A peak-valley electricity-based storage battery operation optimization method and system

Abstract

The application relates to the technical field of battery optimization, and particularly discloses a storage battery operation optimization method and system based on peak-valley electricity, which comprises the following steps: constructing a multi-dimensional operation characteristic matrix of a target power supply system; performing trend correlation analysis on the multi-dimensional operation characteristic matrix to establish a dynamic mapping mechanism of a peak-valley electricity price period and a storage battery state of charge interval, and obtaining a charging and discharging strategy mode; taking minimization of electricity cost as an optimization target and a battery attenuation rate as a constraint condition, performing multi-target collaborative deduction on the charging and discharging strategy mode to obtain an optimal charging and discharging power curve of the storage battery and a corresponding time period window, generating a charging and discharging control instruction of the storage battery, and delivering the charging and discharging control instruction to a storage battery energy management system; and correcting mapping parameters of the dynamic mapping mechanism according to an actual state of charge feedback value of the storage battery energy management system; and the application can improve the operation optimization efficiency of the storage battery.

Description

Technical Field

[0001] This invention relates to the field of battery optimization technology, and in particular to a method and system for optimizing battery operation based on peak and off-peak electricity. Background Technology

[0002] Currently, most battery charging and discharging control in energy storage systems relies on fixed-time strategies, failing to dynamically adapt to peak-valley electricity prices and load fluctuations. This makes it difficult to flexibly adjust operating status based on real-time electricity prices and demand, resulting in poor electricity cost control and failing to fully realize the economic benefits of off-peak energy storage and peak-peak discharge. Furthermore, existing control schemes do not incorporate battery degradation rate into core constraints, and the settings for charging and discharging power and duration lack scientific basis, easily leading to overcharging or abnormal battery operation, accelerating battery aging and shortening its lifespan.

[0003] Existing battery operation optimization methods mostly adopt a single-objective decision-making model, failing to achieve a synergistic balance between electricity costs and battery life, resulting in low strategy inference accuracy and execution efficiency. Furthermore, strategy parameters are mostly statically configured, unable to adjust mapping relationships in real time based on actual operating data. Over long-term operation, this leads to problems such as deviations in state-of-charge prediction and mismatches between control commands and actual operating conditions, resulting in a continuous decline in overall operation optimization effectiveness and failing to meet the long-term stable, efficient, and economical operation requirements of power supply systems. Summary of the Invention

[0004] In view of the above problems, the purpose of this invention is to provide a battery operation optimization method and system based on peak and off-peak electricity prices, so as to solve the problems that existing battery charging and discharging control strategies cannot dynamically adapt to peak and off-peak electricity prices and load fluctuations, cannot balance the minimization of electricity costs and battery degradation control, and have static and fixed strategy parameters that cannot be corrected in real time based on actual operation feedback, resulting in poor cost control effect, rapid battery life loss and low operation optimization efficiency.

[0005] This invention provides a battery operation optimization method based on peak-valley electricity, comprising: Step 1: Obtain load data, peak-valley electricity price time period division information, and real-time state of charge and cycle life parameters of the target power supply system to construct a multi-dimensional operating characteristic matrix of the target power supply system; Step 2: Perform trend correlation analysis on the multidimensional operating characteristic matrix to establish a dynamic mapping mechanism between peak and valley electricity price periods and battery state of charge intervals, and obtain the charging and discharging strategy mode of the battery in the current period. Step 3: With minimizing electricity costs as the optimization objective and battery degradation rate as the constraint, perform multi-objective collaborative simulation of the charging and discharging strategy mode to obtain the optimal charging and discharging power curve of the battery and the corresponding time window. Step 4: Based on the optimal charge and discharge power curve and time window, generate charge and discharge control commands for the battery and send the charge and discharge control commands to the battery energy management system. Step 5: Based on the actual state-of-charge feedback value of the battery energy management system, correct the mapping parameters of the dynamic mapping mechanism.

[0006] Preferably, in step 1, the process of acquiring load data of the target power supply system, peak-valley electricity price time period division information, and real-time state of charge and cycle life parameters of the battery to construct a multi-dimensional operating characteristic matrix of the target power supply system is as follows: Obtain the load power sampling sequence of the target power supply system within the past preset period, and align the load power sampling sequence with timestamps to obtain the historical load data of the target power supply system; The instantaneous load power value of the target power system at the current moment is collected, and the instantaneous load power value is spliced ​​with the historical load data in time series to obtain the fused time series load dataset of the target power system; Read the start time of the valley period, the start time of the peak period, and the start time of the normal period from the preset peak-valley electricity price time period division table, and generate the electricity price time period identifier of the target power supply system; The battery energy management system's status register is queried to obtain the battery's real-time state of charge percentage and cumulative charge-discharge cycle count. Based on the cumulative charge-discharge cycle count, the battery's cycle life parameters are retrieved from the preset battery life degradation curve. The time-series load dataset, electricity price period identifiers, real-time state of charge percentage values, and cycle lifetime parameters are structured according to a preset matrix dimension to obtain a multi-dimensional operating feature matrix of the target power supply system.

[0007] Preferably, in step 2, the trend correlation analysis of the multidimensional operating characteristic matrix is ​​performed to establish a dynamic mapping mechanism between peak and off-peak electricity price periods and battery state of charge intervals, thereby obtaining the battery's charging and discharging strategy mode for the current period. The process is as follows: Historical load power sequences and historical electricity price time period identifier sequences are extracted from the multidimensional operation feature matrix. The historical load power sequences are then grouped into time periods according to the historical electricity price time period identifier sequences to obtain the load power subsets for valley periods, peak periods, and normal periods. The power distribution statistics of the load power subsets during valley hours, peak hours, and normal hours are calculated respectively to obtain the mean and variance of the load power of the battery in each period. The average load power is correlated with the electricity price level during the corresponding period to identify the correlation pattern between the load power level and the electricity price level, and to obtain the time coupling law of the battery. Based on the time coupling law, the average load power, the variance of load power, and the time period price level code in the preset peak-valley electricity price time period division table are vectorized and concatenated to obtain the load-price correlation feature vector of the battery.

[0008] Preferably, in step 2, the process of performing trend correlation analysis on the multidimensional operating characteristic matrix to establish a dynamic mapping mechanism between peak and off-peak electricity price periods and battery state of charge intervals, and obtaining the charging and discharging strategy mode of the target power supply system in the current period, further includes: Based on the time-period electricity price level code in the load-electricity price correlation feature vector, determine the electricity price time period type of the battery at the current moment. The electricity price time period type includes valley period, peak period and normal period. The real-time state of charge of the battery is compared with a preset state of charge range by a threshold comparison to determine the state of charge range in which the real-time state of charge is located. The state of charge range includes a low charge range, a medium charge range, and a high charge range. Based on the combination of electricity price time period type and state of charge interval, the preset charging and discharging strategy mapping table is queried to obtain the charging and discharging strategy mode corresponding to the battery in the current time period.

[0009] Preferably, in step 3, the process of performing multi-objective collaborative deduction on the charging and discharging strategy mode with the goal of minimizing electricity costs and the constraint of battery degradation rate to obtain the optimal charging and discharging power curve and corresponding time window of the battery is as follows: Based on the charging and discharging strategy mode, determine the set of decision variables for the target power supply system in the current optimization cycle. The set of decision variables includes the battery charging and discharging power value and the charging and discharging duration. Based on the charging and discharging power values, charging and discharging duration, and the grid electricity price data of the battery, the estimated electricity cost within the current optimization cycle is evaluated, and the estimated electricity cost is used as the optimization target. Based on the current cycle life parameters and historical charging and discharging current of the battery, calculate the estimated battery degradation rate within the current optimization cycle; The estimated battery degradation rate is compared with a preset upper limit threshold for battery degradation. When the estimated battery degradation rate is less than the upper limit threshold for battery degradation, the set of decision variables is marked as a feasible solution. Under the constraint of satisfying feasible solutions, the set of decision variables is iteratively optimized to obtain the optimal combination of decision variables that minimizes the estimated cost of electricity. Arrange the charging and discharging power values ​​in the optimal decision variable combination in chronological order to obtain the optimal charging and discharging power curve of the battery. Then, label the start and end times of each time period in the optimal decision variable combination to obtain the time period window of the battery.

[0010] Preferably, the process of calculating the estimated battery degradation rate within the current optimization cycle based on the battery's current cycle life parameters and historical charge / discharge current is as follows: Read the current cycle life parameters from the battery energy management system of the battery. The current cycle life parameters include the current health status index and the cumulative equivalent cycle count of the battery. By integrating the charge and discharge current sequence in the historical charge and discharge current records over time periods, the equivalent ampere-hour throughput of the battery in the current optimization cycle is obtained. Based on the current health status index, cumulative equivalent cycle count, and equivalent ampere-hour throughput, the estimated battery degradation rate within the current optimization cycle is calculated. The formula for calculating the estimated battery degradation rate is as follows:

[0011] in, This is an estimate of the battery degradation rate. This is the initial health status index of the battery. This represents the current health status index. To accumulate the equivalent number of iterations, For equivalent cyclic increments, This represents the current operating temperature of the battery. This refers to the standard operating temperature value for the battery. The preset health status decay weighting coefficient, The preset cyclic stress weighting coefficient, The preset temperature sensitivity index, This is for natural exponent calculations.

[0012] Preferably, in step 4, the process of generating the battery charge / discharge control command based on the optimal charge / discharge power curve and time window is as follows: Power value analysis is performed on the optimal charge and discharge power curve to identify charging periods with positive power values ​​and discharging periods with negative power values; For the charging period, extract the charging power value from the optimal charging and discharging power curve, compare the charging power value with the rated charging power of the battery, and take the smaller value of the two as the charging power setting value. For the discharge period, extract the absolute value of the discharge power value from the optimal charge and discharge power curve, compare the absolute value of the discharge power value with the rated discharge power of the battery and the current load demand power, and take the minimum value of the three as the discharge power setting value. Based on the time period boundary times in the optimal charging and discharging time window, the charging and discharging start and stop time identifiers of the battery are generated. The charging and discharging start and stop time identifiers include the charging start time, charging stop time, discharging start time, and discharging stop time. The charging power setting value, the discharging power setting value, and the charging / discharging start / stop time identifier are encoded and encapsulated according to a preset communication protocol format to obtain the battery charging / discharging control command.

[0013] Preferably, in step 4, the process of sending the charge / discharge control command to the battery energy management system is as follows: The charging and discharging control commands are encapsulated into data frames according to a preset communication protocol format to obtain the battery's command data frames. Transmit the command data frame to the battery energy management system of the battery; The battery energy management system receives and parses command data frames to obtain the charging power setting value, the discharging power setting value, and the charging / discharging start / stop time identifier; The battery energy management system determines the charging and discharging operation type of the battery at the current time based on the charging and discharging start and stop time identifier, and calls the corresponding power control circuit according to the charging and discharging operation type. The charging power setting value or discharging power setting value is written into the power setting register of the power control circuit to drive the battery to perform the corresponding charging and discharging operation according to the set power value.

[0014] Preferably, in step 5, the process of correcting the mapping parameters of the dynamic mapping mechanism based on the actual state-of-charge feedback value of the battery energy management system is as follows: After the charge and discharge control command is executed, the actual state of charge feedback value of the battery and the actual electricity cost data are collected. Extract predicted state of charge (SOC) and predicted electricity cost from the multidimensional operational feature matrix; Calculate the state of charge deviation between the actual state of charge feedback value and the predicted state of charge value, and the electricity cost deviation between the actual electricity cost data and the predicted electricity cost value; By weighting and fusing the state of charge deviation and the electricity cost deviation, a comprehensive mapping deviation index for the dynamic mapping mechanism is obtained. The comprehensive mapping deviation index is compared with the preset deviation tolerance threshold. When the comprehensive mapping deviation index exceeds the deviation tolerance threshold, it is determined that there is a significant mapping deviation in the dynamic mapping mechanism, and the mapping parameter correction operation of the dynamic mapping mechanism is triggered.

[0015] The present invention also provides a battery operation optimization system based on peak and off-peak electricity, the system comprising: The operation feature construction module is used to acquire load data, peak-valley electricity price time period division information, and real-time state of charge and cycle life parameters of the target power supply system in order to construct a multi-dimensional operation feature matrix of the target power supply system. The strategy pattern establishment module is used to perform trend correlation analysis on the multi-dimensional operating characteristic matrix to establish a dynamic mapping mechanism between peak and valley electricity price periods and battery state of charge intervals, thereby obtaining the charging and discharging strategy pattern of the battery in the current period. The charge and discharge optimization simulation module is used to perform multi-objective collaborative simulation of the charge and discharge strategy mode with the goal of minimizing electricity cost and the constraint of battery degradation rate, so as to obtain the optimal charge and discharge power curve of the battery and the corresponding time window. The instruction generation and distribution module is used to generate charge and discharge control instructions for the battery based on the optimal charge and discharge power curve and time window, and to distribute the charge and discharge control instructions to the battery energy management system. The mapping parameter correction module is used to correct the mapping parameters of the dynamic mapping mechanism based on the actual state of charge feedback value of the battery energy management system.

[0016] As can be seen from the above technical solution, the present invention provides a battery operation optimization method and system based on peak and valley electricity prices. By constructing a multi-dimensional operation feature matrix and establishing a dynamic mapping mechanism between peak and valley electricity prices and battery state of charge, and combining the goal of minimizing electricity costs with the constraint of battery degradation rate, multi-objective optimization deduction is performed. This can accurately generate the optimal charging and discharging power curve and control commands, and at the same time, the mapping parameters are corrected in real time based on actual operation feedback. This effectively improves the battery operation optimization efficiency, significantly reduces electricity costs, slows down battery degradation, extends battery life, and achieves synergistic optimization of the economic operation of the power supply system and the health status of the battery. Attached Figure Description

[0017] Other objects and results of the invention will become more apparent and readily understood by referring to the following description taken in conjunction with the accompanying drawings, and with a more complete understanding of the invention. In the drawings: Figure 1 This is a flowchart illustrating a battery operation optimization method based on peak-valley electricity according to an embodiment of the present invention. Figure 2 This is a functional block diagram of a battery operation optimization system based on peak-valley electricity according to an embodiment of the present invention; Detailed Implementation Existing battery charging and discharging control strategies cannot dynamically adapt to peak and off-peak electricity prices and load fluctuations, and it is difficult to balance minimizing electricity costs with controlling battery degradation. Furthermore, the strategy parameters are static and fixed, and cannot be corrected in real time based on actual operational feedback, resulting in poor cost control, rapid battery life loss, and low operational optimization efficiency.

[0018] To address the aforementioned problems, this invention provides a method and system for optimizing battery operation based on peak and off-peak electricity. The specific embodiments of this invention will be described in detail below with reference to the accompanying drawings.

[0019] To illustrate the battery operation optimization method and system based on peak-valley electricity provided by this invention, Figure 1 An exemplary illustration is provided for a battery operation optimization method based on peak-valley electricity according to an embodiment of the present invention; Figure 2 An exemplary illustration is provided for a battery operation optimization system based on peak-valley electricity according to an embodiment of the present invention.

[0020] The following description of exemplary embodiments is merely illustrative and is in no way intended to limit the invention or its application or use. Techniques and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques and equipment should be considered part of the specification.

[0021] Reference Figure 1 The diagram shown is a flowchart illustrating a battery operation optimization method based on peak-valley electricity consumption according to an embodiment of the present invention. In this embodiment, the battery operation optimization method based on peak-valley electricity consumption includes: Step 1: Obtain load data, peak-valley electricity price time period division information, and real-time state of charge and cycle life parameters of the target power supply system to construct a multi-dimensional operating characteristic matrix of the target power supply system; In this embodiment of the invention, in step 1, the process of acquiring load data of the target power supply system, peak-valley electricity price time period division information, and real-time state of charge and cycle life parameters of the battery to construct a multi-dimensional operating characteristic matrix of the target power supply system is as follows: Obtain the load power sampling sequence of the target power supply system within the past preset period, and align the load power sampling sequence with timestamps to obtain the historical load data of the target power supply system; The instantaneous load power value of the target power system at the current moment is collected, and the instantaneous load power value is spliced ​​with the historical load data in time series to obtain the fused time series load dataset of the target power system; Read the start time of the valley period, the start time of the peak period, and the start time of the normal period from the preset peak-valley electricity price time period division table, and generate the electricity price time period identifier of the target power supply system; The battery energy management system's status register is queried to obtain the battery's real-time state of charge percentage and cumulative charge-discharge cycle count. Based on the cumulative charge-discharge cycle count, the battery's cycle life parameters are retrieved from the preset battery life degradation curve. The time-series load dataset, electricity price period identifiers, real-time state of charge percentage values, and cycle lifetime parameters are structured according to a preset matrix dimension to obtain a multi-dimensional operating feature matrix of the target power supply system.

[0022] From the historical monitoring database of the target power supply system, all load power sampling records within a complete preset period are extracted using data query commands. These records are stored in the form of data pairs, each containing a timestamp accurate to milliseconds and a corresponding active power value. The acquired load power sampling sequence is then timestamped. Specifically, a fixed time interval, such as five minutes, is defined as the alignment benchmark. Each data point in the sequence is then iterated through, and the difference between each original timestamp and this fixed time interval is calculated. The timestamp is adjusted to the time point of the grid with the smallest difference. All data points whose timestamps are now identical due to the adjustment retain the one with the latest power value. Finally, a historical load data sequence of the target power supply system is generated, with absolutely uniform time intervals and data points arranged strictly at equal intervals according to time.

[0023] The instantaneous power values ​​in the electrical circuits are read in real time at a sampling frequency of once per second using smart meters or power transmitters installed on the AC side of the target power supply system. This value is then transmitted to the central processing unit via a communication module. The instantaneous load power value at the current moment is concatenated with the historical load data obtained in the previous step. This process involves creating a new array, copying all power values ​​from the historical load data sequence into the array in their original chronological order, and then adding the latest instantaneous power value as a new element to the end of the array. Simultaneously, a precise timestamp synchronized with the current system time is generated and associated with this power value. This creates a fused time-series load dataset for the target power supply system that extends continuously from the historical data to the latest moment.

[0024] Open a structured data file from the system's configuration file storage path. This file contains a preset peak-valley electricity price period division table, which explicitly defines the specific clock times for the start times of valley, peak, and normal periods in tabular form. Generate an electricity price period identifier based on the current system time. The process involves parsing the hour and minute values ​​of the current time, comparing them sequentially with the values ​​of the three start times in the electricity price division table, determining which start time the current time falls after and before the next start time, thus identifying its electricity price period category, and outputting a preset, unique symbol code bound to the category as an identifier.

[0025] Using industrial communication protocols such as Modbus or CAN, a read command is sent to the specified register address of the battery energy management system. From the returned data frame, two status variables are parsed and obtained: a real-time state-of-charge percentage (SOP) value representing the remaining battery capacity, and an integer representing the cumulative number of charge-discharge cycles, recording the battery's complete charge-discharge history. Using the obtained cumulative charge-discharge cycle count as a lookup key, a precise search is performed on a battery life degradation curve stored in a data table. This data table establishes a mapping from cycle count to battery capacity degradation coefficient. By matching the cycle count, the corresponding record in the table is found, and the predefined battery health index in that record is read. This index is the battery cycle life parameter obtained from the query.

[0026] Based on a pre-defined multidimensional matrix structure template in the system configuration, all feature data are filled into fixed positions. Specifically, a blank matrix with a fixed number of rows and columns is first allocated. The power values ​​from the integrated time-series load dataset are then filled into the first few rows of the first column of the matrix in chronological order. Next, the code values ​​of the electricity price period identifiers are filled into the specified rows immediately following the first column of the matrix. Then, the real-time state of charge percentage values ​​are filled into the next row in decimal form. Finally, the cycle lifetime parameters are filled into the last row of the first column of the matrix. Through this strict mapping of order and position, all heterogeneous data are integrated into a target power supply system multidimensional operating feature matrix with a unified format and fixed structure.

[0027] Step 2: Perform trend correlation analysis on the multidimensional operating characteristic matrix to establish a dynamic mapping mechanism between peak and valley electricity price periods and battery state of charge intervals, and obtain the charging and discharging strategy mode of the battery in the current period. In this embodiment of the invention, in step 2, the trend correlation analysis of the multidimensional operating feature matrix is ​​performed to establish a dynamic mapping mechanism between peak and valley electricity price periods and battery state of charge intervals, thereby obtaining the battery's charging and discharging strategy mode for the current period. The process is as follows: Historical load power sequences and historical electricity price time period identifier sequences are extracted from the multidimensional operation feature matrix. The historical load power sequences are then grouped into time periods according to the historical electricity price time period identifier sequences to obtain the load power subsets for valley periods, peak periods, and normal periods. The power distribution statistics of the load power subsets during valley hours, peak hours, and normal hours are calculated respectively to obtain the mean and variance of the load power of the battery in each period. The average load power is correlated with the electricity price level during the corresponding period to identify the correlation pattern between the load power level and the electricity price level, and to obtain the time coupling law of the battery. Based on the time coupling law, the average load power, the variance of load power, and the time period price level code in the preset peak-valley electricity price time period division table are vectorized and concatenated to obtain the load-price correlation feature vector of the battery.

[0028] In step 2, the process of performing trend correlation analysis on the multidimensional operating characteristic matrix to establish a dynamic mapping mechanism between peak and off-peak electricity price periods and battery state of charge intervals, thereby obtaining the charging and discharging strategy mode of the target power supply system in the current period, further includes: Based on the time-period electricity price level code in the load-electricity price correlation feature vector, determine the electricity price time period type of the battery at the current moment. The electricity price time period type includes valley period, peak period and normal period. The real-time state of charge of the battery is compared with a preset state of charge range by a threshold comparison to determine the state of charge range in which the real-time state of charge is located. The state of charge range includes a low charge range, a medium charge range, and a high charge range. Based on the combination of electricity price time period type and state of charge interval, the preset charging and discharging strategy mapping table is queried to obtain the charging and discharging strategy mode corresponding to the battery in the current time period.

[0029] Historical load power sequences are extracted from the first column of the multidimensional operating feature matrix. This sequence consists of a series of previously filled power values. Historical electricity price time period identifier sequences are extracted from the rows corresponding to the electricity price time period identifiers in the same matrix. These sequences are time period codes corresponding to the historical power points. The historical load power sequences are then grouped into time periods according to the historical electricity price time period identifier sequences. Specifically, starting from the first data point, each power value is placed into three different sets based on its corresponding time period identifier code: power values ​​corresponding to valley periods are placed in the valley period load power subset, power values ​​corresponding to peak periods are placed in the peak period load power subset, and power values ​​corresponding to normal periods are placed in the normal period load power subset.

[0030] Statistical calculations are performed on the off-peak load power subset. All power values ​​within this subset are summed, and the result is divided by the total number of power values ​​in the subset. The result is the average load power of the battery during the off-peak period. Next, the difference between each power value in this subset and the average load power during the off-peak period is calculated. Each difference is squared, and the sum of all squared results is divided by the total number of power values. The result is the variance of the battery's load power during the off-peak period. The same calculation process is applied to the peak-hour load power subset and the flat-hour load power subset, ultimately yielding the average load power and variance for the battery during off-peak, peak, and flat-hour periods, respectively.

[0031] Retrieve specific electricity price levels for each historical time period from system storage. Compare the calculated average load power during off-peak hours with the corresponding electricity price level, then compare the average load power during peak hours with the corresponding electricity price level, and finally compare the average load power during normal hours with the corresponding electricity price level. Identify the correlation pattern between load power levels and electricity price levels by comparing the order of the average values ​​and the order of the electricity price levels for the three time periods, observing whether they show a consistent trend, and recording this observed pattern as the battery's time coupling law.

[0032] The predefined time-of-use price level codes are extracted from the preset peak-valley electricity price period division table. For example, number 1 represents the price level during the valley period, number 3 represents the price level during the peak period, and number 2 represents the price level during the normal period. Based on the time coupling law, the calculated average load power, variance of load power, and price level code during the valley period are grouped together and sequentially concatenated into a longer data string. In the same way, the three data points for the peak period and the three data points for the normal period are also concatenated into data strings. Finally, these three data strings are concatenated end to end in the order of valley period, normal period, and peak period to form a longer load-price correlation feature vector of the battery containing all information.

[0033] The electricity price time period code representing the current moment is extracted from the load-electricity price correlation feature vector. This time period electricity price level code is compared with the preset level standard to determine the electricity price time period type of the battery at the current moment. Specifically, if the code value is one, it is determined to be a valley time period type; if the code value is three, it is determined to be a peak time period type; and if the code value is two, it is determined to be a normal time period type.

[0034] The real-time state of charge (SOC) percentage of the battery is obtained from the battery energy management system. Preset SOC range thresholds are read from the configuration file; for example, the low SOC range is 0% to 30%, the medium SOC range is 30% to 80%, and the high SOC range is 80% to 100%. The real-time SOC percentage is compared sequentially with these thresholds to determine the SOC range. Specifically, if the SOC value is greater than or equal to 0% and less than 30%, it is determined to be in the low SOC range; if it is greater than or equal to 30% and less than 80%, it is determined to be in the medium SOC range; and if it is greater than or equal to 80%, it is determined to be in the high SOC range.

[0035] Prepare a pre-defined charging / discharging strategy mapping table. The row headings represent different electricity price periods, and the column headings represent different states of charge (SOC) ranges. The table cells contain the specific charging / discharging strategy mode names. Based on the determined electricity price period type and SOC range, query the charging / discharging strategy mapping table. Specifically, first locate the row containing the corresponding electricity price period type in the table's row headings, then locate the column containing the corresponding SOC range in the column headings. The text content stored in the cell where the row and column intersect represents the battery's charging / discharging strategy mode for the current period.

[0036] Step 3: With minimizing electricity costs as the optimization objective and battery degradation rate as the constraint, perform multi-objective collaborative simulation of the charging and discharging strategy mode to obtain the optimal charging and discharging power curve of the battery and the corresponding time window. In this embodiment of the invention, in step 3, the charging and discharging strategy mode is subjected to multi-objective collaborative deduction with the goal of minimizing electricity costs and the constraint of battery degradation rate, to obtain the optimal charging and discharging power curve and corresponding time window of the battery. The process is as follows: Based on the charging and discharging strategy mode, determine the set of decision variables for the target power supply system in the current optimization cycle. The set of decision variables includes the battery charging and discharging power value and the charging and discharging duration. Based on the charging and discharging power values, charging and discharging duration, and the grid electricity price data of the battery, the estimated electricity cost within the current optimization cycle is evaluated, and the estimated electricity cost is used as the optimization target. Based on the current cycle life parameters and historical charging and discharging current of the battery, calculate the estimated battery degradation rate within the current optimization cycle; The estimated battery degradation rate is compared with a preset upper limit threshold for battery degradation. When the estimated battery degradation rate is less than the upper limit threshold for battery degradation, the set of decision variables is marked as a feasible solution. Under the constraint of satisfying feasible solutions, the set of decision variables is iteratively optimized to obtain the optimal combination of decision variables that minimizes the estimated cost of electricity. Arrange the charging and discharging power values ​​in the optimal decision variable combination in chronological order to obtain the optimal charging and discharging power curve of the battery. Then, label the start and end times of each time period in the optimal decision variable combination to obtain the time period window of the battery.

[0037] The process of calculating the estimated battery degradation rate within the current optimization cycle based on the battery's current cycle life parameters and historical charge / discharge current is as follows: Read the current cycle life parameters from the battery energy management system of the battery. The current cycle life parameters include the current health status index and the cumulative equivalent cycle count of the battery. By integrating the charge and discharge current sequence in the historical charge and discharge current records over time periods, the equivalent ampere-hour throughput of the battery in the current optimization cycle is obtained. Based on the current health status index, cumulative equivalent cycle count, and equivalent ampere-hour throughput, the estimated battery degradation rate within the current optimization cycle is calculated, where, The formula for calculating the estimated battery degradation rate is as follows:

[0038] in, This is an estimate of the battery degradation rate. This is the initial health status index of the battery. This represents the current health status index. To accumulate the equivalent number of iterations, For equivalent cyclic increments, This represents the current operating temperature of the battery. This refers to the standard operating temperature value for the battery. The preset health status decay weighting coefficient, The preset cyclic stress weighting coefficient, The preset temperature sensitivity index, This is for natural exponent calculations.

[0039] The specific text instructions for the charging and discharging strategy mode are analyzed. These instructions define, with explicit keywords, that the battery should be charged during off-peak hours, discharged during peak hours, and maintain power balance during normal hours. Based on these instructions, a specific charging and discharging power value is assigned to each time segment within the current optimization cycle, in 15-minute increments. The sign of the power value clearly indicates charging or discharging, and its absolute value indicates the power intensity. Simultaneously, a specific duration for the charging and discharging action is assigned to each 15-minute time segment, and this duration is an integer multiple of 15 minutes. The charging and discharging power values ​​and corresponding durations of all time segments are combined to form a set of decision variables that includes all adjustable variables.

[0040] Detailed grid electricity price data for every 15-minute period within the current optimization cycle are obtained from the real-time database of the electricity metering subsystem of the target power supply system. Each entry in the list contains a start time point and the corresponding unit kilowatt-hour electricity price. For each time segment in the decision variable set, the charging and discharging power value set for that time segment is converted to kilowatts and multiplied by the value of its charging and discharging duration converted to hours to obtain the total amount of electricity absorbed or released by the battery in kilowatt-hours during that time period. Then, this total amount of electricity is multiplied by the unit kilowatt-hour electricity price corresponding to that time period to calculate the electricity cost for that time segment. Positive values ​​represent expenditure costs, and negative values ​​represent revenues. The electricity costs of all time segments within the current optimization cycle are algebraically summed, and the resulting algebraic sum is the estimated electricity cost for the current optimization cycle. This specific monetary value is established as the optimization objective to be minimized.

[0041] A read request message is sent to the holding register at address 0X040A of the battery energy management system via the Modbus RTU protocol through the RS485 communication bus. This message contains the starting address and data length. After receiving the request, the battery energy management system returns a data frame containing register data. From the data frame, an integer value representing the percentage of the battery's current available capacity relative to its initial capacity, stored in register address 0X040A, is parsed out. This value is the current health status index. At the same time, an integer value recording the total number of complete charge and discharge cycles in the battery's history is parsed out from register address 0X0411. This value is the cumulative equivalent cycle count. These two integer values ​​read from the specified register address together constitute the current cycle life parameter.

[0042] The charging and discharging current history recorded in the current optimization cycle is read from the local monitoring data storage area of ​​the battery. This record stores a series of signed current measurement values ​​at a fixed sampling interval of one second. The sign indicates the current direction. The charging and discharging current sequence is integrated over time to obtain the equivalent ampere-hour throughput. Specifically, each current measurement value in the array is multiplied by its corresponding one-second sampling time interval to obtain the amount of charge flowing through the battery in ampere-hours per second. Then, the charge amounts in all seconds in the current optimization cycle are algebraically summed. The final algebraic sum is the equivalent ampere-hour throughput of the battery in the current optimization cycle.

[0043] The estimated battery degradation rate equals the sum of the health state degradation term and the cycle stress temperature correction term, divided by the sum of the cumulative equivalent cycle count and the equivalent cycle increment. The health state degradation term is obtained by multiplying a preset health state degradation weighting coefficient by the difference between the initial health state index and the current health state index. The cycle stress temperature correction term is obtained by multiplying a preset cycle stress weighting coefficient by the equivalent cycle increment, and then multiplying by the result of the natural exponent calculation. The natural exponent calculation uses the natural constant as the base, and the exponent is the product of a preset temperature sensitivity index and the absolute values ​​of the current operating temperature and the standard operating temperature. This comprehensive assessment of the battery degradation rate takes into account the actual degradation of the battery's health state, the cumulative effect of cycle stress, and the accelerating effect of temperature conditions.

[0044] In the calculation of the battery degradation rate estimate, all parameters are obtained through real-time acquisition and historical data storage from the vehicle's battery management system. The initial health state index is determined by the ratio of the battery's factory-calibrated rated capacity to its standard test capacity. This value is calibrated before the battery is put into use and remains constant throughout its lifespan. The current health state index is calculated by the ratio of the battery's current usable capacity to its rated capacity. It is updated through capacity calibration tests periodically performed by the battery management system, which must be completed under standard operating conditions to ensure data accuracy. The cumulative equivalent cycle count is obtained by the battery management system through equivalent conversion of each charge-discharge cycle. The conversion process uses the standard charge-discharge depth as a benchmark, uniformly converting charge-discharge cycles at different depths into standard cycle counts. This value is continuously accumulated and updated as the battery is used. The equivalent cycle increment is the additional equivalent cycle count added between two health state index detection cycles, calculated from the charge-discharge cycles within that cycle according to the same conversion rules. The current temperature operating condition value is acquired in real-time by the temperature sensor built into the battery management system, with the acquisition frequency consistent with the health state index detection frequency. The standard operating temperature is the standard test environment temperature specified during battery design, calibrated by the battery manufacturer, and remains constant throughout the entire process. The health condition degradation weighting coefficient, cyclic stress weighting coefficient, and temperature sensitivity index are all preset fixed coefficients, obtained by the battery manufacturer through extensive bench aging tests. Once determined, these coefficients remain constant throughout the entire process without any dynamic adjustment. The natural index calculation uses standard mathematical operations, with the natural constant as the base and the product of the absolute value of the temperature difference and the temperature sensitivity index as the exponent.

[0045] As the cumulative equivalent cycle count of the battery increases, the estimated battery degradation rate generally shows an upward trend. As the difference between the current operating temperature and the standard operating temperature increases, the estimated battery degradation rate shows an exponential upward trend. As the rate of decline in the battery health status index increases, the estimated battery degradation rate shows a linear upward trend. As the increment of equivalent cycles increases, the estimated battery degradation rate shows a linear upward trend.

[0046] A preset battery degradation upper limit threshold is read from the system's parameter configuration file. This threshold is a fixed floating-point number representing the maximum allowable decrease in the battery's health status index within a single optimization cycle. The calculated battery degradation rate estimate is directly compared with the preset battery degradation upper limit threshold. When the estimated battery degradation rate is strictly less than the battery degradation upper limit threshold, it is determined that the charging and discharging behavior corresponding to the current decision variable set will not cause the battery's health status to decline too quickly within the cycle, thus marking the decision variable set as a feasible solution.

[0047] Within the discrete set of all labeled feasible solutions, a systematic grid search method is used for iterative optimization. First, all possible combinations of decision variables are generated within the domain of power and duration with a fixed step size. Then, combinations whose power value exceeds the maximum charging and discharging power or whose duration exceeds the remaining cycle time are filtered out based on the physical limits of the battery. For each remaining feasible solution, the corresponding electricity cost estimate is calculated according to the method described in step two. Next, the electricity cost estimates of all feasible solutions are compared, and the electricity cost estimate with the smallest value is found. The specific combination of decision variables corresponding to this smallest value is the optimal combination of decision variables with the smallest electricity cost estimate.

[0048] From the data structure of the optimal decision variable combination, the charging and discharging power values ​​corresponding to each 15-minute time segment are extracted in chronological order from morning to night. These power values ​​are used as the vertical axis, and the start time of each corresponding time segment is used as the horizontal axis. In a rectangular coordinate system, these points are connected by straight line segments to form a continuous broken line. This broken line is the optimal charging and discharging power curve of the battery. At the same time, the precise start and end times of each charging and discharging action are extracted from the optimal decision variable combination. These time points are marked on a horizontal time axis in the form of time labels. The start and end times of each action are connected by a horizontal line segment, and the power value is noted above. All such time intervals marked with power and time period constitute the time period window of the battery.

[0049] Step 4: Based on the optimal charge and discharge power curve and time window, generate charge and discharge control commands for the battery and send the charge and discharge control commands to the battery energy management system. In this embodiment of the invention, the process of generating charge and discharge control commands for the battery based on the optimal charge and discharge power curve and time window in step 4 is as follows: Power value analysis is performed on the optimal charge and discharge power curve to identify charging periods with positive power values ​​and discharging periods with negative power values; For the charging period, extract the charging power value from the optimal charging and discharging power curve, compare the charging power value with the rated charging power of the battery, and take the smaller value of the two as the charging power setting value. For the discharge period, extract the absolute value of the discharge power value from the optimal charge and discharge power curve, compare the absolute value of the discharge power value with the rated discharge power of the battery and the current load demand power, and take the minimum value of the three as the discharge power setting value. Based on the time period boundary times in the optimal charging and discharging time window, the charging and discharging start and stop time identifiers of the battery are generated. The charging and discharging start and stop time identifiers include the charging start time, charging stop time, discharging start time, and discharging stop time. The charging power setting value, the discharging power setting value, and the charging / discharging start / stop time identifier are encoded and encapsulated according to a preset communication protocol format to obtain the battery charging / discharging control command.

[0050] In step 4, the process of sending the charging and discharging control command to the battery energy management system is as follows: The charging and discharging control commands are encapsulated into data frames according to a preset communication protocol format to obtain the battery's command data frames. Transmit the command data frame to the battery energy management system of the battery; The battery energy management system receives and parses command data frames to obtain the charging power setting value, the discharging power setting value, and the charging / discharging start / stop time identifier; The battery energy management system determines the charging and discharging operation type of the battery at the current time based on the charging and discharging start and stop time identifier, and calls the corresponding power control circuit according to the charging and discharging operation type. The charging power setting value or discharging power setting value is written into the power setting register of the power control circuit to drive the battery to perform the corresponding charging and discharging operation according to the set power value.

[0051] The optimal charge / discharge power curve is analyzed by traversing the curve, which consists of time and power data points. The power value of each data point is read sequentially, and the sign of each power value is checked according to mathematical rules. If the power value is greater than zero, the entire time interval corresponding to the data point is marked as a charging period. If the power value is less than zero, the entire time interval corresponding to the data point is marked as a discharging period. If the power value is equal to zero, the point is marked as a resting period. This completes the accurate identification of all charging and discharging periods on the curve.

[0052] For each identified charging period, all corresponding power values ​​within that period are extracted from the data points of the optimal charge / discharge power curve. These power values ​​are then compared one by one with the fixed rated charging power values ​​obtained from the battery technical manual. The smaller value is used: if the extracted charging power value is less than the rated charging power, then the charging power setting value is adopted; if the extracted charging power value is greater than or equal to the rated charging power, then the charging power setting value is adopted. In this way, a charging power setting value limited by the rated power is calculated for each charging period.

[0053] For each identified discharge period, all negative power values ​​corresponding to that period are extracted from the data points of the optimal charge and discharge power curve. First, the absolute value of each negative power value is calculated to obtain a positive absolute discharge power value. Then, this positive absolute discharge power value, the fixed rated discharge power value of the battery, and the current load demand power value read in real time from the target power supply system total load monitoring table are compared together. The minimum value is selected from these three values ​​and determined as the discharge power set value for that discharge period.

[0054] Extract the start and end times of all charging periods from the data structure of the optimal charging / discharging period window. Record the start time of each charging period as the charging start time and the end time as the charging stop time. Similarly, extract the start and end times of all discharging periods. Record the start time of each discharging period as the discharging start time and the end time as the discharging stop time. Group these time information into a structured list according to the categories of charging start time, charging stop time, discharging start time, and discharging stop time. This list is the battery's charging / discharging start and stop time identifier.

[0055] According to the preset MODBUS TCP communication protocol format, the calculated charging power setting value is converted into a hexadecimal integer by multiplying it by a fixed precision coefficient. The discharging power setting value is converted into another hexadecimal integer in the same way. Each time in the charging / discharging start / stop time identifier is converted into the number of seconds starting from 00:00 UTC on January 1, 1970 and encoded into a four-byte hexadecimal long integer. Then, in the order specified by the protocol, the transaction identifier, protocol identifier, length field, unit identifier, function code, data start address, data volume, and all encoded data are concatenated into a continuous byte array. Finally, a cyclic redundancy check code is calculated and added to the end of this byte array to complete the encapsulation of the entire data packet and obtain the battery charging / discharging control command.

[0056] The complete byte array of charge / discharge control instructions is used as application layer data. Based on the previous encapsulation, it is further encapsulated into a complete data frame according to the requirements of the physical transport layer. Specifically, a specific start character for frame synchronization is added to the beginning, followed by the target device address byte of the battery energy management system. Then, the charge / discharge control instruction byte array is placed as the payload. Finally, the cyclic redundancy check (CRC) code of the entire sequence is calculated and appended to the payload, ultimately forming a complete binary data packet containing a frame header, address, data payload, and check code. This data packet is the battery instruction data frame that can be directly transmitted online.

[0057] The command data frame is transmitted to the battery energy management system of the battery through the RS485 serial communication physical line with the baud rate, data bits, stop bits, and parity bits configured. Specifically, the asynchronous transceiver of the main control unit converts each byte of the command data frame into a high-low level signal at a specific baud rate, from the least significant bit to the most significant bit, and sends it sequentially to the differential communication bus. Start bits and stop bits are added before and after each byte to ensure that the bit stream of the entire command data frame is accurately transmitted to the communication interface receiving end of the battery energy management system.

[0058] The battery energy management system continuously monitors the communication bus through its serial communication interface receiver. When the start character of a frame is detected, it begins to receive data and stores the subsequent bit stream byte by byte into the receive buffer. After the end condition of a frame is detected, it first verifies whether the target address matches the local address. Then, it calculates the cyclic redundancy check code of the received data part and compares it with the check code at the end of the frame. After the verification is successful, it extracts the effective load data part, namely the charge and discharge control instruction byte array, from the buffer according to the known frame format. It then decodes the data according to the opposite rules to those of the sending end to restore the original charging power setting value, discharge power setting value, and the specific values ​​and time list of charge and discharge start and stop time markers.

[0059] The control program of the battery energy management system reads the current date and time provided by its internal high-precision real-time clock chip. It compares the current time sequentially with each time in the parsed list of charge / discharge start / stop time identifiers. If the current time is greater than or equal to a charging start time in the list and less than or equal to the corresponding charging stop time, the battery's charge / discharge operation type at the current time is determined to be charging, and the charging power control loop subroutine is called. If the current time is greater than or equal to a discharging start time in the list and less than or equal to the corresponding discharging stop time, the operation type is determined to be discharging, and the discharging power control loop subroutine is called. Then, the parsed charging power setting value or discharging power setting value is written to the power setting register of the corresponding power control loop through the data bus. The output of this register is directly connected to the reference voltage input terminal of the pulse width modulation control chip of the bidirectional DC-DC converter, thereby driving the power converter to adjust its switching duty cycle, so that the battery performs the corresponding charging or discharging operation according to the set power value.

[0060] Step 5: Based on the actual state-of-charge feedback value of the battery energy management system, correct the mapping parameters of the dynamic mapping mechanism.

[0061] In this embodiment of the invention, in step 5, the mapping parameters of the dynamic mapping mechanism are corrected based on the actual state-of-charge feedback value of the battery energy management system, and the process is as follows: After the charge and discharge control command is executed, the actual state of charge feedback value of the battery and the actual electricity cost data are collected. Extract predicted state of charge (SOC) and predicted electricity cost from the multidimensional operational feature matrix; Calculate the state of charge deviation between the actual state of charge feedback value and the predicted state of charge value, and the electricity cost deviation between the actual electricity cost data and the predicted electricity cost value; By weighting and fusing the state of charge deviation and the electricity cost deviation, a comprehensive mapping deviation index for the dynamic mapping mechanism is obtained. The comprehensive mapping deviation index is compared with the preset deviation tolerance threshold. When the comprehensive mapping deviation index exceeds the deviation tolerance threshold, it is determined that there is a significant mapping deviation in the dynamic mapping mechanism, and the mapping parameter correction operation of the dynamic mapping mechanism is triggered.

[0062] At a specific point in time after the charge / discharge control command has been executed and the battery system has entered a steady state, a read command is sent to the battery energy management system to access the storage unit in its status register that stores the current remaining percentage of battery charge. The specific value read is the actual state-of-charge feedback value. Simultaneously, the total energy consumed or fed into the grid during the entire charge / discharge cycle is read from a high-precision smart meter connected to the grid via a communication gateway. This energy value is then multiplied by the tiered electricity price for the corresponding time period to calculate the total cost. This calculated total cost is the actual electricity cost data.

[0063] From the multidimensional operational feature matrix data structure stored in system memory, locate the specific column used to store predicted values, such as the last column of the matrix. Extract the estimated state of charge (SOC) value corresponding to the end of the current cycle, which is pre-calculated and stored when generating charge / discharge control commands. This estimated value is the predicted SOC value. From another specific column of the same multidimensional operational feature matrix, such as the second-to-last column, extract the estimated total electricity cost for the current cycle, which is pre-calculated and stored at the same time. This estimated value is the predicted electricity cost value.

[0064] The state of charge (SCC) deviation between the actual SCC feedback value and the predicted SCC value is calculated by subtracting the predicted SCC value from the actual SCC feedback value, resulting in a positive or negative difference. The absolute value of this difference is then calculated to eliminate the sign, and the resulting non-negative value is the SCC deviation. Similarly, the electricity cost deviation between the actual and predicted electricity cost values ​​is calculated by subtracting the predicted electricity cost value from the actual electricity cost value, resulting in a positive or negative difference. The absolute value of this difference is then calculated to eliminate the sign, and the resulting non-negative value is the electricity cost deviation.

[0065] The state-of-charge (SOC) deviation and the electricity cost deviation are weighted and fused. Specifically, the influence weight coefficients assigned to the SOC deviation (real numbers between zero and one) are read from the system configuration file, along with another influence weight coefficient assigned to the electricity cost deviation. The SOC deviation value is multiplied by its corresponding weight coefficient to obtain a weighted SOC deviation component, and the electricity cost deviation value is multiplied by its corresponding weight coefficient to obtain a weighted electricity cost deviation component. Finally, the weighted SOC deviation component and the weighted electricity cost deviation component are arithmetically added together. The sum is the comprehensive mapping deviation index of the dynamic mapping mechanism.

[0066] The system reads the preset deviation tolerance threshold from the threshold setting section of the system configuration file. This threshold is a predefined positive real number. The calculated comprehensive mapping deviation index is then arithmetically compared with the preset deviation tolerance threshold; specifically, it checks whether the comprehensive mapping deviation index is greater than the deviation tolerance threshold. When the comprehensive mapping deviation index exceeds the deviation tolerance threshold, the system logic determines that the dynamic mapping mechanism has a significant mapping deviation and triggers a mapping parameter correction operation. This triggering action executes a predefined parameter adaptive adjustment function, which adjusts the parameters of the internal mapping table based on the magnitude of the deviation index.

[0067] As can be seen from the above embodiments, the battery operation optimization method based on peak-valley electricity provided by the present invention constructs a multi-dimensional operation feature matrix and establishes a dynamic mapping mechanism between peak-valley electricity prices and battery state of charge. Combining the goal of minimizing electricity costs and the constraint of battery degradation rate, multi-objective optimization deduction is performed. This method can accurately generate the optimal charging and discharging power curve and control commands. At the same time, the mapping parameters are corrected in real time based on actual operation feedback, which effectively improves the battery operation optimization efficiency, significantly reduces electricity costs, slows down the battery degradation rate, extends the battery life, and achieves synergistic optimization of the economic operation of the power supply system and the health status of the battery.

[0068] like Figure 2 The diagram shown is a functional block diagram of a battery operation optimization system 100 based on peak-valley electricity provided in an embodiment of the present invention, including an operation feature construction module 101, a strategy mode establishment module 102, a charge-discharge optimization deduction module 103, an instruction generation and issuance module 104, and a mapping parameter correction module 105.

[0069] In this embodiment, the functions of each module are as follows: The operation feature construction module 101 is used to acquire load data of the target power supply system, peak and valley electricity price time period division information, and real-time state of charge and cycle life parameters of the battery, so as to construct a multi-dimensional operation feature matrix of the target power supply system. The strategy pattern establishment module 102 is used to perform trend correlation analysis on the multi-dimensional operating feature matrix to establish a dynamic mapping mechanism between peak and valley electricity price periods and battery charge state intervals, so as to obtain the charging and discharging strategy pattern of the battery in the current period. The charging and discharging optimization simulation module 103 is used to perform multi-objective collaborative simulation of the charging and discharging strategy mode with the goal of minimizing electricity cost and the constraint of battery degradation rate, so as to obtain the optimal charging and discharging power curve of the battery and the corresponding time window. The instruction generation and distribution module 104 is used to generate charging and discharging control instructions for the battery based on the optimal charging and discharging power curve and time window, and to distribute the charging and discharging control instructions to the battery energy management system. The mapping parameter correction module 105 is used to correct the mapping parameters of the dynamic mapping mechanism based on the actual state of charge feedback value of the battery energy management system.

[0070] As can be seen from the above embodiments, the battery operation optimization system based on peak-valley electricity provided by the present invention constructs a multi-dimensional operation feature matrix and establishes a dynamic mapping mechanism between peak-valley electricity prices and battery state of charge. Combining the goal of minimizing electricity costs and the constraint of battery degradation rate, it performs multi-objective optimization deduction, which can accurately generate the optimal charging and discharging power curve and control commands. At the same time, it corrects the mapping parameters in real time based on actual operation feedback, effectively improving the battery operation optimization efficiency, significantly reducing electricity costs, slowing down the battery degradation rate, extending the battery life, and realizing the coordinated optimization of economic operation of the power supply system and battery health status.

[0071] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0072] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for optimizing battery operation based on peak-valley electricity, characterized in that, The method includes: Step 1: Obtain load data, peak-valley electricity price time period division information, and real-time state of charge and cycle life parameters of the target power supply system to construct a multi-dimensional operating characteristic matrix of the target power supply system; Step 2: Perform trend correlation analysis on the multidimensional operating characteristic matrix to establish a dynamic mapping mechanism between peak and valley electricity price periods and battery state of charge intervals, and obtain the charging and discharging strategy mode of the battery in the current period. Step 3: With minimizing electricity costs as the optimization objective and battery degradation rate as the constraint, perform multi-objective collaborative simulation of the charging and discharging strategy mode to obtain the optimal charging and discharging power curve of the battery and the corresponding time window. Step 4: Based on the optimal charge and discharge power curve and time window, generate charge and discharge control commands for the battery and send the charge and discharge control commands to the battery energy management system. Step 5: Based on the actual state-of-charge feedback value of the battery energy management system, correct the mapping parameters of the dynamic mapping mechanism.

2. The battery operation optimization method based on peak-valley electricity as described in claim 1, characterized in that, In step 1, the process of acquiring load data, peak-valley electricity price time period division information, and real-time state of charge and cycle life parameters of the target power supply system to construct a multi-dimensional operating characteristic matrix of the target power supply system is as follows: Obtain the load power sampling sequence of the target power supply system within the past preset period, and align the load power sampling sequence with timestamps to obtain the historical load data of the target power supply system; The instantaneous load power value of the target power system at the current moment is collected, and the instantaneous load power value is spliced ​​with the historical load data in time series to obtain the fused time series load dataset of the target power system; Read the start time of the valley period, the start time of the peak period, and the start time of the normal period from the preset peak-valley electricity price time period division table, and generate the electricity price time period identifier of the target power supply system; The battery energy management system's status register is queried to obtain the battery's real-time state of charge percentage and cumulative charge-discharge cycle count. Based on the cumulative charge-discharge cycle count, the battery's cycle life parameters are retrieved from the preset battery life degradation curve. The time-series load dataset, electricity price period identifiers, real-time state of charge percentage values, and cycle lifetime parameters are structured according to a preset matrix dimension to obtain a multi-dimensional operating feature matrix of the target power supply system.

3. The battery operation optimization method based on peak-valley electricity as described in claim 2, characterized in that, In step 2, trend correlation analysis is performed on the multidimensional operating characteristic matrix to establish a dynamic mapping mechanism between peak and off-peak electricity price periods and battery state of charge intervals, thereby obtaining the battery's charging and discharging strategy mode for the current period. The process is as follows: Historical load power sequences and historical electricity price time period identifier sequences are extracted from the multidimensional operation feature matrix. The historical load power sequences are then grouped into time periods according to the historical electricity price time period identifier sequences to obtain the load power subsets for valley periods, peak periods, and normal periods. The power distribution statistics of the load power subsets during valley hours, peak hours, and normal hours are calculated respectively to obtain the mean and variance of the load power of the battery in each period. The average load power is correlated with the electricity price level during the corresponding period to identify the correlation pattern between the load power level and the electricity price level, and to obtain the time coupling law of the battery. Based on the time coupling law, the average load power, the variance of load power, and the time period price level code in the preset peak-valley electricity price time period division table are vectorized and concatenated to obtain the load-price correlation feature vector of the battery.

4. The battery operation optimization method based on peak-valley electricity as described in claim 3, characterized in that, In step 2, the process of performing trend correlation analysis on the multidimensional operating characteristic matrix to establish a dynamic mapping mechanism between peak and off-peak electricity price periods and battery state of charge intervals, thereby obtaining the charging and discharging strategy mode of the target power supply system in the current period, further includes: Based on the time-period electricity price level code in the load-electricity price correlation feature vector, determine the electricity price time period type of the battery at the current moment. The electricity price time period type includes valley period, peak period and normal period. The real-time state of charge of the battery is compared with a preset state of charge range by a threshold comparison to determine the state of charge range in which the real-time state of charge is located. The state of charge range includes a low charge range, a medium charge range, and a high charge range. Based on the combination of electricity price time period type and state of charge interval, the preset charging and discharging strategy mapping table is queried to obtain the charging and discharging strategy mode corresponding to the battery in the current time period.

5. The battery operation optimization method based on peak-valley electricity as described in claim 1, characterized in that, In step 3, with minimizing electricity costs as the optimization objective and battery degradation rate as the constraint, a multi-objective collaborative deduction of the charging and discharging strategy mode is performed to obtain the optimal charging and discharging power curve and corresponding time window of the battery. The process is as follows: Based on the charging and discharging strategy mode, determine the set of decision variables for the target power supply system in the current optimization cycle. The set of decision variables includes the battery charging and discharging power value and the charging and discharging duration. Based on the charging and discharging power values, charging and discharging duration, and the grid electricity price data of the battery, the estimated electricity cost within the current optimization cycle is evaluated, and the estimated electricity cost is used as the optimization target. Based on the current cycle life parameters and historical charging and discharging current of the battery, calculate the estimated battery degradation rate within the current optimization cycle; The estimated battery degradation rate is compared with a preset upper limit threshold for battery degradation. When the estimated battery degradation rate is less than the upper limit threshold for battery degradation, the set of decision variables is marked as a feasible solution. Under the constraint of satisfying feasible solutions, the set of decision variables is iteratively optimized to obtain the optimal combination of decision variables that minimizes the estimated cost of electricity. Arrange the charging and discharging power values ​​in the optimal decision variable combination in chronological order to obtain the optimal charging and discharging power curve of the battery. Then, label the start and end times of each time period in the optimal decision variable combination to obtain the time period window of the battery.

6. The battery operation optimization method based on peak-valley electricity as described in claim 5, characterized in that, The process of calculating the estimated battery degradation rate within the current optimization cycle based on the battery's current cycle life parameters and historical charge / discharge current is as follows: Read the current cycle life parameters from the battery energy management system of the battery. The current cycle life parameters include the current health status index and the cumulative equivalent cycle count of the battery. By integrating the charge and discharge current sequence in the historical charge and discharge current records over time periods, the equivalent ampere-hour throughput of the battery in the current optimization cycle is obtained. Based on the current health status index, cumulative equivalent cycle count, and equivalent ampere-hour throughput, the estimated battery degradation rate within the current optimization cycle is calculated. The formula for calculating the estimated battery degradation rate is as follows: in, This is an estimate of the battery degradation rate. This is the initial health status index of the battery. This represents the current health status index. To accumulate the equivalent number of iterations, For equivalent cyclic increments, This represents the current operating temperature of the battery. This refers to the standard operating temperature value for the battery. The preset health status decay weighting coefficient, The preset cyclic stress weighting coefficient, The preset temperature sensitivity index, This is for natural exponent calculations.

7. The battery operation optimization method based on peak-valley electricity as described in claim 1, characterized in that, In step 4, the process of generating charge and discharge control commands for the battery based on the optimal charge and discharge power curve and time window is as follows: Power value analysis is performed on the optimal charge and discharge power curve to identify charging periods with positive power values ​​and discharging periods with negative power values; For the charging period, extract the charging power value from the optimal charging and discharging power curve, compare the charging power value with the rated charging power of the battery, and take the smaller value of the two as the charging power setting value. For the discharge period, extract the absolute value of the discharge power value from the optimal charge and discharge power curve, compare the absolute value of the discharge power value with the rated discharge power of the battery and the current load demand power, and take the minimum value of the three as the discharge power setting value. Based on the time period boundary times in the optimal charging and discharging time window, the charging and discharging start and stop time identifiers of the battery are generated. The charging and discharging start and stop time identifiers include the charging start time, charging stop time, discharging start time, and discharging stop time. The charging power setting value, the discharging power setting value, and the charging / discharging start / stop time identifier are encoded and encapsulated according to a preset communication protocol format to obtain the battery charging / discharging control command.

8. The battery operation optimization method based on peak-valley electricity as described in claim 1, characterized in that, In step 4, the process of sending the charging and discharging control command to the battery energy management system is as follows: The charging and discharging control commands are encapsulated into data frames according to a preset communication protocol format to obtain the battery's command data frames. Transmit the command data frame to the battery energy management system of the battery; The battery energy management system receives and parses command data frames to obtain the charging power setting value, the discharging power setting value, and the charging / discharging start / stop time identifier; The battery energy management system determines the charging and discharging operation type of the battery at the current time based on the charging and discharging start and stop time identifier, and calls the corresponding power control circuit according to the charging and discharging operation type. The charging power setting value or discharging power setting value is written into the power setting register of the power control circuit to drive the battery to perform the corresponding charging and discharging operation according to the set power value.

9. The battery operation optimization method based on peak-valley electricity as described in claim 1, characterized in that, In step 5, the mapping parameters of the dynamic mapping mechanism are corrected based on the actual state-of-charge feedback value of the battery energy management system. The process is as follows: After the charge and discharge control command is executed, the actual state of charge feedback value of the battery and the actual electricity cost data are collected. Extract predicted state of charge (SOC) and predicted electricity cost from the multidimensional operational feature matrix; Calculate the state of charge deviation between the actual state of charge feedback value and the predicted state of charge value, and the electricity cost deviation between the actual electricity cost data and the predicted electricity cost value; By weighting and fusing the state of charge deviation and the electricity cost deviation, a comprehensive mapping deviation index for the dynamic mapping mechanism is obtained. The comprehensive mapping deviation index is compared with the preset deviation tolerance threshold. When the comprehensive mapping deviation index exceeds the deviation tolerance threshold, it is determined that there is a significant mapping deviation in the dynamic mapping mechanism, and the mapping parameter correction operation of the dynamic mapping mechanism is triggered.

10. A battery operation optimization system based on peak-valley electricity, characterized in that, For implementing the battery operation optimization method based on peak-valley electricity as described in any one of claims 1-9, the system comprises: The operation feature construction module is used to acquire load data, peak-valley electricity price time period division information, and real-time state of charge and cycle life parameters of the target power supply system in order to construct a multi-dimensional operation feature matrix of the target power supply system. The strategy pattern establishment module is used to perform trend correlation analysis on the multi-dimensional operating characteristic matrix to establish a dynamic mapping mechanism between peak and valley electricity price periods and battery state of charge intervals, thereby obtaining the charging and discharging strategy pattern of the battery in the current period. The charge and discharge optimization simulation module is used to perform multi-objective collaborative simulation of the charge and discharge strategy mode with the goal of minimizing electricity cost and the constraint of battery degradation rate, so as to obtain the optimal charge and discharge power curve of the battery and the corresponding time window. The instruction generation and distribution module is used to generate charge and discharge control instructions for the battery based on the optimal charge and discharge power curve and time window, and to distribute the charge and discharge control instructions to the battery energy management system. The mapping parameter correction module is used to correct the mapping parameters of the dynamic mapping mechanism based on the actual state of charge feedback value of the battery energy management system.

Images