Battery charge state estimation method and device, equipment and storage medium

Abstract

The invention relates to the technical field of batteries, and discloses a battery charge state estimation method and device, equipment and a storage medium. The method comprises the following steps: constructing a fractional order equivalent circuit model of a to-be-tested battery; constructing a first state equation set according to the fractional order equivalent circuit model, and discretizing the first state equation set to obtain a second state equation set; performing parameter identification on the fractional order equivalent circuit model by adopting a particle swarm optimization algorithm; and estimating the state of charge of the to-be-tested battery by adopting a self-adaptive extended Kalman filtering algorithm according to the second state equation set after the parameters are identified. According to the method, the equivalent circuit model is constructed by adopting the fractional order model, the fractional order model can better reflect the nonlinear characteristic of the battery, so that the precision of the circuit model is improved, and on the basis of the established equivalent circuit model, the state of charge is estimated in combination with the adaptive fractional order extended Kalman filtering algorithm, so that the accuracy of the circuit model is improved. Therefore, the accuracy of state-of-charge estimation can be improved.

Description

Technical Field

[0001] The present application relates to the field of battery technology, and in particular to a battery state of charge estimation method, device, equipment and storage medium. Background Art

[0002] With the gradual development of the new energy industry, more attention has been paid to the performance parameters of lithium batteries, among which the parameters that are often concerned are: the state of charge (SOC) and state of health (SOH) of lithium batteries. Accurate and fast estimation of SOC is particularly important, and it is also the difficulty and focus that needs to be overcome urgently. The battery state of charge reflects the remaining power of the battery, and its accuracy will affect the performance of the battery and the user's experience of using the product.

[0003] Currently, research on SOC is also very extensive. By establishing a suitable equivalent circuit model, SOC can be estimated based on the established equivalent circuit model. However, the current lithium battery model uses an integer-order model, which is not only limited in accuracy, but also has strong nonlinear characteristics. The integer-order model cannot well reflect the nonlinear characteristics of the battery, and thus cannot guarantee the accuracy of the model, resulting in low accuracy of the state of charge estimation. Summary of the invention

[0004] In view of this, in order to solve the technical problem of low accuracy of battery state of charge estimation results in the prior art, the present application provides a battery state of charge estimation method, device, equipment and storage medium.

[0005] In a first aspect, the present application provides a battery state of charge estimation method, comprising:

[0006] Construct a fractional-order equivalent circuit model of the battery to be tested;

[0007] Constructing a first state equation group according to the fractional-order equivalent circuit model, and discretizing the first state equation group to obtain a second state equation group;

[0008] Using a particle swarm optimization algorithm to perform parameter identification on the fractional-order equivalent circuit model;

[0009] An adaptive extended Kalman filter algorithm is used to estimate the state of charge of the battery to be tested according to the second state equation group after parameter identification.

[0010] In an optional implementation, the adopting a particle swarm optimization algorithm to perform circuit parameter identification on the fractional-order equivalent circuit model includes:

[0011] A population is defined, and the position and velocity of each particle in the population are initialized; each particle represents a circuit parameter to be identified;

[0012] A fitness function is defined, and according to the fitness function, the fitness value of each particle is iteratively calculated to determine the optimal position of each particle;

[0013] The optimal position of each particle is taken as the optimal solution of each circuit parameter.

[0014] In an optional implementation, the fitness function is expressed as:

[0015]

[0016] In the formula, J represents fitness; N data Indicates the length of all data, U ce Indicates the actual terminal voltage value, U L represents the terminal voltage value estimated by the fractional-order equivalent circuit model, and i represents the i-th data.

[0017] In an optional embodiment, the iterative calculation of the fitness value of each particle according to the fitness function to determine the optimal position of each particle includes:

[0018] According to the preset speed update formula and position update formula, the position and speed of each particle are updated in each iteration;

[0019] Calculating the fitness value of each particle according to the position and speed of each particle after each update;

[0020] According to the fitness value after each iteration, the current individual optimal position and the global optimal position of each particle are updated until a preset iteration stop condition is reached.

[0021] In an optional embodiment, updating the current individual optimal position and the global optimal position of each particle according to the fitness value after each iteration includes:

[0022] If it is determined that the fitness value of each particle in this iteration is less than the fitness value obtained in the previous iteration, the position of each particle in this iteration is used as its current individual optimal position;

[0023] If it is determined that the fitness values ​​of all particles in the population in this iteration are less than the fitness values ​​obtained in the previous iteration, the positions of the particles in the population in this iteration are correspondingly used as the current global optimal positions.

[0024] In an optional implementation manner, the speed update formula and the position update formula are expressed as:

[0025]

[0026] In the formula, k represents the number of iterations of the kth time, rand1 and rand2 represent random numbers between [0, 1], c1 and c2 represent learning factors, and They represent the velocity and position of particle i in dimension d at the kth iteration, pbest id represents the individual optimal position of particle i in the dth dimension, gbest d represents the global optimal position of all particles in the dth dimension.

[0027] In an optional implementation manner, the adaptive extended Kalman filter algorithm is used, and the state of charge of the battery to be tested is estimated according to the second state equation group after the identification parameters, including:

[0028] Initialize the state variables and error covariance, and predict the state variables and error covariance at the target time;

[0029] Calculating the Kalman gain according to the state variable and the error covariance at the target moment;

[0030] updating an error value according to the state variable at the target moment, and updating the state variable and error covariance at the target moment according to the error value and the Kalman gain;

[0031] The state of charge of the battery to be tested is estimated based on the updated state variables and error covariance at the target moment and the second state equation group.

[0032] In an optional implementation manner, before estimating the state of charge of the battery to be tested based on the updated state variable and error covariance at the target time and the second state equation group, the method further includes:

[0033] The window estimation method is used to calculate the new information real-time estimation function;

[0034] Using the new information real-time estimation function, the covariance of the state noise and the observation noise in the adaptive extended Kalman filter algorithm is updated;

[0035] The expression of the innovation real-time estimation function is:

[0036]

[0037] In the formula, H k Represents the new information real-time estimation function, M represents the window size in the window estimation principle, E represents the expectation, K represents the current number of iterations, and i represents the i-th iteration.

[0038] In an optional implementation, the fractional-order equivalent circuit model includes: an equivalent battery, a first resistor, and an RC unit connected in series in sequence;

[0039] The RC unit includes a second resistor and a capacitor connected in parallel;

[0040] One end of the RC unit is connected to the positive electrode of the equivalent battery, and the other end of the RC unit is connected to the positive electrode input terminal;

[0041] The negative electrode of the equivalent battery is connected to the negative electrode output terminal.

[0042] In an optional implementation, the expression of the second state equation group is:

[0043]

[0044] In the formula, k-1 is the order of the previous iteration process, k is the order of the current iteration process, T is the sampling time, and L is the data length.

[0045] In an optional implementation, the expression of the first state equation group is:

[0046]

[0047] Where, R1 represents the second resistor; C1 represents the capacitor; α represents the order of the capacitor and 0≤α≤1; U1 represents the polarization voltage across the capacitor C1; dt represents the differential of the charge and discharge time; U L represents the battery terminal voltage, that is, the output voltage of the equivalent battery; I represents the output current of the equivalent battery; U oc represents the open circuit voltage value; R0 represents the first resistor; U1 represents the battery terminal voltage; SOC(t) represents the state of charge of the equivalent battery at time t; SOC(0) represents the initial value of the lithium battery state of charge; Q N Indicates the rated capacity of the equivalent battery.

[0048] In a second aspect, the present application provides a battery state of charge estimation device, comprising:

[0049] A model building module, used to build a fractional-order equivalent circuit model of the battery to be tested;

[0050] An equation construction module, used for constructing a first state equation group according to the fractional-order equivalent circuit model, and discretizing the first state equation group to obtain a second state equation group;

[0051] An equation updating module, used for performing parameter identification on the fractional-order equivalent circuit model by using a particle swarm optimization algorithm;

[0052] The estimation module is used to estimate the state of charge of the battery to be tested by using an adaptive extended Kalman filter algorithm and according to the second state equation group after the parameters are identified.

[0053] In a third aspect, the present application provides a computer device, comprising a processor and a memory, wherein the memory stores a computer program, and the processor is used to execute the computer program to implement the aforementioned battery state of charge estimation method.

[0054] In a fourth aspect, the present application provides a computer storage medium storing a computer program, which, when executed on a processor, implements the aforementioned battery state of charge estimation method.

[0055] The embodiments of the present application have the following beneficial effects:

[0056] The embodiment of the present application provides a battery state of charge estimation method, including: constructing a fractional-order equivalent circuit model of the battery to be tested; constructing a first state equation group according to the fractional-order equivalent circuit model, and discretizing the first state equation group to obtain a second state equation group; using a particle swarm optimization algorithm to identify parameters of the fractional-order equivalent circuit model; using an adaptive extended Kalman filter algorithm, and estimating the state of charge of the battery to be tested according to the second state equation group after identifying the parameters. The embodiment of the present application uses a fractional-order model to construct an equivalent circuit model. Compared with an integer-order model, the fractional-order model can better reflect the nonlinear characteristics of the battery, thereby improving the accuracy of the circuit model, and on the basis of the established equivalent circuit model, the adaptive fractional-order extended Kalman filter algorithm is combined to estimate the state of charge, which improves the accuracy of the state of charge estimation while also improving the anti-noise performance during the estimation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solution of the present application, the following is a brief introduction to the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present application and should not be regarded as limiting the scope of protection of the present application. For ordinary technicians in this field, other related drawings can also be obtained based on these drawings without creative work.

[0058] Figure 1 A first flow chart of a method for estimating a battery state of charge in an embodiment of the present application is shown;

[0059] Figure 2 A structural schematic diagram of a fractional-order equivalent circuit model of a battery to be tested in an embodiment of the present application is shown;

[0060] Figure 3A second flow chart of the battery state of charge estimation method in an embodiment of the present application is shown;

[0061] Figure 4 A structural schematic diagram of a battery state of charge estimation device in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0062] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments.

[0063] The components of the embodiments of the present application generally described and shown in the drawings herein may be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0064] Hereinafter, the terms "including", "having" and their cognates, which may be used in various embodiments of the present application, are intended only to indicate specific features, numbers, steps, operations, elements, components, or combinations of the foregoing items, and should not be understood as first excluding the existence of one or more other features, numbers, steps, operations, elements, components, or combinations of the foregoing items or adding the possibility of one or more features, numbers, steps, operations, elements, components, or combinations of the foregoing items.

[0065] Furthermore, the terms “first”, “second”, “third”, etc. are merely used for distinguishing descriptions and are not to be understood as indicating or implying relative importance.

[0066] Unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meanings as those generally understood by those skilled in the art to which the various embodiments of the present application belong. The terms (such as those defined in generally used dictionaries) will be interpreted as having the same meanings as the contextual meanings in the relevant technical field and will not be interpreted as having idealized meanings or overly formal meanings unless clearly defined in the various embodiments of the present application.

[0067] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0068] The fractional-order equivalent circuit model is a mathematical model used to describe the characteristics of lithium batteries. It uses fractional-order calculus to more accurately simulate the dynamic behavior of lithium batteries.

[0069] Particle Swarm Optimization (PSO) is a heuristic search algorithm. In the parameter identification process, PSO can help find the most appropriate model parameters so that the model's prediction results are as close as possible to the actual experimental data.

[0070] The adaptive iterative extended Kalman filter algorithm (AIEK) is a filtering algorithm whose purpose is to gradually adapt to different states and environments through an iterative process, thereby optimizing the filtering effect.

[0071] The embodiment of the present application provides a method for estimating the state of charge of a battery. First, a fractional-order equivalent circuit model of the battery is established, and then discretization and model parameter identification are performed in sequence, and then the state of charge estimation is performed in combination with an adaptive Kalman filter algorithm, so that a relatively accurate state of charge result can be obtained. That is, this embodiment introduces a fractional-order equivalent circuit model and an adaptive extended Kalman filter algorithm to improve the accuracy and robustness of the battery state of charge estimation.

[0072] For example, Figure 1 As shown, the battery state of charge estimation method may include the following steps:

[0073] S110, constructing a fractional-order equivalent circuit model of the battery to be tested.

[0074] In this embodiment, if Figure 2 As shown, a first-order RC model is used to establish an equivalent circuit model of the battery to be tested. The fractional-order equivalent circuit model includes an equivalent battery, a first resistor, and an RC unit connected in series in sequence; one end of the RC unit is connected to the positive electrode of the equivalent battery, the other end of the RC unit is connected to the positive electrode input terminal, and the negative electrode of the equivalent battery is connected to the negative electrode output terminal.

[0075] Furthermore, the RC unit includes a second resistor and a capacitor connected in parallel; wherein a first parallel end of the second resistor and the capacitor is connected to the positive electrode of the equivalent battery, and a second parallel end of the second resistor and the capacitor is connected to the positive electrode input end.

[0076] It should be noted that the conventional integer-order model is generally used to construct the equivalent circuit model in the prior art, but the integer-order model cannot fully capture the nonlinear characteristics of the battery. The fractional-order model is a mathematical model that uses fractional-order calculus (non-integer-order differentials and integrals) to more accurately describe the dynamic behavior of the battery, that is, to better reflect the nonlinear characteristics of the battery.

[0077] Based on this, the fractional-order equivalent circuit model obtained by this embodiment based on the fractional-order model can describe the static and dynamic characteristics of the battery to be tested through components such as resistors and capacitors, so as to effectively reflect the internal mechanism of the battery, and the structure of the constructed equivalent circuit model is simple and convenient for calculation; and because the first-order RC unit circuit structure can more accurately reflect the static and dynamic characteristics of the battery, this embodiment uses a first-order RC unit circuit to perform the equivalent of the battery to be tested. Among them, this embodiment can also select a second-order RC unit circuit or a multi-order RC unit circuit to perform the equivalent of the battery to be tested based on actual needs, and this embodiment is not limited to this.

[0078] S120, constructing a first set of state equations according to the fractional-order equivalent circuit model, and discretizing the first set of state equations to obtain a second set of state equations.

[0079] Firstly, this embodiment constructs a first state equation group for describing the battery state based on the fractional-order equivalent circuit model and Kirchhoff's voltage law.

[0080] Among them, the Kirchhoff voltage law is to define the algebraic sum of all voltage drops equal to zero in a closed loop to solve the voltage and current distribution in the circuit. That is, based on the fractional-order equivalent circuit model, Kirchhoff's voltage law is used to obtain the equations of the circuit model about voltage and current, and then the first state equation group is obtained. Among them, the first state equation group is a state equation group in the continuous time domain.

[0081] For example, the expression of the first state equation group is as follows:

[0082]

[0083] Where, R1 represents the second resistor; C1 represents the capacitor; α represents the order of the capacitor and 0≤α≤1; U1 represents the polarization voltage across the capacitor C1; dt represents the differential of the charge and discharge time; U L represents the battery terminal voltage, that is, the output voltage of the equivalent battery; I represents the output current of the equivalent battery; U oc represents the open circuit voltage value; R0 represents the first resistor; U1 represents the battery terminal voltage; SOC(t) represents the state of charge of the equivalent battery at time t; SOC(0) represents the initial value of the lithium battery state of charge; Q N Indicates the rated capacity of the equivalent battery.

[0084] The order α represents the degree to which the behavior of the constant phase angle element is close to that of an ideal capacitor, and reflects the complexity of the response of the constant phase angle element to signals of different frequencies.

[0085] Then, the first state equation group in the continuous time domain is converted into the second state equation group in the discrete time domain for subsequent calculation and simulation.

[0086] Among them, this embodiment can use the Euler method or other numerical integration methods to perform discretization processing, and this embodiment is not limited to this.

[0087] For example, the expression of the second state equation group is as follows:

[0088]

[0089] In the formula, k-1 is the order of the previous iteration process, k is the order of the current iteration process, T is the sampling time, and L is the data length.

[0090] Furthermore, the definition coefficients of the second state equation group are:

[0091]

[0092] In the formula, A is the state transfer matrix, which describes the linear relationship between the state variables and time; B is the input matrix, which describes the influence of the input on the state variables; C k is the process noise.

[0093] Among them, the definition coefficient describes the change pattern of the battery state variable over time, which directly affects the estimation accuracy of the battery state. The definition coefficient can be adjusted subsequently through the adaptive Kalman filter algorithm, which can better adapt to the working state and environmental changes of the battery and improve the accuracy of state estimation.

[0094] S130, using a particle swarm optimization algorithm to identify circuit parameters of the fractional-order equivalent circuit model.

[0095] In this embodiment, a particle swarm optimization algorithm is used to perform online identification on the fractional-order equivalent circuit model to obtain circuit parameters of the equivalent circuit model; wherein the circuit parameters include a first resistor R0, a second resistor R1, a capacitor C1, and an order α of the capacitor C1.

[0096] Then, the actual charge and discharge data and open-circuit voltage data of the battery to be tested are obtained in real time, and the identified circuit parameters are compared with the actual charge and discharge data and open-circuit voltage data to verify the accuracy of the equivalent circuit model.

[0097] For reference, first, a population is defined in a d-dimensional space, wherein the population includes a number of particles, and each particle represents a circuit parameter to be identified.

[0098] Then, each particle is randomly initialized, that is, the speed and position of each particle are initialized; wherein the position of each particle represents the current parameter value corresponding to the particle (that is, the predicted value of a circuit parameter in the equivalent circuit model), and the speed represents the rate of change of the parameter value.

[0099] For example, the velocity of a single particle i (denoted as v i ) is:

[0100] v i =(v i1 ,v i2 ,…v iD );

[0101] And the position of a single particle i (denoted as x i ) is:

[0102] x i =(x i1 ,x i2 ,…x iD ).

[0103] Next, a fitness function is defined, which is used to evaluate the fitness of each particle (denoted as J); and the fitness of each particle reflects the error between the predicted value and the actual value of the equivalent circuit model; among them, the smaller the fitness value of a single particle, the higher the fitness of the parameter value of the particle, indicating that the difference between the parameter value currently selected by the particle and the actual value is smaller; conversely, the larger the fitness value, the lower the fitness, indicating that the difference is greater.

[0104] Similarly, the fitness function can be used to calculate the fitness value of a population, thereby evaluating the fitness of a population. Among them, evaluating the fitness of a population is to evaluate the fitness of each parameter value combination corresponding to a group of particles.

[0105] For example, the expression of the fitness function is as follows:

[0106]

[0107] Where N data Indicates the length of all data, U ce Indicates the actual terminal voltage value, U L It represents the terminal voltage value estimated by the equivalent circuit model (ie, the predicted value), and i represents the i-th data.

[0108] It can be understood that the fitness function is to find the minimum value of the sum of the squares of the difference between the terminal voltage predicted by the equivalent circuit model and the actual terminal voltage, and then obtain an optimal solution. Among them, the terminal voltage predicted by the equivalent circuit model can be determined according to the position value selected by each particle each time and the aforementioned equivalent circuit model; in addition, this embodiment can also measure and obtain the actual terminal voltage corresponding to the actual circuit environment where the parameter value of the corresponding circuit parameter is the position value of the corresponding particle; the predicted terminal voltage and the actual terminal voltage are substituted into the above fitness function, and the fitness value of the target particle under the current position value can be calculated.

[0109] Then, a total number of iterations or a fitness value of the population reaching a fitness threshold is preset as an iteration stop condition. The specific value of the total number of iterations or the fitness threshold can be set according to actual needs, and this embodiment does not limit this.

[0110] Furthermore, according to the different positions and speeds set for the particles, a fitness function is used to iteratively calculate multiple fitness values ​​corresponding to each particle; and, according to the fitness value after each iteration, the individual optimal position of each particle and the global optimal position corresponding to a population are updated.

[0111] Among them, the individual optimal position (denoted as pbest) is the optimal position of each particle (that is, the parameter combination that satisfies the highest fitness of a single particle); the global optimal position (denoted as gbest) is the optimal position of all particles (that is, the parameter combination that satisfies the highest fitness of all particles).

[0112] For example, if the fitness value of particle i in this iteration is smaller than the fitness value obtained in the previous iteration, the position of particle i in this iteration is taken as its current individual optimal position, and so on, until the individual optimal position corresponding to the minimum fitness value is obtained.

[0113] Then, according to the individual best optimal position of each particle after each iteration and the global optimal position of all particles, the global optimal position corresponding to all particles (denoted as gBest) is updated.

[0114] For example, if the global fitness value corresponding to all particles in this iteration is smaller than the fitness value obtained in the previous iteration, the position of each particle in this iteration is taken as the current global optimal position, and so on, until the global optimal position corresponding to the minimum fitness value is obtained.

[0115] It can be understood that, for example, if a single particle C1 represents a capacitor C1, and as of this iteration, if the current position value of the particle C1 (recorded as ) and the actual terminal voltage value, when substituted into the fitness function, the lowest fitness value is obtained, indicating that the current position of particle C1 is The fitness is the highest, and the particle C1 can be set to the current position As the individual optimal position of capacitor C1.

[0116] Similarly, if the fitness value corresponding to the current position combination of a group of particles (the position combination includes a position value currently set for all particles in a group of particles) is the lowest, it means that the fitness of the current position set for the population is the highest.

[0117] In some examples, during multiple iterations, the present embodiment may also iteratively update the speed and position of each particle according to the iterative fitness value, the individual optimal position obtained after each iteration, and the global optimal position until the calculated fitness value is minimized, thereby obtaining the individual optimal position of each particle and the global optimal position of the population.

[0118] For reference, Figure 3 As shown, in this embodiment, the process of iteratively updating the speed and position of each particle to obtain the individual optimal position and the global optimal position may include the following steps:

[0119] S310, updating the position and speed of each particle in each iteration according to a preset speed update formula and position update formula.

[0120] S320, calculating the fitness value of each particle according to the position and speed of each particle after each update.

[0121] S330, updating the current individual optimal position and the global optimal position of each particle according to the fitness value after each iteration, until a preset iteration stop condition is reached.

[0122] It can be understood that in this embodiment, the speed update formula and the position update formula are preset, so that the position and speed of each particle in each iteration after initialization are updated by the preset speed update formula and the position update formula.

[0123] The expressions used to describe velocity updates and position updates are as follows:

[0124]

[0125] In the formula, k represents the number of iterations of the kth time, rand1 and rand2 represent random numbers between [0, 1], c1 and c2 represent learning factors, and They represent the velocity and position of particle i in dimension d at the kth iteration, pbestid represents the individual optimal position of particle i in the dth dimension, gbest d represents the global optimal position of all particles in the dth dimension.

[0126] It can be understood that the fitness value of each particle and the fitness value of the population are iteratively calculated through the fitness function, and in this process the value of the position and speed of each particle is updated, and the number of iterations and the fitness value obtained in each iteration are recorded.

[0127] Furthermore, after multiple iterations, if it is determined that the current number of iterations is equal to the preset total number of iterations, or the population fitness value obtained by the current iteration reaches the preset fitness threshold, the iteration is stopped and the optimal position is output.

[0128] This embodiment uses the fitness function to iterate the fitness value of each particle so that it can converge to a standard convergence value, thereby obtaining an optimal fitness, ensuring that the error between the predicted value and the actual value of the equivalent circuit model is minimized, so as to improve the accuracy of the charge state subsequently calculated by the predicted value of the equivalent circuit model.

[0129] S140, using an adaptive extended Kalman filter algorithm and estimating the state of charge of the battery to be tested according to a second state equation group after parameter identification.

[0130] Furthermore, the present embodiment can also estimate the state of charge of the battery to be tested by using an adaptive extended Kalman filter algorithm in combination with a second state equation group after identifying the circuit parameters. The adaptive extended Kalman filter algorithm is implemented based on Kalman filtering, wherein the Kalman filter introduces a state space equation into the filter, iterates the state vector at the current moment according to the state vector prediction value at the previous moment and the output prediction value at the next moment, updates the error covariance at the same time, and corrects the coefficient gain, gradually correcting the state vector so that the prediction result approaches the true value.

[0131] For reference, this embodiment first initializes the state variable (denoted as x) and the error covariance (denoted as P) in the state space equation according to prior knowledge or experimental data; wherein the state space equation is the aforementioned second state equation group, and the state variable is the charge state of the battery to be tested.

[0132] Among them, the initialization expression of the state variable is as follows:

[0133]

[0134] In the formula, represents the initialized state variable, x0 represents the state variable at the current moment, T represents transposition, and E represents expectation; represents the initialization error covariance.

[0135] Then, the state variables (denoted as ); the calculation expression corresponding to the prediction process of the state variable is as follows:

[0136]

[0137] In the formula, α represents the order of the aforementioned fractional-order equivalent circuit model.

[0138] And, the error covariance of the prediction k time (denoted as ), where the calculation expression corresponding to the prediction process of the error covariance is as follows:

[0139]

[0140] Next, update the error value (denoted as e k ), where the error update expression is as follows:

[0141]

[0142] In the formula, y k represents the observation of Kalman filter.

[0143] Furthermore, according to the covariance matrix predicted at time k and measurement noise (denoted as R k ), calculate the Kalman gain (denoted as K k ).

[0144]

[0145] Next, the state variables are updated according to the Kalman gain:

[0146]

[0147] And, update the error covariance:

[0148]

[0149] In one example, in the actual process of the extended Kalman filter, the estimation accuracy of the algorithm is easily affected by the noise covariance of the state equation and the observation equation. Since the error covariance of the traditional Kalman filter is often set through human experience during estimation, when the noise information changes, the estimation accuracy of the Kalman filter will decrease. Therefore, in order to better estimate the state of charge, this embodiment proposes a method of combining new information with Kalman filtering, and uses the new information real-time estimation function to update the covariance of state noise and observation noise in real time to improve the estimation accuracy of the battery state of charge; and in the process of estimating the state of charge through the adaptive extended Kalman filter algorithm, this embodiment can also further improve the anti-noise performance and reduce noise interference, thereby effectively improving the accuracy of the state of charge estimation.

[0150] In this embodiment, the window estimation method is used to obtain the new information real-time estimation function, wherein the new information real-time estimation function (denoted as H k ) is:

[0151]

[0152] Where M represents the window size in the window estimation principle, K represents the current iteration number, and i represents the i-th iteration.

[0153] Based on the innovation real-time estimation function, the adaptive estimation of the covariance of state noise and observation noise is obtained:

[0154]

[0155]

[0156] In the formula, Q k represents the process noise covariance, R k represents the measurement noise covariance.

[0157] It can be understood that the present embodiment pre-initializes the state variables and error covariance, then predicts the a priori state variables and error covariance, and calculates the error value, calculates the Kalman gain based on the a priori state variables and error covariance, and calculates the a posteriori state variables and error covariance based on the Kalman gain and error value; repeats the above steps to iteratively calculate the estimated state variables that are more in line with the actual values, that is, outputs the estimated state of charge of the battery to be tested. In the above iterative process, the new information real-time estimation function is also combined to update the state noise and observation noise covariance in the adaptive Kalman filtering process to improve the accuracy of the output state variable estimate.

[0158] It should be noted that the extended Kalman filter algorithm is also used in the prior art to estimate the state of charge, but the extended Kalman filter cannot be adaptive to cope with different working conditions, and has poor robustness and weak noise resistance. Based on this, this embodiment combines a fractional-order equivalent circuit model and introduces a new information real-time estimation function to realize an adaptive extended Kalman filter, so that the technical solution of this embodiment can adaptively cope with different working conditions, so as to improve the robustness and noise resistance of the state of charge estimation process, and can obtain a more accurate state of charge estimation value.

[0159] This embodiment improves the accuracy of lithium battery modeling by constructing a fractional-order equivalent circuit model, thereby solving the problems of filter divergence and low accuracy of battery state estimation by the extended Kalman filter method. The dynamic response of the circuit model accurately describes the concentration polarization, electrochemical polarization and ohmic polarization reaction inside the battery. Compared with the traditional integer-order equivalent circuit model, it has the ability to better fit the battery test data, and provides a model basis for battery state estimation, thereby improving the accuracy and robustness of battery state joint estimation; and then the battery state is jointly estimated according to the adaptive extended Kalman filter method to make a more comprehensive estimate of the battery state; in addition, the noise covariance is adaptively adjusted in the extended Kalman filter by combining the new information real-time estimation function, thereby improving the accuracy and robustness of state estimation, avoiding the influence of filter divergence and model error, and the estimation result under dynamic conditions can quickly converge to the vicinity of the true value, thereby improving the accuracy and reliability of charge state estimation.

[0160] See also Figure 4 This embodiment also provides a battery state of charge estimation device, the device comprising:

[0161] A model building module 410 is used to build a fractional-order equivalent circuit model of the battery to be tested;

[0162] An equation construction module 420, configured to construct a first state equation group according to the fractional-order equivalent circuit model, and discretize the first state equation group to obtain a second state equation group;

[0163] An equation updating module 430 is used to perform parameter identification on the fractional-order equivalent circuit model using a particle swarm optimization algorithm;

[0164] The estimation module 440 is used to estimate the state of charge of the battery to be tested by using an adaptive extended Kalman filter algorithm and according to the second state equation group after parameter identification.

[0165] It can be understood that the battery state of charge estimation device of this embodiment corresponds to the battery state of charge estimation method of the above embodiment, and the options in the above embodiment are also applicable to this embodiment, so they will not be described repeatedly here.

[0166] The present application also provides a computer device. Exemplarily, the computer device includes a processor and a memory, wherein the memory stores a computer program, and the processor runs the computer program to enable the computer device to execute the functions of each module in the above-mentioned battery state of charge estimation method or the above-mentioned battery state of charge estimation device.

[0167] Among them, the processor can be an integrated circuit chip with signal processing capabilities. The processor can be a general-purpose processor, including a central processing unit (CPU), a graphics processing unit (GPU) and a network processor (NP), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or at least one of other programmable logic devices, discrete gates or transistor logic devices, and discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc., which can implement or execute the disclosed methods, steps and logic block diagrams in the embodiments of the present application.

[0168] The memory may be, but is not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), etc. The memory is used to store a computer program, and the processor may execute the computer program accordingly after receiving an execution instruction.

[0169] The present application also provides a computer storage medium for storing the computer program used in the above-mentioned computer device. The computer storage medium may be a readable storage medium, or a non-volatile storage medium or a volatile storage medium. For example, the computer storage medium may include, but is not limited to, various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0170] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and structure diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in an alternative implementation, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the structure diagram and / or the flow diagram, and the combination of boxes in the structure diagram and / or the flow diagram, can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0171] In addition, the functional modules or units in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0172] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, 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, which is stored in a storage medium and includes several instructions for a computer device (which can be a smart phone, a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application.

[0173] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A method for estimating a battery state of charge, characterized in that: include: Construct a fractional-order equivalent circuit model of the battery to be tested; Constructing a first state equation group according to the fractional-order equivalent circuit model, and discretizing the first state equation group to obtain a second state equation group; Using a particle swarm optimization algorithm to perform parameter identification on the fractional-order equivalent circuit model; An adaptive extended Kalman filter algorithm is used to estimate the state of charge of the battery to be tested according to the second state equation group after parameter identification.

2. The battery state of charge estimation method according to claim 1, characterized in that: The method of using a particle swarm optimization algorithm to identify circuit parameters of the fractional-order equivalent circuit model includes: A population is defined, and the position and velocity of each particle in the population are initialized; each particle represents a circuit parameter to be identified; A fitness function is defined, and according to the fitness function, the fitness value of each particle is iteratively calculated to determine the optimal position of each particle; The optimal position of each particle is taken as the optimal solution of each circuit parameter.

3. The method for estimating the battery state of charge according to claim 2, characterized in that: The expression of the fitness function is: In the formula, J represents fitness; N data Indicates the length of all data, U ce Indicates the actual terminal voltage value, U L represents the terminal voltage value estimated by the fractional-order equivalent circuit model, and i represents the i-th data.

4. The method for estimating the battery state of charge according to claim 2, characterized in that: The iterative calculation of the fitness value of each particle according to the fitness function to determine the optimal position of each particle includes: According to the preset speed update formula and position update formula, the position and speed of each particle are updated in each iteration; Calculating the fitness value of each particle according to the position and speed of each particle after each update; According to the fitness value after each iteration, the current individual optimal position and the global optimal position of each particle are updated until a preset iteration stop condition is reached.

5. The method for estimating the battery state of charge according to claim 4, characterized in that: The updating of the current individual optimal position and the global optimal position of each particle according to the fitness value after each iteration includes: If it is determined that the fitness value of each particle in this iteration is less than the fitness value obtained in the previous iteration, the position of each particle in this iteration is used as its current individual optimal position; If it is determined that the fitness values ​​of all particles in the population in this iteration are less than the fitness values ​​obtained in the previous iteration, the positions of the particles in the population in this iteration are correspondingly used as the current global optimal positions.

6. The method for estimating the battery state of charge according to claim 4, characterized in that: The expressions of the speed update formula and the position update formula are: In the formula, k represents the number of iterations of the kth time, rand1 and rand2 represent random numbers between [0, 1], c1 and c2 represent learning factors, and They represent the velocity and position of particle i in dimension d at the kth iteration, pbest id represents the individual optimal position of particle i in the dth dimension, gbest d represents the global optimal position of all particles in the dth dimension.

7. The battery state of charge estimation method according to claim 1, characterized in that: The method adopts the adaptive extended Kalman filter algorithm and estimates the state of charge of the battery to be tested according to the second state equation group after identifying the parameters, including: Initialize the state variables and error covariance, and predict the state variables and error covariance at the target time; Calculating the Kalman gain according to the state variable and the error covariance at the target moment; updating an error value according to the state variable at the target moment, and updating the state variable and error covariance at the target moment according to the error value and the Kalman gain; The state of charge of the battery to be tested is estimated based on the updated state variables and error covariance at the target moment and the second state equation group.

8. The method for estimating the battery state of charge according to claim 7, characterized in that: Before estimating the state of charge of the battery to be tested based on the updated state variable and error covariance at the target time and the second state equation group, the method further includes: The window estimation method is used to calculate the new information real-time estimation function; Using the new information real-time estimation function, the covariance of the state noise and the observation noise in the adaptive extended Kalman filter algorithm is updated; The expression of the innovation real-time estimation function is: In the formula, H k Represents the new information real-time estimation function, M represents the window size in the window estimation principle, E represents the expectation, K represents the current number of iterations, and i represents the i-th iteration.

9. The battery state of charge estimation method according to claim 1, characterized in that: The fractional-order equivalent circuit model comprises: an equivalent battery, a first resistor and an RC unit connected in series in sequence; The RC unit includes a second resistor and a capacitor connected in parallel; One end of the RC unit is connected to the positive electrode of the equivalent battery, and the other end of the RC unit is connected to the positive electrode input terminal; The negative electrode of the equivalent battery is connected to the negative electrode output terminal.

10. The battery state of charge estimation method according to claim 9, characterized in that: The expression of the second state equation group is: In the formula, k-1 is the order of the previous iteration process, k is the order of the current iteration process, T is the sampling time, and L is the data length.

11. The battery state of charge estimation method according to claim 9, characterized in that: The expression of the first state equation group is: Where, R1 represents the second resistor; C1 represents the capacitor; α represents the order of the capacitor and 0≤α≤1; U1 represents the polarization voltage across the capacitor C1; dt represents the differential of the charge and discharge time; U L represents the battery terminal voltage, that is, the output voltage of the equivalent battery; I represents the output current of the equivalent battery; U oc represents the open circuit voltage value; R0 represents the first resistor; U1 represents the battery terminal voltage; SOC(t) represents the state of charge of the equivalent battery at time t; SOC(0) represents the initial value of the lithium battery state of charge; Q N Indicates the rated capacity of the equivalent battery.

12. A battery state of charge estimation device, characterized in that: include: A model building module, used to build a fractional-order equivalent circuit model of the battery to be tested; An equation construction module, used for constructing a first state equation group according to the fractional-order equivalent circuit model, and discretizing the first state equation group to obtain a second state equation group; An equation updating module, used for performing parameter identification on the fractional-order equivalent circuit model by using a particle swarm optimization algorithm; The estimation module is used to estimate the state of charge of the battery to be tested by using an adaptive extended Kalman filter algorithm and according to the second state equation group after the parameters are identified.

13. A computer device, characterized in that: The computer device comprises a processor and a memory, wherein the memory stores a computer program, and the processor is configured to execute the computer program to implement the battery state of charge estimation method according to any one of claims 1 to 11.

14. A computer storage medium, characterized in that: The computer program is stored therein, and when the computer program is executed on a processor, the battery state of charge estimation method according to any one of claims 1 to 11 is implemented.

Images