A new energy power station intelligent charging and discharging control method based on battery health status

Abstract

The present application discloses an intelligent charging and discharging control method for a new energy power station based on the health status of a battery, which relates to the field of new energy power technology. The method collects the energy storage data of the battery, trains a battery health status assessment model, takes the energy storage data of the battery as input data, and outputs the SOH value; collects the SOH value, charge and discharge rate, temperature and discharge depth of the battery during the charge and discharge cycle, trains a battery life prediction model, and predicts the remaining cycle life of the battery; defines optimization objectives and constraints, and solves the optimal balanced charge and discharge strategy; predicts the SOH value of the battery in real time, calculates the SOH difference, and when the SOH difference is greater than a preset difference threshold, triggers the optimization of the next round of balanced charge and discharge strategy; lists batteries with SOH values ​​less than the retirement threshold as retired batteries, performs static and charge and discharge experimental tests on each retired battery, calculates the percentage of available capacity of each retired battery, and matches it with a cascade utilization scenario.

Description

Technical Field

[0001] The present application relates to the field of new energy power technology, and in particular to a method for intelligent charging and discharging control of a new energy power station based on battery health status. Background Art

[0002] With the vigorous development of the global new energy industry, electric vehicles and electric energy storage have been widely used. The health status monitoring and life prediction of power batteries and energy storage batteries have become the focus of the industry, and higher requirements have been placed on the battery management system BMS.

[0003] A Chinese patent with the authorization announcement number CN117674369B discloses a new energy storage battery pack charging and discharging safety management system, which relates to the field of intelligent energy management technology. The system prevents the problem of inconsistent responses of battery packs in different charging pile areas by regulating the response time, ensuring that the battery packs work together and reducing the risk of performance degradation and safety hazards. Battery pack degradation regulation extends the service life of the battery pack and reduces the safety risks caused by battery aging. Real-time load regulation avoids the operation of the battery pack under high load conditions and reduces the risk of overload. Environmental regulation reduces the risk of battery packs working in extreme environments through intelligent monitoring. These control strategies not only enable the system to respond to potential risks in a timely manner, but also improve its overall safety and reliability by optimizing the operating status of the battery pack. The system's intelligent management and timely intervention effectively protect the battery pack from damage and provide a comprehensive solution for the safe operation of new energy storage battery packs.

[0004] The existing technology does not take into account the differences in characteristic parameters such as capacity and internal resistance between battery cells. The difference in battery status leads to unbalanced performance of the battery pack, making it difficult to maximize the service life of the battery pack. Summary of the invention

[0005] The present application aims to solve at least one of the technical problems in the related art to a certain extent. To this end, one purpose of the present application is to propose a method for intelligent charging and discharging control of a new energy power station based on the health status of the battery, thereby realizing intelligent charging and discharging control of the battery in the new energy power station.

[0006] One aspect of the present application provides a method for intelligent charging and discharging control of a new energy power station based on battery health status, comprising:

[0007] Step S100: collecting energy storage data of batteries of different types and different use stages, wherein the energy storage data includes capacity, internal resistance, self-discharge rate and number of cycles, training a battery health status assessment model, using the battery energy storage data as input data, and outputting a SOH value;

[0008] Step S200: collecting the SOH value, charge and discharge rate, temperature and discharge depth of the battery during the charge and discharge cycle, constructing a battery training sample for training a battery life prediction model, and predicting the remaining cycle life of the battery based on the trained battery life prediction model;

[0009] Step S300: defining optimization objectives and constraints, wherein the optimization objectives include minimizing the battery pack capacity decay rate, minimizing the battery pack SOH difference, minimizing the battery pack remaining cycle life standard deviation and maximizing the battery pack energy utilization rate, establishing an optimization model, and solving for the optimal balanced charge and discharge strategy;

[0010] Step S400: executing a balanced charge and discharge strategy and predicting the SOH value of each battery in real time, calculating the SOH difference of the current battery pack, and triggering the optimization of the next round of balanced charge and discharge strategy when the SOH difference is greater than a preset difference threshold;

[0011] Step S500: Batteries with SOH values ​​less than the retirement threshold are listed as retired batteries, and each retired battery is subjected to static and charge-discharge experimental tests, and the percentage of available capacity of each retired battery is calculated to match the cascade utilization scenario for it;

[0012] The specific method of collecting energy storage data of batteries of different types and different use stages, wherein the energy storage data includes capacity, internal resistance, self-discharge rate and number of cycles, training a battery health status assessment model, and using the battery energy storage data as input data to output the SOH value is as follows:

[0013] Step S110: applying a multi-parameter fusion Kalman filter algorithm to calculate the capacity and internal resistance of the battery;

[0014] Step S120: measuring the self-discharge rate of the battery when the battery is at rest, and reading the number of cycles recorded by the BMS;

[0015] Step S130: Integrate the capacity, internal resistance, self-discharge rate and cycle number into the energy storage data of the battery, collect the energy storage data of batteries of different types and different use stages, and mark the corresponding energy storage data with its SOH value. The SOH value is used as a health status label to construct a training sample. , each training sample includes the battery energy storage data And the corresponding SOH value ;

[0016] Step S140: Select Gaussian kernel as kernel function and construct SVM regression model , the expression of the SVM regression model is: , and the SVM regression model satisfies , where w is the weight vector, is the bias term, For the general Functions mapped to high-dimensional feature spaces, Tolerance error, Indicates The SOH value predicted for the input data;

[0017] Step S150: Introducing slack variables and , constructing an objective function and constraints, wherein the objective function includes an L2 norm minimization objective function of the weight vector w and an error penalty function between the predicted SOH value and the true SOH value;

[0018] Step S160: transforming the objective function into a dual problem;

[0019] Step S170: Input the training sample into the SVM regression model, solve the dual problem, and obtain the optimal solution of the Lagrange multiplier and , the optimal solution of the bias term, and the final SVM regression model is obtained according to the optimal solution;

[0020] Step S180: using mean square error as the loss function of the SVM regression model, optimizing the model parameters, obtaining the final battery health status assessment model, and using the trained battery health status assessment model to predict the battery SOH value;

[0021] The specific method of applying the multi-parameter fusion Kalman filter algorithm to calculate the capacity and internal resistance of the battery is:

[0022] Step S111: Select the capacity and internal resistance of the battery as state variables and construct the state vector ;

[0023] The state vector is ,in, is the capacity of the battery at time k, is the internal resistance of the battery at time k;

[0024] Step S112: Based on the state vector Establish the state equation and establish the observation quantity and the state vector The observation variable equation between the observation variables is the terminal voltage and temperature of the battery, and the observation variable equation includes the terminal voltage observation equation and the temperature observation equation;

[0025] The equation of state is: , where A is the state transfer matrix, B is the input matrix, is the state vector at time k-1, is the charge and discharge current, is the process noise;

[0026] The equation of the port voltage observation equation is: ,in, is the open circuit voltage function, is the charge and discharge current at time k. When positive, discharge is negative, is the port voltage;

[0027] The equation of the temperature observation equation is: ,in, is the battery temperature, is the ambient temperature, is the temperature rise coefficient;

[0028] Step S113: Given an initial state estimate and the error covariance matrix ;

[0029] Step S114: Read the charge and discharge current and observation quantity collected by the BMS, and for each sampling time k, obtain the state vector predicted at time k according to the state equation , the predicted state vector Including forecasted capacity and the predicted internal resistance ;

[0030] Step S115: Using the state transfer matrix at time k-1 and the error covariance matrix , predict the error covariance matrix at time k ;

[0031] Step S116: Based on the maximum capacity of the battery and charge and discharge current , respectively calculate the Jacobian matrix of the port voltage observation equation and the Jacobian matrix of the temperature observation equation , calculate the Kalman gain matrix based on the Jacobian matrix ;

[0032] Step S117: Use the observed value and the Kalman gain matrix to correct the predicted state vector and calculate the state vector at the current k moment , and get the error covariance matrix at time k after the update ;

[0033] Step S118: Output the calculated state vector at time k , get the battery capacity at time k and internal resistance ;

[0034] The SOH value, charge and discharge rate, temperature and discharge depth of the battery during the charge and discharge cycle are collected to construct a battery training sample for training the battery life prediction model. The specific method for predicting the remaining cycle life of the battery based on the trained battery life prediction model is as follows:

[0035] Step S210: Perform a complete charge and discharge cycle on the battery until its capacity drops to the battery retirement standard value, and record the number of cycles from the beginning to the actual retirement of the battery, which is used as the remaining cycle life of the corresponding battery;

[0036] Step S220: During the battery charge and discharge cycle, each charge and discharge cycle is performed, recording the battery's SOH value, charge and discharge rate, temperature, and discharge depth at the end of the cycle, wherein the charge and discharge rate is a ratio of the charge and discharge current to the rated capacity of the battery, the temperature is collected by a temperature sensor, and the discharge depth is a ratio of the battery discharge capacity to the rated capacity of the battery;

[0037] Step S230: Arrange the SOH values, charge and discharge rates, temperatures and discharge depths of the historical batteries in chronological order to obtain a SOH sequence, a charge and discharge rate sequence, a temperature sequence and a discharge depth sequence;

[0038] Step S240: Preset the time step length La, and use the sliding window method to obtain input sequences of battery training samples on the SOH sequence, charge and discharge rate sequence, temperature sequence, and discharge depth sequence respectively;

[0039] Step S250: For each input sequence, the corresponding sample label is the remaining cycle life from the end of the current time window to the actual retirement of the battery. Based on the input sequence and the corresponding remaining cycle life, a battery training sample is constructed;

[0040] Step S260: Using the LSTM model as the initial model for training the battery life prediction model, using the input sequence as input data, and the corresponding remaining cycle life as output data, the LSTM model is trained using the battery training sample, and the loss function is the mean square error between the predicted remaining cycle life and the actual remaining cycle life. Minimizing the value of the loss function is used as the training goal. When the loss function reaches convergence, the training is completed;

[0041] Step S270: using the trained battery life prediction model to predict the remaining cycle life of the current battery;

[0042] The specific definition method of the optimization target is:

[0043] Step S310: According to the number of battery cycles , SOH value , charge and discharge rate and temperature , calculate the capacity decay rate of the battery , calculate the capacity decay rate of the battery pack according to the capacity and capacity decay rate of each battery ;

[0044] The calculation formula for the capacity decay rate of the battery is: , where represents the capacity decay rate of the hth battery at the current time k, represents the number of cycles of the hth battery at the current time k, represents the SOH value of the hth battery at the current time k, represents the charge and discharge rate of the hth battery at the current time k, represents the temperature of the hth battery at the current time k, , , , , are the fitting coefficients of the hth battery;

[0045] The calculation formula for the capacity decay rate of the battery pack is: , where represents the capacity decay rate of the battery pack at the current time k, is the number of batteries in the battery pack, represents the capacity of the hth battery at the current time k;

[0046] Step S320: Take minimizing the capacity decay rate of the battery pack as the first optimization objective to obtain the first optimization objective function ;

[0047] The function expression of the first optimization objective function is: , where and are the start and end times of the optimization time domain respectively;

[0048] Step S330: Calculate the SOH difference of the battery pack according to the SOH value of the battery , take minimizing the SOH difference of the battery pack as the second optimization objective to obtain the second optimization objective function ;

[0049] The calculation formula for the SOH difference is: , where is the average SOH of the battery pack;

[0050] The function expression of the second optimization objective function is: ;

[0051] Step S340: Calculate the standard deviation of the remaining cycle life of the battery pack according to the remaining cycle life of the battery , taking minimizing the standard deviation of the remaining cycle life of the battery pack as the third optimization objective, the third optimization objective function is obtained ;

[0052] The calculation formula for the standard deviation of the remaining cycle life of the battery pack is: ,in, represents the standard deviation of the remaining cycle life of the battery pack at the current time k, is the remaining cycle life of the hth battery at the current time k, represents the mean remaining cycle life of battery pack k at the current moment;

[0053] The calculation formula for the mean value of the remaining cycle life of the battery pack is: ;

[0054] The function expression of the third optimization objective function is: ;

[0055] Step S350: Based on the nominal capacity of the battery , Maximum allowed charging voltage and minimum discharge voltage Calculate the maximum available energy of a battery under rated conditions , according to the battery charge and discharge power Calculate the energy utilization of the battery in the optimized time domain based on the maximum available energy of the battery under rated conditions , and get the battery pack energy utilization rate , taking maximizing the battery pack energy utilization as the fourth optimization goal, the fourth optimization objective function is obtained ;

[0056] The maximum available energy of the battery under rated conditions is calculated as follows: ,in, , , are the nominal capacity, the maximum allowed charging voltage and the minimum allowed discharging voltage of the hth battery respectively;

[0057] The calculation formula of the energy utilization rate of the battery in the optimized time domain is: ,in, is the charge and discharge power of the hth battery at the current time k;

[0058] The calculation formula of the battery pack energy utilization rate is: ;

[0059] The calculation formula of the fourth optimization objective function is: ;

[0060] The specific definition method of the constraint condition is:

[0061] Step S360: taking the charge and discharge power of the battery in the optimization time domain as the decision variable, integrating the first optimization objective function, the second optimization objective function, the third optimization objective function, and the fourth optimization objective function as the optimization objective, defining constraints including upper and lower limit constraints of charge and discharge power, state of charge range constraints, and total power balance constraints, and establishing an optimization model;

[0062] The constraint conditions specifically include: the expression of the upper and lower limits of the charge and discharge power constraints is: ,in, and They represent the minimum charge and discharge power and the maximum charge and discharge power allowed by the current SOH value, respectively, and nh is the total number of batteries; the expression of the state of charge range constraint is: ,in, , Represent the minimum state of charge value and the maximum state of charge value, respectively. represents the state of charge value of the hth battery at time k; the state of charge value is obtained based on the charge and discharge current and capacity based on the differential equation, and the differential equation is: ,in, is the terminal voltage of the hth battery at time k, is the charge and discharge current of the hth battery at time k, is the capacity of the hth battery at time k; the expression of the total power balance constraint is: ,in, is the external load power demand, a positive value indicates discharge, and a negative value indicates charge;

[0063] The specific method for solving the optimal balanced charge and discharge strategy is:

[0064] Step S370: using an improved non-dominated sorting genetic algorithm to solve the optimization target and obtain a Pareto optimal solution set, where each optimal solution includes a time series of the charge and discharge power of the battery at each moment in the optimization time domain;

[0065] Step S380: Selecting an optimal solution from the Pareto optimal solution set as the current balanced charge and discharge strategy;

[0066] The specific method of executing the balanced charge and discharge strategy and predicting the SOH value of each battery in real time, calculating the SOH difference of the current battery group, and triggering the optimization of the next round of balanced charge and discharge strategy when the SOH difference is greater than the preset difference threshold is as follows:

[0067] Step S410: executing a balanced charge and discharge strategy, monitoring the battery's port voltage, charge and discharge current, temperature, self-discharge rate, and cycle number in real time through the BMS, predicting the SOH value of each battery, and calculating the SOH difference of the battery pack;

[0068] Step S420: Preset a difference threshold. When the SOH difference is greater than the preset difference threshold, the next round of optimization of the balanced charge and discharge strategy is triggered, and a new optimal balanced charge and discharge strategy is generated based on the updated SOH value and the remaining cycle life of the battery;

[0069] The specific method of classifying batteries with SOH values ​​less than the retirement threshold as retired batteries, performing static and charge-discharge experimental tests on each retired battery, calculating the percentage of available capacity of each retired battery, and matching it with the cascade utilization scenario is as follows:

[0070] Step S510: preset a battery retirement threshold, compare the SOH value of each battery with the retirement threshold, list the batteries with SOH values ​​less than the retirement threshold as retired batteries, and remove the retired batteries;

[0071] Step S520: Perform a 72h static self-discharge test on the retired battery and measure the battery port voltage at 0h and 72h respectively. , , calculate the voltage drop ;

[0072] Step S530: Perform a charge and discharge test on the retired battery and record the discharge capacity of the retired battery ;

[0073] Step S540: Based on the discharge capacity and rated capacity of the retired battery The ratio of , calculate the depth of discharge of retired batteries ;

[0074] Step S550: The cascade utilization scenarios include home backup power scenarios, low-speed electric vehicle scenarios, and uninterruptible power supply scenarios. >0.8 and voltage drop When the retired batteries are allocated to the home backup power scenario, And the voltage drop When the retired batteries are allocated to the low-speed electric vehicle scenario, And the voltage drop When the battery is used, the retired batteries are allocated to the uninterruptible power supply scenario; the remaining retired batteries are recycled.

[0075] One aspect of the present application provides a new energy power station intelligent charging and discharging control device based on battery health status, including:

[0076] The battery health prediction module is used to collect energy storage data of batteries of different types and different use stages, including capacity, internal resistance, self-discharge rate and number of cycles, train a battery health status assessment model, use the battery energy storage data as input data, and output the SOH value;

[0077] The remaining life prediction module is used to collect the SOH value, charge and discharge rate, temperature and discharge depth of the battery during the charge and discharge cycle, build battery training samples for training the battery life prediction model, and predict the remaining cycle life of the battery based on the trained battery life prediction model;

[0078] The balancing strategy optimization module is used to define optimization objectives and constraints. The optimization objectives include minimizing the battery pack capacity decay rate, minimizing the battery pack SOH difference, minimizing the battery pack remaining cycle life standard deviation and maximizing the battery pack energy utilization rate, establishing an optimization model, and solving the optimal balancing charge and discharge strategy;

[0079] The strategy optimization trigger module is used to execute the balanced charge and discharge strategy and predict the SOH value of each battery in real time, calculate the SOH difference of the current battery pack, and trigger the optimization of the next round of balanced charge and discharge strategy when the SOH difference is greater than the preset difference threshold;

[0080] The retired battery processing module is used to list batteries with SOH values ​​less than the retirement threshold as retired batteries, perform static and charge-discharge experimental tests on each retired battery, calculate the percentage of available capacity of each retired battery, and match it with a cascade utilization scenario.

[0081] One aspect of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in a method for intelligent charging and discharging control of a new energy power station based on battery health status are implemented.

[0082] One aspect of the present application provides a readable storage medium, which stores a computer program, and the computer program is suitable for loading by a processor to execute steps in a method for intelligent charging and discharging control of a new energy power station based on battery health status.

[0083] The intelligent charging and discharging control method for a new energy power station based on the battery health status proposed in this application has the following advantages over the prior art:

[0084] This application innovatively uses a multi-parameter fusion Kalman filter algorithm to calculate the capacity and internal resistance of the battery, improves the accuracy of capacity and internal resistance estimation, uses a SVM regression model to train the battery health status assessment model, introduces slack variables and kernel functions, and improves the accuracy of SOH prediction.

[0085] This application uses the LSTM model to train a battery life prediction model, which can effectively capture the characteristics of battery status changes over time and predict the remaining cycle life of the battery.

[0086] This application proposes a multi-objective optimization model for minimizing the battery pack capacity decay rate, minimizing the battery pack SOH difference, minimizing the standard deviation of the battery pack remaining cycle life and maximizing the battery pack energy utilization rate. By optimizing the balanced charge and discharge strategy, the aging of the battery pack can be delayed, the SOH and life differences between batteries can be reduced, and the energy utilization efficiency of the battery pack can be improved.

[0087] This application introduces an SOH difference threshold. When the SOH difference exceeds the threshold, the re-optimization of the balanced charge and discharge strategy is triggered, thereby realizing adaptive update of the strategy.

[0088] This application proposes a retired battery classification method based on voltage drop and discharge depth, which can be matched to different utilization scenarios such as home backup power supply, low-speed electric vehicles, uninterruptible power supply, etc., to achieve full life cycle management of batteries. BRIEF DESCRIPTION OF THE DRAWINGS

[0089] Figure 1 A method flow chart of a method for intelligent charging and discharging control of a new energy power station based on battery health status provided in this application;

[0090] Figure 2 A flow chart of the method for defining the optimization target provided in this application;

[0091] Figure 3 A flow chart of the method for allocating tiered utilization scenarios provided for this application;

[0092] Figure 4 A functional module diagram of an intelligent charging and discharging control device for a new energy power station based on battery health status provided in this application. DETAILED DESCRIPTION

[0093] In order to better understand the present application, various aspects of the present application will be described in more detail with reference to the accompanying drawings. It should be understood that these detailed descriptions are only descriptions of exemplary embodiments of the present application and do not limit the scope of the present application in any way. Throughout the specification, the same reference numerals refer to the same elements. The expression "and / or" includes any and all combinations of one or more of the associated listed items.

[0094] In the accompanying drawings, for ease of illustration, the sizes, dimensions, and shapes of the elements have been slightly adjusted. The accompanying drawings are for illustrative purposes only and are not drawn to an exact scale. As used herein, terms such as "substantially", "about", and similar terms are used as terms indicating approximation, rather than terms indicating degree, and are intended to account for the inherent deviations in measured or calculated values that would be recognized by a person of ordinary skill in the art. Additionally, in this application, the order in which the steps of the processes are described does not necessarily represent the order in which these processes occur in actual operation, unless otherwise clearly specified or derivable from the context.

[0095] It should also be understood that expressions such as "comprising", "including", "having", "containing", and / or "including having" are open-ended rather than closed-ended expressions in this specification, which means that the stated features, elements, and / or components exist, but do not exclude the existence of one or more other features, elements, components, and / or combinations thereof. In addition, when an expression such as "at least one of..." appears after a list of listed features, it modifies the entire list of features, rather than just a single element in the list. Furthermore, when describing the embodiments of this application, the use of "may" means "one or more embodiments of this application". And the term "exemplary" is intended to refer to an example or illustration.

[0096] Unless otherwise defined, all terms used herein (including engineering terms and scientific and technical terms) have the same meaning as commonly understood by a person of ordinary skill in the art to which this application pertains. It should also be understood that, unless clearly stated in this application, words defined in a commonly used dictionary should be interpreted as having a meaning consistent with their meaning in the context of the relevant art, and should not be interpreted in an idealized or overly formal sense.

[0097] It should be noted that, without conflict, the embodiments in this application and the features in the embodiments may be combined with each other. The following will describe this application in detail with reference to the accompanying drawings and in combination with the embodiments.

[0098] Example 1

[0099] As Figure 1 shown, a new energy power station intelligent charge and discharge control method based on the battery health state provided by this application includes:

[0100] Step S100: Collect the energy storage data of batteries of different types and at different usage stages. The energy storage data includes capacity, internal resistance, self-discharge rate, and number of cycles, and train a battery health state evaluation model. Using the energy storage data of the battery as input data, output the SOH value.

[0101] The specific method of collecting energy storage data of batteries of different types and different use stages, wherein the energy storage data includes capacity, internal resistance, self-discharge rate and number of cycles, training a battery health status assessment model, and using the battery energy storage data as input data to output the SOH value is as follows:

[0102] Step S110: Apply a multi-parameter fusion Kalman filter algorithm to calculate the capacity and internal resistance of the battery.

[0103] The specific method of applying the multi-parameter fusion Kalman filter algorithm to calculate the capacity and internal resistance of the battery is:

[0104] Step S111: Select the capacity and internal resistance of the battery as state variables and construct the state vector ;

[0105] The state vector is ,in, is the capacity of the battery at time k, is the internal resistance of the battery at time k;

[0106] Step S112: Based on the state vector Establish the state equation and establish the observation quantity and the state vector The observation variable equation between the observation variables is the terminal voltage and temperature of the battery, and the observation variable equation includes the terminal voltage observation equation and the temperature observation equation;

[0107] The equation of state is: , where A is the state transfer matrix, B is the input matrix, is the state vector at time k-1, is the charge and discharge current, is the process noise;

[0108] The state transfer matrix, input matrix, and process noise are set by those skilled in the art based on historical data and experience;

[0109] The state equation is used to describe the dynamic evolution law of the state vector;

[0110] The state transfer matrix A is used to describe the self-evolution law of capacity and internal resistance, that is, how the state vector at the current moment depends on the state vector at the previous moment, and is obtained by those skilled in the art based on the empirical formula of the battery attenuation model and historical data fitting;

[0111] The input matrix B is used to characterize the effect of charge and discharge current on capacity and internal resistance, and is constructed according to a thermodynamic model;

[0112] The charge and discharge current Directly measured by current sensor;

[0113] The process noise follows a Gaussian distribution with a mean of 0 and a covariance matrix of Q, where Q is equal to , and is the process noise variance;

[0114] Exemplarily, the state vector of the battery obeys an exponential decay model, and the capacity and internal resistance at time k, and the capacity and internal resistance at time k-1 satisfy the state transfer equation: , ,in, and are the attenuation factors of capacity and internal resistance, , are the influence coefficients of charge and discharge current on capacity and internal resistance, is the charge and discharge current, , is the process noise, , are the capacity and internal resistance at time k-1 respectively;

[0115] According to the above two state transfer equations, the state transfer matrix A is extracted to obtain: , extract the input matrix B to get: ;

[0116] The values ​​of the attenuation factors of the capacity and internal resistance, and the coefficients of influence of the charge and discharge current on the capacity and internal resistance are determined by those skilled in the art based on experience.

[0117] The observed variable equation is used to describe the relationship between the state vector and the observed quantity;

[0118] The observed variable equation is: ,in, is the observed quantity, is a nonlinear observation function used to map the state vector to the observed quantity, is the observation noise;

[0119] The observable refers to a physical quantity that can be directly measured and carries the information of the state vector;

[0120] The observation noise follows a Gaussian distribution with a mean of 0 and a covariance of R, where R is equal to , and is the observation noise variance, the specific value of which is determined by those skilled in the art based on historical data and experience;

[0121] When the observed quantity is the terminal voltage of the battery, the terminal voltage observation equation between the terminal voltage and the capacity and internal resistance is established according to the battery equivalent circuit model;

[0122] The equation of the port voltage observation equation is: ,in, is the open circuit voltage function, is the charge and discharge current at time k. When positive, discharge is negative, is the port voltage;

[0123] When the observed quantity is the temperature of the battery, a temperature observation equation between the battery temperature and the battery internal resistance is established;

[0124] The equation of the temperature observation equation is: ,in, is the battery temperature, is the ambient temperature, is the temperature rise coefficient;

[0125] Preferably, the ambient temperature is 25°C. The value is 0.1;

[0126] Step S113: Given an initial state estimate and the error covariance matrix ;

[0127] The initial state estimation value refers to the assumed initial capacity and initial internal resistance of the battery;

[0128] Step S114: Read the charge and discharge current and observation quantity collected by the BMS, and for each sampling time k, obtain the state vector predicted at time k according to the state equation , the predicted state vector Including forecasted capacity and the predicted internal resistance ;

[0129] The state vector predicted at time k is: ,in, is the state vector predicted at time k-1;

[0130] Step S115: Using the state transfer matrix at time k-1 and the error covariance matrix , predict the error covariance matrix at time k ;

[0131] The prediction formula of the error covariance matrix at the k moment is: , where T is the transpose of the matrix, is the covariance matrix , Represents the predicted value of the error covariance matrix at time k;

[0132] Step S116: Based on the maximum capacity of the battery and charge and discharge current , calculate the Jacobian matrix of the port voltage observation equation respectively and the Jacobian matrix of the temperature observation equation , calculate the Kalman gain matrix based on the Jacobian matrix ;

[0133] The calculation formula of the Jacobian matrix of the port voltage observation equation is: ;

[0134] The calculation formula of the Jacobian matrix of the temperature observation equation is: ;

[0135] The calculation formula of the Kalman gain matrix is: ;

[0136] Step S117: Use the observed value and the Kalman gain matrix to correct the predicted state vector and calculate the state vector at the current time k , and get the error covariance matrix at time k after update ;

[0137] The calculation formula of the state vector at the current k moment is: ;

[0138] The calculation formula of the error covariance matrix at the kth moment after the update is: , where I is the identity matrix;

[0139] Step S118: Output the calculated state vector at time k , get the battery capacity at time k and internal resistance .

[0140] Furthermore, steps S114 to S118 are repeated at each sampling moment to update the capacity and internal resistance of the battery in real time, and the capacity and internal resistance are transmitted to the battery health status assessment model to realize the battery health management function;

[0141] Step S120: measuring the self-discharge rate of the battery when the battery is at rest, and reading the number of cycles recorded by the BMS;

[0142] Step S130: Integrate the capacity, internal resistance, self-discharge rate and cycle number into the energy storage data of the battery, collect the energy storage data of batteries of different types and different use stages, and mark the corresponding energy storage data with its SOH value. The SOH value is used as a health status label to construct a training sample. , each training sample includes the energy storage data of the battery And the corresponding SOH value ;

[0143] The health status label is determined by experts based on experience and has a value of 0 to 100%;

[0144] Step S140: Select Gaussian kernel as kernel function and construct SVM regression model , the expression of the SVM regression model is: , and the SVM regression model satisfies , where w is the weight vector, is the bias term, For the general Functions mapped to high-dimensional feature spaces, Tolerance error, Indicates The SOH value predicted for the input data;

[0145] Indicates that the error between the predicted SOH value and the actual SOH value is less than or equal to the tolerance error;

[0146] Step S150: Introducing slack variables and , constructing an objective function and constraints, wherein the objective function includes an L2 norm minimization objective function of the weight vector w and an error penalty function between the predicted SOH value and the true SOH value;

[0147] The functional expression of the objective function is: , C is the penalty coefficient, N represents the total number of training samples, is the slack variable of the i-th training sample, corresponding to the case where the SOH value of the predicted value is larger than the actual SOH value. is the slack variable of the i-th training sample, corresponding to the case where the SOH value of the predicted value is smaller than the actual SOH value. is the objective function for minimizing the L2 norm of the weight vector w, is the error penalty function;

[0148] The penalty coefficient is used to control the weight of the error between the predicted SOH value and the true SOH value in the objective function, and is set by those skilled in the art based on experience;

[0149] The constraints are: , , ;

[0150] Step S160: transforming the objective function into a dual problem;

[0151] The formulation of the dual problem includes:

[0152] ,

[0153] ,

[0154] st

[0155] ;

[0156] in, , is the Lagrange multiplier corresponding to the i-th training sample, respectively and Correspondingly, is a kernel function used to calculate the inner product of the i-th training sample and the j-th training sample in high-dimensional space. Representation Optimization , To maximize the value of the expression;

[0157] Step S170: Input the training sample into the SVM regression model, solve the dual problem, and obtain the optimal solution of the Lagrange multiplier and , the optimal solution of the bias term, and the final SVM regression model is obtained according to the optimal solution;

[0158] The final SVM regression model expression is: , where x is the input sample, is the optimal solution for the bias term, and the calculation formula is: ,in, is the optimal solution of the Lagrange multiplier corresponding to the jth training sample;

[0159] Step S180: Use the mean square error as the loss function of the SVM regression model, optimize the model parameters, obtain the final battery health status assessment model, and use the trained battery health status assessment model to predict the battery SOH value.

[0160] Step S200: Collect the SOH value, charge and discharge rate, temperature and discharge depth of the battery during the charge and discharge cycle, construct a battery training sample for training a battery life prediction model, and predict the remaining cycle life of the battery based on the trained battery life prediction model.

[0161] The SOH value, charge and discharge rate, temperature and discharge depth of the battery during the charge and discharge cycle are collected to construct a battery training sample for training the battery life prediction model. The specific method for predicting the remaining cycle life of the battery based on the trained battery life prediction model is as follows:

[0162] Step S210: Perform a complete charge and discharge cycle on the battery until its capacity drops to the battery retirement standard value, and record the number of cycles from the beginning to the actual retirement of the battery, which is used as the remaining cycle life of the corresponding battery;

[0163] Step S220: During the battery charge and discharge cycle, each charge and discharge cycle is performed, recording the battery's SOH value, charge and discharge rate, temperature, and discharge depth at the end of the cycle, wherein the charge and discharge rate is a ratio of the charge and discharge current to the rated capacity of the battery, the temperature is collected by a temperature sensor, and the discharge depth is a ratio of the battery discharge capacity to the rated capacity of the battery;

[0164] Step S230: Arrange the SOH values, charge and discharge rates, temperatures and discharge depths of the historical batteries in chronological order to obtain a SOH sequence, a charge and discharge rate sequence, a temperature sequence and a discharge depth sequence;

[0165] Step S240: Preset the time step length La, and use the sliding window method to obtain input sequences of battery training samples on the SOH sequence, charge and discharge rate sequence, temperature sequence, and discharge depth sequence respectively;

[0166] The input sequence of the battery training sample is ,in, is the g-th value in the SOH sequence, charge and discharge rate sequence, temperature sequence or discharge depth sequence;

[0167] The value of the time step length is set by those skilled in the art based on experience;

[0168] Step S250: For each input sequence, the corresponding sample label is the remaining cycle life from the end of the current time window to the actual retirement of the battery. Based on the input sequence and the corresponding remaining cycle life, a battery training sample is constructed;

[0169] Step S260: Using the LSTM model as the initial model for training the battery life prediction model, using the input sequence as input data, and the corresponding remaining cycle life as output data, the LSTM model is trained using the battery training sample, and the loss function is the mean square error between the predicted remaining cycle life and the actual remaining cycle life. Minimizing the value of the loss function is used as the training goal. When the loss function reaches convergence, the training is completed;

[0170] Step S270: using the trained battery life prediction model to predict the remaining cycle life of the current battery.

[0171] Step S300: define optimization objectives and constraints, wherein the optimization objectives include minimizing the battery pack capacity decay rate, minimizing the battery pack SOH difference, minimizing the battery pack remaining cycle life standard deviation and maximizing the battery pack energy utilization, establishing an optimization model, and solving to obtain the optimal balanced charge and discharge strategy.

[0172] The specific definition method of the optimization target is:

[0173] Step S310: According to the number of battery cycles , SOH value , Charge and discharge rate and temperature , calculate the capacity decay rate of the battery , according to the capacity and capacity decay rate of each battery, calculate the capacity decay rate of the battery pack ;

[0174] The calculation formula of the capacity decay rate of the battery is: ,in, represents the capacity decay rate of the hth battery at the current time k, represents the number of cycles of the hth battery at the current time k, represents the SOH value of the hth battery at the current time k, represents the charge and discharge rate of the hth battery at the current time k, represents the temperature of the hth battery at the current time k, , , , , is the fitting coefficient of the hth battery;

[0175] The number of cycles refers to the number of times the battery has been charged and discharged;

[0176] The fitting coefficient of the battery is set by those skilled in the art based on experience;

[0177] The calculation formula of the battery pack capacity attenuation rate is: ,in, represents the capacity decay rate of the battery pack at the current time k, is the number of batteries in the battery pack, represents the capacity of the hth battery at the current time k;

[0178] Step S320: Minimize the battery pack capacity decay rate As the first optimization objective, the first optimization objective function is obtained ;

[0179] The functional expression of the first optimization objective function is: ,in, and They are the start and end times of the optimized time domain respectively;

[0180] Step S330: Calculate the SOH difference of the battery pack according to the SOH value of the battery , with minimizing the SOH difference of the battery pack as the second optimization goal, the second optimization objective function is obtained ;

[0181] The calculation formula of the SOH difference is: ,in, is the average SOH value of the battery pack;

[0182] The function expression of the second optimization objective function is: ;

[0183] Under balanced charge and discharge conditions, the SOH difference of the battery pack should be minimized;

[0184] Step S340: Calculate the standard deviation of the remaining cycle life of the battery pack according to the remaining cycle life of the battery , taking minimizing the standard deviation of the remaining cycle life of the battery pack as the third optimization objective, the third optimization objective function is obtained ;

[0185] The calculation formula for the standard deviation of the remaining cycle life of the battery pack is: ,in, represents the standard deviation of the remaining cycle life of the battery pack at the current time k, is the remaining cycle life of the hth battery at the current time k, represents the mean remaining cycle life of battery pack k at the current moment;

[0186] The calculation formula for the mean value of the remaining cycle life of the battery pack is: ;

[0187] The function expression of the third optimization objective function is: ;

[0188] Step S350: Based on the nominal capacity of the battery , Maximum allowed charging voltage and minimum discharge voltage Calculate the maximum available energy of a battery under rated conditions , according to the battery charge and discharge power Calculate the energy utilization of the battery in the optimized time domain based on the maximum available energy of the battery under rated conditions , and get the battery pack energy utilization rate , taking maximizing the battery pack energy utilization as the fourth optimization goal, the fourth optimization objective function is obtained ;

[0189] The calculation formula for the maximum available energy of the battery under rated conditions is: ,in, , , are the nominal capacity, the maximum allowed charging voltage and the minimum allowed discharging voltage of the hth battery respectively;

[0190] The calculation formula of the energy utilization rate of the battery in the optimized time domain is: ,in, is the charge and discharge power of the hth battery at the current time k;

[0191] The calculation formula of the battery pack energy utilization rate is: ;

[0192] The calculation formula of the fourth optimization objective function is: ;

[0193] like Figure 2 As shown, it is a flow chart of the method for defining the optimization target provided in this application.

[0194] The specific definition method of the constraint condition is:

[0195] Step S360: taking the charge and discharge power of the battery in the optimization time domain as the decision variable, integrating the first optimization objective function, the second optimization objective function, the third optimization objective function, and the fourth optimization objective function as the optimization objective, defining constraints including upper and lower limit constraints of charge and discharge power, state of charge range constraints, and total power balance constraints, and establishing an optimization model;

[0196] The constraint conditions specifically include: the expression of the upper and lower limit constraints of the charge and discharge power is: ,in, and They represent the minimum charge and discharge power and the maximum charge and discharge power allowed by the current SOH value, respectively, and nh is the total number of batteries; the expression of the state of charge range constraint is: ,in, , Represent the minimum state of charge value and the maximum state of charge value, respectively. represents the state of charge value of the hth battery at time k; the state of charge value is obtained based on the charge and discharge current and capacity based on the differential equation, and the differential equation is: ,in, is the terminal voltage of the hth battery at time k, is the charge and discharge current of the hth battery at time k, is the capacity of the hth battery at time k; the expression of the total power balance constraint is: ,in, is the external load power demand, a positive value indicates discharging, and a negative value indicates charging.

[0197] The specific method for solving the optimal balanced charge and discharge strategy is:

[0198] Step S370: using an improved non-dominated sorting genetic algorithm to solve the optimization target and obtain a Pareto optimal solution set, where each optimal solution includes a time series of the charge and discharge power of the battery at each moment in the optimization time domain;

[0199] The specific method of using the improved non-dominated sorting genetic algorithm to solve the optimization target and obtain the Pareto optimal solution set is:

[0200] Step S371: Initialize the population, randomly generate NH individuals, each of which consists of 2nh genes, corresponding to the time series of the charge and discharge power of nh batteries, and the gene values ​​meet the constraints;

[0201] Step S372: for each individual, discretize the integral terms of the four optimization objective functions, convert them into a finite term summation problem to calculate the fitness, use the penalty function method to integrate the constraints into the optimization objectives, and impose penalties on individuals that violate the constraints;

[0202] Step S373: using binary tournament selection, simulated binary crossover and polynomial mutation, select Nh individuals from the parent generation to generate offspring;

[0203] Step S374: merge the parent generation and the offspring generation, and select the best performing NH individuals from them to form a new generation population;

[0204] Step S375: If the maximum number of iterations is reached, the current Pareto optimal solution set is output and the algorithm ends, otherwise it returns to step S372.

[0205] Step S380: Selecting an optimal solution from the Pareto optimal solution set as the current balanced charge and discharge strategy;

[0206] The selection of an optimal solution from the Pareto optimal solution set may be performed by a person skilled in the art according to actual needs.

[0207] Step S400: executing a balanced charge and discharge strategy and predicting the SOH value of each battery in real time, calculating the SOH difference of the current battery pack, and triggering the optimization of the next round of balanced charge and discharge strategy when the SOH difference is greater than a preset difference threshold;

[0208] The specific method of executing the balanced charge and discharge strategy and predicting the SOH value of each battery in real time, calculating the SOH difference of the current battery group, and triggering the optimization of the next round of balanced charge and discharge strategy when the SOH difference is greater than the preset difference threshold is as follows:

[0209] Step S410: executing a balanced charge and discharge strategy, monitoring the battery's port voltage, charge and discharge current, temperature, self-discharge rate, and cycle number in real time through the BMS, predicting the SOH value of each battery, and calculating the SOH difference of the battery pack;

[0210] Step S420: Preset a difference threshold. When the SOH difference is greater than the preset difference threshold, the next round of optimization of the balanced charge and discharge strategy is triggered, and a new optimal balanced charge and discharge strategy is generated based on the updated SOH value and the remaining cycle life of the battery;

[0211] The difference threshold is set by those skilled in the art based on experience.

[0212] Step S500: Batteries with SOH values ​​less than the retirement threshold are listed as retired batteries, and each retired battery is subjected to static and charge-discharge experimental tests, and the percentage of available capacity of each retired battery is calculated to match the cascade utilization scenario for it;

[0213] The specific method of classifying batteries with SOH values ​​less than the retirement threshold as retired batteries, performing static and charge-discharge experimental tests on each retired battery, calculating the percentage of available capacity of each retired battery, and matching it with the cascade utilization scenario is as follows:

[0214] Step S510: preset a battery retirement threshold, compare the SOH value of each battery with the retirement threshold, list the batteries with SOH values ​​less than the retirement threshold as retired batteries, and remove the retired batteries;

[0215] Preferably, the retirement threshold is 0.6;

[0216] Step S520: Perform a 72h static self-discharge test on the retired battery and measure the battery port voltage at 0h and 72h respectively. , , calculate the voltage drop ;

[0217] The voltage drop calculation formula is: ;

[0218] Step S530: Perform a charge and discharge test on the retired battery and record the discharge capacity of the retired battery ;

[0219] The calculation formula of the discharge capacity of the battery is: ,in, Indicates the constant discharge current of the retired battery during the discharge process, It means that the integration of the constant discharge current over the entire discharge time is the discharge capacity of the retired battery;

[0220] Step S540: Based on the discharge capacity and rated capacity of the retired battery The ratio of , calculate the depth of discharge of retired batteries ;

[0221] The calculation formula of the discharge depth is: ;

[0222] Step S550: The cascade utilization scenarios include a home backup power supply scenario, a low-speed electric vehicle scenario, and an uninterruptible power supply scenario. When the discharge depth φ>0.8 and the voltage drop ΔV<5%, the retired battery is allocated to the home backup power supply scenario; when 0.7<φ≤0.8 and the voltage drop ΔV<10%, the retired battery is allocated to the low-speed electric vehicle scenario; when 0.6<φ≤0.7 and the voltage drop ΔV<15%, the retired battery is allocated to the uninterruptible power supply scenario; the remaining retired batteries are recycled.

[0223] like Figure 3 As shown, it is a flow chart of the tiered utilization scenario allocation method provided in this application.

[0224] The above step S500 takes energy conservation and environmental protection into consideration, recycles and reuses retired batteries, and tailors cascade utilization scenarios for batteries of different health levels.

[0225] Example 2

[0226] like Figure 4 As shown, a new energy power station intelligent charging and discharging control device based on battery health status provided by the present application includes:

[0227] The battery health prediction module is used to collect energy storage data of batteries of different types and different use stages, including capacity, internal resistance, self-discharge rate and number of cycles, train a battery health status assessment model, use the battery energy storage data as input data, and output the SOH value;

[0228] The remaining life prediction module is used to collect the SOH value, charge and discharge rate, temperature and discharge depth of the battery during the charge and discharge cycle, build battery training samples for training the battery life prediction model, and predict the remaining cycle life of the battery based on the trained battery life prediction model;

[0229] The balancing strategy optimization module is used to define optimization objectives and constraints. The optimization objectives include minimizing the battery pack capacity decay rate, minimizing the battery pack SOH difference, minimizing the battery pack remaining cycle life standard deviation and maximizing the battery pack energy utilization rate, establishing an optimization model, and solving the optimal balancing charge and discharge strategy;

[0230] The strategy optimization trigger module is used to execute the balanced charge and discharge strategy and predict the SOH value of each battery in real time, calculate the SOH difference of the current battery group, and trigger the optimization of the next round of balanced charge and discharge strategy when the SOH difference is greater than the preset difference threshold;

[0231] The retired battery processing module is used to list batteries with SOH values ​​less than the retirement threshold as retired batteries, conduct static and charge-discharge experimental tests on each retired battery, calculate the percentage of available capacity of each retired battery, and match it with a cascade utilization scenario.

[0232] Example 3

[0233] According to another aspect of the present application, an electronic device is also provided. The electronic device may include one or more processors and one or more memories. The memories store computer-readable codes, and when the computer-readable codes are executed by one or more processors, a new energy power station intelligent charging and discharging control method based on battery health status as described above may be executed.

[0234] The method or system according to the implementation manner of the present application can also be implemented with the aid of the architecture of an electronic device. The electronic device may include a bus, one or more CPUs, a read-only memory (ROM), a random access memory (RAM), a communication port connected to a network, an input / output component, a hard disk, and the like. A storage device in an electronic device, such as a ROM or a hard disk, can store a method for intelligent charging and discharging control of a new energy power station based on the battery health status provided in the present application. A method for intelligent charging and discharging control of a new energy power station based on the battery health status may, for example, include: collecting energy storage data of batteries of different types and at different stages of use, the energy storage data including capacity, internal resistance, self-discharge rate, and number of cycles, training a battery health status assessment model, using the battery energy storage data as input data, and outputting an SOH value; collecting the SOH value, charge and discharge rate, temperature, and discharge depth of the battery during the charge and discharge cycle, constructing a battery training sample for training a battery life prediction model, and predicting the remaining cycle life of the battery based on the trained battery life prediction model; defining optimization objectives and constraints, the optimization objectives including minimum The optimization model is established to optimize the capacity decay rate of the battery pack, minimize the SOH difference of the battery pack, minimize the standard deviation of the remaining cycle life of the battery pack, and maximize the energy utilization rate of the battery pack, and the optimal balanced charge and discharge strategy is obtained; the balanced charge and discharge strategy is executed and the SOH value of each battery is predicted in real time, and the SOH difference of the current battery pack is calculated. When the SOH difference is greater than the preset difference threshold, the optimization of the next round of balanced charge and discharge strategy is triggered; the batteries with SOH values ​​less than the retirement threshold are listed as retired batteries, and each retired battery is subjected to static and charge and discharge experimental tests, and the percentage of available capacity of each retired battery is calculated to match the cascade utilization scenario. Furthermore, the electronic device may also include a user interface. Of course, when implementing different devices, one or more components in the electronic device may be omitted according to actual needs.

[0235] Example 4

[0236] According to one embodiment of the present application, a readable storage medium is also provided. Computer-readable instructions are stored on the computer-readable storage medium. When the computer-readable instructions are executed by the processor, a method for intelligent charging and discharging control of a new energy power station based on the battery health status according to an embodiment of the present application described with reference to the above figures can be executed. The storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache). Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0237] In addition, according to the implementation mode of the present application, the process described in the above method flow chart can be implemented as a computer software program. For example, the present application provides a non-temporary machine-readable storage medium, the non-temporary machine-readable storage medium stores machine-readable instructions, the machine-readable instructions can be executed by a processor to execute instructions corresponding to the method steps provided in the present application, for example: collecting energy storage data of batteries of different types and different use stages, the energy storage data including capacity, internal resistance, self-discharge rate and number of cycles, training a battery health status assessment model, using the battery energy storage data as input data, and outputting the SOH value; collecting the SOH value, charge and discharge rate, temperature and discharge depth of the battery during the charge and discharge cycle, constructing a battery training sample for training a battery life prediction model, and predicting the remaining life of the battery based on the trained battery life prediction model. Remaining cycle life; defining optimization objectives and constraints, the optimization objectives include minimizing the battery pack capacity attenuation rate, minimizing the battery pack SOH difference, minimizing the battery pack remaining cycle life standard deviation and maximizing the battery pack energy utilization rate, establishing an optimization model, and solving the optimal balanced charge and discharge strategy; executing the balanced charge and discharge strategy and predicting the SOH value of each battery in real time, calculating the SOH difference of the current battery pack, and when the SOH difference is greater than the preset difference threshold, triggering the optimization of the next round of balanced charge and discharge strategy; listing batteries with SOH values ​​less than the retirement threshold as retired batteries, performing static and charge and discharge experimental tests on each retired battery, and calculating the available capacity percentage of each retired battery to match it with the cascade utilization scenario. When the computer program is executed by the central processing unit (CPU), the above functions defined in the method of this application are executed.

[0238] The methods, apparatuses, and devices of the present application may be implemented in many ways. For example, the methods, apparatuses, and devices of the present application may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present application are not limited to the order specifically described above unless otherwise specified. In addition, in some embodiments, the present application may also be implemented as programs recorded in a recording medium, which include machine-readable instructions for implementing the method according to the present application. Therefore, the present application also covers recording media storing programs for executing the method according to the present application.

[0239] In addition, the parts of the above-mentioned technical solutions provided in the embodiments of the present application that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0240] The specific implementation modes as described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation mode of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for intelligent charging and discharging control of a new energy power station based on battery health status, characterized in that: include: Collect energy storage data of batteries of different types and at different usage stages, including capacity, internal resistance, self-discharge rate and number of cycles, train a battery health status assessment model, use the battery energy storage data as input data, and output the SOH value; Collect the SOH value, charge and discharge rate, temperature and discharge depth of the battery during the charge and discharge cycle, build battery training samples for training the battery life prediction model, and predict the remaining cycle life of the battery based on the trained battery life prediction model; Define optimization objectives and constraints, where the optimization objectives include minimizing the battery pack capacity decay rate, minimizing the battery pack SOH difference, minimizing the battery pack remaining cycle life standard deviation, and maximizing the battery pack energy utilization rate, establish an optimization model, and solve for the optimal balanced charge and discharge strategy; Execute the balanced charge and discharge strategy and predict the SOH value of each battery in real time, calculate the SOH difference of the current battery group, and trigger the optimization of the next round of balanced charge and discharge strategy when the SOH difference is greater than the preset difference threshold; Batteries with SOH values ​​lower than the retirement threshold are listed as retired batteries. Each retired battery is subjected to static and charge-discharge experimental tests, and the available capacity percentage of each retired battery is calculated to match it with a cascade utilization scenario.

2. A method for intelligent charging and discharging control of a new energy power station based on battery health status as claimed in claim 1, characterized in that: The specific method of collecting energy storage data of batteries of different types and different use stages, wherein the energy storage data includes capacity, internal resistance, self-discharge rate and number of cycles, training a battery health status assessment model, and using the battery energy storage data as input data to output the SOH value is as follows: Apply multi-parameter fusion Kalman filter algorithm to calculate the capacity and internal resistance of the battery; Measure the self-discharge rate of the battery when it is at rest, and read the number of cycles recorded by the BMS; The capacity, internal resistance, self-discharge rate and number of cycles are integrated into the energy storage data of the battery. The energy storage data of batteries of different types and different usage stages are collected, and the corresponding energy storage data is marked with its SOH value. The SOH value is used as a health status label to construct training samples. , each training sample includes the battery energy storage data And the corresponding SOH value ; Select Gaussian kernel as kernel function and build SVM regression model , the expression of the SVM regression model is: , and the SVM regression model satisfies , where w is the weight vector, is the bias term, For the general Functions mapped to high-dimensional feature spaces, Tolerance error, Indicates The SOH value predicted for the input data; Introducing slack variables and , constructing an objective function and constraints, wherein the objective function includes an L2 norm minimization objective function of the weight vector w and an error penalty function between the predicted SOH value and the true SOH value; Transform the objective function into a dual problem; Input the training samples into the SVM regression model, solve the dual problem, and obtain the optimal solution of the Lagrange multiplier and , the optimal solution of the bias term, and the final SVM regression model is obtained according to the optimal solution; The mean square error is used as the loss function of the SVM regression model to optimize the model parameters and obtain the final battery health status assessment model. The trained battery health status assessment model is used to predict the battery SOH value.

3. A method for intelligent charging and discharging control of a new energy power station based on battery health status as claimed in claim 2, characterized in that: The specific method of applying the multi-parameter fusion Kalman filter algorithm to calculate the capacity and internal resistance of the battery is: Select the capacity and internal resistance of the battery as state variables and construct the state vector ; The state vector is ,in, is the capacity of the battery at time k, is the internal resistance of the battery at time k; Based on the state vector Establish the state equation and establish the observation quantity and the state vector The observation variable equation between the observation variables is the terminal voltage and temperature of the battery, and the observation variable equation includes the terminal voltage observation equation and the temperature observation equation; Given an initial state estimate and the error covariance matrix ; Read the charge and discharge current and observation quantity collected by BMS. For each sampling time k, the state vector predicted at time k is obtained according to the state equation. , the predicted state vector Including forecasted capacity and the predicted internal resistance ; Using the state transfer matrix at time k-1 and the error covariance matrix , predict the error covariance matrix at time k ; Based on the maximum capacity of the battery and charge and discharge current , respectively calculate the Jacobian matrix of the port voltage observation equation and the Jacobian matrix of the temperature observation equation , calculate the Kalman gain matrix based on the Jacobian matrix ; The predicted state vector is corrected using the observations and the Kalman gain matrix to calculate the state vector at the current time k , and get the error covariance matrix at time k after the update ; Output the calculated state vector at time k , get the battery capacity at time k and internal resistance .

4. A method for intelligent charging and discharging control of a new energy power station based on battery health status as claimed in claim 3, characterized in that: The SOH value, charge and discharge rate, temperature and discharge depth of the battery during the charge and discharge cycle are collected to construct a battery training sample for training the battery life prediction model. The specific method for predicting the remaining cycle life of the battery based on the trained battery life prediction model is as follows: Perform a complete charge and discharge cycle on the battery until its capacity drops to the battery retirement standard value, and record the number of cycles from the beginning to the actual retirement of the battery. This number of cycles is used as the remaining cycle life of the corresponding battery; During the battery charge and discharge cycle, each charge and discharge cycle is performed to record the battery's SOH value, charge and discharge rate, temperature and discharge depth at the end of the cycle. The charge and discharge rate is the ratio of the charge and discharge current to the rated capacity of the battery. The temperature is collected by a temperature sensor. The discharge depth is the ratio of the battery discharge capacity to the rated capacity of the battery. The SOH value, charge and discharge rate, temperature and discharge depth of the historical battery are arranged in chronological order to obtain the SOH sequence, charge and discharge rate sequence, temperature sequence and discharge depth sequence; The time step length La is preset, and the sliding window method is used to obtain the input sequence of battery training samples on the SOH sequence, charge and discharge rate sequence, temperature sequence and discharge depth sequence respectively; For each input sequence, the corresponding sample label is the remaining cycle life from the end of the current time window to the actual retirement of the battery. Based on the input sequence and the corresponding remaining cycle life, a battery training sample is constructed; The LSTM model is used as the initial model for training the battery life prediction model, the input sequence is used as the input data, and the corresponding remaining cycle life is used as the output data. The LSTM model is trained using the battery training samples. The loss function is the mean square error between the predicted remaining cycle life and the actual remaining cycle life. The training goal is to minimize the value of the loss function. When the loss function reaches convergence, the training is completed. The trained battery life prediction model is used to predict the remaining cycle life of the current battery.

5. The intelligent charging and discharging control method of a new energy power station based on battery health status as claimed in claim 4, characterized in that: The specific definition method of the optimization target is: According to the number of battery cycles , SOH value , Charge and discharge rate and temperature , calculate the capacity decay rate of the battery , according to the capacity and capacity decay rate of each battery, calculate the battery pack capacity decay rate ; To minimize the battery capacity decay rate As the first optimization objective, the first optimization objective function is obtained ; Calculate the SOH difference of the battery pack based on the SOH value of the battery , with minimizing the SOH difference of the battery pack as the second optimization goal, the second optimization objective function is obtained ; Calculate the standard deviation of the remaining cycle life of the battery pack based on the remaining cycle life of the battery , taking minimizing the standard deviation of the remaining cycle life of the battery pack as the third optimization objective, the third optimization objective function is obtained ; According to the nominal capacity of the battery , Maximum allowed charging voltage and minimum discharge voltage Calculate the maximum available energy of a battery under rated conditions , according to the battery charging and discharging power Calculate the energy utilization of the battery in the optimized time domain based on the maximum available energy of the battery under rated conditions , and get the battery pack energy utilization rate , taking maximizing the battery pack energy utilization as the fourth optimization goal, the fourth optimization objective function is obtained .

6. A method for intelligent charging and discharging control of a new energy power station based on battery health status as claimed in claim 5, characterized in that: The specific definition method of the constraint condition is: The charge and discharge power of the battery in the optimization time domain is used as the decision variable, the first optimization objective function, the second optimization objective function, the third optimization objective function, and the fourth optimization objective function are integrated as the optimization objective, and the constraints including the upper and lower limits of the charge and discharge power, the state of charge range constraints, and the total power balance constraints are defined to establish an optimization model; The constraint conditions specifically include: the expression of the upper and lower limits of the charge and discharge power constraints is: ,in, and They represent the minimum charge and discharge power and the maximum charge and discharge power allowed by the current SOH value, respectively, and nh is the total number of batteries; the expression of the state of charge range constraint is: ,in, , Represent the minimum state of charge value and the maximum state of charge value, respectively. represents the state of charge value of the hth battery at time k; the state of charge value is obtained based on the charge and discharge current and capacity based on the differential equation, and the differential equation is: ,in, is the terminal voltage of the hth battery at time k, is the charge and discharge current of the hth battery at time k, is the capacity of the hth battery at time k; the expression of the total power balance constraint is: ,in, is the external load power demand, a positive value indicates discharging, and a negative value indicates charging.

7. A method for intelligent charging and discharging control of a new energy power station based on battery health status as claimed in claim 6, characterized in that: The specific method of classifying batteries with SOH values ​​less than the retirement threshold as retired batteries, performing static and charge-discharge experimental tests on each retired battery, calculating the percentage of available capacity of each retired battery, and matching it with the cascade utilization scenario is as follows: Preset the retirement threshold of the battery, compare the SOH value of each battery with the retirement threshold, list the batteries with SOH values ​​less than the retirement threshold as retired batteries, and remove the retired batteries; The retired battery was subjected to a 72h static self-discharge test, and the battery port voltage was measured at 0h and 72h respectively. , , calculate the voltage drop ; Conduct charge and discharge tests on retired batteries and record the discharge capacity of retired batteries ; Based on the discharge capacity and rated capacity of retired batteries The ratio of , calculate the depth of discharge of retired batteries ; The cascade utilization scenarios include home backup power supply scenarios, low-speed electric vehicle scenarios, and uninterruptible power supply scenarios. >0.8 and voltage drop When the retired batteries are allocated to the home backup power scenario, And the voltage drop When the retired batteries are allocated to the low-speed electric vehicle scenario, And the voltage drop When the battery is used, the retired batteries are allocated to the uninterruptible power supply scenario; the remaining retired batteries are recycled.

8. An intelligent charge and discharge control device for a new energy power station based on the battery health status, which is implemented based on an intelligent charge and discharge control method for a new energy power station based on the battery health status as described in any one of claims 1 to 7, characterized in that: include: The battery health prediction module is used to collect energy storage data of batteries of different types and different use stages, including capacity, internal resistance, self-discharge rate and number of cycles, train a battery health status assessment model, use the battery energy storage data as input data, and output the SOH value; The remaining life prediction module is used to collect the SOH value, charge and discharge rate, temperature and discharge depth of the battery during the charge and discharge cycle, build battery training samples for training the battery life prediction model, and predict the remaining cycle life of the battery based on the trained battery life prediction model; The balancing strategy optimization module is used to define optimization objectives and constraints. The optimization objectives include minimizing the battery pack capacity decay rate, minimizing the battery pack SOH difference, minimizing the battery pack remaining cycle life standard deviation and maximizing the battery pack energy utilization rate, establishing an optimization model, and solving the optimal balancing charge and discharge strategy; The strategy optimization trigger module is used to execute the balanced charge and discharge strategy and predict the SOH value of each battery in real time, calculate the SOH difference of the current battery group, and trigger the optimization of the next round of balanced charge and discharge strategy when the SOH difference is greater than the preset difference threshold; The retired battery processing module is used to list batteries with SOH values ​​less than the retirement threshold as retired batteries, perform static and charge-discharge experimental tests on each retired battery, calculate the percentage of available capacity of each retired battery, and match it with a cascade utilization scenario.

9. An electronic device, characterized in that: It includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps in the intelligent charging and discharging control method of a new energy power station based on the battery health status as described in any one of claims 1 to 7.

10. A readable storage medium, characterized in that: The readable storage medium stores a computer program, which is suitable for loading by a processor to execute the steps in a new energy power station intelligent charging and discharging control method based on battery health status as described in any one of claims 1 to 7.

Images