydropower unit to obtain the joint output command of the grid connection point;

[0075] Step S6.2: Collect the actual output power of the grid connection point and calculate the tracking error between the combined output command of the grid connection point and the actual output power of the grid connection point;

[0076] Step S6.3: Determine whether the tracking error is less than the AGC assessment dead zone. If it is less, maintain the current grid connection point joint output command.

[0077] Step S6.4: If the tracking error is not less than the AGC assessment dead zone, perform closed-loop correction on the grid connection point joint output command according to the proportional-integral control law to obtain the corrected joint output command.

[0078] Step S6.5: Determine whether the modified combined output command exceeds the output constraints of the hydropower unit and the rated power constraints of the energy storage battery. If it does not exceed, output it directly; if it exceeds, output it after limiting it according to the constraint boundary.

[0079] Step S6.6: Decompose the modified and limited combined output command into the active power command of each operating hydropower unit and the charging and discharging power command of the energy storage battery to obtain the grid-connected regulation execution signal for the entire station.

[0080] Beneficial effects of steps S6.1 to S6.6:

[0081] This series of steps addresses the technical problems of insufficient power tracking accuracy at the grid connection point and the lack of closed-loop correction and constraint limiting in the combined power output command of the water-storage system. By superimposing the real-time compensation power command of the energy storage and the target output of the hydropower unit after constraint correction, a combined power output command is generated. A proportional-integral control law is used for closed-loop correction and constraint limiting to achieve high-precision tracking of the grid connection point power and safe output of the station-wide regulation execution signal.

[0082] Specifically, step S6.1 involves superimposing the real-time energy storage compensation power command with the target output of the hydropower units after constraint correction to obtain the joint output command at the grid connection point, thereby achieving unified and coordinated output of hydropower and storage power. Step S6.2 involves collecting the actual output power at the grid connection point and calculating the tracking error between it and the joint output command to quantify the power tracking deviation. Step S6.3 involves comparing the tracking error with the AGC assessment dead zone and maintaining the current command when the error meets the accuracy requirements to avoid unnecessary adjustment actions. Step S6.4 involves performing closed-loop correction of the joint output command according to the proportional-integral control law when the tracking error exceeds the assessment dead zone to eliminate the steady-state deviation of power tracking. Step S6.5 involves judging and limiting the output and rated power constraints of the corrected joint output command to ensure that the execution command is within a safe range. Step S6.6 involves decomposing the corrected and limited joint output command into the active power command of each operational hydropower unit and the charging and discharging power command of the energy storage battery, and outputting a directly executable grid-connected regulation execution signal for the entire station.

[0083] To achieve the above objectives, this application also provides the following technical solutions:

[0084] A water-storage integrated AGC phased response system is provided, wherein the water-storage integrated AGC phased response system is applied to the water-storage integrated AGC phased response method as described above, and the water-storage integrated AGC phased response system includes:

[0085] The standardized phased control parameter set acquisition module is used to acquire real-time operation data of the water-storage combined system and historical operation data of the unit, and input them into the water-power-storage coupling model for parameter identification. The water-power-storage coupling model after parameter identification is then input into the NSGA-II multi-objective genetic algorithm to solve the standardized phased control parameter set.

[0086] The effective AGC event feature sequence acquisition module is used to identify instructions from the standardized phased control parameter set and perform threshold judgment calculations on the real-time AGC power instructions to obtain a complete and labeled effective AGC event feature sequence.

[0087] The hydropower storage tiered power response benchmark value acquisition module is used to perform three-stage collaborative allocation of the effective AGC event feature sequence and the standardized phased control parameter set, and divides the rapid response stage, energy maintenance stage and precise adjustment stage according to the time sequence to obtain the hydropower storage tiered power response benchmark value.

[0088] The constrained correction target output acquisition module is used to perform vibration zone constraint correction on the hydropower storage layered power response benchmark value and the vibration zone constraint parameters in the standardized phased control parameter set, so as to complete the unit output safety side locking and the crossing trajectory amplitude limit calculation, and obtain the constrained correction target output of the hydropower unit.

[0089] The real-time energy storage compensation power command acquisition module is used to perform energy storage power matching calculation between the target output of the hydropower unit after the constraint correction and the energy storage recovery parameters in the standardized phased control parameter set, and simultaneously couple the energy storage state of charge boundary constraint and the active recovery calculation of the state of charge to obtain the real-time energy storage compensation power command.

[0090] The whole-station grid-connected regulation execution signal acquisition module is used to superimpose the real-time energy storage compensation power command with the target output of the constrained hydropower unit to generate a joint output command at the grid connection point and perform power closed-loop tracking calculation to obtain the whole-station grid-connected regulation execution signal.

[0091] To achieve the above objectives, this application also provides the following technical solutions:

[0092] An electronic device includes a processor and a memory coupled to the processor, the memory storing program instructions executable by the processor; when the processor executes the program instructions stored in the memory, it implements the water storage combined AGC phased response method as described above.

[0093] To achieve the above objectives, this application also provides the following technical solutions:

[0094] A computer-readable storage medium storing program instructions, which, when executed by a processor, enable the implementation of the aforementioned water-storage combined AGC phased response method. Attached Figure Description

[0095] Figure 1 This is a schematic flowchart illustrating the steps of an embodiment of the water storage combined AGC phased response method of this application;

[0096] Figure 2 This is a diagram illustrating the AGC response effect during the dry season of an embodiment of the water storage combined AGC phased response method of this application;

[0097] Figure 3 This is a diagram showing the active power output curve during the dry season of an embodiment of the water storage combined AGC phased response method of this application;

[0098] Figure 4 This is a comparison chart of the SOC curves of various methods during the dry season of one embodiment of the water storage combined AGC phased response method of this application;

[0099] Figure 5 This is a diagram illustrating the AGC event response during the dry season of one embodiment of the water storage combined AGC phased response method of this application;

[0100] Figure 6 This is a diagram illustrating the AGC response effect during the high-water season of an embodiment of the water-storage combined AGC phased response method of this application;

[0101] Figure 7 This is a diagram showing the active power output curve during the high-water season of an embodiment of the water-storage combined AGC phased response method of this application;

[0102] Figure 8 This is a comparison chart of the SOC curves of various methods during the high-water season of one embodiment of the water-storage combined AGC phased response method of this application;

[0103] Figure 9 This is a diagram illustrating the AGC event response during the high-water season, based on an embodiment of the water-storage combined AGC phased response method of this application.

[0104] Figure 10 This is a schematic diagram of the functional modules of an embodiment of a water storage combined AGC phased response system according to this application;

[0105] Figure 11 This is a schematic diagram of the structure of an embodiment of the electronic device of this application;

[0106] Figure 12 This is a schematic diagram of the structure of one embodiment of the storage medium of this application. Detailed Implementation

[0107] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0108] The terms "first," "second," and "third" in this application are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first," "second," or "third" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified. All directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of this application are only used to explain the relative positional relationships and movements between components in a specific orientation (e.g., as shown in the figures). If the specific orientation changes, the directional indications also change accordingly. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive having. For example, a process, method, system, product, or device having a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0109] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be present in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0110] It should be noted that, due to the limited types and number of symbols or letters that can represent specific meanings, for embodiments with many formulas or codes, there may be situations where symbols or letters cannot meet the usage requirements. Therefore, the interpretation of formula symbols in the steps or sub-steps of the embodiments is only valid for the current step or sub-step.

[0111] If the same symbol has different interpretations in different steps or sub-steps, the interpretation in the current step or sub-step shall prevail; if the same symbol appears in different steps or sub-steps, but no interpretation is given in subsequent steps or sub-steps after its first appearance, the interpretation in the first step or sub-step shall be used.

[0112] For example, "i=1,2,...,n" is a writing convention and a well-known meaning. If "i=1,2,...,n" has already appeared once in an embodiment and needs to be used again in formulas with different scenarios and meanings in the future, then the meaning of "i=1,2,...,n" should be interpreted according to the corresponding formulas. If "i=1,2,...,n" is not used for the first time and is changed to a symbol or letter with a non-well-known meaning, such as "p=1,2,...,q", to avoid repetition, it is more likely to cause confusion, ambiguity, and unclear problems.

[0113] like Figure 1 As shown, this embodiment provides an example of a phased response method for a combined water and storage AGC system. In this embodiment, the phased response method for a combined water and storage AGC system includes the following steps:

[0114] Step S1: Obtain real-time operation data of the water-storage combined system and historical operation data of the unit, and input them into the water-power-storage coupling model for parameter identification. Then, input the parameter-identified water-power-storage coupling model into the NSGA-II multi-objective genetic algorithm to solve it, and obtain a standardized staged control parameter set.

[0115] Furthermore, step S1 specifically includes the following steps:

[0116] Step S1.1: Collect the plant-wide AGC commands of the water-storage combined system, the actual output of each operating hydropower unit, the guide vane opening, the unit's operating status, the energy storage battery power, and the energy storage battery's state of charge, and integrate them to obtain real-time operating data.

[0117] Preferably, the core of this step is to collect and integrate the real-time operating status data of the water-storage combined system. By connecting the power plant monitoring system and the energy storage converter communication interface, the system can obtain the AGC commands of the entire plant and the operating data of each device in real time, providing a real-time data foundation for the subsequent construction of the hydropower-storage coupling model and parameter identification.

[0118] Among them, the AGC instructions for the entire plant are issued by the power dispatch center to the power plant AGC host, with a sampling period of 1 to 5 seconds and an instruction resolution of 0.1 megawatts, including the target power value and the rate of change limit.

[0119] The actual output of each hydropower unit in operation is collected by the unit's outlet power transmitter. The transmitter's accuracy level is not lower than 0.2, and the collection cycle is synchronized with the AGC command. The sum of the output of each unit minus the plant's power consumption is the total power supplied to the grid.

[0120] The guide vane opening is acquired by a servo displacement sensor with an accuracy of no less than 0.1%. The guide vane opening range is 0% to 100%, where 0% corresponds to fully closed and 100% corresponds to fully open. The guide vane opening change rate is constrained by the safety guide vane action rate.

[0121] The unit commissioning status includes three states: operation, shutdown, and maintenance. Units that are only in operation are included in the AGC regulation range. The commissioning status is collected through the switch signals of the power plant monitoring system.

[0122] Specifically, the power of the energy storage battery is collected through the output power of the energy storage converter. The charging and discharging direction is represented by positive values ​​for discharging and negative values ​​for charging. The collection period is no more than 100 milliseconds, and the power measurement accuracy is no less than 0.5%.

[0123] Specifically, the state of charge (SOC) of the energy storage battery is estimated using the ampere-hour integration method or the Kalman filter method, with an estimation accuracy of no less than 2%, a sampling period of 1 second, and an SOC range of 0% to 100%.

[0124] Preferably, the sampling period for AGC commands across the entire plant is 2 seconds; the sampling period for actual unit output is 1 second; the sampling period for guide vane opening is 500 milliseconds; the sampling period for energy storage battery power is 100 milliseconds; the sampling period for energy storage battery state of charge is 1 second; and all data sources are integrated into a real-time operating data vector after being arranged according to a unified timestamp. Where n is the number of generating units in operation. This is the status indicator for the operation of the i-th generating unit.

[0125] For example, during the dry season, two generating units of a hydropower station are operating. At time YYYY-MM-DD-HH:MM:SS-001, the following data was collected: AGC command for the entire plant: 97.0 MW; actual output of Unit 1: 48.2 MW; actual output of Unit 2: 46.8 MW; guide vane opening of Unit 1: 62.3%; guide vane opening of Unit 2: 60.1%; both units are in operation; energy storage battery power: -2.5 MW (charging); energy storage battery state of charge: 47%. This data is then integrated to obtain a real-time operating data vector. .

[0126] Step S1.2: Collect the equivalent time constant of the hydropower unit governor, the equivalent time constant of the guide vane actuator, the equivalent time constant of the water flow inertia, the boundary of the unit's vibration zone, the rated power of the energy storage battery, and the rated energy of the energy storage battery, and integrate them to obtain the historical operation dataset of the unit.

[0127] Preferably, the core of this step is to collect and integrate the parameters of the hydropower unit and the energy storage battery equipment. By consulting the equipment nameplate, factory test report and historical operation records, all the parameters required for the model are obtained, providing a static parameter basis for subsequent model construction.

[0128] Among them, the equivalent time constant of the speed governor It reflects the dynamic response characteristics of the speed governor control loop, and is related to the speed governor PID parameters and the characteristics of the hydraulic servo mechanism. The typical range is 0.05 seconds to 0.2 seconds.

[0129] Among them, the equivalent time constant of the guide vane actuator This reflects the inertial delay of the guide vane servo mechanism from receiving the command to completing the opening adjustment. It is related to the hydraulic system volume, piston area, and oil pressure, and typically ranges from 2 to 10 seconds.

[0130] Among them, the equivalent time constant of water flow inertia This reflects the inertial delay of water flow caused by the water hammer effect, and is related to the length, cross-sectional area, and head of the water intake pipe. The typical range is 0.5 to 4 seconds. System stability requirements... The ratio is in the range of 0.1 to 0.5.

[0131] The boundary of the unit's vibration zone is determined based on unit model tests or long-term operational vibration monitoring data, using the output range as the basis. This indicates that the unit's vibration amplitude exceeds the permissible threshold within this range.

[0132] Among them, for the rated power of energy storage batteries With the rated energy of the energy storage battery The parameters are determined by the nameplate parameters of the energy storage system equipment. Determines the maximum charge and discharge power of the energy storage battery. Determine the usable energy capacity of the energy storage battery.

[0133] Preferably, the equivalent time constant of the speed governor The value ranges from 0.05 seconds to 0.2 seconds; the equivalent time constant of the guide vane actuator. The value ranges from 2 seconds to 10 seconds; the equivalent time constant of water flow inertia. The value ranges from 0.5 seconds to 4 seconds; rated power of the energy storage battery. The rated capacity of the energy storage battery is 10 MW to 30 MW, depending on the power station configuration. Based on the power plant configuration, 5 MWh to 20 MWh were selected; all parameters were integrated into a historical operating dataset of the unit. .

[0134] For example, the equipment parameter ledger of Unit 2 of a hydropower station records the equivalent time constant of the governor. Seconds, equivalent time constant of guide vane actuator Second, equivalent time constant of water flow inertia Seconds, generator vibration zone boundary [40,55] MW, energy storage system nameplate parameters, energy storage battery rated power Megawatts, rated energy of energy storage batteries Megawatt-hours, integrated to obtain the unit's historical operating dataset .

[0135] Step S1.3: Input the real-time operation data and the historical operation dataset of the unit into the hydropower storage coupling model. The governor, guide vane actuator and water flow inertial dynamic process of the hydropower unit are equivalently described by the series inertial link to obtain the model to be identified.

[0136] Preferably, the core of this step is to construct a hydropower-storage coupling model, which uses a series inertial link to provide an equivalent description of the multi-stage dynamic process of the hydropower unit, and combines the power response link of the energy storage battery to form an overall model of the hydropower-storage combined system, providing a model framework for subsequent parameter identification.

[0137] For the speed controller component, the transfer function is: ,in is the equivalent time constant of the speed governor, and s is the Laplace operator. This part describes the dynamic process of the speed governor outputting the guide vane opening control signal after receiving the power command.

[0138] For the guide vane actuator, the transfer function is: ,in The equivalent time constant of the guide vane actuator describes the inertial delay in which the servo mechanism converts the opening command into the actual guide vane opening.

[0139] For the inertial component of the water flow, an elastic water hammer model is adopted, and the transfer function is: ,in The equivalent time constant of water flow inertia describes the dynamic relationship between water flow changes caused by water hammer effect and mechanical power output. The negative term in the numerator reflects the initial reverse response characteristics of water hammer effect.

[0140] Among them, for the mechanical power output of the unit, ,in For guide vane opening, The three components, connected in series, constitute a complete dynamic channel for the hydropower unit from guide vane commands to mechanical power output.

[0141] For the power response stage of the energy storage battery, the transfer function is: ,in This is the response time constant of the energy storage converter, typically ranging from 0.02 seconds to 0.1 seconds. This component describes the rapid response process of the energy storage converter from receiving a power command to actually outputting power.

[0142] Among them, for the grid connection point power, ,in The actual output power of the energy storage battery is the sum of the unit's mechanical power and the energy storage battery's output power. The model to be identified consists of all the above components.

[0143] Preferably, the response time constant of the energy storage converter The value ranges from 0.02 seconds to 0.1 seconds; the parameters to be identified in the model include the equivalent time constant of the speed governor. Equivalent time constant of guide vane actuator and the equivalent time constant of water flow inertia The model inputs are guide vane opening command and energy storage power command, and the output is grid connection point power.

[0144] For example, when constructing a model to be identified for a hydropower station, the initial parameters are taken from the equipment ledger values. Second, Second, Second, At 10:00 pm, the model input guide vane opening command jumped from 60% to 70%, while the energy storage power command remained at 0 MW. The simulation showed that the grid connection point power experienced a reverse overshoot of about -1.2 MW within 2 seconds after the step jump, and then stabilized to the target output within 8 seconds. This response curve, compared with the actual unit test record, will be used for subsequent parameter identification.

[0145] Step S1.4: Use hydropower station test data or historical operation data to identify the parameters of the model to be identified and verify the accuracy of the model to be identified, so as to obtain the hydropower-storage coupling model after parameter identification.

[0146] Preferably, the core of this step is to complete the parameter identification and accuracy verification of the model to be identified, and to minimize the mean square error between the model output and the actual output by using the least squares method or particle swarm optimization algorithm, so that the model approximates the response characteristics of the actual working condition, and provides a reliable simulation model for subsequent multi-objective optimization.

[0147] For the identification algorithm, the least squares method or particle swarm optimization algorithm is adopted, and the identification objective function is the mean square error between the model output and the actual output. Where N is the number of sampling points, Output power for the model. This represents the actual power.

[0148] Among them, the identification parameters include the governor's equivalent time constant. Equivalent time constant of guide vane actuator and the equivalent time constant of water flow inertia The data segment to be identified is the running interval containing the step change of the AGC instruction, and the data segment length is not less than 120 seconds.

[0149] For model accuracy verification, the coefficient of determination is used. ,in The actual average power is required. If the verification does not meet the requirements, the identification data segment will be replaced and re-identified.

[0150] Preferably, the length of the identification data segment is between 120 and 300 seconds; the sampling interval is 1 second; the particle swarm optimization algorithm parameters are: population size 50, inertia weight 0.7, self-learning factor 1.5, swarm learning factor 1.5, maximum number of iterations 200; and the determination coefficient threshold is [not specified]. .

[0151] For example, the No. 2 generating unit of a certain hydropower station used AGC operation data in the YYYY-MM-DD time period for identification. The data segment length was 180 seconds, the sampling interval was 1 second, and the identification result was... Second, Second, Seconds, the coefficient of determination between the model output and the actual output on the validation data segment. To meet the accuracy requirements, a coupled hydropower energy storage model with parameter identification was obtained.

[0152] Step S1.5: Input the hydropower storage coupling model into the NSGA-II multi-objective genetic algorithm, and use the safe guide vane action rate, the exit threshold of the rapid response stage, the re-injection ratio threshold of the precise regulation stage, and the compensation coefficient of the precise regulation stage as optimization variables to obtain the first set of optimization variables.

[0153] Preferably, the core of this step is to extract and set the range of optimization variables directly related to the stage response, extract the core parameters of the control strategy from the hydropower-storage coupling model as optimization variables, and provide decision variables for subsequent multi-objective optimization.

[0154] Among them, the operating rate of the safety guide vane , defined as the maximum allowable rate of change of guide vane opening, in percentage per second, reflects the maximum allowable ramp rate of the unit, with an optimization range of 0.5% / second to 3.0% / second.

[0155] Among them, the exit threshold for the rapid response phase It is defined as the ratio of the change in the output of the hydropower unit to the change in the amplitude of the AGC command. When the ratio reaches the threshold, the fast response phase is exited. The optimization range is 30% to 70%.

[0156] Among them, the re-investment ratio threshold for precise adjustment stage , defined as the output completion ratio of the hydropower unit. When the ratio of the actual output change to the target change reaches this threshold, the energy maintenance stage is exited and the precise adjustment stage is entered. The optimization range is 75% to 95%.

[0157] Among them, the compensation coefficient for the precise adjustment stage , defined as the proportion of the energy storage battery's compensation power to the residual error at the grid connection point, determines the strength of the energy storage battery in eliminating residual errors during the precise adjustment stage, with an optimization range of 0.3 to 0.9.

[0158] Preferably, the first set of optimization variables is represented as follows: Safety guide vane operating rate The optimization range is 0.5% / second to 3.0% / second; the exit threshold for the fast response phase. The optimization range is 30% to 70%; the threshold for precise adjustment of the reinvestment ratio is also included. The optimization range is 75% to 95%; the compensation coefficient is adjusted during the precise adjustment stage. The optimization range is from 0.3 to 0.9.

[0159] For example, the initial parameters of a hydropower station under dry season conditions are taken as follows: / Second, , , The first set of optimization variables In the NSGA-II algorithm, it is used as part of the initial population for iterative optimization.

[0160] Step S1.6: Using the vibration zone safety side lock-in duration, energy storage state of charge recovery gain, and energy storage state of charge recovery power maximum amplitude as optimization variables, the second set of optimization variables is obtained.

[0161] Preferably, the core of this step is to extract and set the range of optimization variables related to vibration zone constraints and energy storage recovery, and to incorporate key parameters that affect unit safety and energy storage availability into the optimization framework to achieve structured management of optimization variables.

[0162] Among them, the locking time on the safe side of the vibration zone The minimum time a unit needs to remain on the safe side after completing a vibration zone crossing is defined as the minimum time that the unit must remain on the safe side to prevent the unit from re-entering the vibration zone during the lockout period. The optimization range is 30 seconds to 180 seconds.

[0163] Among them, for the energy storage state of charge recovery gain , defined as the mapping coefficient from SOC deviation to recovery power, determines the response speed of SOC recovery. A larger value results in faster recovery but may cause power fluctuations; the optimization range is 0.1 to 2.0.

[0164] Among them, the maximum amplitude of the energy storage state of charge recovery power. Defined as the maximum allowable regulation power during SOC recovery, expressed as a percentage of the energy storage rated power, it limits the impact of recovery operations on the grid connection point power, with an optimization range of 5% to 30% of the energy storage rated power.

[0165] Preferably, the second set of optimization variables is represented as follows: ; Vibration zone safety side locking duration The optimization range is 30 seconds to 180 seconds; energy storage state of charge recovery gain. The optimization range is 0.1 to 2.0; the maximum amplitude of energy storage state-of-charge recovery power. The optimization range is 5% to 30% of the rated power of the energy storage.

[0166] For example, the initial parameters of a certain hydropower station are set to... Second, , That is, 3 megawatts, the second set of optimization variables. Together with the first set of optimization variables, they constitute the complete decision variable space of the NSGA-II algorithm.

[0167] Step S1.7: Using the maximization of AGC comprehensive regulation performance index, the minimization of energy storage state of charge deviation, the minimization of the cumulative residence time in the vibration zone of the unit, and the minimization of the controlled crossing number of the vibration zone as objective functions, a multi-objective optimization model is constructed for the first set of optimization variables and the second set of optimization variables.

[0168] Preferably, the core of this step is to construct the objective function of the multi-objective optimization model, incorporating the three core indicators of AGC response performance, energy storage availability, and unit safety into a unified optimization framework, thus providing an objective space for subsequent Pareto solutions.

[0169] Among them, the comprehensive regulation performance index K of AGC is calculated with reference to the power grid AGC performance evaluation standard. ,in To adjust speed indicators, To adjust accuracy indicators, As a response time metric, a larger K value indicates better AGC response performance.

[0170] Among them, for the energy storage state of charge deviation Defined as the average absolute deviation between the SOC and the reference value within the evaluation period. Where T is the evaluation period, This is a reference value for the state of charge. The smaller the value, the less the energy storage deviates from the median range.

[0171] Among them, the cumulative residence time in the unit's vibration zone Where M is the number of times the object falls into the vibration zone during the evaluation period. Let j be the duration of the stay. The smaller the value, the better the safety of the generator unit.

[0172] Among them, the number of controlled crossings of the vibration zone Defined as the total number of times the unit crosses from one safe side of the vibration zone to the other safe side within the assessment period. The fewer the values, the smoother the power output adjustment path of the unit.

[0173] For a multi-objective optimization model, it is represented as: The first term being negative indicates maximizing K, and there are mutual constraints among the four objective functions.

[0174] Preferably, the weight of the AGC comprehensive regulation performance index Weight 0.45 Weight 0.30 Weight 0.25; Evaluation period T is 24 hours; State of charge reference value Take 50%; the multi-objective optimization model has no constraints, and the constraints are applied in step S1.8.

[0175] For example, during the 24-hour dry season assessment period of a certain hydropower station, the simulation under the initial parameters yielded K=2.86. , minute, Next, target vector The NSGA-II algorithm converges each objective function toward the Pareto front by iteratively optimizing the first and second sets of optimization variables.

[0176] Step S1.8: Apply energy storage power constraints, energy storage state of charge constraints, hydropower output constraints, guide vane safety opening and closing constraints, and vibration zone crossing time constraints to the multi-objective optimization model, and solve and output the Pareto solution set.

[0177] Preferably, the core of this step is to apply constraints to the multi-objective optimization model and solve it using NSGA-II. By applying five types of engineering constraints during the optimization iteration, the feasibility of the solution is guaranteed, and the Pareto front solution set is output for subsequent selection of representative parameter sets.

[0178] Among them, regarding the energy storage power constraint, The charging and discharging power of energy storage batteries must not exceed the rated power.

[0179] Among them, for the energy storage state of charge constraints, ,in Typically, 10% to 20% is used. Typically, 80% to 90% is used to prevent overcharging and over-discharging of energy storage batteries.

[0180] Among them, the constraints on hydropower output, ,in and These are the minimum and maximum technical outputs of the unit, determined by the unit's operating characteristic curves.

[0181] Among them, the safety opening and closing constraints of the guide vanes, ,in The maximum permissible operating rate of the guide vane is usually set to 3% / second to prevent excessive water hammer pressure caused by excessively fast guide vane operation.

[0182] Among them, the vibration zone travel time constraint, ,in The maximum permissible vibration zone crossing time is typically set at 60 seconds, which limits the duration the unit spends in the vibration zone.

[0183] For the NSGA-II algorithm parameters, the population size is 200, the crossover probability is 0.9, the mutation probability is 0.1, and the maximum number of iterations is 300 generations. Tournament selection, simulated binary crossover, and polynomial mutation operators are used.

[0184] Preferably, the energy storage state of charge limit Take 15%; Upper limit of energy storage state of charge Take 85%; maximum permissible operating rate of the guide vane. Take 3% / second; upper limit of vibration zone crossing time. Use 60 seconds; NSGA-II population size 200; crossover probability 0.9; mutation probability 0.1; maximum number of iterations 300.

[0185] For example, under the dry season operating conditions of two generating units at a hydropower station, the NSGA-II algorithm outputs a Pareto solution set of 127 non-dominated solutions after 300 iterations. Among them, the AGC comprehensive regulation performance index K ranges from 2.91 to 3.12, and the SOC deviation is... The range is 3.1% to 6.8%, and the cumulative residence time in the vibration zone is... The timeframe is 18 to 55 minutes, with controlled number of crossings. The range is from 3 to 9, and each solution has a trade-off relationship with different objectives.

[0186] Step S1.9: Select representative parameter sets from the Pareto solution set that prioritize AGC response performance and have energy storage state of charge deviation less than a preset deviation threshold, and format them as a standardized phased control parameter set.

[0187] Preferably, the core of this step is to complete the screening and formatting of the Pareto solution set, select representative parameter sets from non-dominated solutions according to priority rules, integrate the seven optimization variables and three energy storage parameters into a standardized phased control parameter set in a unified format, and output parameter configurations that can be directly used for online control.

[0188] Specifically, the selection criteria first involve selecting the top 20% of solutions based on the AGC comprehensive regulation performance index K, and then selecting the SOC deviation from the selection results. Less than the preset deviation threshold If multiple solutions simultaneously satisfy the conditions, then the cumulative residence time in the vibration zone is selected. The smallest solution is taken as the representative parameter set.

[0189] Among them, for the preset deviation threshold The value of 5% indicates that only solutions with an average SOC deviation of no more than 5% are included as candidates.

[0190] Specifically, for formatting, the seven optimization variables of the representative parameter set are integrated with the energy storage state of charge reference value and upper and lower limits into a standardized phased control parameter set. .

[0191] Preferably, a preset deviation threshold is used. Take 5%; the top 20% of the AGC comprehensive regulation performance indicators are selected; state of charge reference value Take 50%; State of charge limit Take 15%; Upper limit of state of charge Take 85%.

[0192] For example, under the dry season operating conditions of two generating units at a hydropower station, among the 127 non-dominated solutions in the Pareto solution set, there are 25 solutions with K ranking in the top 20%, of which the SOC deviation is significant. There are 8 solutions with a rate less than 5%, and the cumulative residence time in the vibration zone is selected. The minimum solution corresponds to the parameter: / Second, , , , Second, , Megawatts, formatted as a standardized, phased set of control parameters. .

[0193] Beneficial effects of steps S1.1 to S1.9:

[0194] This series of steps addresses the technical problem of relying on manual experience to tune control parameters in combined hydropower and energy storage AGC systems, making it difficult to balance multiple performance indicators. By collecting real-time operating data and historical operating datasets of the units, a coupled hydropower and energy storage model is constructed. The NSGA-II multi-objective genetic algorithm is then used to perform multi-objective optimization with AGC comprehensive regulation performance, energy storage state of charge deviation, cumulative residence time in the vibration zone, and controlled crossing times as objective functions. This achieves systematic tuning and global optimal configuration of control parameters, enabling a balance between response performance, energy storage availability, and unit safety in the phased response strategy.

[0195] Specifically, steps S1.1 and S1.2 collect real-time operating data and historical operating datasets of the unit, respectively, to provide a complete data foundation for model construction and parameter identification; step S1.3 uses a series inertial link to equivalently describe the multi-stage dynamic process of the hydropower unit, establishing a model reflecting the coupling characteristics of hydropower and energy storage; step S1.4 uses experimental data or historical operating data to identify parameters and verify the accuracy of the model to be identified, improving the model's approximation of actual operating conditions; steps S1.5 and S1.6 divide the optimization variables into a first optimization variable group directly related to the stage response and a group related to vibration zone locking and S... The second set of optimization variables related to OC recovery is used to achieve structured management of optimization variables; Step S1.7 constructs a multi-objective optimization model with four performance indicators as objective functions, incorporating AGC response performance, energy storage availability and unit safety into a unified optimization framework; Step S1.8 ensures the engineering feasibility of the optimization results by applying constraints on energy storage power, state of charge, hydropower output, guide vane safe opening and closing and vibration zone crossing time; Step S1.9 selects representative parameter sets from the Pareto solution set through preset deviation thresholds and formats them to output a standardized phased control parameter set that can be directly used for online control.

[0196] Step S2 involves identifying the standardized phased control parameter set and performing threshold determination calculations on the real-time AGC power commands to obtain a complete and valid AGC event feature sequence.

[0197] Furthermore, step S2 specifically includes the following steps:

[0198] Step S2.1: Extract the effective AGC instruction change threshold and minimum duration from the standardized phased control parameter set.

[0199] Preferably, the core of this step is to extract the event recognition threshold, and obtain two judgment parameters, the effective AGC instruction change threshold and the minimum duration, from the standardized phased control parameter set, so as to provide a quantitative basis for subsequent candidate AGC event marking and duration determination.

[0200] Among them, the effective AGC command change threshold It is defined as the percentage of the change in AGC power command relative to the plant's rated capacity. When the absolute value of the command change exceeds this threshold, it is considered a valid adjustment signal. The typical value range is 1% to 5%.

[0201] Among them, for the minimum duration The value is defined as the shortest time during which the change in AGC power command exceeds a threshold. The validity of the event is confirmed when the change remains in the same direction and does not return to zero within this time. The typical value range is 5 to 15 seconds.

[0202] Preferably, the effective AGC command change threshold Take 3%; minimum duration Take 10 seconds; both parameters are stored in a standardized phased control parameter set. In the middle, the index is and .

[0203] For example, a hydropower station with a total rated capacity of 200 megawatts has standardized, phased control parameters centralized. Corresponding to an absolute value of 6 megawatts A valid AGC event is defined as an AGC power command change exceeding 6 megawatts and lasting for 10 seconds or more.

[0204] Step S2.2: Obtain the real-time AGC power command at the current sampling time and calculate the change in the real-time AGC power command relative to the previous sampling time.

[0205] Preferably, the core of this step is to complete the acquisition and change calculation of real-time AGC power commands, eliminate the impact of communication jitter through moving average filtering, quantify the magnitude and direction of the current adjustment demand, and provide input for subsequent threshold comparison.

[0206] For real-time AGC power command acquisition, the sampling period is... Get , where k is the sampling sequence number, and the sampling period is consistent with the sampling period of the whole plant AGC instruction in step S1.1.

[0207] Among them, for the AGC power command change, A positive value indicates an upward adjustment instruction, and a negative value indicates a downward adjustment instruction.

[0208] Specifically, for the moving average filtering, the original AGC power command sequence is preprocessed using a moving average with a sliding window length L = 3 to 5 sampling points. The filtered command... This eliminates spurious changes caused by communication jitter.

[0209] Preferably, the sampling period The duration is 2 seconds; the moving average filter window length L is 3 sampling points; the change is calculated based on the filtered instruction sequence, i.e. .

[0210] For example, a hydropower station has a sampling period of 2 seconds and a filtering window with 3 sampling points. The original AGC commands at three consecutive sampling times are 95.0, 97.2, and 98.1 MW. After filtering... megawatts, a moment ago Megawatts, change If the power consumption is less than 6 megawatts, event recognition will not be triggered.

[0211] Step S2.3: Determine whether the change amount is greater than the effective AGC instruction change threshold. If it is greater, mark it as a candidate AGC event.

[0212] Preferably, the core of this step is to complete the threshold comparison of the change in AGC power command and the marking of candidate events. Candidate AGC events are screened by comparing the absolute value of the change with the effective AGC command change threshold, and small fluctuations with insufficient amplitude are eliminated.

[0213] Among them, for the comparison conditions, the absolute value of the change in AGC power command is used. Threshold for changes in valid AGC instructions The corresponding absolute values ​​are compared.

[0214] Among them, for candidate AGC event flags, when When this time occurs, mark it as the start time of the candidate AGC event. ,in Record the adjustment direction for the plant's rated capacity. and the magnitude of the change in instructions .

[0215] Among them, for the non-triggered case, when If no event recognition is triggered, continue monitoring the next sampling time.

[0216] Preferably, the rated capacity of the entire plant Determined based on the power plant's installed capacity; effective AGC command change threshold. Take 3%; adjust direction +1 indicates an upward adjustment, and -1 indicates a downward adjustment.

[0217] For example, a certain hydropower station has a total rated capacity of 200 megawatts. The corresponding absolute value of the threshold is 6 megawatts. This corresponds to the change in the AGC command after filtering at sampling time YYYY-MM-DD-HH:MM:SS-001. megawatt-hours, Mark the start time of candidate AGC events Adjust direction Instruction change amplitude Megawatts.

[0218] Step S2.4: Determine whether the duration of the candidate AGC event is not less than the minimum duration. If it is satisfied, it is identified as a valid AGC event.

[0219] Preferably, the core of this step is to determine the duration of candidate AGC events. By continuously monitoring the direction and duration of the change, candidate events that meet the minimum duration requirement are upgraded to valid AGC events, thus filtering out short-term pulse interference.

[0220] Among them, for the calculation of duration, ,in For the current moment, This represents the start time of the candidate AGC event.

[0221] Among them, regarding the upgrade conditions, when Furthermore, if the change in AGC power command remains in the same direction during the duration, the candidate AGC event will be upgraded to a valid AGC event.

[0222] Among them, regarding the conditions for revocation, if in If the change in the internal AGC power command is reversed or returns to zero, the candidate AGC event flag is removed, and the system re-enters the monitoring state.

[0223] Preferably, minimum duration Take 10 seconds; the allowable range of variation during the duration is... Within this range, it is considered to maintain the same direction.

[0224] For example, the start time of a candidate AGC event YYYY-MM-DD-HH:MM:SS-001, adjust the direction. Minimum duration Seconds, from For the first 5 consecutive sampling periods (2 seconds each, 10 seconds in total), the AGC command changes were all positive and the cumulative amplitude was 12.0 MW. Second This event is upgraded to a valid AGC event.

[0225] Step S2.5: Record the start time, adjustment direction, command change amplitude, initial output of hydropower unit, and initial state of charge of energy storage battery of valid AGC events to obtain a complete set of valid AGC event feature sequences.

[0226] Preferably, the core of this step is to record and encapsulate the effective AGC event feature information. By recording the event start time, adjustment direction, command change amplitude, and initial running state, complete initial state information is provided for the subsequent three-stage collaborative allocation.

[0227] Among them, for the start time of the event, Accurate to the second.

[0228] Regarding adjusting the direction, +1 indicates an upward adjustment and -1 indicates a downward adjustment.

[0229] Among them, for the magnitude of the instruction change, This represents the cumulative change in AGC power command from the start of the event to the confirmation time. ,in This is the moment the event is confirmed.

[0230] Among them, for the initial output of the hydropower unit, Where n is the number of operating units, For the i-th unit in The actual output at any given moment.

[0231] Among them, the initial state of charge of the energy storage battery, The SOC estimation module in step S1.1 provides the SOC estimation.

[0232] Among them, for feature sequence encapsulation, the effective AGC event feature sequence is represented as follows: .

[0233] Preferably, the event start time is accurate to the second; adjust direction. Take +1 or -1; Instruction change amplitude The unit is megawatt; initial output of hydropower units. The unit is megawatts; initial state of charge of the energy storage battery. The unit is percentage.

[0234] For example, the start time of a valid AGC event record. Adjusting direction Command change amplitude Initial output of megawatt-class hydroelectric units Megawatts (48.2 MW and 36.8 MW for the two units), initial state of charge of energy storage batteries Feature sequence .

[0235] Beneficial effects of steps S2.1 to S2.5:

[0236] This series of steps addresses the technical problem of frequent short-term power fluctuations triggering responses in hydropower storage integrated AGC systems, leading to frequent start-stop adjustments of hydropower storage. By using a dual criterion of effective AGC command change threshold and minimum duration, real-time AGC power commands are filtered to identify genuine power adjustment events and record complete feature information. This achieves accurate identification of effective AGC events and effective filtering of short-term fluctuations, providing a reliable event triggering basis for subsequent three-stage coordinated allocation.

[0237] Specifically, step S2.1 extracts the effective AGC command change threshold and minimum duration from the standardized phased control parameter set to provide a quantitative basis for event identification; step S2.2 quantifies the amplitude characteristics of the current adjustment demand by calculating the change in real-time AGC power command relative to the previous sampling time; step S2.3 filters candidate AGC events by comparing the change with the effective AGC command change threshold, eliminating minor fluctuations with insufficient amplitude; step S2.4 further filters short-term pulse interference by determining the minimum duration, ensuring that only continuous power changes are identified as effective AGC events; and step S2.5 provides complete initial state information for the three-stage coordinated allocation by recording the event start time, adjustment direction, command change amplitude, initial output of the hydropower unit, and initial state of charge of the energy storage battery.

[0238] Step S3 involves a three-stage collaborative allocation of the effective AGC event characteristic sequence and the standardized phased control parameter set, dividing the sequence into a rapid response phase, an energy maintenance phase, and a precise adjustment phase, to obtain the stratified power response benchmark value for hydropower energy storage.

[0239] Furthermore, step S3 specifically includes the following steps:

[0240] Step S3.1: Extract the exit threshold for the rapid response stage, the reinvestment ratio threshold for the precise adjustment stage, and the compensation coefficient for the precise adjustment stage from the standardized phased control parameter set.

[0241] Preferably, the core of this step is to extract the three-stage control parameters, obtain the stage switching threshold and compensation coefficient from the standardized stage control parameter set, and provide a quantitative basis for the timing division and power allocation of the three stages of rapid response, energy maintenance and precise regulation.

[0242] Among them, the exit threshold for the rapid response phase The value is defined as the ratio of the change in hydropower unit output to the change in AGC command amplitude. When the change in hydropower unit output reaches this ratio, the rapid response phase is terminated. This threshold determines when the energy storage battery switches from full compensation to suspension.

[0243] Among them, the re-investment ratio threshold for precise adjustment stage The threshold is defined as the ratio of the actual output of the hydropower unit to the target output. When the ratio of the actual output change of the hydropower unit to the target output change reaches this threshold, the unit exits the energy maintenance phase and enters the precision adjustment phase. This threshold determines when the energy storage battery is put back into fine adjustment.

[0244] Among them, the compensation coefficient for the precise adjustment stage Defined as the proportion of the energy storage battery's compensation power to the residual error at the grid connection point, it determines the extent to which the energy storage battery eliminates residual errors during the precise adjustment phase. The larger the residual error, the faster it is eliminated, but the more energy storage capacity is consumed.

[0245] Preferably, the exit threshold for the rapid response phase. Take 50%; threshold for reinvestment ratio during precise adjustment phase. Take 85%; compensation coefficient for precise adjustment stage Set to 0.6; the three parameters are stored in the standardized phased control parameter set. middle.

[0246] For example, a standardized phased control parameter set for a hydropower station , When the AGC command changes amplitude In the megawatt-hour (MWH) phase, the exit condition for the rapid response phase is when the change in the output of the hydropower unit reaches 6.0 MW, and the exit condition for the energy maintenance phase is when the output completion rate of the hydropower unit reaches 85%, i.e., 10.2 MW.

[0247] Step S3.2: Determine the event start time and adjustment direction based on the effective AGC event characteristic sequence and enter the rapid response stage. Use the energy storage battery to compensate for the grid connection point power error according to the rated power and use the hydropower unit to adjust the output power to the target according to the allowable guide vane action rate.

[0248] Preferably, the core of this step is to complete the power allocation execution of the rapid response phase, start the rapid response according to the effective AGC event characteristic sequence, use the rapid response capability of the energy storage battery to fully compensate for the power error at the grid connection point, and at the same time, the hydropower unit climbs the slope to the target output according to the safe guide vane action rate to make up for the dynamic lag of hydropower.

[0249] For event initiation, based on the valid AGC event characteristic sequence The start time of the event and adjust direction Initiate the rapid response phase.

[0250] Among them, for the grid connection point power error, ,in The actual power at the grid connection point. This refers to the actual total output of the hydropower unit. This represents the actual output power of the energy storage battery.

[0251] Among them, for energy storage battery power commands, ,in As a limiting function, the energy storage battery prioritizes compensating for grid connection point power errors according to rated power during the fast response phase.

[0252] Among them, for the regulation of hydropower units, the target output of hydropower units ,in The rated ramp time for the generator unit is determined by the safe guide vane operating rate. Adjust the force towards the target.

[0253] Preferably, the unit's rated ramp time The energy storage battery bears the majority of the power error during the fast response phase; limiting function. .

[0254] For example, a valid AGC event , megawatts megawatts megawatts / second, rated power of a single unit of 80 MW, power error at the grid connection point at the start of the fast response phase. Megawatts, energy storage battery power command The hydropower units are climbing towards the 97.0 MW target at a rate of 1.2 MW / s.

[0255] Step S3.3: Determine whether the output change of each operating hydropower unit has reached the exit threshold of the rapid response stage. If it has, exit the rapid response stage and obtain the rapid response stage end signal.

[0256] Preferably, the core of this step is to complete the exit determination of the rapid response phase. By calculating the proportion of the change in the output of the hydropower unit to the change in the amplitude of the AGC command in real time and comparing it with the exit threshold, the timing of the energy storage battery switching from full compensation to suspension is determined, and a phase switching signal is generated.

[0257] Among them, the change in the output of the hydropower unit, Where n is the number of operating units, Let i be the output of the i-th generator unit at the start of the event.

[0258] Among them, the exit criteria are as follows: Record the end time of the fast response phase when the conditions are met. It also generates a signal indicating the end of the fast response phase.

[0259] Preferably, the exit threshold for the rapid response phase. Take 50%; Exit time Accurate to the second; the end signal includes a timestamp and the current output status of the hydropower unit.

[0260] For example, a valid AGC event megawatts The exit condition is that the change in the output of the hydropower unit reaches 6.0 MW, at the start time of the event. The output of the two generating units, YYYY-MM-DD-HH:MM:SS-001, increased from 48.2 MW to 51.2 MW and from 36.8 MW to 39.8 MW respectively after 5 seconds. megawatts ,Record It also generates a signal indicating the end of the fast response phase.

[0261] Step S3.4: Based on the end signal of the rapid response phase, the energy maintenance phase begins, in which the hydropower unit undertakes the main power regulation task and the energy storage battery stops the conventional power compensation.

[0262] Preferably, the core of this step is to complete the power allocation execution of the energy maintenance phase. The hydropower unit continues to adjust the output to the target according to the safe guide vane operation rate to undertake the main adjustment task. The energy storage battery stops the conventional power compensation and reserves the capacity for subsequent precise adjustment, so as to achieve a reasonable allocation of energy storage capacity.

[0263] For hydropower unit regulation, the hydropower unit continues to operate at the safe guide vane operating rate. It exerts force to adjust towards the target and undertakes the main adjustment task of AGC commands.

[0264] Among them, for energy storage battery power commands, The energy storage battery stops regular power compensation and enters standby mode.

[0265] Specifically, for the phase objective, the main part of the AGC command is completed by utilizing the continuous regulation capability of the hydropower unit, while reserving energy storage capacity for the elimination of residual errors in the subsequent precise regulation phase.

[0266] Preferably, the energy storage battery power command during the energy maintenance phase Hydropower units continue to operate as scheduled. / second adjustment; the energy storage battery remains in standby monitoring mode.

[0267] For example, the end of the rapid response phase The total output of the hydropower units is 91.0 MW. After entering the energy maintenance phase, the two units continue to climb towards the target output of 97.0 MW at a rate of 1.2 MW / s. The power command of the energy storage battery drops from 12.0 MW to 0 MW and enters standby mode.

[0268] Step S3.5: Calculate the output completion ratio of the hydropower unit and determine whether the output completion ratio of the hydropower unit is greater than the re-start ratio threshold of the precise adjustment stage. If it is greater, exit the energy maintenance stage and obtain the end signal of the energy maintenance stage.

[0269] Preferably, the core of this step is to determine the exit from the energy maintenance phase. By calculating the output completion ratio of the hydropower unit in real time and comparing it with the re-entry ratio threshold, the timing for the energy storage battery to be re-entered for fine adjustment is determined, thereby achieving an accurate switch from the energy maintenance phase to the precise adjustment phase.

[0270] Among them, the percentage of hydropower unit output completed, This reflects the proportion of the regulation that the hydropower unit has completed relative to the target regulation.

[0271] Among them, the exit criteria are as follows: When the conditions are met, record the end time of the energy maintenance phase. And generate an end signal for the energy maintenance phase.

[0272] Preferably, the threshold for precise adjustment of the reinvestment ratio is used. Take 85%; Exit time Accurate to the second.

[0273] For example, a valid AGC event megawatts megawatts The exit condition is that the total output of the hydropower units reaches [a certain value]. Megawatts; after approximately 3.5 seconds, the combined output of the two units increased from 91.0 megawatts to 95.2 megawatts. ,Record And generate an end signal for the energy maintenance phase.

[0274] Step S3.6: Based on the end signal of the energy maintenance stage, enter the precise adjustment stage, and eliminate the residual power error at the grid connection point by using the energy storage battery according to the compensation coefficient of the precise adjustment stage.

[0275] Preferably, the core of this step is to complete the power distribution execution of the precise adjustment stage, the energy storage battery eliminates the residual power error at the grid connection point according to the compensation coefficient, the hydropower unit maintains its current output basically unchanged, and improves the power tracking accuracy of AGC.

[0276] Among them, the residual power error at the grid connection point, This reflects the remaining power deviation of the hydropower unit after the main regulation is completed.

[0277] Among them, for energy storage battery power commands, ,in To precisely adjust the stage compensation coefficient.

[0278] In the case of hydropower unit output, the current output is kept basically unchanged during the precise adjustment phase, and the energy storage battery performs fine adjustment to eliminate residual errors.

[0279] Preferably, the compensation coefficient of the precise adjustment stage is used. Set to 0.6; after residual error is eliminated, the energy storage battery power command is returned to zero; end of the precise adjustment phase. This is the moment when the absolute value of the residual error is less than the AGC assessment dead zone.

[0280] For example, the end of the energy maintenance phase. The total output of the hydropower units is 95.2 MW, the AGC target is 97.0 MW, and the residual power error is... Megawatts, energy storage battery power command The power generation capacity of the hydropower unit remains unchanged at 95.2 MW, while the energy storage battery discharges at 1.08 MW to compensate for residual errors.

[0281] Step S3.7: Integrate the target output sequence of hydropower units and the power sequence of energy storage batteries in the rapid response stage, energy maintenance stage, and precise regulation stage to obtain the stratified power response benchmark value of hydropower energy storage.

[0282] Preferably, the core of this step is to complete the time-series splicing and encapsulation of the three-stage power sequence, splicing the target output sequence of the hydropower unit and the power sequence of the energy storage battery in the three stages of rapid response, energy maintenance and precise adjustment along the time axis, and outputting a complete layered power response benchmark value for subsequent vibration zone constraint correction.

[0283] Among them, for the target output sequence of hydropower units, It is spliced ​​together according to the three-stage time sequence.

[0284] Among them, for the power sequence of energy storage batteries, It is spliced ​​together according to the three-stage time sequence.

[0285] Among them, for the hierarchical power response reference value, ,in This is to determine the end time of the precise adjustment phase.

[0286] Preferably, the three-stage sequences are seamlessly spliced ​​along the timeline; baseline value It includes the target output of the hydropower unit and the power command of the energy storage battery at each sampling time; the sequence length is equal to the value from... arrive The total number of sampling points.

[0287] For example, a valid AGC event from arrive The process lasts for 20 seconds, with a sampling period of 2 seconds and a total of 10 sampling points. The fast response phase lasts 5 seconds (3 sampling points), the energy maintenance phase lasts 3.5 seconds (2 sampling points), and the precise adjustment phase lasts 11.5 seconds (5 sampling points). The tiered power response reference value is also included. Includes target output commands for 10 hydropower units and power commands for energy storage batteries, for example megawatts Megawatts.

[0288] Beneficial effects of steps S3.1 to S3.7:

[0289] This series of steps addresses the technical problems of existing hydropower-storage combined AGC strategies not being designed in stages according to the response process time scale and the fixed power division between hydropower and storage leading to the inability to fully leverage complementary advantages. By dividing the time sequence into three stages—prioritizing energy storage compensation in the rapid response stage, reserving capacity of hydropower main regulation and energy storage in the energy maintenance stage, and eliminating residual errors by proportion of energy storage in the precise regulation stage—this achieves optimized power division between hydropower and energy storage in different response stages, improves the overall regulation performance of AGC, and rationally allocates energy storage capacity consumption.

[0290] The process involves several steps: Step S3.1 extracts three-stage control parameters from a standardized set of phased control parameters, providing a quantitative basis for phase division and switching; Step S3.2, in the rapid response phase, prioritizes compensation for grid connection point power errors using energy storage batteries and adjusts the hydropower units according to the allowable guide vane movement rate, utilizing the rapid response capability of energy storage to compensate for hydropower dynamic lag; Step S3.3, by determining whether the hydropower unit output change reaches the rapid response phase exit threshold, timely exit from the rapid response phase and phase switching are achieved; Step S3.4, in the energy maintenance phase, hydropower units undertake the main power regulation task, and energy storage batteries cease routine power compensation, reserving energy storage capacity for subsequent precise regulation; Step S3.5, by comparing the hydropower unit output completion ratio with the re-injection ratio threshold, accurate switching from the energy maintenance phase to the precise regulation phase is achieved; Step S3.6, in the precise regulation phase, energy storage batteries eliminate residual grid connection point power errors according to the compensation coefficient, improving AGC power tracking accuracy; and Step S3.7, by integrating the three-stage hydropower unit target output sequence and the energy storage battery power sequence, a complete layered power response benchmark value is output for subsequent vibration zone constraint correction.

[0291] Step S4 involves performing vibration zone constraint correction on the stratified power response benchmark value of hydropower storage and the vibration zone constraint parameters in the standardized phased control parameter set to complete the unit output safety side locking and the crossing trajectory amplitude limit calculation, thereby obtaining the target output of the hydropower unit after constraint correction.

[0292] Furthermore, step S4 specifically includes the following steps:

[0293] Step S4.1: Extract the vibration zone boundary, vibration zone safety margin, and vibration zone safety side locking duration from the standardized phased control parameter set.

[0294] Preferably, the core of this step is to extract the constraint parameters of the vibration zone. The vibration zone boundary, safety margin, and safety side locking duration are obtained from the standardized phased control parameter set, providing a quantitative basis for subsequent safety side boundary calculation and constraint correction.

[0295] Among them, for the boundary of the vibration zone This indicates the output range within which the unit is not allowed to operate for extended periods. Within this range, the unit's vibration amplitude exceeds the allowable threshold. The boundary value is determined based on unit model tests or long-term operation vibration monitoring data.

[0296] Among them, the safety margin of the vibration zone This indicates a safety margin reserved outside the boundary of the vibration zone to prevent the unit from accidentally entering the edge of the vibration zone due to power output fluctuations. It is usually taken as 2% to 8% of the rated power of the unit.

[0297] Among them, the locking time on the safe side of the vibration zone This indicates the minimum time the unit must remain on the safe side after completing the crossing, to prevent the unit from repeatedly entering the vibration zone during the lockout period.

[0298] Preferably, the safety margin of the vibration zone Take 3 MW; Vibration zone safety side lock-in time Take 90 seconds; both parameters are stored in a standardized phased control parameter set. middle.

[0299] For example, the vibration zone boundary of Unit 2 of a certain hydropower station [40,55] MW, safety margin Megawatts, safety-side lockout duration Seconds, from the standardized phased control parameter set The extracted data is then used for subsequent safety side boundary calculations.

[0300] Step S4.2: Calculate the lower safety boundary and upper safety boundary of the vibration zone for each operational hydropower unit based on the vibration zone boundary and the vibration zone safety margin.

[0301] Preferably, the core of this step is to calculate the safety side boundary. By superimposing a safety margin outside the boundary of the vibration zone, the allowable operating range of the unit is determined, and the output range within which the unit can operate safely for a long period of time is clarified.

[0302] Among them, for the lower safe side boundary of the vibration zone, This indicates the safe operating boundary below the vibration zone.

[0303] Among them, for the safe side boundary of the vibration zone, This indicates the safe operating boundary on the upper side of the vibration zone.

[0304] Among them, the permissible operating range of the unit, ,in To minimize the technical output of the unit, To maximize the unit's technical output, the unit is only permitted to operate for extended periods within two safety-side zones.

[0305] Preferably, the lower safety side boundary of the vibration zone ; Upper safety boundary of the vibration zone Each operating unit calculates its own safety boundary.

[0306] For example, the vibration zone boundary of Unit 2 of a certain hydropower station [40,55] MW, safety margin Megawatts, lower safe boundary of the vibration zone Megawatts, upper safe boundary of the vibration zone Megawatts, the permissible operating range of the unit is Megawatts, of which 20 megawatts is the minimum technical output and 80 megawatts is the maximum technical output.

[0307] Step S4.3: Compare the stratified power response benchmark value of hydropower storage with the lower safe side boundary and upper safe side boundary of the vibration zone of each operational hydropower unit, and determine whether the target output of the hydropower unit falls into the vibration zone.

[0308] Preferably, the core of this step is to compare and determine the target output with the safety side boundary. The target output of each operating unit is extracted from the hierarchical power response benchmark value and compared with the safety side boundary of the unit at each time step, providing a basis for subsequent branch processing.

[0309] Among them, the comparison method is based on the stratified power response benchmark value of hydropower energy storage. Extract the target output of each operational hydropower unit. .

[0310] Among them, regarding the judgment conditions, If the conditions are met, the target output of the unit falls into the vibration zone; otherwise, it does not fall into the vibration zone.

[0311] In the comparison process, each operating unit was compared one by one, and the unit number and corresponding time of falling into the vibration zone were recorded.

[0312] Preferably, the comparison granularity is determined for each unit at each sampling time; the recorded content includes the unit number, timestamp, target output value, and determination result.

[0313] For example, at a certain moment, the target output of Unit 1 is 35.0 MW, the target output of Unit 2 is 56.0 MW, and the safety boundary of Unit 1 is [30,53] MW. Established, Unit 1 falls into the vibration zone; Unit 2 safety side boundary [37,58] MW, The system was established, and Unit 2 also fell into the vibration zone. The unit numbers and the time of the incident were recorded.

[0314] Step S4.4: If the target output of the hydropower unit does not fall into the vibration zone, then execute according to the original target output to obtain the first constraint correction result.

[0315] Preferably, the core of this step is to complete the constraint correction for units that have not fallen into the vibration zone. No adjustments are made to units whose target output is within the safe side range. The original target output is executed according to the stratified power response benchmark value to avoid unnecessary output adjustments.

[0316] Among them, for the correction rule, the first constraint correction result That is, the unit executes according to the original target output sequence.

[0317] Specifically, this step is executed when step S4.3 determines that the unit's target output has not fallen into the vibration zone, without changing the output plan.

[0318] Preferably, the result of the first constraint correction is completely consistent with the original target output; no additional adjustment is introduced.

[0319] For example, at a certain moment, the target output of Unit 1 is 65.0 MW, and the safety boundary is [30, 53] MW. That is, the target output is within the upper safe zone and does not fall into the vibration zone, and the first constraint correction result is... Megawatts, delivering power according to the original target.

[0320] Step S4.5: If the target output of the hydropower unit falls into the vibration zone, the target output of the unit is redistributed between the lower safety side and the upper safety side of the vibration zone, and the scheme with the smallest deviation from the average distribution and fewer switching times of the safety side is selected to obtain the second constraint correction result.

[0321] Preferably, the core of this step is to redistribute the output of the unit that falls into the vibration zone, redistributing the target output to the lower or upper safe side of the vibration zone, with the optimization goal of minimizing deviation from the average distribution and reducing the number of safe side switching times, while taking into account power balance and safe side switching frequency.

[0322] Specifically, for the redistribution of targets, the target output of the units is redistributed to... or No output is retained within the vibration zone.

[0323] Among them, for the goal of equal distribution, , where n is the number of generating units in operation.

[0324] Among them, for the deviation calculation, the lower deviation is... Upper deviation Choose the safe side with the smaller deviation as the target output for correction.

[0325] In handling multiple units simultaneously falling into the vibration zone, the optimization objective is to minimize the total number of safety-side switching operations, and a greedy strategy is adopted to allocate the units one by one.

[0326] Preferably, the second constraint correction result Multiple units are allocated using a greedy strategy; deviation calculation is based on the plant-wide average allocation target.

[0327] For example, if at a certain moment the target outputs of two generating units are 35.0 MW and 56.0 MW respectively, both falling within the vibration zone, and the target outputs are evenly distributed... Megawatts, Unit 1 safety side boundary [30,53] megawatts, , Select the upper safety side of 53.0 MW; Unit 2 safety side boundary [37,58] MW, The lower safety side is selected as 37.0 MW, and the second constraint correction result is 53.0 MW for Unit 1 and 37.0 MW for Unit 2.

[0328] Step S4.6: If the hydropower unit needs to cross from one side of the vibration zone to the other safe side, a constrained climbing trajectory is generated according to the safe operating rate of the guide vanes. A limiting calculation is performed on the climbing trajectory to obtain the third constraint correction result.

[0329] Preferably, the core of this step is to generate and limit the controlled crossing trajectory in the vibration zone, generate the climbing trajectory from the current safe side to the opposite side according to the guide vane safe operating rate, and during the crossing, the energy storage battery enters a full error compensation state to make up for the power gap, ensuring the safety and stability of the crossing process.

[0330] Among them, regarding the climbing rate constraint, ,in For the safety of the guide vane's operating speed, Let be the rated power of the i-th unit.

[0331] For the constrained climbing trajectory, starting from the current safe side boundary, the trajectory crosses the vibration zone at the maximum permissible climbing rate, stops upon reaching the opposite safe side boundary, and the crossing time is specified. .

[0332] Specifically, regarding energy storage compensation, the energy storage battery enters a state of full error compensation during the crossing process. This will compensate for the power shortfall caused by changes in unit output during the trip.

[0333] Preferably, the operating rate of the safety guide vane The speed is set to 1.5% / second; the climbing trajectory traverses at a constant speed at the maximum permissible rate; the result of the third constraint correction. This is the sequence of climbing trajectories after amplitude limiting.

[0334] For example, a certain unit has a rated power of 80 MW, a safety guide vane operating rate of 1.5% / second, and a maximum ramp rate. MW / s, safety side boundary [37,58] MW, ramp from the lower safety side 37 MW to 58 MW, traverse time Seconds, during the crossing, the energy storage battery fully compensates for the power gap caused by changes in the unit's output.

[0335] Step S4.7: After the unit completes the vibration zone crossing, it remains locked on the current safe side of the continuous vibration zone for a specified period of time. If a new AGC command does not require crossing to another safe side within the locked period, the unit is prohibited from re-entering the vibration zone.

[0336] Preferably, the core of this step is to complete the execution of the safety side locking mechanism. After the unit completes the crossing, it remains on the current safety side for a set time. During the locking period, it is prohibited to adjust towards the vibration zone to prevent the unit from repeatedly entering the vibration zone. If a new AGC command requires crossing, the lock is released in advance and the crossing is re-executed.

[0337] Specifically, the lock-on start time is the time when the unit reaches the opposite safety boundary. .

[0338] Among them, the duration of the lock is During the locked period The target output of the internal unit is clamped to the current safe side boundary value.

[0339] Specifically, regarding the lock release conditions, if a new AGC command is received during the lock-up period and the command requires the unit to cross to another safe side, the lock-up is released in advance and the crossing trajectory generation in step S4.6 is re-executed; if the direction of the new AGC command is consistent with the current safe side, the output adjustment is performed normally.

[0340] Preferably, the locking duration on the safe side of the vibration zone Take 90 seconds; during the lockout period, the unit's target output clamp is located at the safe side boundary value; the lockout is only released in advance when the new AGC instruction requires it to be crossed.

[0341] For example, a certain unit completes the crossover from 37 MW on the lower safety side to 58 MW on the upper safety side at time YYYY-MM-DD-HH:MM:SS-001, locking the start time. The lockout lasts for 90 seconds. During the lockout period, the unit's target output remains unchanged at 58 MW. If a new AGC command is received after 30 seconds requiring the unit to descend to the lower safety side, the lockout will be released in advance and the crossing trajectory generation will be re-executed.

[0342] Step S4.8: Integrate the first constraint correction result, the second constraint correction result, and the third constraint correction result to obtain the target output of the hydropower unit after constraint correction.

[0343] Preferably, the core of this step is to integrate the correction results of the three types of constraints, and to merge the correction results of the three cases of not falling into the vibration zone, falling into the vibration zone and redistribution, and controlled crossing according to the unit number and time, and output the target output sequence of the hydropower unit after vibration zone correction for subsequent energy storage power matching.

[0344] Among them, for the integration method, the target output of the hydropower unit after constraint correction ,in The first constraint correction result, the second constraint correction result, or the third constraint correction result are taken according to the unit status.

[0345] For state determination, the current state of each unit is determined at each sampling time: if it has not fallen into the vibration zone, the first constraint correction result is taken; if it has fallen into the vibration zone and is redistributed, the second constraint correction result is taken; if it is crossing, the third constraint correction result is taken.

[0346] Preferably, the integrated target output sequence covers from arrive All sampling times; each unit corresponds to only one constraint correction result at each time point; the integrated output is used as the input for subsequent energy storage power matching calculation.

[0347] For example, at a certain moment, Unit 1, which is not in the vibration zone, is calculated as 65.0 MW under the first constraint correction, while Unit 2, which is in the vibration zone, is calculated as 37.0 MW under the second constraint correction. After constraint correction, the target output of the hydropower units is... Megawatts.

[0348] Beneficial effects of steps S4.1 to S4.8:

[0349] This series of steps addresses the technical problem that the target output path of hydropower units is prone to falling into the vibration zone and that existing solutions lack a systematic avoidance and controlled crossing mechanism. By calculating the safety side boundary through the vibration zone boundary and safety margin, when the unit's target output falls into or crosses the vibration zone, the output is redistributed, the climbing trajectory is limited, and the safety side is locked. This enables the unit to actively avoid and control the crossing of the vibration zone, reduce the risk of equipment fatigue damage, and ensure the safe operation of the unit.

[0350] Specifically, step S4.1 extracts the vibration zone boundary, safety margin, and safety-side locking duration from the standardized phased control parameter set to provide quantitative parameters for constraint correction; step S4.2 calculates the upper and lower safety-side boundaries of each operating unit based on the vibration zone boundary and safety margin to clarify the safe output range allowed for unit operation; step S4.3 compares the layered power response benchmark value with the safety-side boundary to determine whether the unit's target output falls within the vibration zone, providing a basis for subsequent branch processing; step S4.4 executes the original target if the target output does not fall within the vibration zone to avoid unnecessary output adjustments; step S4.5 executes the target output... When the unit enters the vibration zone, the target output of the unit is redistributed between the upper and lower safety sides, and the scheme with the smallest deviation from the average distribution and fewer switching times is selected, taking into account both power balance and safety side switching frequency; Step S4.6 When the unit needs to cross the vibration zone, a constrained climbing trajectory is generated according to the guide vane safe operating rate and amplitude limiting calculation is performed to control the safety and stability of the crossing process; Step S4.7 After the crossing is completed, the safety side is locked for a set duration to prevent the unit from repeatedly entering the vibration zone during the lock period; Step S4.8 By integrating the three types of constraint correction results, the target output of the hydropower unit after vibration zone correction is output for subsequent energy storage power matching.

[0351] Step S5: Perform energy storage power matching calculation on the target output of the hydropower unit after constraint correction and the energy storage recovery parameters in the standardized phased control parameter set, and simultaneously couple the energy storage charge state boundary constraint and charge state active recovery calculation to obtain the energy storage real-time compensation power command.

[0352] Furthermore, step S5 specifically includes the following steps:

[0353] Step S5.1: Extract the upper limit of energy storage state of charge, the lower limit of energy storage state of charge, the reference value of energy storage state of charge, the energy storage state of charge recovery gain, and the maximum value of energy storage state of charge recovery power from the standardized phased control parameter set.

[0354] Preferably, the core of this step is to extract the energy storage control parameters, obtaining five parameters from the standardized phased control parameter set: upper and lower limits of SOC, reference value, recovery gain, and maximum recovery power, providing parameter basis for subsequent energy storage power matching and SOC recovery calculation.

[0355] Among them, the upper limit of the state of charge of energy storage This indicates the maximum permissible state of charge of the energy storage battery. Charging is prohibited when this value is exceeded to prevent overcharging.

[0356] Among them, the state of charge limit of energy storage This indicates the minimum permissible state of charge of the energy storage battery. Discharge operations are prohibited when the charge level is below this value to prevent over-discharge.

[0357] Among them, the reference value for the state of charge of energy storage , which represents the ideal state of charge level that the energy storage battery is expected to maintain. The goal of the SOC recovery operation is to bring the actual SOC back to this value.

[0358] Among them, for the energy storage state of charge recovery gain , defined as the mapping coefficient from SOC deviation to recovery power, determines the response speed of SOC recovery. The larger the value, the faster the recovery, but it may cause power fluctuations.

[0359] Among them, the maximum amplitude of the energy storage state of charge recovery power. Defined as the maximum allowable regulation power during SOC recovery, expressed as a percentage of the energy storage's rated power, limiting the impact of recovery operations on the grid connection point's power.

[0360] Preferably, the upper limit of the energy storage state of charge. Take 85%; Energy storage state of charge limit Take 15%; Energy storage state of charge reference value Take 50%; Energy storage state of charge recovery gain Take 0.8; Maximum amplitude of energy storage state-of-charge recovery power. Take 3 megawatts.

[0361] For example, a standardized phased control parameter set for a hydropower station , , , , Megawatts, from The extracted data is then used for subsequent energy storage power matching and SOC recovery calculations.

[0362] Step S5.2: Calculate the grid connection point power error based on the target output of the hydropower unit after constraint correction and the AGC command of the whole plant, and use the grid connection point power error as the AGC response energy storage power command.

[0363] Preferably, the core of this step is to complete the real-time calculation of the grid connection point power error and the conversion of the energy storage power command. The difference between the target output of the hydropower unit after constraint correction and the AGC command of the whole plant is used as the AGC response energy storage power command, so as to realize the real-time determination of the energy storage compensation amount.

[0364] Among them, for the grid connection point power error, ,in For the entire plant's AGC instructions, To constrain the target output of the hydropower unit after the correction.

[0365] Among them, for AGC to respond to energy storage power commands, A positive value indicates that the energy storage battery is insufficient to compensate for the power shortage during discharge, while a negative value indicates that the energy storage battery is absorbing excess power during charging.

[0366] Preferably, the sampling period for the grid connection point power error is consistent with the sampling period for the AGC command; in subsequent steps, the AGC responds to the energy storage power command and selects candidate values ​​based on the system status.

[0367] For example, at a certain moment, the plant's AGC command is 97.0 MW, and after constraint correction, the target output of the hydropower unit is 95.2 MW, with a power error at the grid connection point. Megawatts, AGC responds to energy storage power commands Megawatts (MW) means that the energy storage battery needs to discharge 1.8 MW to compensate for the power deficit.

[0368] Step S5.3: Determine whether there is a valid AGC event or a controlled crossing of the unit's vibration zone. If so, use the AGC response energy storage power command as a candidate value for real-time energy storage compensation power.

[0369] Preferably, the core of this step is to determine the priority of AGC response and select candidate values. When the system is in a valid AGC event or a controlled crossing of the vibration zone, the AGC response energy storage power command is used as a candidate value to ensure that the adjustment task has a higher priority than SOC recovery.

[0370] Specifically, the determination criteria include checking whether the current moment is within the valid AGC event activation interval. Or the vibration zone is controlled to cross the interval .

[0371] Specifically, regarding the selection of candidate values, when the judgment conditions are met, the candidate values ​​for real-time energy storage compensation power are selected. This means that the energy storage battery will prioritize performing the AGC response compensation task.

[0372] Regarding the meaning of priority, AGC response and vibration zone crossing compensation have higher priority than SOC recovery. When both are present, SOC recovery is suspended or reduced.

[0373] Preferably, the determination criteria include two types: effective AGC event activation state and controlled crossing of the vibration zone; if either condition is met, the AGC response energy storage power command is selected as a candidate value.

[0374] For example, at a certain moment, within the effective AGC event activation range, the AGC responds to the energy storage power command. Megawatts, candidate power values ​​for real-time energy storage compensation The megawatt-level SOC operation was suspended.

[0375] Step S5.4: If there is no valid AGC event and no controlled crossing of the unit vibration zone, calculate the deviation between the current state of charge of the energy storage battery and the reference value of the energy storage state of charge.

[0376] Preferably, the core of this step is to calculate the SOC deviation during the static stable period. During the static stable period without effective AGC events and without vibration zone crossing, the deviation between the current SOC of the energy storage battery and the reference value is quantified, providing input for the subsequent SOC recovery trigger determination.

[0377] Specifically, for determining the static stability period, the current moment is not within the effective AGC event activation range. It is not within the controlled crossing zone of the vibration zone.

[0378] Among them, for SOC deviation calculation, Where SOC(t) is the current state of charge of the energy storage battery. This is a reference value for the state of charge (SOC) of energy storage. A positive value indicates that the SOC is too high and needs to be discharged to revert to its previous state, while a negative value indicates that the SOC is too low and needs to be charged to revert to its previous state.

[0379] Preferably, the SOC deviation calculation sampling period is 1 second; the sign of the deviation value determines the direction of the subsequent recovery power.

[0380] For example, the state of charge of an energy storage battery at a certain moment Reference values SOC deviation This indicates that the SOC is too low and needs to be recharged.

[0381] Step S5.5: Determine whether the deviation is greater than the recovery dead zone. If it is, generate the SOC recovery power plan based on the energy storage state of charge recovery gain and the maximum amplitude of the energy storage state of charge recovery power.

[0382] Preferably, the core of this step is to complete the SOC recovery trigger determination and recovery power plan generation. The determination of whether recovery is triggered is made by comparing the deviation and the recovery dead zone. After triggering, the recovery power is calculated based on the recovery gain and deviation and the maximum amplitude constraint is applied to generate the SOC recovery power plan.

[0383] Among them, for the recovery dead zone threshold Defined as the minimum deviation range that will not trigger SOC recovery, this avoids frequent triggering of recovery operations near the reference value of SOC, with a typical value range of 2% to 5%.

[0384] Among them, regarding the triggering conditions, When the condition is met, the SOC recovery operation is triggered.

[0385] Among them, for the calculation of recovery power, The negative sign here makes the direction of the recovery power opposite to the direction of the deviation. The percentage is converted to a dimensionless coefficient by dividing by 100 to represent the rated energy storage capacity.

[0386] Among them, for the maximum amplitude constraint, The recovery power is limited to the maximum allowable amplitude.

[0387] Preferably, the dead zone threshold is recovered. Take 3%; recovery gain in the recovery power calculation formula Maximum recovery power Megawatts.

[0388] For example, SOC deviation Restoring dead zones , Trigger recovery, restore power Megawatts (charging direction). No limiting required, SOC power recovery plan Megawatts.

[0389] Step S5.6: Adjust the total target output of the hydropower unit according to the SOC recovery power plan to make the state of charge of the energy storage battery return to the reference value of the energy storage state of charge, and obtain the SOC recovery energy storage power command.

[0390] Preferably, the core of this step is to adjust the output of the hydropower unit based on the SOC recovery power plan. By adjusting the total target output of the hydropower unit, the energy storage battery performs charging or discharging operations, so that the SOC returns to the reference value, thereby achieving active maintenance of energy storage availability.

[0391] Specifically, regarding the adjustment of hydropower unit output, when the SOC is low and charging is required, This means increasing the output of the hydropower unit to provide charging power for the energy storage battery; when the SOC is too high and discharge is required... This means reducing the output of the hydropower unit to allow the energy storage battery to discharge.

[0392] Specifically, for the energy storage battery power command, the energy storage battery power command during SOC recovery... This means that the energy storage battery performs charging and discharging operations in the opposite direction to the power recovery plan.

[0393] Regarding the SOC regression effect, the SOC of the energy storage battery gradually increases during charging and gradually decreases during discharging, eventually trending towards... return.

[0394] Preferably, the power output adjustment of the hydropower unit is equal to the planned value of the SOC recovery power; the charging and discharging power of the energy storage battery is equal to the negative of the planned recovery power value; and the power output of the hydropower unit after adjustment must meet the unit output constraints.

[0395] For example, SOC power recovery plan MW (charging direction), after constraint correction, the target output of the hydropower unit is 95.2 MW. Megawatts, energy storage battery power command At 1.28 MW (charging), the energy storage battery gradually returns to 50% SOC from 42%.

[0396] Step S5.7: Arbitrate the priority of the SOC recovery power command and the AGC response power command. When the priority of the AGC response is increased, reduce or suspend the SOC recovery power to obtain the candidate value of the real-time compensation power of the energy storage.

[0397] Preferably, the core of this step is to complete the priority arbitration between AGC response and SOC recovery. When a new valid AGC event is triggered, the priority of AGC response is increased, the current SOC recovery power is reduced or suspended, and the priority execution of AGC adjustment tasks is guaranteed.

[0398] Specifically, for the arbitration logic, it monitors in real time whether there is a valid AGC event or controlled crossing of the vibration zone. When an AGC event is triggered, the candidate value of the real-time energy storage compensation power is switched from the SOC recovery power to the AGC response energy storage power command.

[0399] Specifically, for the reduction strategy, if the AGC response to the energy storage power command is in the same direction as the SOC recovery power, they are executed together; if they are in opposite directions, the SOC recovery power is paused, and the AGC response is executed first.

[0400] Specifically, for pause and resume, if the SOC deviation still exceeds the recovery dead zone after the AGC event ends, the SOC recovery operation will be automatically resumed.

[0401] Preferably, the arbitration priority is ranked as follows: AGC response is higher than SOC recovery; SOC recovery is superimposed when the directions are consistent and paused when the directions are opposite; SOC recovery determination is automatically resumed after the AGC event ends.

[0402] For example, during SOC recovery, the energy storage battery is charging at 1.28 MW. At this time, a new valid AGC event is triggered, and the AGC response energy storage power command is to discharge at 2.0 MW, in the opposite direction to the SOC recovery charging direction. SOC recovery is then paused, and the energy storage real-time compensation power candidate value switches to... Megawatt (discharge), if the SOC deviation still exceeds the recovery dead zone after the AGC event ends, the SOC recovery operation will resume.

[0403] Step S5.8: Apply the energy storage rated power constraint, the energy storage state of charge upper limit constraint, and the energy storage state of charge lower limit constraint to the energy storage real-time compensation power candidate value to obtain the energy storage real-time compensation power command.

[0404] Preferably, the core of this step is to complete the safety constraint limit of the candidate value of the real-time compensation power of the energy storage, and to ensure the safe operation of the energy storage battery by applying rated power constraints and SOC upper and lower limit constraints, and output the final real-time compensation power command of the energy storage.

[0405] Among them, for rated power constraints, The charging and discharging power of energy storage batteries must not exceed the rated power.

[0406] Among them, for the SOC upper limit constraint, when Charging is prohibited at this time. If the candidate value is the charging direction (negative value), it is set to zero.

[0407] Among them, for the lower bound constraint of SOC, when Discharge operation is prohibited at this time. If the candidate value is the discharge direction (positive value), it is set to zero.

[0408] Preferably, rated power constraint Megawatts; SOC limit SOC lower limit After constraint limiting, output the real-time energy storage compensation power command. .

[0409] For example, candidate values ​​for real-time compensation power of energy storage Megawatts (discharge), energy storage rated power 20 megawatts. The rated power constraint is met. , Meets the lower limit constraint of SOC, and provides real-time power compensation command for energy storage. Megawatts.

[0410] Beneficial effects of steps S5.1 to S5.8:

[0411] This series of steps addresses the technical problems of the lack of an active recovery mechanism for the state of charge (SOC) of energy storage batteries and the tendency for the SOC to deviate from the median region for a long time after continuous AGC events, leading to a decrease in availability. By performing energy storage power matching during the AGC response, generating a recovery power plan based on the SOC deviation during the static steady period, and adjusting the hydropower output to bring the SOC back to the reference value, and by simultaneously coupling the state of charge boundary constraints and priority arbitration, coordinated control of AGC response and SOC recovery is achieved, thereby improving the availability of continuous frequency regulation of energy storage.

[0412] Specifically, step S5.1 extracts the upper and lower limits of the energy storage state of charge (SOC), reference value, recovery gain, and maximum recovery power from the standardized phased control parameter set, providing parameter basis for power matching and SOC recovery; step S5.2 calculates the grid connection point power error based on the target output of the hydropower unit after constraint correction and the AGC command of the entire plant, and uses the power error as the AGC response energy storage power command to achieve real-time determination of the energy storage compensation amount; step S5.3 uses the AGC response energy storage power command as a candidate value when there is a valid AGC event or controlled crossing of the vibration zone, ensuring that the adjustment priority is higher than SOC recovery; step S5.4 calculates the deviation between the current SOC of the energy storage battery and the reference value during the static stability period, quantifying the SOC deviation. Step S5.5: Determine whether SOC recovery is triggered by comparing the deviation with the recovery dead zone, and generate an SOC recovery power plan based on the recovery gain and the maximum recovery power when triggered; Step S5.6: Adjust the total target output of the hydropower unit according to the SOC recovery power plan to make the state of charge of the energy storage battery return to the reference value, thereby achieving active maintenance of energy storage availability; Step S5.7: Arbitrate the priority between the SOC recovery energy storage power command and the AGC response energy storage power command, and reduce or suspend the SOC recovery power when the priority of the AGC response is increased to ensure the priority execution of the regulation task; Step S5.8: Apply rated power and upper and lower limit constraints of state of charge to the candidate values ​​of real-time compensation power for energy storage to ensure the safe operation of energy storage.

[0413] Step S6: The real-time energy storage compensation power command and the target output of the hydropower unit after constraint correction are superimposed to generate the joint output command of the grid connection point and the power closed-loop tracking calculation is performed to obtain the grid connection regulation execution signal of the whole station.

[0414] Furthermore, step S6 specifically includes the following steps:

[0415] Step S6.1: The real-time energy storage compensation power command is superimposed with the target output of the hydropower unit after constraint correction to obtain the joint output command of the grid connection point.

[0416] Preferably, the core of this step is to complete the superposition and synthesis of the combined output of water and storage, adding the real-time compensation power command of energy storage with the target output of the hydropower unit after constraint correction, generating the combined output command of the grid connection point, and realizing the unified and coordinated output of water and storage power.

[0417] Among them, for the calculation of combined output commands, ,in To constrain the target output of the hydropower unit after the correction, This is a real-time power compensation command for energy storage.

[0418] Regarding the meaning of the command, the joint output command indicates the total target value of the power output by the hydropower unit and the energy storage battery to the grid connection point, with a positive value indicating the power output to the grid.

[0419] Preferably, the sampling period of the joint output command is consistent with the sampling period of the AGC command; the joint output command serves as a reference target for subsequent closed-loop correction.

[0420] For example, at a certain moment, after constraint correction, the target output of the hydropower unit is 95.2 MW, the real-time compensation power command for energy storage is 2.0 MW, and the joint output command of the grid connection point is... Megawatts.

[0421] Step S6.2: Collect the actual output power of the grid connection point and calculate the tracking error between the combined output command of the grid connection point and the actual output power of the grid connection point.

[0422] Preferably, the core of this step is to collect the actual power of the grid connection point and calculate the tracking error. By collecting the measured power value of the grid connection point in real time and comparing it with the joint output command, the power tracking deviation is quantified, providing error input for subsequent closed-loop correction.

[0423] Specifically, the actual output power at the grid connection point is acquired through a power transmitter at the grid connection point. The transmitter accuracy class is not lower than 0.2, and the acquisition cycle is synchronized with the combined output command.

[0424] Among them, for tracking error calculation, A positive value indicates that the actual power is lower than the commanded target, and a negative value indicates that the actual power is higher than the commanded target.

[0425] Preferably, the power transmitter at the grid connection point has a data acquisition cycle of 1 second; the tracking error calculation is performed synchronously with the combined output command.

[0426] For example, at a certain moment, the joint output command of the grid-connected point is 97.2 MW, and the actual output power of the grid-connected point is 96.5 MW, resulting in a tracking error. Megawatts indicates that the actual power output is 0.7 megawatts lower than the mandated target.

[0427] Step S6.3: Determine whether the tracking error is less than the AGC assessment dead zone. If it is less, maintain the current grid connection point joint output command.

[0428] Preferably, the core of this step is to compare and determine the tracking error with the AGC assessment dead zone. When the tracking error is within the assessment dead zone range, the power tracking accuracy is considered to meet the requirements, and the current command is maintained without adjustment to avoid unnecessary adjustment actions.

[0429] Among them, the AGC assessment dead zone This is defined as the allowable power tracking error range in the power grid AGC performance assessment. No points will be deducted within this range, and it is usually taken as 0.5% to 2% of the plant's rated capacity.

[0430] Among them, regarding the judgment conditions, When the conditions are met, the current joint output command will remain unchanged.

[0431] Preferably, the AGC assessment dead zone Take 1% of the plant's rated capacity; during maintenance, the combined output command and the power command of each piece of equipment remain unchanged.

[0432] For example, the plant's rated capacity is 200 megawatts, and there is an AGC assessment dead zone. Megawatts, tracking error megawatts The current combined output command for the grid connection point remains unchanged at 97.2 MW.

[0433] Step S6.4: If the tracking error is not less than the AGC assessment dead zone, perform closed-loop correction on the combined output command of the grid connection point according to the proportional-integral control law to obtain the corrected combined output command.

[0434] Preferably, the core of this step is to complete the closed-loop correction of the combined output command. When the tracking error exceeds the assessment dead zone, the proportional-integral control law is used to eliminate the steady-state deviation of power tracking and improve the power tracking accuracy at the grid connection point.

[0435] For proportional-integral control laws, ,in For proportional gain, For integral gain, This is for tracking error.

[0436] Among them, for the revised combined output command, The proportional-integral control quantity is superimposed on the original combined output command.

[0437] Specifically, for integral anti-saturation, when the cumulative value of the integral term exceeds the rated power of the energy storage, a limit is applied to prevent overshoot caused by integral saturation.

[0438] Preferably, proportional gain Take 0.5; Integral gain Take 0.1; the limit range of the integral term is... .

[0439] For example, tracking error The MW exceeds the assessment dead zone by 2.0 MW, and the proportional gain... Integral gain proportional term Megawatts, with the integral term cumulative value assumed to be 0.6 megawatts, control quantity Megawatts, revised joint output command Megawatts.

[0440] Step S6.5: Determine whether the modified combined output command exceeds the output constraints of the hydropower unit and the rated power constraints of the energy storage battery. If it does not exceed, output directly; if it exceeds, output after limiting according to the constraint boundaries.

[0441] Preferably, the core of this step is to complete the safety constraint limit of the combined output command after correction. By judging and limiting the output constraint of the hydropower unit and the rated power constraint of the energy storage, the execution command is ensured to be within a safe range, preventing the correction amount from exceeding the equipment capacity.

[0442] Among them, the output constraint of hydropower units, ,in To contribute the least amount of technical effort to the entire factory To contribute the most technical expertise to the entire factory.

[0443] Among them, regarding the rated power constraint of energy storage, The charging and discharging power of energy storage batteries must not exceed the rated power.

[0444] Among them, for the amplitude limiting method, when the modified combined output command exceeds... The clamp is at this upper limit; when it is below... The clamp is located at this lower limit.

[0445] Preferably, the output constraint of the hydropower unit is calculated based on the sum of the minimum and maximum technical output of all operating units in the plant; the rated power constraint of the energy storage unit is... Megawatts; Combined power output command after limiting .

[0446] For example, the revised combined output command is 99.3 MW, the maximum technical output of the entire plant is 160 MW (two 80 MW units), the rated energy storage capacity is 20 MW, and the upper limit of combined output is 160 + 20 = 180 MW. The constraints are met; the minimum technical output of the entire plant is 40 MW, and the lower limit of the combined output is 40-20=20 MW. Constraints are met, and combined output command is given after amplitude limiting. Megawatts.

[0447] Step S6.6: Decompose the modified and limited combined output command into the active power command of each operating hydropower unit and the charging and discharging power command of the energy storage battery to obtain the grid-connected regulation execution signal for the entire station.

[0448] Preferably, the core of this step is to decompose the combined output command and output the execution signal. The combined output command after the limit is decomposed into the active power command of each hydropower unit in operation according to the unit capacity ratio. The remaining part is used as the charging and discharging power command of the energy storage battery, and the output is a grid-connected regulation execution signal that can be directly executed.

[0449] Among them, the decomposition of active power commands for hydropower units ,in The rated power of the i-th unit is allocated proportionally to the unit capacity.

[0450] Among them, for the charging and discharging power commands of energy storage batteries, The sum of the combined output command and the hydropower unit command is the energy storage battery power command.

[0451] Among them, for the whole station grid-connected regulation execution signal, It includes active power commands for each operational hydropower unit and charging / discharging power commands for energy storage batteries.

[0452] Preferably, the hydropower units are decomposed according to their capacity ratio; the power command of the energy storage battery is the difference between the combined output command and the hydropower unit command; the execution signal includes the power commands of all operating units.

[0453] For example, the combined output command after the limit is 99.3 MW, the target output of the hydropower unit after constraint correction is 95.2 MW, the rated power of both units is 80 MW, and the active power command of Unit 1 is... Megawatts, Active Power Command for Unit 2 Megawatts, energy storage battery charging and discharging power command Megawatts, grid-connected regulation execution signal for the entire station .

[0454] Beneficial effects of steps S6.1 to S6.6:

[0455] This series of steps addresses the technical problems of insufficient power tracking accuracy at the grid connection point and the lack of closed-loop correction and constraint limiting in the combined power output command of the water-storage system. By superimposing the real-time compensation power command of the energy storage and the target output of the hydropower unit after constraint correction, a combined power output command is generated. A proportional-integral control law is used for closed-loop correction and constraint limiting to achieve high-precision tracking of the grid connection point power and safe output of the station-wide regulation execution signal.

[0456] Specifically, step S6.1 involves superimposing the real-time energy storage compensation power command with the target output of the hydropower units after constraint correction to obtain the joint output command at the grid connection point, thereby achieving unified and coordinated output of hydropower and storage power. Step S6.2 involves collecting the actual output power at the grid connection point and calculating the tracking error between it and the joint output command to quantify the power tracking deviation. Step S6.3 involves comparing the tracking error with the AGC assessment dead zone and maintaining the current command when the error meets the accuracy requirements to avoid unnecessary adjustment actions. Step S6.4 involves performing closed-loop correction of the joint output command according to the proportional-integral control law when the tracking error exceeds the assessment dead zone to eliminate the steady-state deviation of power tracking. Step S6.5 involves judging and limiting the output and rated power constraints of the corrected joint output command to ensure that the execution command is within a safe range. Step S6.6 involves decomposing the corrected and limited joint output command into the active power command of each operational hydropower unit and the charging and discharging power command of the energy storage battery, and outputting a directly executable grid-connected regulation execution signal for the entire station.

[0457] Figures 2 to 4 The paper provides the full-day AGC response, power distribution, and SOC curves of the three methods during the dry season.

[0458] Among them, M1 (fixed ratio allocation method): each valid AGC command is shared by hydropower and energy storage in a fixed ratio; no parameter optimization is performed, and no active SOC recovery or controlled crossing of the vibration zone is implemented.

[0459] M2 (Hydropower Priority Capacity Deficit Compensation Method): Hydropower priority response; energy storage compensates for capacity deficits only when the change in command exceeds the preset safety dynamic capacity of hydropower.

[0460] M3 (the method of this invention): adopts a phased response of rapid response, energy maintenance and precise regulation; combined with controlled rapid crossing of the vibration zone; and in the static steady period, the SOC is brought back to near the reference median by water storage synergy.

[0461] Specifically, M1 gradually consumes electricity with continuous response; M2 only puts energy storage into operation when hydropower capacity is insufficient, and basically maintains the initial low SOC on a typical day; M3 uses the interval of AGC command to implement constrained SOC recovery, so that energy storage returns to near the median value.

[0462] Figure 5 The diagram shows the AGC event response for a certain AGC event. It can be seen that the energy storage cooperates with the hydropower in stages to respond to AGC. The energy storage first compensates for the dynamic lag of the hydropower, and then the hydropower follows at a safe rate.

[0463] Specifically, the key indicators are shown in Table 1 below.

[0464] Table 1. Comparison of Key Indicators During the Dry Season:

[0465] M1 1.624 0.406 1.000 0.648 0.987 0.206 0.187 0.00% 384 0.4444% M2 1.348 0.239 1.000 0.318 2.307 0.200 0.200 0.00% 384 0.4444% M3 1.957 0.458 1.000 0.903 0.690 0.584 0.594 91.53% 22 0.0255%

[0466] The comparison shows that M3's comprehensive performance index K is 0.903, higher than M1's 0.648 and M2's 0.318. Based on the benchmark electricity market price, the daily equivalent assessment cost is reduced by approximately 30.1% compared to M1 and by approximately 70.1% compared to M2. Simultaneously, the cumulative dwell time in the vibration zone decreased from 384s to 22s. M3's average SOC is 0.584, and its end-of-day value is 0.594. It spends 91.53% of the day within the median range of [0.4, 0.6], indicating its ability to continuously participate in subsequent AGC and respond to power system emergencies.

[0467] Figures 6 to 8 It provides full-day AGC response, power allocation, and SOC curves for three methods under the following scenarios: four units are put into operation during the high-water season, more effective AGC events occur, and continuous regulation capability is tested.

[0468] Figure 9 This is a diagram showing the AGC event response for a specific AGC event.

[0469] As shown in the figure, the SOC of M1 is close to the lower limit for a long time; although M2 does not consume energy storage, its SOC remains at a low level of 0.20; while M3 recovers its SOC within the static stability window.

[0470] Specifically, the key indicators are shown in Table 2.

[0471] Table 2 Comparison of Key Indicators During High Water Season:

[0472] M1 1.053 0.386 1.000 0.380 3.726 0.108 0.134 0.00% 2887 3.3414% M2 0.854 0.240 1.000 0.192 6.941 0.200 0.200 0.00% 2887 3.3414% M3 1.646 0.460 1.000 0.734 3.326 0.565 0.607 91.28% 98 0.1134%

[0473] The comparison shows that M3's K is 0.734, higher than M1's 0.380 and M2's 0.192. Based on the benchmark electricity market price, the daily equivalent assessment cost is reduced by approximately 10.7% and 52.1% compared to M1 and M2, respectively. M1's average SOC is only 0.108, with a day-end value of 0.134, indicating that a fixed-ratio allocation will keep energy storage close to its lower limit for a long time. M2 is in a low SOC state, lacking the available discharge capacity required for subsequent emergency load increases. M3's average SOC is 0.565, with a day-end value of 0.607, and a median dwell time of 91.28%. It rapidly crosses the vibration zone, with a cumulative dwell time of only 98 seconds, significantly less than M1 and M2. Therefore, M3 transforms the unavoidable crossing into a short-term, constrained, safe crossing.

[0474] In summary, the two typical daily implementation examples demonstrate that the method of the present invention can synergistically improve existing water-storage coordinated AGC control systems by addressing issues such as "hydraulic response lag, long-term low SOC operation of energy storage, excessive dwell time in vibration zones, and insufficient continuous response capability." Specifically, at the initial stage of an effective AGC command, energy storage is used to quickly compensate for the lag in hydropower regulation, improving the tracking performance of combined output and reducing daily equivalent assessment costs. When hydropower output needs to cross vibration zones, controlled rapid crossing is implemented based on the safe ramp rate and maximum crossing time, preventing the unit from continuously operating within the vibration zone. During the AGC command interval, water-storage power coordination restores the energy storage SOC from a low initial value to near the reference median value, preserving subsequent discharge and charging margins. Compared with the fixed ratio allocation method M1 and the hydropower priority capacity deficit compensation method M2, the method of the present invention achieves higher comprehensive regulation performance index K, lower daily equivalent assessment cost, shorter cumulative residence time in the vibration zone, and a SOC median zone residence rate of over 91% during both dry and wet seasons. It effectively solves the problem of long-term deviation of energy storage SOC from the reference state and decline in continuous frequency regulation capability caused by traditional water-storage joint operation methods.

[0475] In summary, the overall beneficial effects of steps S1 to S6 in this embodiment are as follows:

[0476] This method addresses the technical problems of existing hydropower-storage combined AGC control strategies, such as the lack of phased design according to the response process time scale, the lack of active recovery mechanism for energy storage state of charge, and the lack of systematic avoidance measures for unit vibration zones. It adopts the NSGA-II multi-objective genetic algorithm to optimize the standardized phased control parameter set offline. Through the coordinated allocation of three stages—rapid response, energy maintenance, and precise regulation—it achieves power division optimization between hydropower and energy storage at different time scales. Simultaneously, it couples vibration zone constraint correction and active recovery calculation of energy storage state of charge, thereby achieving the technical effects of improving the overall regulation performance of AGC, enhancing the continuous availability of energy storage, and reducing the dwell time in the unit vibration zone.

[0477] In step S1, the parameters of the hydropower-storage coupling model are identified and solved using the NSGA-II multi-objective genetic algorithm. This outputs a standardized set of phased control parameters that takes into account AGC response performance, state of charge deviation, and residence time in the vibration zone, thus achieving systematic tuning of the control parameters. Step S2 identifies valid AGC events by performing threshold judgment calculations on real-time AGC power commands, eliminating short-term fluctuation interference and improving the accuracy of subsequent response triggering. Step S3 divides the time sequence into three stages: fast response, energy maintenance, and precise adjustment. This allows the energy storage battery to prioritize compensating for the dynamic lag of hydropower during the fast response stage and retain capacity during the energy maintenance stage. In the precise adjustment stage, residual errors are eliminated to fully leverage the complementary advantages of water and storage. Step S4, through vibration zone constraint correction and safety side locking, achieves controlled crossing and trajectory limiting when the unit's target output path crosses the vibration zone, reducing the risk of equipment fatigue damage. Step S5, through energy storage power matching calculation, synchronously couples state of charge boundary constraints and active recovery calculation, and performs priority arbitration between AGC response and SOC recovery, enabling the energy storage state of charge to return to the reference value and improving the availability of continuous frequency regulation. Step S6, through power closed-loop tracking calculation, performs closed-loop correction and constraint limiting on the joint output command, improving the power tracking accuracy at the grid connection point.

[0478] like Figure 10 As shown, this embodiment provides an example of a water-storage combined AGC phased response system. In this embodiment, the water-storage combined AGC phased response system is applied to the water-storage combined AGC phased response method as described in the above embodiment.

[0479] Specifically, the hydropower-storage combined AGC phased response system includes a standardized phased control parameter set acquisition module 1, which is electrically or communicatively connected in sequence; an effective AGC event characteristic sequence acquisition module 2; a hydropower-storage tiered power response benchmark value acquisition module 3; a target output acquisition module for hydropower units after constraint correction 4; a real-time energy storage compensation power command acquisition module 5; and a whole-station grid-connected regulation execution signal acquisition module 6.

[0480] The standardized phased control parameter set acquisition module 1 is used to acquire real-time operation data of the hydro-storage combined system and historical operation datasets of the units, and input them into the hydro-storage coupling model for parameter identification. The parameter-identified hydro-storage coupling model is then input into the NSGA-II multi-objective genetic algorithm for solution, resulting in the standardized phased control parameter set. The effective AGC event feature sequence acquisition module 2 is used to identify instructions from the standardized phased control parameter set and perform threshold judgment calculations on real-time AGC power instructions, obtaining a fully labeled effective AGC event feature sequence. The hydro-storage tiered power response benchmark value acquisition module 3 is used to perform three-stage collaborative allocation of the effective AGC event feature sequence and the standardized phased control parameter set, dividing it into a rapid response stage, an energy maintenance stage, and a precise adjustment stage according to time sequence, to obtain the hydro-storage tiered power response benchmark value. The target output acquisition module 4 for the constrained hydropower unit is used to perform vibration zone constraint correction on the benchmark value of the stratified power response of hydropower energy storage and the vibration zone constraint parameters in the standardized phased control parameter set, so as to complete the unit output safety side locking and the crossing trajectory amplitude limit calculation, and obtain the target output of the constrained hydropower unit. The real-time energy storage compensation power command acquisition module 5 is used to perform energy storage power matching calculation on the target output of the constrained hydropower unit and the energy storage recovery parameters in the standardized phased control parameter set, and simultaneously couple the energy storage state of charge boundary constraint and the active state of charge recovery calculation to obtain the real-time energy storage compensation power command. The whole station grid connection regulation execution signal acquisition module 6 is used to superimpose the real-time energy storage compensation power command and the target output of the constrained hydropower unit to generate the grid connection point joint output command and perform power closed-loop tracking calculation to obtain the whole station grid connection regulation execution signal.

[0481] like Figure 11 As shown, the electronic device 7 includes a processor 71 and a memory 72 coupled to the processor 71.

[0482] The memory 72 stores program instructions for implementing the water storage combined AGC phased response method of any of the above embodiments.

[0483] The processor 71 is used to execute program instructions stored in the memory 72 to perform a phased response of the water storage combined AGC.

[0484] The processor 71 can also be referred to as a CPU (Central Processing Unit). The processor 71 may be an integrated circuit chip with signal processing capabilities. The processor 71 can also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. A general-purpose processor can be a microprocessor or any conventional processor.

[0485] Furthermore, Figure 12 This is a schematic diagram of the structure of a storage medium according to an embodiment of this application. See also: Figure 12 In this embodiment of the application, the storage medium 8 stores program instructions 81 capable of implementing all the above methods. These program instructions 81 can be stored in the storage medium in the form of a software product, including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks, or terminal devices such as computers, servers, mobile phones, and tablets.

[0486] In the embodiments provided in this application, it should be understood that the disclosed systems, methods, and approaches can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For instance, the phased response of the water storage combined AGC is only one logical function of the water storage combined AGC phased response. In actual implementation, there may be other water storage combined AGC phased response methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection of the system or unit may be electrical, mechanical, signal, or other forms.

[0487] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units described above can be implemented in hardware or as software functional units. The above are merely embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made based on the description and drawings of this application, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A phased response method for water storage combined with AGC, characterized in that, The water-storage combined AGC phased response method includes: Step S1: Obtain real-time operation data of the water-storage combined system and historical operation data of the unit, and input them into the water-power-storage coupling model for parameter identification. Input the parameter-identified water-power-storage coupling model into the NSGA-II multi-objective genetic algorithm for solution to obtain a standardized staged control parameter set. Step S2: Perform instruction recognition on the standardized phased control parameter set and perform threshold determination calculation on the real-time AGC power instruction to obtain a complete and valid AGC event feature sequence. Step S3: Perform three-stage collaborative allocation of the effective AGC event feature sequence and the standardized phased control parameter set, and divide the sequence into a rapid response stage, an energy maintenance stage, and a precise adjustment stage to obtain the hydropower storage stratified power response benchmark value. Step S4: Perform vibration zone constraint correction on the reference value of the stratified power response of the hydropower storage and the vibration zone constraint parameters in the standardized phased control parameter set to complete the unit output safety side locking and the crossing trajectory amplitude limit calculation, and obtain the target output of the hydropower unit after constraint correction. Step S5: Perform energy storage power matching calculation on the target output of the hydropower unit after constraint correction and the energy storage recovery parameters in the standardized phased control parameter set, and simultaneously couple the energy storage state of charge boundary constraint and the active recovery calculation of the state of charge to obtain the real-time energy storage compensation power command. Step S6: The real-time energy storage compensation power command is superimposed with the target output of the constrained hydropower unit to generate a joint output command for the grid connection point, and a power closed-loop tracking calculation is performed to obtain the grid connection regulation execution signal for the entire station.

2. The water-storage combined AGC phased response method according to claim 1, characterized in that, Step S1: Obtain real-time operating data of the hydro-storage combined system and historical operating data of the units, and input them into the hydro-storage coupling model for parameter identification. Then, input the parameter-identified hydro-storage coupling model into the NSGA-II multi-objective genetic algorithm to solve for the standardized staged control parameter set, including: Step S1.1: Collect the plant-wide AGC commands of the water storage combined system, the actual output of each operating hydropower unit, the guide vane opening, the unit's operating status, the energy storage battery power, and the energy storage battery's state of charge, and integrate them to obtain real-time operating data; Step S1.2: Collect the equivalent time constant of the hydropower unit speed governor, the equivalent time constant of the guide vane actuator, the equivalent time constant of the water flow inertia, the boundary of the unit vibration zone, the rated power of the energy storage battery, and the rated energy of the energy storage battery, and integrate them to obtain the historical operation dataset of the unit; Step S1.3: Input the real-time operation data and the historical operation dataset of the unit into the hydropower storage coupling model, and use the series inertial link to equivalently describe the governor, guide vane actuator and water flow inertial dynamic process of the hydropower unit to obtain the model to be identified; Step S1.4: Use hydropower station test data or historical operation data to identify the parameters of the model to be identified and verify the accuracy of the model to be identified, so as to obtain the hydropower storage coupling model after parameter identification. Step S1.5: Input the hydropower storage coupling model into the NSGA-II multi-objective genetic algorithm, and use the safe guide vane action rate, the exit threshold of the rapid response stage, the re-injection ratio threshold of the precise adjustment stage, and the compensation coefficient of the precise adjustment stage as optimization variables to obtain the first set of optimization variables; Step S1.6: Using the vibration zone safety side lock-in duration, energy storage state of charge recovery gain, and energy storage state of charge recovery power maximum amplitude as optimization variables, the second set of optimization variables is obtained; Step S1.7: Using the maximization of AGC comprehensive regulation performance index, the minimization of energy storage state of charge deviation, the minimization of the cumulative residence time in the vibration zone of the unit, and the minimization of the controlled crossing number of the vibration zone as objective functions, a multi-objective optimization model is constructed for the first set of optimization variables and the second set of optimization variables. Step S1.8: Apply energy storage power constraints, energy storage state of charge constraints, hydropower output constraints, guide vane safety opening and closing constraints, and vibration zone crossing time constraints to the multi-objective optimization model, and solve and output the Pareto solution set. Step S1.9: Select representative parameter groups from the Pareto solution set that preferentially meet the AGC response performance and whose energy storage state of charge deviation is less than a preset deviation threshold, and format them into a standardized phased control parameter set.

3. The water-storage combined AGC phased response method according to claim 1, characterized in that, Step S2 involves identifying instructions from the standardized phased control parameter set and performing threshold determination calculations on the real-time AGC power instructions to obtain a fully labeled valid AGC event feature sequence, including: Step S2.1: Extract the effective AGC instruction change threshold and minimum duration from the standardized phased control parameter set; Step S2.2: Obtain the real-time AGC power command at the current sampling time, and calculate the change in the real-time AGC power command relative to the previous sampling time; Step S2.3: Determine whether the change amount is greater than the change dead zone of the valid AGC instruction. If it is greater, mark it as a candidate AGC event. Step S2.4: Determine whether the duration of the candidate AGC event is not less than the minimum duration. If the minimum duration is met, the event is identified as a valid AGC event. Step S2.5: Record the start time, adjustment direction, command change amplitude, initial output of the hydropower unit, and initial state of charge of the energy storage battery of the effective AGC event to obtain a complete and labeled feature sequence of effective AGC events.

4. The water-storage combined AGC phased response method according to claim 1, characterized in that, Step S3 involves performing a three-stage collaborative allocation between the effective AGC event feature sequence and the standardized phased control parameter set, dividing the sequence into a rapid response stage, an energy maintenance stage, and a precise adjustment stage, to obtain the tiered power response benchmark value for hydropower storage, including: Step S3.1: Extract the exit threshold of the rapid response stage, the re-investment ratio threshold of the precise adjustment stage, and the compensation coefficient of the precise adjustment stage from the standardized phased control parameter set. Step S3.2: Determine the event start time and adjustment direction based on the effective AGC event characteristic sequence and enter the rapid response stage. Use the energy storage battery to compensate for the grid connection point power error according to the rated power and use the hydro-generator unit to adjust the power output to the target according to the allowable guide vane action rate. Step S3.3: Determine whether the output change of each operating hydropower unit reaches the exit threshold of the rapid response stage. If it does, exit the rapid response stage and obtain the rapid response stage end signal. Step S3.4: According to the signal indicating the end of the rapid response phase, the energy maintenance phase begins, in which the hydropower unit undertakes the main power regulation task and the energy storage battery stops the conventional power compensation. Step S3.5: Calculate the output completion ratio of the hydropower unit and determine whether the output completion ratio of the hydropower unit is greater than the re-start ratio threshold of the precise adjustment stage. If it is greater, exit the energy maintenance stage and obtain the energy maintenance stage end signal. Step S3.6: According to the end signal of the energy maintenance stage, enter the precise adjustment stage, and eliminate the residual power error at the grid connection point by means of the energy storage battery according to the compensation coefficient of the precise adjustment stage; Step S3.7: Integrate the target output sequence of hydropower units and the power sequence of energy storage batteries in the rapid response stage, energy maintenance stage, and precise regulation stage to obtain the stratified power response benchmark value of hydropower energy storage.

5. The water-storage combined AGC phased response method according to claim 1, characterized in that, Step S4 involves performing vibration zone constraint correction on the stratified power response benchmark value of the hydropower storage system and the vibration zone constraint parameters in the standardized phased control parameter set to complete the unit output safety side locking and trajectory limiting calculation, thereby obtaining the target output of the hydropower unit after constraint correction, including: Step S4.1: Extract the vibration zone boundary, vibration zone safety margin, and vibration zone safety side locking duration from the standardized phased control parameter set; Step S4.2: Calculate the lower safety boundary and upper safety boundary of the vibration zone for each operational hydropower unit based on the vibration zone boundary and vibration zone safety margin. Step S4.3: Compare the reference value of the stratified power response of the hydropower storage with the lower safe side boundary and the upper safe side boundary of the vibration zone of each operational hydropower unit, and determine whether the target output of the hydropower unit falls into the vibration zone. Step S4.4: If the target output of the hydropower unit does not fall into the vibration zone, then execute according to the original target output to obtain the first constraint correction result; Step S4.5: If the target output of the hydropower unit falls into the vibration zone, the target output of the unit is redistributed between the lower safety side and the upper safety side of the vibration zone, and the scheme with the smallest deviation from the average distribution and fewer switching times of the safety side is selected to obtain the second constraint correction result. Step S4.6: If the hydropower unit needs to cross from one side of the vibration zone to the other safe side, a constrained climbing trajectory is generated according to the safe operating rate of the guide vanes. A limiting calculation is performed on the climbing trajectory to obtain the third constraint correction result. Step S4.7: After the unit completes the crossing of the vibration zone, it remains on the current safe side for the duration of the vibration zone safety side lock. If a new AGC command does not require crossing to another safe side within the lock duration, the unit is prohibited from re-entering the vibration zone. Step S4.8: Integrate the first constraint correction result, the second constraint correction result, and the third constraint correction result to obtain the target output of the hydropower unit after constraint correction.

6. The water-storage combined AGC phased response method according to claim 1, characterized in that, Step S5 involves performing energy storage power matching calculations on the constrained target output of the hydropower unit and the energy storage recovery parameters in the standardized phased control parameter set, and simultaneously coupling the energy storage state-of-charge boundary constraints and active state-of-charge recovery calculations to obtain real-time energy storage compensation power commands, including: Step S5.1: Extract the upper limit of energy storage state of charge, the lower limit of energy storage state of charge, the reference value of energy storage state of charge, the recovery gain of energy storage state of charge, and the maximum value of the recovery power of energy storage state of charge from the standardized phased control parameter set. Step S5.2: Calculate the grid connection point power error based on the target output of the hydropower unit after constraint correction and the AGC command of the whole plant, and use the grid connection point power error as the AGC response energy storage power command; Step S5.3: Determine whether there is a valid AGC event or a controlled crossing of the unit's vibration zone. If so, use the AGC response energy storage power command as a candidate value for real-time energy storage compensation power. Step S5.4: If there is no valid AGC event and no controlled crossing of the unit vibration zone, calculate the deviation between the current state of charge of the energy storage battery and the reference value of the energy storage state of charge. Step S5.5: Determine whether the deviation is greater than the recovery dead zone. If it is, generate a SOC recovery power plan based on the energy storage state of charge recovery gain and the maximum value of the energy storage state of charge recovery power. Step S5.6: Adjust the total target output of the hydropower unit according to the SOC recovery power plan, so that the state of charge of the energy storage battery returns to the reference value of the energy storage state of charge, and obtain the SOC recovery energy storage power command. Step S5.7: Arbitrate the priority of the SOC recovery energy storage power command and the AGC response energy storage power command. When the priority of the AGC response is increased, reduce or suspend the SOC recovery power to obtain the candidate value of real-time energy storage compensation power. Step S5.8: Apply the energy storage rated power constraint, the energy storage state of charge upper limit constraint, and the energy storage state of charge lower limit constraint to the energy storage real-time compensation power candidate value to obtain the energy storage real-time compensation power command.

7. The water-storage combined AGC phased response method according to claim 1, characterized in that, Step S6: The real-time energy storage compensation power command is superimposed with the constrained corrected target output of the hydropower unit to generate a joint output command for the grid connection point, and power closed-loop tracking calculation is performed to obtain the grid connection regulation execution signal for the entire station, including: Step S6.1: The real-time energy storage compensation power command is superimposed with the target output of the constrained hydropower unit to obtain the joint output command of the grid connection point; Step S6.2: Collect the actual output power of the grid connection point and calculate the tracking error between the combined output command of the grid connection point and the actual output power of the grid connection point; Step S6.3: Determine whether the tracking error is less than the AGC assessment dead zone. If it is less, maintain the current grid connection point joint output command. Step S6.4: If the tracking error is not less than the AGC assessment dead zone, perform closed-loop correction on the grid connection point joint output command according to the proportional-integral control law to obtain the corrected joint output command. Step S6.5: Determine whether the modified combined output command exceeds the output constraints of the hydropower unit and the rated power constraints of the energy storage battery. If it does not exceed, output it directly; if it exceeds, output it after limiting it according to the constraint boundary. Step S6.6: Decompose the modified and limited combined output command into the active power command of each operating hydropower unit and the charging and discharging power command of the energy storage battery to obtain the grid-connected regulation execution signal for the entire station.

8. A water-storage integrated AGC phased response system, wherein the water-storage integrated AGC phased response system is applied to the water-storage integrated AGC phased response method as described in any one of claims 1 to 7, characterized in that, The water storage combined AGC phased response system includes: The standardized phased control parameter set acquisition module is used to acquire real-time operation data of the water-storage combined system and historical operation data of the unit, and input them into the water-power-storage coupling model for parameter identification. The water-power-storage coupling model after parameter identification is then input into the NSGA-II multi-objective genetic algorithm to solve the standardized phased control parameter set. The effective AGC event feature sequence acquisition module is used to identify instructions from the standardized phased control parameter set and perform threshold judgment calculations on the real-time AGC power instructions to obtain a complete and labeled effective AGC event feature sequence. The hydropower storage tiered power response benchmark value acquisition module is used to perform three-stage collaborative allocation of the effective AGC event feature sequence and the standardized phased control parameter set, and divides the rapid response stage, energy maintenance stage and precise adjustment stage according to the time sequence to obtain the hydropower storage tiered power response benchmark value. The constrained correction target output acquisition module is used to perform vibration zone constraint correction on the hydropower storage layered power response benchmark value and the vibration zone constraint parameters in the standardized phased control parameter set, so as to complete the unit output safety side locking and the crossing trajectory amplitude limit calculation, and obtain the constrained correction target output of the hydropower unit. The real-time energy storage compensation power command acquisition module is used to perform energy storage power matching calculation between the target output of the hydropower unit after the constraint correction and the energy storage recovery parameters in the standardized phased control parameter set, and simultaneously couple the energy storage state of charge boundary constraint and the active recovery calculation of the state of charge to obtain the real-time energy storage compensation power command. The whole-station grid-connected regulation execution signal acquisition module is used to superimpose the real-time energy storage compensation power command with the target output of the constrained hydropower unit to generate a joint output command at the grid connection point and perform power closed-loop tracking calculation to obtain the whole-station grid-connected regulation execution signal.

9. An electronic device, characterized in that, The method includes a processor and a memory coupled to the processor, the memory storing program instructions executable by the processor; when the processor executes the program instructions stored in the memory, it implements the water-storage combined AGC phased response method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores program instructions that, when executed by a processor, enable the water-storage combined AGC phased response method as described in any one of claims 1 to 7.

Images