A dynamic adjustment method of an electricity storage system

Abstract

The application provides a kind of power storage system dynamic adjustment method, comprising: real-time acquisition of the operating state parameters of power storage system, power grid operating parameter and electric power market transaction parameter;Based on the above parameters, a multi-objective optimization function is established, and the battery life extension target, power grid stability target and economic benefit target are fused to realize collaborative optimization;Based on the function, generate charge-discharge control strategy, use rolling optimization framework periodic update strategy and embed fuzzy decision tree to discretize continuous variables into executable instructions, and trigger frequency modulation auxiliary service when power grid frequency exceeds limit;Finally, execute the strategy through power conversion device, use feedforward feedback compound control and carrier phase-shift technology to suppress harmonic interference, realize the balanced goal of life extension, power grid support strengthening and market revenue maximization.

Description

Technical Field

[0001] This invention relates to the field of energy storage control, and more particularly to a dynamic adjustment method for an energy storage system. Background Technology

[0002] Energy storage systems are increasingly used in grid peak shaving, frequency regulation, and electricity market transactions. However, traditional regulation methods struggle to coordinate and optimize system lifespan, grid stability, and economic benefits. Existing technologies often handle battery aging constraints, grid fluctuation responses, and market trading strategies independently, leading to accelerated aging of energy storage systems during frequent charging and discharging, or delayed responses when grid frequency exceeds limits. Furthermore, they fail to dynamically capture arbitrage opportunities arising from electricity price fluctuations. Especially under the backdrop of electricity market reforms, the complex coupling between real-time price jumps, ancillary service compensation mechanisms, and load demand response further exacerbates the difficulty of multi-objective coordinated optimization, necessitating a global optimization method that integrates equipment condition monitoring, grid dynamic response, and market decision-making. Summary of the Invention

[0003] This invention proposes a dynamic adjustment method for an energy storage system, comprising: S1. Real-time acquisition of operating status parameters of the energy storage system, power grid operating parameters, and electricity market transaction parameters; S2. Establish a multi-objective optimization function based on operating status parameters, power grid operating parameters, and electricity market transaction parameters; S3. Generating charging and discharging control strategies for energy storage systems based on multi-objective optimization functions; S4. Power regulation by executing the charge and discharge control strategy through the power conversion device of the energy storage system.

[0004] Specifically, collecting the operating status parameters of the energy storage system includes: obtaining operating status parameters through the management unit of the energy storage system; synchronizing grid operating parameters from the grid dispatch interface; and parsing the power trading data stream to extract power market trading parameters.

[0005] Specifically, establishing the multi-objective optimization function includes: An objective function for extending the lifespan of an energy storage system is constructed, with the cycle number, internal resistance change, and temperature effect among the operating parameters as aging constraints. The aging constraints limit the maximum depth of charge and discharge and the rate of charge / discharge. A target function for improving power grid stability is constructed, and a dynamic weight model is established based on frequency deviation, voltage fluctuation, and load forecast in the power grid operating parameters. The dynamic weight model increases the weight coefficient when the frequency exceeds the limit. Construct an economically optimal objective function, and combine it with real-time electricity price, service compensation price, and response incentives from electricity market trading parameters to construct a revenue function; The Pareto front algorithm is used to normalize the lifetime extension objective function, the power grid stability objective function, and the economic objective function into a cost function.

[0006] Specifically, the generation charge-discharge control strategy includes: A rolling optimization framework is adopted to periodically update the strategy. When a preset trigger condition is met, the multi-objective optimization function is re-solved. The trigger condition includes a time interval threshold and a sudden power grid state event. Embed the maximum acceptable charge / discharge rate threshold derived from operating state parameters into the charge / discharge power constraint; When the frequency deviation in the power grid operating parameters exceeds the safety limit, a frequency regulation ancillary service strategy is generated first, and the power output priority is allocated according to the frequency regulation compensation unit price in the power market transaction parameters. The continuous policy variables are discretized into an executable instruction set by using a fuzzy decision tree. The instruction set includes constant current charging, constant voltage charging, constant power discharging, and emergency power support operation modes.

[0007] Specifically, the discretization using fuzzy decision trees includes: Establish a rule base for selecting charging and discharging modes, with the rule antecedent being the Cartesian product of the grid frequency deviation rate and the electricity price fluctuation range; Design a membership function to quantify the impact of operational state parameters on rule weights; The output command includes a timestamp and a confidence level assessment.

[0008] Specifically, the normalization using the Pareto front algorithm includes: The quantitative indicator for the life extension target is defined as the normalized value of the inverse of the cumulative change in aging factors; The power grid stability target is quantified as a weighted sum of the integral square of the frequency deviation and the peak-to-valley difference of the voltage fluctuation; The economic objective is transformed into the algebraic sum of the charging and discharging price difference revenue and service revenue per unit time period; A genetic algorithm with an elitist retention strategy is used to solve the three-dimensional target space; The solution that is closest to the ideal point is selected as the cost function output.

[0009] Specifically, the parsing of the power transaction data stream includes: Subscribe to the nodal marginal electricity price data stream in the electricity spot market and the service market clearing result data stream; Establish a data validity verification mechanism that triggers a manual confirmation process when a price jump exceeds a threshold. Different time-scale parameters are integrated through a time series alignment module, and a sliding window interpolation method is used to compensate for acquisition delay. Feature extraction is performed on the demand response incentive signals in the electricity market trading parameters to identify eligible load transfer periods.

[0010] Specifically, the power regulation performed by the power conversion device includes: The charge / discharge control strategy is converted into a modulation signal sequence for the power conversion device; The DC bus voltage fluctuation is adjusted by using a feedforward-feedback composite control. The harmonic content of the grid-connected current is suppressed by carrier phase shifting technology.

[0011] The present invention also proposes a computer-readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the aforementioned dynamic adjustment method for the energy storage system.

[0012] This invention also proposes a dynamic adjustment system for an energy storage system, comprising: The parameter acquisition module is used to acquire real-time operating status parameters of the energy storage system, grid operating parameters, and electricity market transaction parameters. The optimization calculation module is used to establish a multi-objective optimization function based on the above parameters and generate a charge-discharge control strategy. The power execution module, which includes a power conversion device and a drive circuit, is used to convert the strategy into power regulation commands and execute them.

[0013] This invention overcomes the limitations of traditional single-objective optimization by achieving triple synergistic optimization through multi-dimensional parameter fusion and a dynamic decision-making mechanism. Specifically, it includes: dynamically limiting charge / discharge rates using aging constraints, and extending battery life by combining temperature and internal resistance changes; automatically prioritizing stability objectives when grid frequency exceeds limits using a dynamic weighting model, enhancing system anti-disturbance capabilities; constructing an economic objective function based on real-time electricity prices and service compensation, simultaneously optimizing price arbitrage and ancillary service revenue. A rolling optimization framework with periodic updates is employed to respond to sudden grid events and embed fuzzy decision trees to discretize commands, improving control real-time performance. The power conversion device suppresses harmonic interference through composite modulation technology, ensuring power quality. Ultimately, it achieves a balance between extending the lifespan of the energy storage system, enhancing grid support capabilities, and maximizing market returns. Attached Figure Description

[0014] Figure 1 This is a flowchart illustrating a dynamic adjustment method for an energy storage system proposed in this invention. Detailed Implementation

[0015] refer to Figure 1 This invention proposes a dynamic adjustment method for an energy storage system, comprising: S1. Real-time acquisition of operating status parameters of the energy storage system, power grid operating parameters, and electricity market transaction parameters.

[0016] S2. Establish a multi-objective optimization function based on operating status parameters, power grid operating parameters, and electricity market transaction parameters.

[0017] S3. Generate charging and discharging control strategies for energy storage systems based on multi-objective optimization functions.

[0018] S4. Power regulation by executing the charge and discharge control strategy through the power conversion device of the energy storage system.

[0019] Specifically, the energy storage system is an integrated device for storing and releasing electrical energy; the dynamic regulation method is a control process that optimizes the behavior of the energy storage system based on real-time conditions; the operating status parameters are indicators that reflect the internal condition of the energy storage system; the grid operating parameters are data that describe the current operating state of the grid; the electricity market trading parameters are economic information related to electricity trading; the multi-objective optimization function is a mathematical function that simultaneously optimizes multiple objectives; the charging and discharging control strategy is a set of instructions that determines the charging or discharging mode of the energy storage system; the power conversion device is a hardware device that converts electrical energy into its form; and the power regulation is the process of adjusting the output or input of electrical energy.

[0020] Specifically, the system first collects real-time operating status parameters of the energy storage system, including the battery's state of charge and temperature changes; simultaneously, it collects grid operating parameters, such as grid frequency fluctuations and voltage levels; and obtains electricity market transaction parameters, such as real-time electricity prices and incentive compensation information. Then, based on these parameters, a multi-objective optimization function is established, integrating lifetime extension, grid stability, and economic benefit objectives. Based on this function, a charge / discharge control strategy is generated, specifying the charging rate or discharging power of the energy storage system. Finally, the strategy is executed through a power conversion device, which converts the control signals into actual power output. Optionally, data synchronization from the management unit is used when collecting parameters; aging constraints are considered when establishing the function; a maximum charge / discharge rate threshold is embedded when generating the strategy; feedforward feedback composite control technology is used to suppress voltage fluctuations when performing power regulation; or carrier phase-shifting technology is used to reduce harmonic interference; for example, power conversion is implemented in inverter hardware.

[0021] Furthermore, the collection of operating status parameters of the energy storage system specifically includes: obtaining operating status parameters through the management unit of the energy storage system; synchronizing grid operating parameters from the grid dispatch interface; and parsing the power trading data stream to extract power market trading parameters.

[0022] Specifically, the management unit is the data processing component inside the energy storage system; the grid dispatch interface is the communication channel connecting the grid system; the power trading data stream is the sequence of trading information transmitted in the power market; and the parsing is the process of analyzing the data stream to extract key parameters.

[0023] Specifically, the method includes: acquiring operating status parameters through the management unit of the energy storage system, with the unit monitoring the battery's internal resistance and cycle count; simultaneously synchronizing grid operating parameters from the grid dispatch interface, which receives grid frequency deviation and load forecast data; and parsing the power trading data stream to extract power market trading parameters, identifying real-time electricity prices and service compensation prices during the parsing process; optionally, the management unit uses sensors to collect temperature impact data; the grid dispatch interface communicates with the grid center through a standard protocol; feature extraction algorithms are used to identify load transfer periods when parsing the data stream; or a data validity verification mechanism is established to detect price jumps; for example, triggering a manual confirmation process when prices are abnormal; a time series alignment module is used to integrate information from different time scales when synchronizing parameters; or a sliding window interpolation method is applied to compensate for data acquisition delays; the entire process ensures parameter accuracy and real-time performance, providing reliable input for subsequent optimization.

[0024] Furthermore, the establishment of the multi-objective optimization function specifically includes: constructing an objective function for extending the lifespan of the energy storage system, using the number of cycles, internal resistance changes, and temperature effects among the operating state parameters as aging constraints, which limit the maximum charge / discharge depth and rate; constructing an objective function for improving grid stability, establishing a dynamic weight model based on frequency deviation, voltage fluctuations, and load forecasting among the grid operating parameters, with the dynamic weight model increasing weight coefficients when the frequency exceeds the limit; constructing an economically optimal objective function, combining real-time electricity prices, service compensation prices, and response incentives among the electricity market trading parameters to construct a revenue function; and normalizing the lifespan extension objective function, grid stability objective function, and economic objective function into a cost function using the Pareto front algorithm.

[0025] Specifically, the lifespan extension objective function is the optimization objective of extending the lifespan of the energy storage system; the aging constraint condition is the boundary condition that limits system aging; the cycle count is the cumulative number of battery charge and discharge cycles; the internal resistance change is the variation of the battery's internal resistance; the temperature effect is the effect of temperature on battery performance; the maximum charge and discharge depth is the upper limit of the allowable charge and discharge capacity; the rate is the charge and discharge rate; the grid stability improvement objective function is the optimization objective of enhancing grid reliability; the frequency deviation is the deviation of the grid frequency; the voltage fluctuation is the fluctuation of voltage; the load forecast is the data for predicting grid load; the dynamic weight model is the mathematical model for adjusting the target weights; the weight coefficient is the priority factor for each objective; the frequency exceeding the limit is the frequency exceeding the safe range; the economic optimization objective function is the optimization objective of maximizing economic benefits; the real-time electricity price is the current electricity price; the service compensation price is the reward for ancillary services; the response incentive is the reward for demand response; the revenue function is the function for calculating economic benefits; the Pareto front algorithm is the mathematical method for multi-objective optimization; normalization is the process of unifying the target dimensions; and the cost function is the function of the comprehensive optimization results.

[0026] Specifically, the method includes: constructing an objective function for extending the lifespan of the energy storage system, which uses the number of cycles and changes in internal resistance as aging constraints to limit the maximum depth of charge and discharge and the charging / discharging rate to prevent battery degradation; simultaneously constructing an objective function for improving grid stability, which establishes a dynamic weight model based on frequency deviation and voltage fluctuations, and the model automatically increases the weight coefficients to prioritize grid stability when the frequency exceeds the limit; and constructing an economically optimal objective function, which combines real-time electricity prices and response incentives to construct a revenue function to maximize the revenue from the charge-discharge price difference; and normalizing the above objective functions into a cost function using the Pareto front algorithm, with the algorithm balancing lifespan extension and stability. The evaluation process includes economic objectives; optionally, the strengthening of aging constraints due to temperature effects can be considered when constructing lifetime objectives; dynamic weights can be adjusted using load forecast data in stability objectives; service compensation prices can be integrated into economic objectives to calculate revenue; or a genetic algorithm with an elite retention strategy can be used to solve the optimization space; the algorithm selects the solution closest to the ideal point as the output; for example, the normalized value of the inverse of the cumulative change in aging factors can be used to quantify lifetime objectives; the weighted sum of the integral square of frequency deviation and the peak-to-valley difference of voltage fluctuations can be used to quantify stability objectives; and the algebraic sum of the charge-discharge price difference revenue and service revenue per unit time period can be used to quantify economic objectives; the entire process ensures multi-objective collaborative optimization.

[0027] Furthermore, the generation of the charging and discharging control strategy specifically includes: periodically updating the strategy using a rolling optimization framework; resolving the multi-objective optimization function when preset triggering conditions are met, the triggering conditions including time interval thresholds and sudden grid state events; embedding the maximum acceptable charging and discharging rate threshold derived from operating state parameters into the charging and discharging power constraints; when the frequency deviation in the grid operating parameters exceeds the safety limit, prioritizing the generation of frequency regulation ancillary service strategies, and allocating power output priority according to the frequency regulation compensation unit price in the electricity market trading parameters; discretizing the continuous strategy variables into an executable instruction set through a fuzzy decision tree, the instruction set including constant current charging, constant voltage charging, constant power discharging, and emergency power support operation modes.

[0028] Specifically, the rolling optimization framework is the optimization structure of the periodically updated strategy; the triggering condition is the condition for initiating re-optimization; the time interval threshold is a fixed time period; the power grid state change event is an abnormal change in the power grid; the charging and discharging power constraint is the rule for limiting power; the maximum acceptable charging and discharging rate threshold is the upper limit of the allowed rate; the frequency deviation exceeding the safety limit is a frequency danger state; the frequency regulation ancillary service strategy is a control scheme for stabilizing the power grid frequency; the frequency regulation compensation unit price is the rate of return for ancillary services; the power output priority is the order of power allocation; the fuzzy decision tree is a decision model for handling uncertainty; discretization is the conversion of continuous values ​​into discrete instructions; the executable instruction set is a set of operation commands; constant current charging is a fixed current charging mode; constant voltage charging is a fixed voltage charging mode; constant power discharging is a stable power discharging mode; and the emergency power support operation mode is a power output method with rapid response.

[0029] Specifically, the method includes: employing a rolling optimization framework to periodically update the charging and discharging control strategy; the framework re-solves the multi-objective optimization function when triggering conditions are met, including time interval thresholds or sudden grid state events; embedding a maximum acceptable charging and discharging rate threshold into the charging and discharging power constraints, the threshold being derived from operating state parameters to protect the system; when the frequency deviation exceeds the safety limit, prioritizing the generation of frequency regulation ancillary service strategies, the strategies allocating power output priority based on the frequency regulation compensation unit price to maximize revenue; discretizing continuous strategy variables into an executable instruction set through a fuzzy decision tree, the instruction set covering modes such as constant current charging or constant voltage charging; optionally, the rolling optimization framework uses fixed-period triggering updates; considering the impact of internal resistance changes when embedding constraints; combining load forecast data when generating frequency regulation strategies; the fuzzy decision tree outputs instructions based on a rule base; or the instructions are accompanied by timestamps and confidence assessment values; for example, the constant power discharge mode is activated during peak electricity prices; the emergency power support operation mode responds to grid faults; the entire process achieves dynamic adjustment and execution of the strategy.

[0030] Furthermore, the discretization via fuzzy decision tree specifically includes: establishing a rule base for selecting charging and discharging modes, where the antecedent of the rules is the Cartesian product of the grid frequency deviation rate and the electricity price fluctuation range; designing a membership function to quantify the impact of operating state parameters on rule weights; and outputting instructions with timestamps and confidence assessment values.

[0031] Specifically, the fuzzy decision tree is a decision structure based on fuzzy logic; discretization is the process of transforming continuous variables; the charging / discharging mode selection rule base is a database storing decision rules; the rule antecedent is the condition part of the rule; the power grid frequency deviation rate is the proportion of frequency deviation; the electricity price fluctuation range is the range of electricity price changes; the Cartesian product is the combination of conditions; the membership function is the function that quantifies fuzzy membership; the rule weight is the priority of the rule; the timestamp is the time stamp of the instruction; and the confidence score is the score of the instruction's reliability.

[0032] Specifically, the method includes: discretization through fuzzy decision trees; firstly, establishing a rule base for selecting charging and discharging modes, with the rule antecedent being the Cartesian product of the grid frequency deviation rate and the electricity price fluctuation range to cover multiple scenarios; designing a membership function to quantify the impact of operating state parameters on rule weights, with the function converting parameters such as temperature into weight adjustment factors; outputting instructions with timestamps and confidence assessment values, the timestamp recording the generation time and the confidence assessment value reflecting the reliability of the decision; optionally, the rule base is generated based on historical data training; the membership function handles parameters such as internal resistance changes; grid stability requirements are combined when outputting instructions; or fuzzy logic is used to handle uncertainty; for example, combining high frequency deviation and low electricity price ranges to trigger constant power discharge; the confidence assessment value is used for subsequent strategy optimization; the entire process enhances the adaptability and accuracy of the decision.

[0033] Furthermore, the normalization using the Pareto front algorithm specifically includes: defining the quantitative index of the lifespan extension target as the normalized value of the inverse of the cumulative change in the aging factor; quantifying the grid stability target as the weighted sum of the integral square of the frequency deviation and the peak-to-valley difference of the voltage fluctuation; transforming the economic target into the algebraic sum of the charging and discharging price difference revenue and service revenue per unit time period; using a genetic algorithm with an elite retention strategy to solve the three-dimensional target space; and selecting the solution closest to the ideal point as the cost function output.

[0034] Specifically, the Pareto front algorithm is a mathematical method for multi-objective optimization; normalization is a process of unifying dimensions; the quantitative index of the lifespan extension objective is a numerical measure of lifespan; the normalized value of the reciprocal of the cumulative change of the aging factor is an expression for the lifespan index; the quantification of the power grid stability objective is a stability measure; the integral square of the frequency deviation is the cumulative square of the frequency deviation; the peak-to-valley difference of voltage fluctuation is the maximum and minimum voltage difference; the weighted sum is a weighted sum; the transformation of the economic objective is the calculation of economic benefits; the charging and discharging price difference revenue per unit time period is the charging and discharging profit; the service revenue is the ancillary service revenue; the algebraic sum is the sum of addition and subtraction; the genetic algorithm with an elite retention strategy is an optimization solution algorithm; the three-dimensional objective space is the optimization domain of the three objectives; the solution closest to the ideal point is the optimal compromise solution; and the cost function output is the final optimization result.

[0035] Specifically, the method includes: normalizing multiple objectives using the Pareto front algorithm. First, the lifetime extension objective is defined as a normalized value of the inverse of the cumulative change in the aging factor to reflect the degree of system aging; the grid stability objective is quantified as a weighted sum of the integral square of the frequency deviation and the peak-to-valley difference of the voltage fluctuation to assess grid health; the economic objective is transformed into the algebraic sum of the charging and discharging price difference revenue and service revenue per unit time period to calculate economic benefits; a genetic algorithm with an elite retention strategy is used to solve the three-dimensional objective space, the algorithm retains excellent solutions and searches for the Pareto front; the solution closest to the ideal point is selected as the cost function output, which balances the conflicts among the objectives; optionally, the temperature effect is considered when defining the lifetime index; the weight coefficients are adjusted when quantifying the stability objective; real-time electricity price data is integrated when transforming the economic objective; the genetic algorithm uses an iterative optimization process; or the algorithm outputs a multi-dimensional solution set; for example, the ideal point is calculated based on the theoretical optimal value; the cost function is used to generate control strategies; the entire process achieves efficient multi-objective integration.

[0036] Furthermore, the analysis of the electricity trading data stream specifically includes: subscribing to the marginal electricity price data stream of the electricity spot market and the service market clearing result data stream; establishing a data validity verification mechanism, triggering a manual confirmation process when a price jump exceeds a threshold; integrating different time-scale parameters through a time series alignment module, and using a sliding window interpolation method to compensate for acquisition delays; and extracting features from the demand response incentive signals in the electricity market trading parameters to identify eligible load transfer periods.

[0037] Specifically, the analysis of electricity trading data streams involves analyzing trading information; subscription involves receiving data streams; the marginal electricity price data stream of the electricity spot market is a real-time electricity price information sequence; the service market clearing result data stream is service transaction result information; the data validity verification mechanism is a process for verifying data accuracy; price jumps exceeding thresholds refer to price mutation events; the manual confirmation process is a manual intervention step; the time series alignment module is a component for synchronizing time data; different time scale parameters refer to data with inconsistent time; the sliding window interpolation method is a mathematical method for compensating for delays; acquisition delay refers to data acquisition lag; demand response incentive signals refer to demand response reward information; feature extraction refers to identifying key features; and load transfer periods refer to the time periods of adjustable load.

[0038] Specifically, the method includes: parsing electricity trading data streams by first subscribing to the marginal electricity price data stream of the electricity spot market to obtain real-time prices; subscribing to the service market clearing result data stream to obtain service compensation information; establishing a data validity verification mechanism, which triggers a manual confirmation process when price jumps exceed a threshold to ensure data reliability; integrating different time-scaled parameters through a time series alignment module, which uses sliding window interpolation to compensate for acquisition delays and synchronize data; extracting features from demand response incentive signals to identify eligible load transfer periods for optimized response; optionally, subscribing to data streams is implemented using a message queue; the verification mechanism combines historical data comparison; the time series alignment module processes grid operating parameters; the sliding window interpolation method fills in missing values; and machine learning algorithms are applied to feature extraction, such as identifying peak periods for demand response participation. The entire process improves the accuracy and real-time performance of data parsing.

[0039] Furthermore, the power regulation performed by the power conversion device specifically includes: converting the charging and discharging control strategy into a modulation signal sequence of the power conversion device; adjusting the DC bus voltage fluctuation using feedforward feedback composite control; and suppressing the harmonic content of the grid-connected current using carrier phase shifting technology.

[0040] Among them, the power conversion device is specifically an energy conversion device; the power regulation is specifically a power control implementation; the modulation signal sequence is specifically a control command sequence; the feedforward feedback composite control is specifically a control method that combines prediction and correction; the DC bus voltage fluctuation is specifically a DC voltage fluctuation; the carrier phase shifting technology is specifically a technology to reduce harmonics; and the grid-connected current harmonic content is specifically the distortion component of the grid-connected current.

[0041] Specifically, the method includes: Power regulation is performed through a power conversion device. First, the charging / discharging control strategy is converted into a modulation signal sequence for the power conversion device, which drives the device operation. Feedforward-feedback composite control is used to adjust DC bus voltage fluctuations, controlling predicted disturbances and correcting errors to stabilize the voltage. Carrier phase-shifting technology is used to suppress grid-connected current harmonic content, reducing harmonic interference. Optionally, a digital signal processor is used when converting the strategy. Feedforward-feedback composite control is combined with sensor feedback. Carrier phase-shifting technology applies multi-level modulation; for example, the modulation signal sequence controls the inverter switching. Composite control responds to changes in grid parameters. The technology ensures grid-connected power quality. The entire process achieves efficient power execution.

[0042] The present invention also proposes a computer-readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the aforementioned dynamic adjustment method for the energy storage system.

[0043] Specifically, computer-readable storage medium refers to physical or logical media that store data; computer program refers to executable code; processor refers to the core component of a computing device; and implementation refers to the steps of execution.

[0044] Specifically, this includes: storing a computer program on a computer-readable storage medium; when the program is executed by a processor, implementing a dynamic adjustment method for the energy storage system; the method includes collecting operating status parameters and grid parameters; establishing a multi-objective optimization function; generating a charging and discharging control strategy; and executing power regulation; optionally, the medium is a solid-state drive or cloud storage; encoding an optimization algorithm in the program; the processor running a rolling optimization framework; or the program calling a fuzzy decision tree module; for example, the processor executing a Pareto front algorithm to normalize the objective; and the program outputting control commands to the power conversion device; the entire process enables the energy storage system to dynamically respond to changes in the grid and the market.

[0045] This invention also proposes a dynamic adjustment system for an energy storage system, comprising: a parameter acquisition module, an optimization calculation module, and a power execution module. The parameter acquisition module is used to acquire in real time the operating status parameters of the energy storage system, the grid operating parameters, and the electricity market transaction parameters. The optimization calculation module is used to establish a multi-objective optimization function based on the above parameters and generate a charging and discharging control strategy. The power execution module includes a power conversion device and a drive circuit, used to convert the strategy into a power adjustment command and execute it.

[0046] Among them, the dynamic adjustment system of the energy storage system is specifically a combination of hardware and software to implement the adjustment method; the parameter acquisition module is specifically a data acquisition unit; real-time acquisition is specifically real-time collection; the optimization calculation module is specifically a component for processing optimization calculations; the power execution module is specifically a unit for performing power operations; the power conversion device is specifically an energy conversion device; the drive circuit is specifically a circuit of the control device; conversion is specifically signal conversion; and execution is specifically the implementation of operations.

[0047] Specifically, the system includes: a dynamic adjustment system for the energy storage system comprising a parameter acquisition module, which acquires real-time operating status parameters such as battery temperature; acquires grid operating parameters such as frequency deviation; and acquires electricity market trading parameters such as real-time electricity price; an optimization calculation module establishes a multi-objective optimization function based on the parameters, integrating lifetime and stability objectives; and generates a charging and discharging control strategy; a power execution module including a power conversion device and drive circuit, which converts the strategy into power adjustment commands and executes them; optionally, the parameter acquisition module integrates a management unit and an analytical algorithm; the optimization calculation module uses a genetic algorithm to solve the problem; the power execution module applies feedforward feedback control; for example, the drive circuit generates a modulation signal; the device executes a constant current charging mode; and the system achieves dynamic adjustment through module collaboration.

[0048] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A dynamic adjustment method for an energy storage system, characterized in that, include: S1. Real-time acquisition of operating status parameters of the energy storage system, power grid operating parameters, and electricity market transaction parameters; S2. Establish a multi-objective optimization function based on operating status parameters, power grid operating parameters, and electricity market transaction parameters; S3. Generating charging and discharging control strategies for energy storage systems based on multi-objective optimization functions; S4. Power regulation by executing the charge and discharge control strategy through the power conversion device of the energy storage system.

2. The method as described in claim 1, characterized in that, The specific steps for collecting the operating status parameters of the energy storage system include: obtaining operating status parameters through the management unit of the energy storage system; synchronizing grid operating parameters from the grid dispatch interface; and parsing the power trading data stream to extract power market trading parameters.

3. The method as described in claim 1, characterized in that, The establishment of the multi-objective optimization function specifically includes: An objective function for extending the lifespan of an energy storage system is constructed, with the cycle number, internal resistance change, and temperature effect among the operating parameters as aging constraints. The aging constraints limit the maximum depth of charge and discharge and the rate of charge / discharge. A target function for improving power grid stability is constructed, and a dynamic weight model is established based on frequency deviation, voltage fluctuation, and load forecast in the power grid operating parameters. The dynamic weight model increases the weight coefficient when the frequency exceeds the limit. Construct an economically optimal objective function, and combine it with real-time electricity price, service compensation price, and response incentives from electricity market trading parameters to construct a revenue function; The Pareto front algorithm is used to normalize the lifetime extension objective function, the power grid stability objective function, and the economic objective function into a cost function.

4. The method as described in claim 1, characterized in that, The specific generation charge-discharge control strategy includes: A rolling optimization framework is adopted to periodically update the strategy. When a preset trigger condition is met, the multi-objective optimization function is re-solved. The trigger condition includes a time interval threshold and a sudden power grid state event. Embed the maximum acceptable charge / discharge rate threshold derived from operating state parameters into the charge / discharge power constraint; When the frequency deviation in the power grid operating parameters exceeds the safety limit, a frequency regulation ancillary service strategy is generated first, and the power output priority is allocated according to the frequency regulation compensation unit price in the power market transaction parameters. The continuous policy variables are discretized into an executable instruction set by using a fuzzy decision tree. The instruction set includes constant current charging, constant voltage charging, constant power discharging, and emergency power support operation modes.

5. The method as described in claim 4, characterized in that, The discretization using fuzzy decision trees specifically includes: Establish a rule base for selecting charging and discharging modes, with the rule antecedent being the Cartesian product of the grid frequency deviation rate and the electricity price fluctuation range; Design a membership function to quantify the impact of operational state parameters on rule weights; The output command includes a timestamp and a confidence level assessment.

6. The method as described in claim 3, characterized in that, The normalization using the Pareto front algorithm specifically includes: The quantitative indicator for the life extension target is defined as the normalized value of the inverse of the cumulative change in aging factors; The power grid stability target is quantified as a weighted sum of the integral square of the frequency deviation and the peak-to-valley difference of the voltage fluctuation; The economic objective is transformed into the algebraic sum of the charging and discharging price difference revenue and service revenue per unit time period; A genetic algorithm with an elitist retention strategy is used to solve the three-dimensional target space; The solution that is closest to the ideal point is selected as the cost function output.

7. The method as described in claim 2, characterized in that, The parsing of the power transaction data stream specifically includes: Subscribe to the nodal marginal electricity price data stream in the electricity spot market and the service market clearing result data stream; Establish a data validity verification mechanism that triggers a manual confirmation process when a price jump exceeds a threshold. Different time-scale parameters are integrated through a time series alignment module, and a sliding window interpolation method is used to compensate for acquisition delay. Feature extraction is performed on the demand response incentive signals in the electricity market trading parameters to identify eligible load transfer periods.

8. The method as described in claim 1, characterized in that, The power regulation performed by the power conversion device specifically includes: The charge / discharge control strategy is converted into a modulation signal sequence for the power conversion device; The DC bus voltage fluctuation is adjusted by using a feedforward-feedback composite control. The harmonic content of the grid-connected current is suppressed by carrier phase shifting technology.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the dynamic adjustment method of the energy storage system as described in any one of claims 1-8.

10. A dynamic adjustment system for an energy storage system, characterized in that, include: The parameter acquisition module is used to acquire real-time operating status parameters of the energy storage system, grid operating parameters, and electricity market transaction parameters. The optimization calculation module is used to establish a multi-objective optimization function based on the above parameters and generate a charge-discharge control strategy. The power execution module, which includes a power conversion device and a drive circuit, is used to convert the strategy into power regulation commands and execute them.

Images