Intelligent dynamic evaluation method and equipment for safety adjustment boundary of power battery

Abstract

The invention relates to an intelligent dynamic evaluation method and equipment for a safety adjustment boundary of a power battery, and the method comprises the following steps: collecting various test data, and constructing a power battery database; based on the power battery database, establishing a power battery second-order equivalent circuit model in combination with an extended Kalman filtering algorithm, and generating state-of-charge estimation results without considering aging influence under different working conditions; establishing a power battery health state estimation model by combining a neural network algorithm based on the power battery database and the charge state evolution relationship, and generating power battery health state estimation results in different working condition environments; and constructing a multi-constraint power state estimation model considering the state of charge, the state of health and the maximum charging and discharging current, and generating a power battery adjustment safety boundary result. Compared with the prior art, the method has the advantages of being capable of accurately achieving power battery safety adjustment boundary evaluation under multiple types of working conditions, high in reliability and the like.

Description

Technical Field

[0001] The present invention relates to the field of vehicle-grid interaction technology, and in particular to a method and device for intelligent dynamic evaluation of power battery safety adjustment boundaries. Background Art

[0002] With the large-scale grid connection of new energy sources, the problem of safe operation of power grids has become increasingly prominent. Since electric vehicles have the ability to respond quickly, accurately and smoothly to grid regulation, they have gradually become an important resource for the safety support of large urban power grids. In order to respond to the dispatching requirements of the superior power grid, if the output of electric vehicles is not constrained by charging piles, the power batteries are very likely to be overcharged and over-discharged, seriously heated, and accelerate battery aging, and even cause safety accidents. On the other hand, electric vehicle users are not obliged to participate in grid interaction. It is necessary to evaluate the power battery regulation cost and mobilize the enthusiasm of users in the form of economic incentives to ensure the stability of the dispatchable resources of electric vehicles. Therefore, accurate evaluation of the adjustable safety boundary of electric vehicle power batteries and quantification of the regulation cost are crucial to responding to the needs of the superior power grid and providing strong support.

[0003] In the field of power battery safety boundary assessment, most studies only focus on the power characteristics of power batteries at different states of charge on a short time scale, ignoring that cycle aging will reduce the power performance of power batteries. Therefore, current research mainly conducts joint estimation of state of charge and power state, and rarely considers the coupling relationship between state of charge, power state and health state. On the other hand, different operating conditions, such as different temperatures, discharge depths, state of charge mean values, charge and discharge rates, etc., will further aggravate the coupling of the above three states, greatly increase the difficulty of power battery state estimation, and make its safety boundary difficult to quantify. At present, the commonly used fusion equivalent circuit and Kalman filtering methods are difficult to accurately solve the state characteristics of highly nonlinear systems.

[0004] Due to the complexity of the capacity attenuation process and unclear mechanism of power batteries under complex working conditions, there is currently little attention paid to the life attenuation caused by the participation of electric vehicles in grid regulation. There is still a lack of research on the cost quantification of power batteries participating in grid regulation. When participating in grid regulation under the constraints of power battery safety boundaries, how to make the safe and healthy status of power batteries transparent, and how to reasonably compensate participating users are key issues that need to be urgently addressed. Summary of the invention

[0005] One of the purposes of the present invention is to provide a method and device for intelligent dynamic evaluation of the safety adjustment boundary of a power battery in order to solve the technical problem that the existing power battery state estimation method is difficult to adapt to complex working conditions.

[0006] The second purpose of the present invention is to provide a method and device for intelligent dynamic evaluation of the safety regulation boundary of a power battery in order to solve the technical problems that the prior art lacks the ability to evaluate the safety regulation boundary of the power battery, the capacity decay process is difficult to quantify, and the regulation cost evaluation is difficult.

[0007] The purpose of the present invention can be achieved by the following technical solutions:

[0008] A method for intelligent dynamic evaluation of power battery safety regulation boundary comprises the following steps:

[0009] Collect various test data and build a power battery database, including a power battery initial identification data set and a power battery status dynamic evolution data set;

[0010] Based on the power battery database, a second-order equivalent circuit model of the power battery is established in combination with an extended Kalman filter algorithm to generate state of charge estimation results without considering the influence of aging under different operating conditions;

[0011] Based on the relationship between the power battery database and the state of charge evolution, a power battery health state estimation model is established in combination with a neural network algorithm to generate power battery health state estimation results under different working conditions;

[0012] The power battery health state estimation result is input into the power battery second-order equivalent circuit model, a multi-constraint power state estimation model considering the charge state, health state and maximum charge and discharge current is constructed, and a power battery adjustment safety boundary result is generated.

[0013] Furthermore, the test includes a power battery open circuit voltage and state of charge test, a hybrid power pulse test, and a power battery cycle aging test at different temperatures, discharge depths, state of charge averages, and charge and discharge rates. Based on the power battery open circuit voltage and state of charge test and the hybrid power pulse test, an initial identification data set of the power battery under static and pulse response is generated. Based on the power battery cycle aging test at different temperatures, discharge depths, discharge averages, and charge and discharge rates, a dynamic evolution data set of the power battery state under different operating conditions is generated.

[0014] Furthermore, generating the state of charge estimation results without considering the influence of aging under different working conditions specifically includes:

[0015] Based on the initial identification data set of the power battery and the least squares algorithm, a second-order equivalent circuit model of the power battery is constructed and parameter identification is realized;

[0016] The external characteristic data of the power battery under different working conditions are used as the input of the second-order equivalent circuit model of the power battery, and combined with the extended Kalman filter algorithm, the state of charge estimation without considering the influence of battery aging is realized, and the state of charge estimation results without considering the influence of aging under different working conditions are generated.

[0017] Furthermore, the least squares algorithm is a weighted adaptive least squares method.

[0018] Furthermore, generating the power battery health status estimation results under different working conditions specifically includes:

[0019] Extracting power battery aging characteristics based on the power battery database and the state of charge evolution relationship;

[0020] A power battery health state estimation model is constructed, and the power battery aging characteristics are used as input to generate power battery health state estimation results under different working conditions. The power battery health state estimation model is a back-propagation neural network data-driven model.

[0021] Furthermore, when the back-propagation neural network data-driven model is trained, the weights and bias values ​​of the network are automatically updated through the back-propagation algorithm based on the training set, the model estimation performance is evaluated based on the test set using the root mean square error and determination coefficient indicators, and the network super parameters are adjusted based on the model estimation performance evaluation results.

[0022] Furthermore, the expression of the power battery adjustment safety margin is:

[0023]

[0024] in, and are the peak discharge current and peak charge current at time k respectively; I max and I min They are the maximum discharge current and minimum charge current specified in the factory design of the power battery; and are the peak discharge current and peak charge current under the state of charge constraint; P min and P max They are the minimum charging power and maximum discharging power limits set during power battery design and application; and are the peak discharge power and peak charging power estimated by the multi-constrained dynamic continuous peak power estimation method at time k; U k (·) represents the battery terminal voltage at the peak charge (discharge) current at time k.

[0025] Furthermore, the method further comprises:

[0026] The power battery capacity attenuation rate is used to measure the life loss of the power battery participating in regulation. Based on the power battery investment and operation and maintenance costs, cost estimation results under different capacity attenuation rates are generated.

[0027] Furthermore, the specific process of generating cost estimation results under different capacity decay rates includes:

[0028] Fit the relationship between the power battery capacity attenuation rate and the discharge rate, discharge depth and temperature, generate a power battery capacity attenuation model suitable for different adjustment behaviors, and calculate the capacity attenuation rate of the power battery reference cycle number;

[0029] The power battery investment and operation and maintenance costs corresponding to different power battery capacity attenuations are apportioned, and a power battery cost evaluation model suitable for different regulation behaviors is established. The power battery cost evaluation model is based on the ratio of the capacity attenuation caused by different regulation behaviors to the benchmark value, as well as the mapping relationship between the power battery capacity attenuation and the power battery investment and maintenance costs, and the regulation cost of the power battery participating in grid regulation is obtained.

[0030] The present invention also provides an electronic device, comprising one or more processors, a memory and one or more programs stored in the memory, wherein the one or more programs include instructions for executing the intelligent dynamic assessment method for the safety adjustment boundary of a power battery as described above.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] 1. The present invention carries out open circuit voltage and state of charge test and hybrid power pulse characteristic test to generate initial identification data sets of power batteries under static and pulse response; carries out power battery cycle aging experiments considering different temperatures, discharge depths, state of charge mean values, and charge and discharge rates to generate dynamic evolution data sets of power battery states under corresponding working conditions; based on the initial identification data set of power batteries and the extended Kalman filter algorithm, a second-order equivalent circuit model of power batteries is established to realize state of charge estimation without considering the influence of battery aging; based on the dynamic evolution data of power battery states under different working conditions, the neural network data-driven method is combined to realize aging state estimation considering different working conditions; analyzes the coupling relationship between state of charge and health state to realize dynamic update of aging parameters of second-order equivalent circuit model of power batteries, constructs a multi-constraint power state estimation model of power batteries considering maximum charge and discharge current, state of charge, and health state, and realizes intelligent dynamic evaluation of power battery safety regulation boundaries adapted to multiple types of working conditions. The present invention can accurately realize the evaluation of power battery safety regulation boundaries under multiple types of working conditions through the above process, with high reliability.

[0033] 2. The present invention is based on the intelligent dynamic evaluation results of the safety regulation boundary of power batteries. It adopts the capacity decay rate to measure the life loss of power batteries participating in regulation under different working conditions. Based on the investment and maintenance costs of power batteries, an empirical model of capacity decay rate and regulation cost is generated to achieve quantification of regulation cost considering the service life of power batteries, improve the observability and controllability of power batteries, and is of great significance for encouraging electric vehicle users to participate in grid regulation and supporting safe and stable operation of grids.

[0034] 3. The present invention can provide new ideas for the development of multi-state coupling estimation and capacity attenuation quantification technology for power batteries under complex working conditions, and provide theoretical support for the safety and reliability of electric vehicles participating in grid regulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 is a flow chart of the steps of the method of the present invention;

[0036] Figure 2 It is the cycle aging curve of the power battery at different discharge rates;

[0037] Figure 3 It is the comparison curve between the estimated SOC value and the actual value at different discharge rates of the power battery;

[0038] Figure 4 It is a comparison curve between the estimated value and the actual value of the health status of the power battery at different discharge rates;

[0039] Figure 5 It is the peak power curve of charge and discharge of power battery at different aging degrees;

[0040] Figure 6 Fitting surface of power battery capacity attenuation, temperature, and charge / discharge rate;

[0041] Figure 7 It is the fitting surface of power battery capacity attenuation, discharge depth and charge and discharge rate. DETAILED DESCRIPTION

[0042] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0043] Example 1

[0044] refer to Figure 1 As shown, this embodiment provides a method for intelligent dynamic evaluation of power battery safety regulation boundary, including:

[0045] Step S1, collecting various test data and constructing a power battery database, including a power battery initial identification data set and a power battery state dynamic evolution data set.

[0046] Specifically, the tests of this embodiment include open circuit voltage and state of charge test of the power battery, hybrid power pulse test and power battery cycle aging test under different temperatures, discharge depths, state of charge averages and charge and discharge rates. Based on the open circuit voltage and state of charge test of the power battery and the hybrid power pulse test, the initial identification data set of the power battery under static and pulse response is generated. Based on the power battery cycle aging test under different temperatures, discharge depths, discharge averages and charge and discharge rates, the dynamic evolution data set of the power battery state under different operating conditions is generated.

[0047] Figure 2 The figure shows the cycle aging curve of the power battery at different discharge rates, including 0.5C, 1C, 2C and 3C discharge rates. It can be seen from the figure that with the increase of the number of cycles, the aging of the power battery is aggravated, and the greater the discharge rate, the faster the power battery ages.

[0048] Step S2: Based on the power battery database, a second-order equivalent circuit model of the power battery is established in combination with an extended Kalman filter algorithm to generate state of charge estimation results without considering the influence of aging under different operating conditions, specifically including:

[0049] S201: Based on the power battery initial identification data set and the least squares algorithm, complete the second-order equivalent circuit model parameter identification of the power battery.

[0050] 1) Use one ohm internal resistor and two RC circuits to construct the Thevenin equivalent circuit model of the power battery, and its mathematical expression is:

[0051]

[0052] Among them, E is the electromotive force of the power battery, in V; I(t) is the current, in A; U(t) is the terminal voltage, in V; I(t) is the current, in A; R0, R1, and R2 are the ohmic internal resistance, the charged polarization internal resistance, and the concentration polarization internal resistance, respectively, in Ω; U1 and U2 are the charged polarization internal resistance voltage and the concentration polarization internal resistance voltage, respectively, in V; C1 and C2 are the charged polarization capacitance and the concentration polarization capacitance, respectively, in F; SOC(t) represents the state of charge at time t.

[0053] Based on the definition of power battery state of charge (SOC) and the second-order equivalent circuit model, the state equation and observation equation of the power battery model are obtained:

[0054]

[0055] U(kT)=f(SOC(kT))-U1(kT)-U2(kT)-R0i(kT) (3)

[0056] Among them, τ1=R1C1 and τ2=R2C2, the unit is s; T is the sampling period, the unit is s; η is the charge and discharge efficiency; C N It is the rated power battery capacity in Ah.

[0057] 2) In the above second-order RC equivalent model, the OCV-SOC relationship is obtained by fitting the open circuit voltage and state of charge data obtained by S101. On this basis, the parameters of the battery equivalent model are identified by weighted adaptive least squares method. The specific process is as follows:

[0058] Perform Laplace transform on formula (1):

[0059]

[0060] Where E(s) is the open circuit voltage and the time domain unit is s. With E(s)-U(s) as input and I(s) as output, the transfer function of the model is calculated as:

[0061]

[0062] Furthermore, based on equation (6), equation (5) is bilinearly transformed to obtain the discrete transfer function (7):

[0063]

[0064] Furthermore, equation (7) is transformed into a difference equation and expressed in vector form as equation (9):

[0065] y(n)=a1y(n-1)+a2y(n-2)+a3I(n)+a4I(n-1)+a5I(n-2) (8)

[0066]

[0067] Among them, a1, a2, a3, a4 and a5 are unknown coefficients. Equation (8) is extended to N dimensions, and the recursive least squares method is used, with the minimum residual sum of squares as the objective function. The recursive equation is shown in equation (11). On this basis, the adaptive weight factor is used to adjust the recursive least squares method, and the weighted adaptive recursive equation is obtained as follows:

[0068]

[0069] Among them, λ(n) is the adaptive weight factor, l and w are adjustable parameters. When the error increases, the weight factor decreases, and when the error decreases, the weight factor increases. The optimal parameters are calculated by formula (11) and substituted into formula (7), and the identification equation of the model is obtained as shown in formula (12). Combined with the above formula, the parameters of the power battery Thevenin equivalent circuit model can be solved:

[0070]

[0071] S202: Using the external characteristic data of the power battery under different working conditions as input and combining it with the extended Kalman filter algorithm, the state of charge estimation without considering the influence of battery aging is realized.

[0072] 1) Construct the power battery state equation and observation equation as follows:

[0073]

[0074] Among them, x n is the state variable of the system at time n; y n is the output variable; w n is the state noise; v n represents the observation noise. On this basis, the nonlinear function f(x n ,u n ) and g(x n ,u n ) is expanded by the first order Taylor as follows:

[0075]

[0076] in, is the system estimate at time n.

[0077] 2) To solve the power battery SOC, the above formula can be further expressed as follows after discretization:

[0078]

[0079] According to formula (16), the power battery SOC value can be recursively estimated by using the Kalman filter.

[0080] Figure 3 The figure shows the comparison curve of the estimated SOC value and the true value at different discharge rates of the power battery, including 0.5C, 1C and 2C discharge rates. It can be seen from the figure that the greater the discharge rate, the faster the power battery SOC decreases; the estimated value and the true value change trend are consistent, and this method has a good SOC estimation performance.

[0081] Step S3: Based on the power battery database and the charge state evolution relationship, a power battery health state estimation model is established in combination with a neural network algorithm to generate power battery health state estimation results under different working conditions, specifically including:

[0082] S301: extracting power battery aging characteristics based on the power battery database and the state of charge evolution relationship;

[0083] S302: Construct a power battery health state estimation model, take the power battery aging characteristics as input, and generate power battery health state estimation results under different working conditions, wherein the power battery health state estimation model is a back propagation neural network data driven model.

[0084] When training a back-propagation neural network data-driven model, the weights and bias values ​​of the network are automatically updated through the back-propagation algorithm based on the training set. The model estimation performance is evaluated using the root mean square error and determination coefficient indicators based on the test set. The network hyperparameters are adjusted based on the model estimation performance evaluation results. Specifically:

[0085] 1) By extracting the battery aging characteristics under different working conditions from the power battery body characteristic database, a back propagation neural network model input data set is constructed and split into a training set and a test set in a ratio of 7:3;

[0086] 2) Initialize the network structure of the backpropagation neural network model, including the number of layers and neurons in the input layer, hidden layer, and output layer, and customize the hyperparameters of the backpropagation neural network model, including the maximum number of iterations, learning rate, and number of hidden layer nodes;

[0087] 3) Use the training set to train the back-propagation neural network, and automatically update the weights and bias values ​​of the network through the back-propagation algorithm to generate the minimum network error;

[0088] 4) Use the test set to test the trained backpropagation neural network, preliminarily generate the estimated value of the health status of the power battery, and evaluate the estimation performance of the backpropagation neural network model based on the root mean square error and determination coefficient indicators;

[0089] 5) The method of manually optimizing the back-propagation neural network is used to adjust the network hyper parameters based on the indicator evaluation results to avoid "underfitting" and "overfitting" of the back-propagation neural network model.

[0090] The optimized back-propagation neural network model is used to evaluate the life attenuation of the power battery under operating conditions of different temperatures, discharge depths, state of charge averages, and charge and discharge rates, and finally an estimated value of the power battery health status is generated.

[0091] Figure 4The comparison curve of the estimated value and the true value of the health state of the power battery at different discharge rates, including 0.5C, 1C, 2C and 3C discharge rates. It can be seen from the figure that the larger the discharge rate, the faster the power battery ages; the estimated value and the true value change trend are consistent, and this method has a good health state estimation performance.

[0092] Step S4: input the power battery health state estimation result into the power battery second-order equivalent circuit model, construct a multi-constraint power state estimation model that considers the state of charge, health state, and maximum charge and discharge current, and generate a power battery regulation safety margin result:

[0093] S401: Based on the current state of charge and health state, update the aging parameters of the second-order equivalent circuit model of the power battery;

[0094] S402: Construct a multi-constraint power state estimation model for a power battery that takes into account maximum charge and discharge current, state of charge, and state of health, and implement dynamic estimation of the power battery's safety regulation boundary.

[0095] In this embodiment, the constraints considered by the multi-constraint power state estimation model include:

[0096] 1) Maximum charge and discharge current constraint: The maximum discharge current I specified by the power battery factory design max , minimum charging current I min , which serves as a constraint to prevent the power battery from operating beyond the tolerance of the energy storage body.

[0097] 2) State of charge constraint: The maximum limit of the state of charge specified by the control system max and the minimum limit z min , the peak current of charge and discharge under the state of charge constraint is calculated as:

[0098]

[0099] Among them, z k is the SOC at time k; η is the power battery charge and discharge efficiency; C a It is the current maximum available capacity of the power battery, in Ah.

[0100] 3) Health status constraint: As aging progresses, the maximum available capacity of the power battery gradually decreases, and the charging and discharging capabilities weaken. It is necessary to update the maximum available capacity C based on the current health status. a .

[0101] Based on the above constraints, the multi-constrained peak dynamic current and power battery peak power are calculated to build the power battery safety adjustable boundary:

[0102]

[0103] in, and are the peak discharge current and peak charge current at time k, in A; P min and P max They are the minimum charging power and maximum discharging power limits set during power battery design and application, in W. and are the peak discharge power and peak charging power estimated by the multi-constrained dynamic continuous peak power estimation method at time k, in W; U k (·) represents the battery terminal voltage at the peak charge (discharge) current at time k.

[0104] Figure 5 The peak power curves of charge and discharge of power batteries at different aging levels include 0.5C, 1C and 2C discharge rates. As can be seen from the figure, the more severe the aging, the smaller the peak power of charge and discharge of the power battery, that is, the smaller the adjustable safety margin.

[0105] refer to Figure 1 As shown, in a preferred embodiment, the above method further includes:

[0106] Step S5: using the power battery capacity attenuation rate as a measure of the life loss of the power battery participating in the regulation, and generating cost estimation results under different capacity attenuation rates based on the power battery investment and operation and maintenance costs, specifically including:

[0107] S501: Fitting the relationship between the power battery capacity attenuation rate and the discharge rate, discharge depth and temperature to generate a power battery capacity attenuation model suitable for different adjustment behaviors.

[0108] 1) Process the battery aging data under multiple working conditions in the power battery characteristic database and calculate the power battery capacity attenuation rate;

[0109] 2) Using the mapping relationship between different working environment influencing factors and the power battery capacity attenuation rate, a power battery capacity attenuation model with temperature, discharge depth, state of charge mean, and charge and discharge rate as dependent variables is fitted to generate the capacity attenuation rate under different power battery regulation behaviors.

[0110] S502: Allocate the power battery investment and operation and maintenance costs corresponding to different power battery capacity decays, and establish a power battery cost evaluation model suitable for different adjustment behaviors.

[0111] 1) Using the relationship between power battery investment, maintenance cost, total capacity and rated life of power batteries, establish a mapping relationship between power battery capacity attenuation and power battery investment and maintenance cost;

[0112] 2) Based on the capacity attenuation rate under different regulation behaviors, the regulation cost of power batteries participating in grid regulation is generated.

[0113]

[0114] Among them, C adjust Indicates the power battery adjustment cost, in yuan; C in Represents the total investment cost of power batteries, in yuan; C main Indicates the power battery operation and maintenance cost, in yuan; Cycle N is the reference cycle number of the power battery under specific working conditions before leaving the factory (which can be further converted into the product service life); De(·) represents the capacity attenuation rate of the power battery under different adjustment behaviors; DOD represents the depth of discharge; C Rate Indicates the charge and discharge rate; T indicates the ambient temperature, in °C; SOC mean De represents the mean value of charge state change; N Indicates the capacity attenuation rate of the power battery under the rated reference cycle number of factory life.

[0115] In step S5, the capacity decay rate of the reference cycle number of the power battery is calculated, and based on this, the ratio of the capacity decay caused by different adjustment behaviors to the reference value is calculated, so as to obtain the adjustment cost of the power battery participating in the grid adjustment based on the mapping relationship between the power battery capacity decay and the power battery investment and maintenance cost.

[0116] Figure 6 The figure shows the power battery capacity attenuation, temperature, and charge / discharge rate fitting surface, where the sampling temperatures include 0°C, 25°C, 45°C, and 65°C, and the sampling charge / discharge rates include 0.5C, 1C, 2C, and 3C. As can be seen from the figure, both too low and too high temperatures will accelerate the power battery capacity attenuation, and the power battery is more suitable for working at an ambient temperature of 20°C; the higher the charge / discharge rate, the greater the power battery capacity attenuation.

[0117] Figure 7 The figure shows the power battery capacity decay, discharge depth, and charge / discharge rate fitting surface, where the sampled discharge depths include 0.2, 0.4, 0.6, 0.8, and 1, and the sampled charge / discharge rates include 0.5C, 1C, 2C, and 3C. As can be seen from the figure, a large discharge depth will accelerate the power battery capacity decay; the higher the charge / discharge rate, the greater the power battery capacity decay. Therefore, the power regulation behavior of a small rate and a small discharge depth helps to extend the service life.

[0118] In the embodiment of the present invention, an initial identification data set of a power battery under static and pulse response is generated by conducting open circuit voltage and state of charge tests and hybrid power pulse characteristic tests. A power battery cycle aging experiment considering different temperatures, discharge depths, state of charge mean values, and charge and discharge rates is carried out to generate a dynamic evolution data set of the power battery state under the corresponding working conditions. Based on the initial identification data set of the power battery and the extended Kalman filter algorithm, a second-order equivalent circuit model of the power battery is established to achieve state of charge estimation without considering the influence of battery aging. Based on the dynamic evolution data of the power battery state, a power battery health state estimation model under different working conditions is fitted, and a neural network data-driven method is used to achieve aging state estimation considering different working conditions. The coupling relationship between the state of charge and the health state is analyzed to achieve dynamic update of the aging parameters of the second-order equivalent circuit model of the power battery, and a multi-constrained power state estimation model of the power battery considering the maximum charge and discharge current, state of charge, and health state is constructed to achieve intelligent dynamic evaluation of the safety adjustment boundary of the power battery adapted to multiple types of working conditions. The capacity decay rate is used to measure the life loss of power batteries participating in regulation under different working conditions. Based on the investment and maintenance costs of power batteries, an empirical model of capacity decay rate and regulation cost is generated to quantify the regulation cost considering the service life of power batteries.

[0119] If the above method is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0120] Example 2

[0121] This embodiment provides an electronic device, including one or more processors, a memory, and one or more programs stored in the memory, wherein the one or more programs include instructions for executing the intelligent dynamic assessment method for the safety adjustment boundary of a power battery as described in Example 1.

[0122] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. The schemes in the embodiments of the present invention may be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0123] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0124] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0125] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0126] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0127] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A method for intelligent dynamic evaluation of power battery safety regulation boundary, characterized in that: The following steps are involved: Collect various test data and build a power battery database, including a power battery initial identification data set and a power battery status dynamic evolution data set; Based on the power battery database, a second-order equivalent circuit model of the power battery is established in combination with an extended Kalman filter algorithm to generate state of charge estimation results without considering the influence of aging under different operating conditions; Based on the relationship between the power battery database and the state of charge evolution, a power battery health state estimation model is established in combination with a neural network algorithm to generate power battery health state estimation results under different operating conditions; The power battery health state estimation result is input into the power battery second-order equivalent circuit model, a multi-constraint power state estimation model considering the charge state, health state and maximum charge and discharge current is constructed, and a power battery adjustment safety boundary result is generated.

2. The intelligent dynamic evaluation method for power battery safety regulation boundary according to claim 1 is characterized in that: The test includes a power battery open circuit voltage and state of charge test, a hybrid power pulse test, and a power battery cycle aging test at different temperatures, discharge depths, state of charge averages, and charge and discharge rates. Based on the power battery open circuit voltage and state of charge test and the hybrid power pulse test, an initial identification data set of the power battery under static and pulse response is generated. Based on the power battery cycle aging test at different temperatures, discharge depths, discharge averages, and charge and discharge rates, a dynamic evolution data set of the power battery state under different operating conditions is generated.

3. The intelligent dynamic evaluation method for power battery safety regulation boundary according to claim 1 is characterized in that: The generating of the state of charge estimation results without considering the influence of aging under different working conditions specifically includes: Based on the initial identification data set of the power battery and the least squares algorithm, a second-order equivalent circuit model of the power battery is constructed and parameter identification is realized; The external characteristic data of the power battery under different working conditions are used as the input of the second-order equivalent circuit model of the power battery, and combined with the extended Kalman filter algorithm, the state of charge estimation without considering the influence of battery aging is realized, and the state of charge estimation results without considering the influence of aging under different working conditions are generated.

4. The intelligent dynamic evaluation method for power battery safety regulation boundary according to claim 3 is characterized in that: The least squares algorithm is a weighted adaptive least squares method.

5. The intelligent dynamic evaluation method for power battery safety regulation boundary according to claim 1 is characterized in that: The generating of the power battery health status estimation results under different working conditions specifically includes: Extracting power battery aging characteristics based on the power battery database and the state of charge evolution relationship; A power battery health state estimation model is constructed, and the power battery aging characteristics are used as input to generate power battery health state estimation results under different working conditions. The power battery health state estimation model is a back-propagation neural network data-driven model.

6. The intelligent dynamic evaluation method for power battery safety regulation boundary according to claim 5 is characterized in that: When the back-propagation neural network data-driven model is trained, the weights and bias values ​​of the network are automatically updated through the back-propagation algorithm based on the training set, the model estimation performance is evaluated based on the test set using the root mean square error and determination coefficient indicators, and the network super parameters are adjusted based on the model estimation performance evaluation results.

7. The intelligent dynamic evaluation method for power battery safety regulation boundary according to claim 1 is characterized in that: The expression of the power battery regulation safety margin is: in, and are the peak discharge current and peak charge current at time k respectively; I max and I min They are the maximum discharge current and minimum charge current specified in the factory design of the power battery; and are the peak discharge current and peak charge current under the state of charge constraint; P min and P max They are the minimum charging power and maximum discharging power limits set during power battery design and application; and are the peak discharge power and peak charging power estimated by the multi-constrained dynamic continuous peak power estimation method at time k; U k (·) represents the battery terminal voltage at the peak charge (discharge) current at time k.

8. The intelligent dynamic evaluation method for power battery safety regulation boundary according to claim 1 is characterized in that: The method further includes: The power battery capacity attenuation rate is used to measure the life loss of the power battery participating in regulation. Based on the power battery investment and operation and maintenance costs, cost estimation results under different capacity attenuation rates are generated.

9. The intelligent dynamic evaluation method for power battery safety regulation boundary according to claim 8 is characterized in that: The specific process of generating the cost estimation results under different capacity decay rates includes: Fit the relationship between the power battery capacity attenuation rate and the discharge rate, discharge depth and temperature, generate a power battery capacity attenuation model suitable for different adjustment behaviors, and calculate the capacity attenuation rate of the power battery reference cycle number; The power battery investment and operation and maintenance costs corresponding to different power battery capacity attenuations are apportioned, and a power battery cost evaluation model suitable for different regulation behaviors is established. The power battery cost evaluation model is based on the ratio of the capacity attenuation caused by different regulation behaviors to the benchmark value, as well as the mapping relationship between the power battery capacity attenuation and the power battery investment and maintenance costs, and the regulation cost of the power battery participating in grid regulation is obtained.

10. An electronic device, characterized in that: It includes one or more processors, a memory and one or more programs stored in the memory, and the one or more programs include instructions for executing the intelligent dynamic assessment method for the safety adjustment boundary of a power battery as described in any one of claims 1-9.

Images