CN113406499A - Lithium battery SOC real-time estimation method and device based on optimized TCN, electronic equipment and storage medium - Google Patents
Lithium battery SOC real-time estimation method and device based on optimized TCN, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN113406499A CN113406499A CN202110724564.XA CN202110724564A CN113406499A CN 113406499 A CN113406499 A CN 113406499A CN 202110724564 A CN202110724564 A CN 202110724564A CN 113406499 A CN113406499 A CN 113406499A
- Authority
- CN
- China
- Prior art keywords
- lithium battery
- estimation model
- soc estimation
- battery soc
- initial
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/382—Arrangements for monitoring battery or accumulator variables, e.g. SoC
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Secondary Cells (AREA)
Abstract
The invention provides a method and a device for estimating the SOC of a lithium battery in real time based on an optimized TCN, an electronic device and a storage medium, wherein the method comprises the following steps: acquiring lithium ion battery data comprising a first data set and a second data set; constructing a first initial lithium battery SOC estimation model, wherein the first initial lithium battery SOC estimation model comprises initial time convolution network parameters; obtaining a first transition lithium battery SOC estimation model based on a genetic algorithm and a first data set, wherein the first transition lithium battery SOC estimation model comprises target time convolution network parameters; constructing a second initial lithium battery SOC estimation model, transferring the target time convolution network parameters to the second initial lithium battery SOC estimation model, and obtaining a pre-training lithium battery SOC estimation model; training the pre-trained lithium battery SOC estimation model through a second data set to obtain a target lithium battery SOC estimation model, and estimating the SOC of the lithium battery through the target lithium battery SOC estimation model. The invention improves the accuracy and the training speed of the SOC estimation of the lithium battery.
Description
Technical Field
The invention relates to the technical field of lithium battery monitoring, in particular to a method and a device for estimating the SOC of a lithium battery in real time based on optimized TCN, electronic equipment and a storage medium.
Background
With the adjustment of national development strategy, the new energy automobile will gradually replace fuel oil automobiles, and lithium batteries are also developed as power sources of the new energy automobile. The improvement of the battery performance becomes the key for realizing breakthrough of the endurance mileage, the safety performance, the service life and the power characteristic of the electric automobile. In order to ensure safe and efficient operation of the lithium battery, the most critical link is a battery management system technology. The method can realize accurate estimation and dynamic monitoring of lithium battery parameters and equalization among battery monomers. However, because the electric automobile has the characteristics of multiple working conditions, variable load, wide speed regulation range and the like, the power battery of the electric automobile shows high nonlinearity and variable flow working characteristics in the using process, so that the accurate estimation of the SoC of the lithium battery has great difficulty.
The existing SoC estimation technology of lithium batteries can be roughly divided into three categories: 1. a method based on battery characteristic parameters; 2. a battery model based solution; 3. based on a data-driven scheme.
However, the prior art has the following problems: 1. method based on battery characteristic parameters an open loop algorithm due to uncertain disturbances and variables such as: temperature, current, etc. may cause uncertainty of SoC, and the calculated SoC value has initial SoC error and accumulated current measurement error, so the accuracy requirement for the sensor is high, and using this method requires full charge and discharge of the lithium battery and regular capacity calibration, which shortens the service life of the lithium battery. 2. The model-based method tries to integrate various factors into a complex mathematical equation to estimate the SoC of the lithium battery, but the characteristics of the lithium battery cannot be completely represented by an equivalent circuit model, an electrochemical model or an electrochemical impedance model; and it is difficult to cover all the usage states of the lithium batteries with a certain specific model, resulting in low estimation accuracy. 3. The neural network established based on the data-driven method has no memory function, and a large amount of manual parameter adjustment is needed during training of the network, so that the training process is too slow.
Disclosure of Invention
In view of the above, it is necessary to provide a method and an apparatus for estimating SOC of a lithium battery in real time based on an optimized TCN, an electronic device and a storage medium, so as to solve the technical problems of inaccurate SOC estimation of the lithium battery and too slow training process in the prior art.
In order to solve the technical problem, the invention provides an optimized TCN-based lithium battery SOC real-time estimation method, which comprises the following steps:
acquiring lithium ion battery data, wherein the lithium ion battery data comprises a first data set and a second data set;
constructing a first initial lithium battery SOC estimation model, wherein the first initial lithium battery SOC estimation model comprises initial time convolution network parameters;
optimizing the first initial lithium battery SOC estimation model based on a genetic algorithm and the first data set to obtain a first transition lithium battery SOC estimation model, wherein the first transition lithium battery SOC estimation model comprises target time convolution network parameters;
constructing a second initial lithium battery SOC estimation model, and transferring the target time convolution network parameters to the second initial lithium battery SOC estimation model by adopting a transfer learning method to obtain a pre-training lithium battery SOC estimation model;
training the pre-trained lithium battery SOC estimation model through the second data set to obtain a target lithium battery SOC estimation model, and estimating the SOC of the lithium battery through the target lithium battery SOC estimation model.
In one possible implementation, the target time convolution network parameters include the number of residual blocks, an input step size, and an expansion factor.
In one possible implementation, the first data set carries measured values; the optimizing the first initial lithium battery SOC estimation model based on a genetic algorithm and a first data set comprises:
setting an initialization population, a cross rate, a variation rate and a maximum iteration number, and carrying out real number coding on initial time convolution network parameters;
inputting the first data set into the first initial lithium battery SOC estimation model to obtain a predicted value;
designing a fitness function, and calculating a fitness value of the population according to the fitness function, the measured value and the predicted value;
judging whether the fitness value of the initialized population meets the requirement, if so, taking the current initial time convolution network parameter as the target time convolution network parameter;
if not, retaining the optimal individuals, and performing selection, crossing and mutation operations on the individuals in the optimal individuals to generate a new population and a transition time convolution network parameter until the times of the selection, crossing and mutation operations are greater than the maximum iteration times, and then taking the transition time convolution network parameter with the minimum fitness value as the target time convolution network parameter.
In one possible implementation, the fitness function is:
wherein F is the fitness value of the individual to which the coefficient is added; n is the number of data in the first data set;
is an actual measurement value; SoCi is a predicted value; k is a coefficient; abs is an absolute value function.
In one possible implementation, the selection operation is selected by roulette, and the selection probability is:
fi＝m/F
in the formula, PiTo select a probability; n is the number of population individuals; m is a weight; f. ofiFitness value of the ith individual; f. ofjIs the fitness value of the jth individual.
In one possible implementation, the interleaving operation is:
in the formula, akjThe j position of the kth individual; a isljIs the jth position of the ith individual; b is a random number between 0 and 1.
In one possible implementation, the mutation operation is:
f(g)＝r2(1-g/Gmax)2
in the formula, amaxAnd aminAre respectively aijUpper and lower thresholds of; r is2Is a random number; g is iteration times; gmaxIs the maximum iteration number;&r is a random number between 0 and 1.
The invention also provides a lithium battery SOC real-time estimation device based on the optimized TCN, which comprises the following components:
the data acquisition unit is used for acquiring lithium ion battery data, and the lithium ion battery data comprises a first data set and a second data set;
the system comprises a first initial model building unit, a second initial model building unit and a third initial model building unit, wherein the first initial model building unit is used for building a first initial lithium battery SOC estimation model which comprises initial time convolution network parameters;
the optimization unit is used for optimizing the first initial lithium battery SOC estimation model based on a genetic algorithm and the first data set to obtain a first transition lithium battery SOC estimation model, and the first transition lithium battery SOC estimation model comprises target time convolution network parameters;
the second initial model building unit is used for building a second initial lithium battery SOC estimation model, and transferring the target time convolution network parameters to the second initial lithium battery SOC estimation model by adopting a transfer learning method to obtain a pre-training lithium battery SOC estimation model;
and the estimation unit is used for training the pre-trained lithium battery SOC estimation model through the second data set to obtain a target lithium battery SOC estimation model, and estimating the SOC of the lithium battery through the target lithium battery SOC estimation model.
The present invention also provides an electronic device comprising a memory and a processor, wherein,
the memory is used for storing programs;
the processor is coupled to the memory and configured to execute the program stored in the memory to implement the steps in the method for estimating SOC of a lithium battery based on optimized TCN in any one of the above implementations.
The invention further provides a computer-readable storage medium for storing a computer-readable program or instruction, where the program or instruction, when executed by a processor, can implement the steps in the method for estimating SOC of a lithium battery in real time based on optimized TCN in any one of the above-mentioned implementation manners.
The beneficial effects of adopting the above embodiment are: according to the method for estimating the SOC of the lithium battery in real time based on the optimized TCN, the first initial SOC estimation model of the lithium battery is optimized through a genetic algorithm, so that the time for adjusting parameters is greatly shortened, and meanwhile, the estimation accuracy is improved; and after the first transition lithium battery SOC estimation model is obtained through a genetic algorithm, the target lithium battery SOC estimation model is obtained through transfer learning to estimate the SOC of the lithium battery, the number of sample data required for constructing the target lithium battery SOC estimation model is greatly reduced, the training process speed is further increased, the SOC of different lithium batteries can be estimated through the transfer learning, and the applicability of the optimized TCN-based lithium battery SOC real-time estimation method is improved.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a schematic flowchart of an embodiment of a method for estimating SOC of a lithium battery in real time based on an optimized TCN according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of an embodiment of a first initial lithium battery SOC estimation model and a second initial lithium battery SOC estimation model provided in an embodiment of the present invention;
FIG. 3 is a flowchart illustrating an embodiment of S103 according to the present invention;
fig. 4 is a schematic structural diagram of an embodiment of an optimized TCN-based real-time estimation apparatus for SOC of a lithium battery according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of an embodiment of an electronic device according to an embodiment of the present invention.
Detailed Description
The technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In the following, the terms "first", "second" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the embodiments of the present application, "a plurality" means two or more unless otherwise specified.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the invention. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein can be combined with other embodiments.
Where soc (state of charge) is a state of charge, which is used to reflect the remaining capacity of the battery, and is numerically defined as the ratio of the remaining capacity to the battery capacity, and is usually expressed in percentage. The value range of the battery charging indicator is 0-1, when the SOC is 0, the battery is completely discharged, and when the SOC is 1, the battery is completely charged.
The invention provides a method and a device for estimating the SOC of a lithium battery in real time based on optimized TCN, electronic equipment and a storage medium, which are respectively explained below.
As shown in fig. 1, a schematic flowchart of an embodiment of a method for estimating SOC of a lithium battery in real time based on optimized TCN according to an embodiment of the present invention is shown, where the method includes:
s101, obtaining lithium ion battery data, wherein the lithium ion battery data comprises a first data set and a second data set;
s102, constructing a first initial lithium battery SOC estimation model, wherein the first initial lithium battery SOC estimation model comprises initial time convolution network parameters;
s103, optimizing a first initial lithium battery SOC estimation model based on a genetic algorithm and a first data set to obtain a first transition lithium battery SOC estimation model, wherein the first transition lithium battery SOC estimation model comprises target time convolution network parameters;
s104, constructing a second initial lithium battery SOC estimation model, and transferring the target time convolution network parameters to the second initial lithium battery SOC estimation model by adopting a transfer learning method to obtain a pre-training lithium battery SOC estimation model;
s105, training the pre-trained lithium battery SOC estimation model through the second data set to obtain a target lithium battery SOC estimation model, and estimating the SOC of the lithium battery through the target lithium battery SOC estimation model.
Compared with the prior art, the method for estimating the SOC of the lithium battery in real time based on the optimized TCN provided by the embodiment of the invention optimizes the first initial SOC estimation model of the lithium battery through a genetic algorithm, thereby greatly reducing the time for adjusting parameters and improving the estimation accuracy; and after the first transition lithium battery SOC estimation model is obtained through a genetic algorithm, the target lithium battery SOC estimation model is obtained through transfer learning to estimate the SOC of the lithium battery, the number of sample data required for constructing the target lithium battery SOC estimation model is greatly reduced, the training process speed is further increased, the SOC of different lithium batteries can be estimated through the transfer learning, and the applicability of the optimized TCN-based lithium battery SOC real-time estimation method is improved.
Specifically, the manner of acquiring the lithium ion battery data in S101 may be: the method comprises the steps of collecting experimental data or design experiments published on a network, building an experimental platform, and obtaining lithium battery data through the experiments.
In some embodiments of the invention, the first data set is experimental data disclosed on the collected network; the second data set is experimental data obtained by building an experimental platform.
Further, in order to unify the data formats of the first data set and the second data set, so as to facilitate subsequent training, in some embodiments of the present invention, before S102 or S103, the first data set and the second data set need to be preprocessed, specifically: and removing sample data obviously not conforming to the actual sample data in the first data set and the second data set, and performing normalization processing on the first data set and the second data set after the sample data obviously not conforming to the actual sample data is removed.
The sample data in the first data set and the second data set are time series data, and when the first data set or the second data set is input into a model for training, the input formats of the first data set and the second data set comprise: number of samples, input step size, and feature quantity.
Further, the first initial lithium battery SOC estimation model and the second initial lithium battery SOC estimation model have the same model structure and are both time convolution network models (TCN), and specifically, the first initial lithium battery SOC estimation model and the second initial lithium battery SOC estimation model are composed of causal convolution, expansion convolution and a plurality of residual blocks.
Specifically, the target time convolution network parameters include the number of residual blocks, the input step size, and the expansion factor. And optimizing the parameters to obtain a first transition lithium battery SOC estimation model.
Wherein, the expansion factor and the number of residual blocks determine the receptive field of the TCN. In some embodiments of the present invention, as shown in fig. 2, each of the first initial lithium battery SOC estimation model and the second initial lithium battery SOC estimation model includes 2 residual blocks, and each of the residual blocks is expanded three times from bottom to top, and the expansion factors are 2, 3, and 5 in sequence.
Further, the first data set carries measured values; as shown in fig. 3, S103 includes:
s301, setting an initialization population, a cross rate, a variation rate and a maximum iteration number, and carrying out real number coding on initial time convolution network parameters;
s302, inputting the first data set into a first initial lithium battery SOC estimation model to obtain a predicted value;
s303, designing a fitness function, and calculating a fitness value of the population according to the fitness function, the measured value and the predicted value;
s304, judging whether the fitness value of the initialized population meets the requirement, if so, taking the current initial time convolution network parameter as a target time convolution network parameter;
and S305, if the target time convolution network parameter does not meet the requirement, reserving the optimal individual, and performing selection, crossing and mutation operations on the individual in the optimal individual to generate a new population and the transition time convolution network parameter until the times of the selection, crossing and mutation operations are greater than the maximum iteration times, wherein the transition time convolution network parameter with the minimum fitness value is the target time convolution network parameter.
Through the setting, the target time convolution network parameters meeting or having the minimum fitness value can be obtained, and therefore the optimization of the first initial lithium battery SOC estimation model is achieved. And the parameters are adjusted through a genetic algorithm, compared with the prior art, manual participation is not needed, and the parameter adjusting speed is high.
Specifically, the fitness function is:
wherein F is the fitness value of the individual to which the coefficient is added; n is the number of data in the first data set;
is an actual measurement value; SoC (system on chip)iIs a predicted value; k is a coefficient; abs is an absolute value function.
Specifically, the selection mode of the selection operation is roulette selection, and the selection probability is as follows:
fi＝m/F
in the formula, PiTo select a probability; n is the number of population individuals; m is a weight; f. ofiFitness value of the ith individual; f. ofjIs the fitness value of the jth individual.
The roulette selection is a playback type random sampling method, individuals are selected to enter the next generation according to probabilities, and the probability of each individual entering the next generation is the ratio of the fitness value of each individual to the fitness values of all the individuals.
Specifically, the interleaving operation is:
in the formula, akjThe j position of the kth individual; a isljIs the jth position of the ith individual; b is a random number between 0 and 1.
The meaning of the above expression is: two individuals were randomly selected, and the same position was randomly selected for crossover.
Specifically, the mutation operation is:
f(g)＝r2(1-g/Cmax)2
in the formula, amaxAnd aminAre respectively aijUpper and lower thresholds of; r is2Is a random number; g is iteration times; gmaxIs the maximum iteration number;&r is a random number between 0 and 1.
Further, to ensure the accuracy of the first transition lithium battery SOC estimation model, in some embodiments of the invention, the first data set includes a first training set and a first test set, the first initial lithium battery SOC estimation model is optimized using the first training set and a genetic algorithm, and the accuracy of the first transition lithium battery SOC estimation model estimation is verified using the first test set.
Similarly, the second data set also includes a second training set and a second test set, and in S105, the pre-trained lithium battery SOC estimation model is trained through the second training set, and the target lithium battery SOC estimation model is verified through the second test set.
Through the arrangement, the accuracy of the SOC estimation model of the target lithium battery on the SOC can be improved.
It should be noted that: in the process of optimizing the first initial lithium battery SCO estimation model through the first training set or training the pre-trained lithium battery SOC estimation model through the second training set, whether the first initial lithium battery SCO estimation model and the pre-trained lithium battery SOC estimation model are trained or not can be judged through judging the evaluation function. Specifically, the evaluation function is Mean Absolute Error (MAE), and the specific formula is not described herein.
In order to better implement the method for estimating the SOC of the lithium battery in real time based on the optimized TCN in the embodiment of the present invention, on the basis of the method for estimating the SOC of the lithium battery in real time based on the optimized TCN, as shown in fig. 4, correspondingly, an embodiment of the present invention further provides a device 400 for estimating the SOC of the lithium battery in real time based on the optimized TCN, including:
a data obtaining unit 401, configured to obtain lithium ion battery data, where the lithium ion battery data includes a first data set and a second data set;
a first initial model building unit 402, configured to build a first initial lithium battery SOC estimation model, where the first initial lithium battery SOC estimation model includes initial time convolution network parameters;
an optimizing unit 403, configured to optimize the first initial lithium battery SOC estimation model based on a genetic algorithm and a first data set, to obtain a first transition lithium battery SOC estimation model, where the first transition lithium battery SOC estimation model includes target time convolution network parameters;
the second initial model building unit 404 is configured to build a second initial lithium battery SOC estimation model, and transfer the target time convolution network parameters to the second initial lithium battery SOC estimation model by using a transfer learning method to obtain a pre-training lithium battery SOC estimation model;
the estimating unit 405 is configured to train the pre-trained lithium battery SOC estimation model through the second data set to obtain a target lithium battery SOC estimation model, and estimate the SOC of the lithium battery through the target lithium battery SOC estimation model.
Here, it should be noted that: the optimized TCN-based real-time estimation apparatus 400 for SOC of a lithium battery according to the foregoing embodiments may implement the technical solutions described in the foregoing method embodiments, and the specific implementation principles of the modules or units may refer to the corresponding contents in the foregoing method embodiments, and are not described herein again.
Fig. 5 shows a schematic structural diagram of an electronic device provided in an embodiment of the present invention. As shown in fig. 5, the electronic device 500 includes a memory 501 and a processor 502. Wherein the memory 501 may be configured to store other various data to support operations on the electronic device. Examples of such data include instructions for any application or method operating on the electronic device. The memory 501 may be implemented by any type or combination of volatile or non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.
The memory 501 is used for storing programs;
the processor 502, coupled to the memory 501, is configured to execute the program stored in the memory 501, so as to:
acquiring lithium ion battery data, wherein the lithium ion battery data comprises a first data set and a second data set;
constructing a first initial lithium battery SOC estimation model, wherein the first initial lithium battery SOC estimation model comprises initial time convolution network parameters;
optimizing the first initial lithium battery SOC estimation model based on a genetic algorithm and a first data set to obtain a first transition lithium battery SOC estimation model, wherein the first transition lithium battery SOC estimation model comprises target time convolution network parameters;
constructing a second initial lithium battery SOC estimation model, and transferring the target time convolution network parameters to the second initial lithium battery SOC estimation model by adopting a transfer learning method to obtain a pre-training lithium battery SOC estimation model;
training the pre-trained lithium battery SOC estimation model through a second data set to obtain a target lithium battery SOC estimation model, and estimating the SOC of the lithium battery through the target lithium battery SOC estimation model.
It should be understood that: the processor 502 may also perform other functions in addition to the above functions when executing the program in the memory 201, see in particular the description of the corresponding method embodiments above.
Further, the type of the electronic device 500 is not particularly limited in the embodiment of the present invention, and the electronic device 500 may be a portable electronic device such as a mobile phone, a tablet computer, a Personal Digital Assistant (PDA), a wearable device, and a laptop computer (laptop). Exemplary embodiments of portable electronic devices include, but are not limited to, portable electronic devices that carry an iOS, android, microsoft, or other operating system. The portable electronic device may also be other portable electronic devices such as laptop computers (laptop) with touch sensitive surfaces (e.g., touch panels), etc. It should also be understood that in other embodiments of the present invention, the electronic device 500 may not be a portable electronic device, but may be a desktop computer having a touch-sensitive surface (e.g., a touch panel).
Accordingly, the present application also provides a computer-readable storage medium, which is used for storing a computer-readable program or instruction, and when the program or instruction is executed by a processor, the program or instruction can implement the method steps or functions provided by the above method embodiments.
Those skilled in the art will appreciate that all or part of the flow of the method implementing the above embodiments may be implemented by a computer program, which is stored in a computer readable storage medium, to instruct related hardware. The computer readable storage medium is a magnetic disk, an optical disk, a read-only memory or a random access memory.
The method, the device, the electronic device and the storage medium for estimating the SOC of the lithium battery based on the optimized TCN provided by the present invention in real time are described in detail above, and a specific example is applied in the present disclosure to explain the principle and the implementation of the present invention, and the description of the above embodiment is only used to help understanding the method and the core idea of the present invention; meanwhile, for those skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (10)
1. A lithium battery SOC real-time estimation method based on optimized TCN is characterized by comprising the following steps:
acquiring lithium ion battery data, wherein the lithium ion battery data comprises a first data set and a second data set;
constructing a first initial lithium battery SOC estimation model, wherein the first initial lithium battery SOC estimation model comprises initial time convolution network parameters;
optimizing the first initial lithium battery SOC estimation model based on a genetic algorithm and the first data set to obtain a first transition lithium battery SOC estimation model, wherein the first transition lithium battery SOC estimation model comprises target time convolution network parameters;
constructing a second initial lithium battery SOC estimation model, and transferring the target time convolution network parameters to the second initial lithium battery SOC estimation model by adopting a transfer learning method to obtain a pre-training lithium battery SOC estimation model;
training the pre-trained lithium battery SOC estimation model through the second data set to obtain a target lithium battery SOC estimation model, and estimating the SOC of the lithium battery through the target lithium battery SOC estimation model.
2. The method of claim 1, wherein the target time convolution network parameters comprise the number of residual blocks, the input step size, and the expansion factor.
3. The method of claim 2, wherein the first data set carries measured values; the optimizing the first initial lithium battery SOC estimation model based on a genetic algorithm and a first data set comprises:
setting an initialization population, a cross rate, a variation rate and a maximum iteration number, and carrying out real number coding on initial time convolution network parameters;
inputting the first data set into the first initial lithium battery SOC estimation model to obtain a predicted value;
designing a fitness function, and calculating a fitness value of the population according to the fitness function, the measured value and the predicted value;
judging whether the fitness value of the initialized population meets the requirement, if so, taking the current initial time convolution network parameter as the target time convolution network parameter;
if not, retaining the optimal individuals, and performing selection, crossing and mutation operations on the individuals in the optimal individuals to generate a new population and a transition time convolution network parameter until the times of the selection, crossing and mutation operations are greater than the maximum iteration times, and then taking the transition time convolution network parameter with the minimum fitness value as the target time convolution network parameter.
4. The method for real-time estimation of SOC of lithium battery based on optimized TCN as claimed in claim 3, wherein the fitness function is:
wherein F is the fitness value of the individual to which the coefficient is added; n is the number of data in the first data set;
is an actual measurement value; SoC (system on chip)iIs a predicted value; k is a coefficient; abs is an absolute value function.
5. The method for real-time estimation of SOC of lithium battery based on optimized TCN as claimed in claim 4, wherein the selection operation is selected by roulette with a probability of:
fi＝m/F
in the formula, PiTo select a probability; n is the number of population individuals; m is a weight; f. ofiFitness value of the ith individual; f. ofjIs the fitness value of the jth individual.
6. The method of claim 4, wherein the interleaving is performed by:
in the formula, akjThe j position of the kth individual; a isljIs the jth position of the ith individual; b is a random number between 0 and 1.
7. The method of claim 4, wherein the mutation operation is to:
f(g)＝r2(1-g/Gmax)2
in the formula, amaxAnd aminAre respectively aijUpper and lower thresholds of; r is2Is a random number; g is iteration times; gmaxIs the maximum iteration number;&r is a random number between 0 and 1.
8. The utility model provides a lithium cell SOC real-time estimation device based on TCN of optimizing which characterized in that includes:
the data acquisition unit is used for acquiring lithium ion battery data, and the lithium ion battery data comprises a first data set and a second data set;
the system comprises a first initial model building unit, a second initial model building unit and a third initial model building unit, wherein the first initial model building unit is used for building a first initial lithium battery SOC estimation model which comprises initial time convolution network parameters;
the optimization unit is used for optimizing the first initial lithium battery SOC estimation model based on a genetic algorithm and the first data set to obtain a first transition lithium battery SOC estimation model, and the first transition lithium battery SOC estimation model comprises target time convolution network parameters;
the second initial model building unit is used for building a second initial lithium battery SOC estimation model, and transferring the target time convolution network parameters to the second initial lithium battery SOC estimation model by adopting a transfer learning method to obtain a pre-training lithium battery SOC estimation model;
and the estimation unit is used for training the pre-trained lithium battery SOC estimation model through the second data set to obtain a target lithium battery SOC estimation model, and estimating the SOC of the lithium battery through the target lithium battery SOC estimation model.
9. An electronic device comprising a memory and a processor, wherein,
the memory is used for storing programs;
the processor, coupled with the memory, is configured to execute the program stored in the memory to implement the steps in the method for real-time estimation of SOC of a TCN-based lithium battery according to any one of the claims 1 to 7.
10. A computer readable storage medium for storing a computer readable program or instructions, which when executed by a processor, can implement the steps of the method for real-time SOC estimation of a TCN-based lithium battery according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110724564.XA CN113406499A (en) | 2021-06-28 | 2021-06-28 | Lithium battery SOC real-time estimation method and device based on optimized TCN, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110724564.XA CN113406499A (en) | 2021-06-28 | 2021-06-28 | Lithium battery SOC real-time estimation method and device based on optimized TCN, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113406499A true CN113406499A (en) | 2021-09-17 |
Family
ID=77680008
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110724564.XA Pending CN113406499A (en) | 2021-06-28 | 2021-06-28 | Lithium battery SOC real-time estimation method and device based on optimized TCN, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113406499A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111220921A (en) * | 2020-01-08 | 2020-06-02 | 重庆邮电大学 | Lithium battery capacity estimation method based on improved convolution-long-and-short-term memory neural network |
| CN111695301A (en) * | 2020-06-16 | 2020-09-22 | 中国科学院深圳先进技术研究院 | Method and device for predicting battery charge state, storage medium and equipment |
| CN112540298A (en) * | 2020-12-26 | 2021-03-23 | 福州大学 | Lithium ion battery SOC estimation method based on deep-migration learning |
-
2021
- 2021-06-28 CN CN202110724564.XA patent/CN113406499A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111220921A (en) * | 2020-01-08 | 2020-06-02 | 重庆邮电大学 | Lithium battery capacity estimation method based on improved convolution-long-and-short-term memory neural network |
| CN111695301A (en) * | 2020-06-16 | 2020-09-22 | 中国科学院深圳先进技术研究院 | Method and device for predicting battery charge state, storage medium and equipment |
| CN112540298A (en) * | 2020-12-26 | 2021-03-23 | 福州大学 | Lithium ion battery SOC estimation method based on deep-migration learning |
Non-Patent Citations (2)
| Title |
|---|
| 于仲安 等: "基于GA-BP神经网络的锂离子电池SOC估计", 《电源技术》 * |
| 梁超: "基于时间卷积网络的锂离子电池SOC估算模型研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107870306A (en) | A kind of lithium battery charge state prediction algorithm based under deep neural network | |
| WO2022253038A1 (en) | Method and system for predicting state of health of lithium battery on basis of elastic network, and device and medium | |
| Sun et al. | A sequential capacity estimation for the lithium-ion batteries combining incremental capacity curve and discrete Arrhenius fading model | |
| CN111680848A (en) | Battery life prediction method based on prediction model fusion and storage medium | |
| CN113671381B (en) | Lithium ion power battery estimation method based on time convolution network | |
| CN110568359A (en) | lithium battery residual life prediction method | |
| CN115291116B (en) | Energy storage battery health state prediction method and device and intelligent terminal | |
| CN113589189B (en) | Lithium battery health condition prediction method and device based on charging and discharging data characteristics | |
| CN110750852B (en) | Method and device for predicting residual service life of supercapacitor and electronic equipment | |
| CN111426960A (en) | Energy storage lithium battery charge state monitoring method and device | |
| CN116449218B (en) | Lithium battery health state estimation method | |
| CN114384435A (en) | WSA-LSTM algorithm-based self-adaptive prediction method for residual service life of new energy automobile power battery | |
| CN115308628A (en) | Battery health state monitoring method | |
| CN117031310A (en) | Method for predicting residual service life of power battery of electric automobile | |
| CN115166532A (en) | Method and device for predicting capacity of nickel-metal hydride battery, electronic device and storage medium | |
| Cao et al. | A flexible battery capacity estimation method based on partial voltage curves and polynomial fitting | |
| Chen et al. | State of health estimation of lithium-ion batteries based on equivalent circuit model and data-driven method | |
| Zhang et al. | Online state-of-health estimation for the lithium-ion battery based on an LSTM neural network with attention mechanism | |
| CN117289167A (en) | Battery remaining life prediction method, device and medium based on multiple neural network | |
| CN116609686B (en) | Battery cell consistency assessment method based on cloud platform big data | |
| US20230349977A1 (en) | Method and apparatus for estimating state of health of battery | |
| CN113406499A (en) | Lithium battery SOC real-time estimation method and device based on optimized TCN, electronic equipment and storage medium | |
| CN116679213A (en) | SOH estimation method for electric vehicle power battery based on integrated deep learning | |
| CN116047339A (en) | Lithium ion battery pack SOC estimation method and device based on thermoelectric coupling model | |
| Romero et al. | Fast charge of Li-ion batteries using a two-layer distributed MPC with electro-chemical and thermal constraints |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210917 |