A method for jointly estimating external thermal resistance and internal and external temperatures of lithium-ion power batteries

Abstract

The present invention provides a method for jointly estimating the external thermal resistance and internal and external temperatures of a lithium-ion power battery, comprising: establishing a battery equivalent thermal model according to the thermal characteristics of the lithium-ion power battery; using an adaptive particle swarm algorithm based on genetic algorithm optimization to identify thermal parameters; establishing a high-gain observer, initializing the high-gain observer with the external thermal resistance in the identified thermal parameters as an initial value, and completing real-time estimation of the external thermal resistance; constructing an adaptive strong tracking extended Kalman filter, inputting the identified thermal parameters and the real-time estimated external thermal resistance into the adaptive strong tracking extended Kalman filter; introducing an improved Sky Eagle optimization algorithm based on multiple strategies to optimize the process noise covariance and the observation noise covariance of the adaptive strong tracking extended Kalman filter, obtaining an optimized filter, and using the optimized filter to estimate the internal and external temperatures of the lithium-ion power battery in real time.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy vehicles, and in particular to a method for jointly estimating the external thermal resistance and internal and external temperatures of a lithium-ion power battery. Background Art

[0002] Electric vehicles have become the direction of automobile development with their energy-saving and environmentally friendly characteristics, and lithium-ion power batteries are widely used in the field of electric vehicles due to their large capacity, high operating voltage, and long life cycle. However, the thermal safety of lithium-ion power batteries is still an important factor restricting the development of electric vehicles. The current battery thermal management system's imperfect monitoring and early warning of battery temperature is an important cause of battery thermal runaway. It is necessary to develop an effective battery temperature estimation method to predict the battery temperature.

[0003] During battery operation, there is a large difference in temperature between the inside and outside of the battery. The internal temperature of the battery is usually higher than the surface temperature. It is not enough to accurately estimate the battery operating status based solely on the external surface temperature. The time-varying nature of the battery's external thermal resistance affects the result of estimating the internal and external temperatures of the battery. Therefore, how to accurately estimate the battery's external thermal resistance and internal and external temperatures is a technical problem that technicians in this field need to solve. Summary of the invention

[0004] The object of the present invention is to provide a method for jointly estimating the external thermal resistance and internal and external temperatures of a lithium-ion power battery, so as to improve the performance and safety of the battery.

[0005] A method for jointly estimating the external thermal resistance and internal and external temperatures of a lithium-ion power battery comprises the following steps:

[0006] Step S1, establishing a battery equivalent thermal model based on Kirchhoff's current law according to the thermal characteristics of the lithium-ion power battery;

[0007] Step S2, based on the battery equivalent thermal model established in step S1, using an adaptive particle swarm algorithm based on genetic algorithm optimization to identify thermal parameters;

[0008] Step S3, based on the battery equivalent thermal model established in step S1, a high-gain observer is established, and the high-gain observer is initialized with the external thermal resistance in the thermal parameters identified in step S2 as an initial value, so as to complete the real-time estimation of the external thermal resistance;

[0009] Step S4, based on the battery equivalent thermal model established in step S1, construct an adaptive strong tracking extended Kalman filter, and input the thermal parameters identified in step S2 and the external thermal resistance estimated in real time in step S3 into the adaptive strong tracking extended Kalman filter;

[0010] Step S5, based on the adaptive strong tracking extended Kalman filter constructed in step S4, introduce the Sky Eagle optimization algorithm based on multi-strategy improvement to optimize the process noise covariance and observation noise covariance of the adaptive strong tracking extended Kalman filter to obtain an optimized filter, and use the optimized filter to estimate the internal and external temperatures of the lithium-ion power battery in real time.

[0011] The method for jointly estimating the external thermal resistance and internal and external temperatures of a lithium-ion power battery provided by the present invention has the following beneficial effects:

[0012] 1. The present invention comprehensively considers the thermal characteristics of lithium-ion power batteries, establishes a battery equivalent thermal model according to Kirchhoff's current law (KCL), and identifies the thermal model parameters through an adaptive particle swarm optimization based on genetic algorithm (adaptive GA-PSO) algorithm. Considering that the external thermal resistance is time-varying and the external thermal resistance is the main thermal parameter affecting the battery temperature estimation result, a high-gain observer is established to estimate the external thermal resistance in real time, and the Sky Eagle optimization algorithm based on multi-strategy improvement is introduced to optimize the adaptive strong tracking extended Kalman filter, thereby completing the real-time estimation of the internal and external temperatures of the lithium-ion power battery;

[0013] 2. Different from the general battery thermal model parameter identification algorithm, the present invention adopts the adaptive GA-PSO algorithm to identify the parameters of the thermal model. The traditional GA algorithm has a diverse population and good globality, but the individual has no memory and slow convergence, while the PSO algorithm has memory and fast convergence but is prone to premature maturity and poor diversity. The present invention adopts the adaptive GA-PSO algorithm, combines the crossover and mutation operators of the GA algorithm with the PSO algorithm, so that the two algorithms can overcome the limitations in performance and achieve complementary advantages, and complete the accurate identification of thermal model parameters;

[0014] 3. Different from the general extended Kalman filter algorithm, the present invention adopts the Sky Eagle optimization algorithm based on multi-strategy improvement to optimize the adaptive strong tracking extended Kalman filter, and estimates the temperature of the established battery thermal model in real time. Considering that the uncertainty of the system thermal model will affect the accuracy of the extended Kalman filter, a strong tracking filter is introduced on the basis of the traditional extended Kalman filter, and the state prediction covariance matrix is ​​adjusted to improve the robustness and anti-interference ability of the filter. In order to prevent the sensor from being continuously disturbed and generating large noise, the new information sequence increases, and the covariance matrix is ​​amplified when passing through the strong tracking extended Kalman filter. The Sage-Husa adaptive filter is introduced to estimate and adjust the sensor noise in real time online, and the Sky Eagle optimization algorithm based on multi-strategy improvement is used to optimize the process noise covariance matrix and the observation noise covariance matrix, thereby improving the estimation accuracy.

[0015] 4. The present invention realizes the joint estimation of the external thermal resistance and the internal and external temperatures of the lithium-ion power battery. The estimation process has a small demand for sensors, which reduces the cost. The estimation process is basically not affected by changes in battery structural parameters and has good adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic flow chart of a method for jointly estimating the external thermal resistance and internal and external temperatures of a lithium-ion power battery of the present invention;

[0017] Figure 2 A comparison chart of the estimated value of the surface temperature of a lithium-ion power battery by the method proposed in the present invention and the estimated value and true value of the surface temperature of a lithium-ion power battery by a traditional algorithm;

[0018] Figure 3 This is a comparison chart of the estimated value of the internal temperature of a lithium-ion power battery by the method proposed in the present invention and the estimated value and true value of the internal temperature of a lithium-ion power battery by a traditional algorithm. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] See also Figure 1 The method for jointly estimating the external thermal resistance and internal and external temperatures of a lithium-ion power battery provided by an embodiment of the present invention includes steps S1 to S5.

[0021] Step S1, according to the thermal characteristics of the lithium-ion power battery, a battery equivalent thermal model is established using Kirchhoff's current law.

[0022] In step S1, a battery equivalent thermal model is established by Kirchhoff's current law (KCL), and the expression of the established battery equivalent thermal model is:

[0023] ;

[0024] make , ,have to:

[0025] ;

[0026] ;

[0027] in, Ra is the external thermal resistance, which is used to simulate the convection cooling of the battery surface; R b is the internal thermal resistance, which is used to simulate the heat exchange between the battery interior and the battery surface; C f is the heat capacity of the housing, C i is the internal heat capacity, T f is the estimated value of the battery surface temperature, T i is the estimated value of the internal temperature of the battery, T e is the ambient temperature, dt Indicates time t The differential of T ie The estimated value of the internal temperature of the battery minus the ambient temperature. T fe is the estimated value of the battery surface temperature minus the ambient temperature, Q c The heat generated by the battery;

[0028] in, Q c The Bernadi heat generation model is used for calculation, and the calculation formula is:

[0029] ;

[0030] ;

[0031] in, I C is the charge and discharge current of the battery, U is the battery terminal voltage, U C is the open circuit voltage of the battery, T C is the battery temperature, Ud C Express U C The differential of dT C Express T C The differential of represents the entropy thermal coefficient.

[0032] In addition, the present invention uses the ampere-hour integration method to define the state of charge of the battery SOC .

[0033] Step S2, based on the battery equivalent thermal model established in step S1, an adaptive particle swarm algorithm based on genetic algorithm optimization is used to identify thermal parameters.

[0034] Wherein, step S2 specifically includes:

[0035] Step S21, determining the thermal parameters to be identified, the thermal parameters to be identified include R a , R b , C f , C i ;

[0036] Step S22, initializing the population of the genetic algorithm and the particle swarm optimization particle swarm, and determining the upper and lower boundaries of the thermal parameters to be identified;

[0037] Step S23, using the fitness function to calculate and evaluate the fitness value of each individual;

[0038] Step S24, using a genetic algorithm to perform crossover and mutation operations to obtain a new population;

[0039] Step S25, updating the speed and position of the new individual in the particle swarm, and recording the historical optimal position of each particle and the optimal position of the swarm;

[0040] Step S26, when the set maximum number of iterations is reached or the set error tolerance range is reached, the algorithm terminates, otherwise returns to step S22 to continue iterative optimization;

[0041] Among them, the fitness function of the adaptive particle swarm algorithm based on genetic algorithm optimization is Function for:

[0042] ;

[0043] in, for k The battery surface temperature measured at all times, for k The battery surface temperature is output by the particle swarm algorithm based on genetic algorithm optimization, which is adaptive at all times. n For the length of time.

[0044] Step S3, based on the battery equivalent thermal model established in step S1, a high gain observer is established, and the high gain observer is initialized with the external thermal resistance in the thermal parameters identified in step S2 as the initial value to complete the real-time estimation of the external thermal resistance.

[0045] It should be pointed out that due to C f andC i Heat capacity represents the heat storage capacity of an object. R b is the internal thermal resistance, which depends on the battery material, R b is a fixed value, C f , C i , R b No real-time estimation is required. External thermal resistance R a It is the main parameter of battery surface convection cooling and the main parameter affecting the estimation result of battery internal and external temperature. Therefore, the present invention only considers the external thermal resistance R a Real-time estimation, the remaining thermal parameter values ​​are fixed.

[0046] Wherein, step S3 specifically includes:

[0047] The battery equivalent thermal model established in step S1 is transformed and discretized using Euler to obtain the following formula:

[0048] ;

[0049] ;

[0050] in, for k +1 time T ie , for k Moment T ie , for k +1 time T fe , for k Moment T fe , is the sampling time interval;

[0051] make , ,have to:

[0052] ;

[0053] in, and It is a custom intermediate quantity;

[0054] Will As a state variable, , design the following high-gain observer to estimate the external thermal resistance in real time:

[0055] ;

[0056] ;

[0057] in, for k The state variable of the high-gain observer at time +1 is, for k The state variable of the high-gain observer at time , L is the high gain observer matrix, T fe As the observed variable of the high-gain observer, for k The actual value of the battery surface temperature at time +1 minus the value of the ambient temperature, C is the observation matrix of the high gain observer, , the external thermal resistance identified in step S2 is used as the initial value and input into the high gain observer to accelerate the convergence speed of the high gain observer, and finally complete the and Real-time estimation, combined with , and the thermal parameters identified in step S2 to complete the real-time estimation of the external thermal resistance.

[0058] Step S4, based on the battery equivalent thermal model established in step S1, construct an adaptive strong tracking extended Kalman filter, and input the thermal parameters identified in step S2 and the external thermal resistance estimated in real time in step S3 into the adaptive strong tracking extended Kalman filter.

[0059] Wherein, step S4 specifically includes:

[0060] The discrete state equation and observation equation of the adaptive strong tracking extended Kalman filter for nonlinear systems are shown as follows:

[0061] ;

[0062] ;

[0063] ;

[0064] ;

[0065] in, for k The state variables of the adaptive strong tracking extended Kalman filter at time +1, , fork The state variables of the adaptive strong tracking extended Kalman filter at time, is the state transfer matrix, is the control matrix, for k The heat generated by the battery at all times, for k The process noise at the moment, for k The observation variables of the adaptive strong tracking extended Kalman filter at time +1, , is the observation matrix of the adaptive strong tracking extended Kalman filter, , for k Observation noise at time +1;

[0066] The state equation and observation equation are introduced into the adaptive strong tracking extended Kalman filter for prediction and update. A fading factor is added to the extended Kalman filter, and the error covariance matrix is ​​corrected to obtain the strong tracking extended Kalman filter. The specific steps are as follows:

[0067] Define the innovation covariance matrix:

[0068] ;

[0069] ;

[0070] in, for k The information covariance matrix at time +1, for k The information covariance matrix at time , for k +1 time strong tracking filter new information, T represents the transpose operation, for k +1 time state estimate, For the forgetting factor, In this embodiment, =0.95;

[0071] Introducing undetermined factors :

[0072] ;

[0073] in, trace To find the trace operator; is a weakening factor used to adjust the measurement noise. In this embodiment, =1.5; for kThe measurement noise covariance at time +1, for k The process noise covariance at time , for k The covariance estimate of the moments;

[0074] Define the fade factor :

[0075] ;

[0076] Introducing the fading factor into the state prediction covariance matrix, we get the modified covariance prediction formula:

[0077] ;

[0078] in, is based on k Predicted at all times k +1 moment covariance;

[0079] In order to prevent the sensor from generating large noise due to continuous interference, which results in the covariance matrix being amplified when passing through the strong tracking extended Kalman filter, the present embodiment introduces Sage-Husa adaptive filtering to estimate and adjust the sensor noise in real time. The specific process is as follows:

[0080] Define the innovation of Sage-Husa adaptive filtering:

[0081] ;

[0082] in, for k +1 time Sage-Husa adaptive filtering new information, To measure the noise expectation k The estimated value of the time, for k The weighting factor of time, , is the forgetting factor of the Sage-Husa adaptive filter, and In this embodiment, =0.96;

[0083] Status prediction:

[0084] ;

[0085] in, Based on k Predicted at all times k +1 time state variables, for k The estimated state value at time is the process noise expectation k Estimated value of the moment;

[0086] Covariance prediction:

[0087]

[0088] in, for k An estimate of the process noise covariance at time t;

[0089] News:

[0090] ;

[0091] Kalman gain:

[0092] ;

[0093] in, for k Kalman gain at time +1, for k An estimate of the measurement noise covariance matrix at time t;

[0094] Finally, the state update and covariance update are calculated to complete the construction of the adaptive strong tracking extended Kalman filter. The expressions for state update and covariance update are:

[0095] ;

[0096] ;

[0097] in, for k +1 time covariance estimate, is the identity matrix;

[0098] Update measurement noise expectation k +1 estimated value at the moment , process noise expectation k +1 estimated value at the moment , k The estimated value of the measurement noise covariance matrix at time +1 , k Estimated value of the process noise covariance at time +1 :

[0099] ;

[0100] ;

[0101] ;

[0102] .

[0103] Step S5, based on the adaptive strong tracking extended Kalman filter constructed in step S4, introduce the Sky Eagle optimization algorithm based on multi-strategy improvement to optimize the process noise covariance and observation noise covariance of the adaptive strong tracking extended Kalman filter to obtain an optimized filter, and use the optimized filter to estimate the internal and external temperatures of the lithium-ion power battery in real time.

[0104] Wherein, step S5 specifically includes:

[0105] In the Tianying optimization algorithm, an improved Tent chaotic map is introduced to generate the initial population to prevent the uneven distribution of the population that may occur during the optimization process, accelerate the convergence speed, improve the convergence accuracy, and prevent the Tent chaotic map from being unstable during the iteration process. In this embodiment, a random factor is added , the expression of the improved Tent chaos map is:

[0106] ;

[0107] in, and Respectively i and i +1 Chaos Sequence of Skyhawk, rand is a random number between 0 and 1, I is the population size, is a constant;

[0108] Mapping the improved Tent chaos to the Eagle population generation formula yields:

[0109] ;

[0110] in, is a chaotic population, and Respectively i The minimum and maximum values ​​of individual eagles;

[0111] In the expansion and exploration, the lack of sufficient communication between the eagle population may lead to insufficient convergence accuracy. This invention introduces the golden sine factor F Update the population position, improve the convergence accuracy, and the improved Tianying optimization algorithm expands the exploration formula as follows:

[0112] ;

[0113] ;

[0114] in, To expand the exploration phase of the improved Tianying optimization algorithm h +1 iteration solution, iter and iter max Represent the current iteration and the maximum number of iterations respectively, For the h The global optimal solution of the iteration is is the average position of all individuals in the current iteration;

[0115] In order to further balance the expansion and reduction of exploration capabilities in the Tianying optimization algorithm, an adaptive weight factor is proposed In the early stage of iteration, let the Eagle move at a faster speed to quickly reach the target value. In the later stage of iteration, let the Eagle move at a lower speed to avoid falling into the local optimum. The calculation formula of the adaptive weight factor is:

[0116] ;

[0117] The improved Tianying optimization algorithm also satisfies the following formula:

[0118] ;

[0119] in, For the h The optimal position of the multi-strategy improved Eagle optimization algorithm of the iteration;

[0120] The process noise covariance of the adaptive strong tracking extended Kalman filter and the measurement noise covariance As the optimization dimension of the improved Tianying optimization algorithm , as shown below:

[0121] ;

[0122] in, diag represents a diagonal matrix, , for The process noise value in for The observation noise value in ;

[0123] Improved fitness function of the Skyhawk optimization algorithm The expression is:

[0124]

[0125] in, for k The actual measured variable value at time, for kThe estimated value output by the adaptive strong tracking extended Kalman filter at time;

[0126] Finally, the optimized filter is used to estimate the internal and external temperatures of the lithium-ion power battery in real time.

[0127] Figure 2 This is a comparison chart of the estimated value of the surface temperature of a lithium-ion power battery by the method proposed in the present invention and the estimated value and true value of the surface temperature of a lithium-ion power battery by the traditional algorithm. Figure 3 This is a comparison chart of the estimated value of the internal temperature of a lithium-ion power battery by the method proposed in the present invention and the estimated value and true value of the internal temperature of a lithium-ion power battery by the traditional algorithm. The traditional algorithm adopts the extended Kalman filter (EKF) algorithm. Figure 2 and Figure 3 It can be seen that the method proposed in the present invention significantly improves the estimation accuracy of the internal and external temperatures of lithium-ion power batteries, wherein the surface temperature estimation error is within 0.1°C, and the internal temperature estimation error is within 0.5°C, which is significantly better than the traditional algorithm.

[0128] In summary, the method for jointly estimating the external thermal resistance and internal and external temperatures of a lithium-ion power battery provided by the present invention has the following beneficial effects:

[0129] 1. The present invention comprehensively considers the thermal characteristics of lithium-ion power batteries, establishes a battery equivalent thermal model according to Kirchhoff's current law (KCL), and identifies the thermal model parameters through an adaptive particle swarm optimization based on genetic algorithm (adaptive GA-PSO) algorithm. Considering that the external thermal resistance is time-varying and the external thermal resistance is the main thermal parameter affecting the battery temperature estimation result, a high-gain observer is established to estimate the external thermal resistance in real time, and the Sky Eagle optimization algorithm based on multi-strategy improvement is introduced to optimize the adaptive strong tracking extended Kalman filter, thereby completing the real-time estimation of the internal and external temperatures of the lithium-ion power battery;

[0130] 2. Different from the general battery thermal model parameter identification algorithm, the present invention adopts the adaptive GA-PSO algorithm to identify the parameters of the thermal model. The traditional GA algorithm has a diverse population and good globality, but the individual has no memory and slow convergence, while the PSO algorithm has memory and fast convergence but is prone to premature maturity and poor diversity. The present invention adopts the adaptive GA-PSO algorithm, combines the crossover and mutation operators of the GA algorithm with the PSO algorithm, so that the two algorithms can overcome the limitations in performance and achieve complementary advantages, and complete the accurate identification of thermal model parameters;

[0131] 3. Different from the general extended Kalman filter algorithm, the present invention adopts the Sky Eagle optimization algorithm based on multi-strategy improvement to optimize the adaptive strong tracking extended Kalman filter, and estimates the temperature of the established battery thermal model in real time. Considering that the uncertainty of the system thermal model will affect the accuracy of the extended Kalman filter, a strong tracking filter is introduced on the basis of the traditional extended Kalman filter, and the state prediction covariance matrix is ​​adjusted to improve the robustness and anti-interference ability of the filter. In order to prevent the sensor from being continuously disturbed and generating large noise, the new information sequence increases, and the covariance matrix is ​​amplified when passing through the strong tracking extended Kalman filter. The Sage-Husa adaptive filter is introduced to estimate and adjust the sensor noise in real time online, and the Sky Eagle optimization algorithm based on multi-strategy improvement is used to optimize the process noise covariance matrix and the observation noise covariance matrix, thereby improving the estimation accuracy.

[0132] 4. The present invention realizes the joint estimation of the external thermal resistance and the internal and external temperatures of the lithium-ion power battery. The estimation process has a small demand for sensors, which reduces the cost. The estimation process is basically not affected by changes in battery structural parameters and has good adaptability.

[0133] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0134] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

Claims

1. A method for jointly estimating the external thermal resistance and internal and external temperatures of a lithium-ion power battery, characterized in that: include: Step S1, establishing a battery equivalent thermal model based on Kirchhoff's current law according to the thermal characteristics of the lithium-ion power battery; Step S2, based on the battery equivalent thermal model established in step S1, using an adaptive particle swarm algorithm based on genetic algorithm optimization to identify thermal parameters; Step S3, based on the battery equivalent thermal model established in step S1, a high-gain observer is established, and the high-gain observer is initialized with the external thermal resistance in the thermal parameters identified in step S2 as an initial value, so as to complete the real-time estimation of the external thermal resistance; Step S4, based on the battery equivalent thermal model established in step S1, construct an adaptive strong tracking extended Kalman filter, and input the thermal parameters identified in step S2 and the external thermal resistance estimated in real time in step S3 into the adaptive strong tracking extended Kalman filter; Step S5, based on the adaptive strong tracking extended Kalman filter constructed in step S4, introducing the Sky Eagle optimization algorithm based on multi-strategy improvement to optimize the process noise covariance and observation noise covariance of the adaptive strong tracking extended Kalman filter to obtain an optimized filter, and using the optimized filter to estimate the internal and external temperatures of the lithium-ion power battery in real time; Wherein, step S3 specifically includes: The battery equivalent thermal model established in step S1 is transformed and discretized using Euler to obtain the following formula: ； ； in, for k +1 time T ie , for k Moment T ie , T ie The estimated value of the internal temperature of the battery minus the ambient temperature. for k +1 time T fe , for k Moment T fe , T fe is the estimated value of the battery surface temperature minus the ambient temperature, is the sampling time interval, R a is the external thermal resistance, R b is the internal thermal resistance, C f is the heat capacity of the housing, C i is the internal heat capacity, Q c The heat generated by the battery; make , ,have to: ； in, and It is a custom intermediate quantity; Will As a state variable, , design the following high-gain observer to estimate the external thermal resistance in real time: ； ； in, for k The state variable of the high-gain observer at time +1 is, for k The state variable of the high-gain observer at time , L is the high gain observer matrix, T fe As the observed variable of the high-gain observer, for k The actual value of the battery surface temperature at time +1 minus the value of the ambient temperature, C is the observation matrix of the high gain observer, The external thermal resistance identified in step S2 is used as the initial value and input into the high gain observer to complete the and Real-time estimation, combined with , and the thermal parameters identified in step S2 to complete the real-time estimation of the external thermal resistance.

2. The method for jointly estimating the external thermal resistance and internal and external temperatures of a lithium-ion power battery according to claim 1, characterized in that: In step S1, the expression of the established battery equivalent thermal model is: ； make , ,have to: ； ； in, T f is the estimated value of the battery surface temperature, T i is the estimated value of the internal temperature of the battery, T e is the ambient temperature, dt Indicates time t The differential of in, Q c The Bernadi heat generation model is used for calculation, and the calculation formula is: ； ； in, I C is the charge and discharge current of the battery, U is the battery terminal voltage, U C is the open circuit voltage of the battery, T C is the battery temperature, Ud C Express U C The differential of dT C Express T C The differential of .

3. The method for jointly estimating the external thermal resistance and internal and external temperatures of a lithium-ion power battery according to claim 2, characterized in that: Step S2 specifically includes: Step S21, determining the thermal parameters to be identified, the thermal parameters to be identified include R a , R b , C f , C i ; Step S22, initializing the population of the genetic algorithm and the particle swarm optimization particle swarm, and determining the upper and lower boundaries of the thermal parameters to be identified; Step S23, using the fitness function to calculate and evaluate the fitness value of each individual; Step S24, using a genetic algorithm to perform crossover and mutation operations to obtain a new population; Step S25, updating the speed and position of the new individual in the particle swarm, and recording the historical optimal position of each particle and the optimal position of the swarm; Step S26, when the set maximum number of iterations is reached or the set error tolerance range is reached, the algorithm terminates, otherwise returns to step S22 to continue iterative optimization; Among them, the fitness function of the adaptive particle swarm algorithm based on genetic algorithm optimization is Function for: ； in, for k The battery surface temperature measured at all times, for k The battery surface temperature is output by the particle swarm algorithm based on genetic algorithm optimization, which is adaptive at all times. n For the length of time.

4. The method for jointly estimating the external thermal resistance and internal and external temperatures of a lithium-ion power battery according to claim 3, characterized in that: Step S4 specifically includes: The discrete state equation and observation equation of the adaptive strong tracking extended Kalman filter for nonlinear systems are shown as follows: ； ； ； ； in, for k The state variables of the adaptive strong tracking extended Kalman filter at time +1, , for k The state variables of the adaptive strong tracking extended Kalman filter at time, is the state transfer matrix, is the control matrix, for k The heat generated by the battery at all times, for k The process noise at the moment, for k The observation variables of the adaptive strong tracking extended Kalman filter at time +1, , is the observation matrix of the adaptive strong tracking extended Kalman filter, , for k Observation noise at time +1; The state equation and observation equation are introduced into the adaptive strong tracking extended Kalman filter for prediction and update. A fading factor is added to the extended Kalman filter, and the error covariance matrix is ​​corrected to obtain the strong tracking extended Kalman filter. The specific steps are as follows: Define the innovation covariance matrix: ； ； in, for k The information covariance matrix at time +1, for k The information covariance matrix at time , for k +1 time strong tracking filter new information, T represents the transpose operation, for k +1 time state estimate, For the forgetting factor; Introducing undetermined factors : ； in, trace To find the trace operator, is a weakening factor, for k The measurement noise covariance at time +1, for k The process noise covariance at time , for k The covariance estimate of the moments; Define the fade factor : ； Introducing the fading factor into the state prediction covariance matrix, we get the modified covariance prediction formula: ； in, is based on k Predicted at all times k +1 moment covariance; Sage-Husa adaptive filtering is introduced to estimate and adjust the sensor noise in real time. The specific process is as follows: Define the innovation of Sage-Husa adaptive filtering: ； in, for k +1 time Sage-Husa adaptive filtering new information, To measure the noise expectation k The estimated value of the time, for k The weighting factor of time, , is the forgetting factor of the Sage-Husa adaptive filter, and ; Status prediction: ； in, Based on k Predicted at all times k +1 time state variables, for k The estimated state value at time is the process noise expectation k Estimated value of the moment; Covariance prediction: in, for k An estimate of the process noise covariance at time t; News: ； Kalman gain: ； in, for k Kalman gain at time +1, for k An estimate of the measurement noise covariance matrix at time t; Finally, the state update and covariance update are calculated to complete the construction of the adaptive strong tracking extended Kalman filter. The expressions for state update and covariance update are: ； ； in, for k +1 time covariance estimate, is the identity matrix; Update measurement noise expectation k +1 estimated value at the moment , process noise expectation k +1 estimated value at the moment , k The estimated value of the measurement noise covariance matrix at time +1 , k Estimated value of the process noise covariance at time +1 : ； ； ； 。 5. The method for jointly estimating the external thermal resistance and internal and external temperatures of a lithium-ion power battery according to claim 4, characterized in that: Step S5 specifically includes: In the Tianying optimization algorithm, the improved Tent chaotic map is introduced to generate the initialization population. The expression of the improved Tent chaotic map is: ； in, and Respectively i and i +1 Chaos Sequence of Skyhawk, rand is a random number between 0 and 1, I is the population size, is a constant; Mapping the improved Tent chaos to the Eagle population generation formula yields: ； in, is a chaotic population, and Respectively i The minimum and maximum values ​​of individual eagles; Introducing the golden sine factor F Update the population position, and the improved Tianying optimization algorithm expands the exploration formula as follows: ； ； in, To expand the exploration phase of the improved Tianying optimization algorithm h +1 iteration solution, iter and iter max Represent the current iteration and the maximum number of iterations respectively, For the h The global optimal solution of the iteration is is the average position of all individuals in the current iteration; The improved Tianying optimization algorithm also satisfies the following formula: ； in, For the h The optimal position of the multi-strategy improved Eagle optimization algorithm of the iteration; The process noise covariance of the adaptive strong tracking extended Kalman filter and the measurement noise covariance As the optimization dimension of the improved Tianying optimization algorithm , as shown below: ； in, diag represents a diagonal matrix, , for The process noise value in for The observed noise value in ; Improved fitness function of the Skyhawk optimization algorithm The expression is: in, for k The actual measured variable value at time, for k The estimated value output by the adaptive strong tracking extended Kalman filter at time; Finally, the optimized filter is used to estimate the internal and external temperatures of the lithium-ion power battery in real time.

Images