CN115407217B - Online estimation method and system for state of charge of lithium battery of electric vehicle - Google Patents
Online estimation method and system for state of charge of lithium battery of electric vehicle Download PDFInfo
- Publication number
- CN115407217B CN115407217B CN202211352798.7A CN202211352798A CN115407217B CN 115407217 B CN115407217 B CN 115407217B CN 202211352798 A CN202211352798 A CN 202211352798A CN 115407217 B CN115407217 B CN 115407217B
- Authority
- CN
- China
- Prior art keywords
- battery
- estimation
- soc
- lithium battery
- model
- 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.)
- Active
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]
- G01R31/382—Arrangements for monitoring battery or accumulator variables, e.g. SoC
- G01R31/3842—Arrangements for monitoring battery or accumulator variables, e.g. SoC combining voltage and current measurements
-
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/60—Other road transportation technologies with climate change mitigation effect
- Y02T10/70—Energy storage systems for electromobility, e.g. batteries
Abstract
The invention relates to the technical field of battery monitoring, in particular to an online estimation method and system for the state of charge of a lithium battery of an electric vehicle, wherein the method comprises the following steps: acquiring battery data of the electric automobile at the current moment; uploading the current moment of battery data to a cloud battery monitoring platform to utilize the cloud battery monitoring platform to realize: based on the battery data at the current moment and the historical moment, carrying out SOC estimation through a stored cloud estimation model to obtain a first SOC estimation value at the next moment, and returning the first SOC estimation value; based on the current at the current moment, carrying out SOC estimation through a vehicle end estimation model to obtain a second SOC estimation value at the next moment; and fusing through a Kalman filter by taking the obtained first SOC estimation value as a measurement value and the second SOC estimation value as an observation value to obtain the SOC estimation value of the lithium battery at the next moment. The method can realize the quick and high-precision SOC estimation of the lithium battery of the electric automobile.
Description
Technical Field
The embodiment of the invention relates to the technical field of battery monitoring, in particular to an online estimation method and system for the state of charge of a lithium battery of an electric vehicle, electronic equipment and a storage medium.
Background
In recent years, electric vehicles have received much attention due to their great advantages in technology, environmental protection, and energy saving. Lithium batteries are widely used as power sources of electric vehicles due to their advantages of high energy density, high voltage, long service life, etc. The State of Charge (SOC) of a lithium battery reflects the remaining capacity of the battery, and is one of important evaluation indexes. Accurate and stable state of charge estimation (also called state of charge prediction) is very important for prolonging the service life of a battery and ensuring safe driving of an electric automobile. However, accurate state of charge estimation for lithium batteries is often difficult due to the effects of temperature, unknown noise, and uncertain outliers. At present, the factors such as calculation power of a vehicle-mounted BMS are limited, the precision and the real-time performance of the lithium battery state of charge estimation are difficult to be considered, and the safety monitoring of the battery and the electric automobile is not facilitated.
Disclosure of Invention
Based on the problem that the precision and the real-time performance of lithium battery state-of-charge estimation are difficult to be considered in the prior art, the embodiment of the invention provides a vehicle-cloud-coordinated electric vehicle lithium battery state-of-charge online estimation method, system, electronic equipment and storage medium, and can realize quick and high-precision electric vehicle lithium battery SOC estimation.
In a first aspect, an embodiment of the present invention provides an online estimation method for a state of charge of a lithium battery of an electric vehicle, including:
acquiring battery data of the electric automobile at the current moment, wherein the battery data comprises current, voltage and temperature of a lithium battery;
uploading the current battery data to a cloud battery monitoring platform to utilize the cloud battery monitoring platform to realize: based on the battery data at the current moment and the historical moment, carrying out SOC estimation through a stored cloud estimation model to obtain a first SOC estimation value at the next moment, and returning the first SOC estimation value;
based on the current at the current moment, carrying out SOC estimation through a vehicle end estimation model to obtain a second SOC estimation value at the next moment; the vehicle-end estimation model is constructed by adopting an ampere-hour integral method;
and fusing through a Kalman filter by taking the obtained first SOC estimation value as a measurement value and the second SOC estimation value as an observation value to obtain the SOC estimation value of the lithium battery at the next moment.
Optionally, the cloud estimation model is a trained deep learning model.
Optionally, the deep learning model is a CNN model, RNN model, LSTM model, or GRU model.
Optionally, the deep learning model is a 1D-CNN model;
each layer of the network hidden layer in the 1D-CNN model comprises a BN layer for normalization, and the gradient descent calculation method adopted in the 1D-CNN model is an adaptive moment estimation method.
Optionally, the deep learning model is trained by:
battery data and corresponding lithium cell SOC of a plurality of different moments under two kind at least operating modes of electric automobile are obtained respectively, reach on uploading to high in the clouds battery monitor platform, in order to utilize high in the clouds battery monitor platform realizes:
constructing a deep learning model;
preprocessing the acquired data;
obtaining a training sample set and a testing sample set based on the preprocessed data; the training sample set comprises battery data and corresponding lithium battery SOC at a plurality of different moments under one working condition, and the testing sample set comprises battery data and corresponding lithium battery SOC at a plurality of different moments under other working conditions;
and training the constructed deep learning model based on the obtained training sample set and the test sample set until the deep learning model converges.
Optionally, the preprocessing the acquired data includes:
and carrying out abnormal data cleaning and fixed interval sampling on each battery data and the corresponding lithium battery SOC.
Optionally, the SOC estimation is performed through a vehicle end estimation model, and the following expression is adopted:
let the current time bekAt the moment of time, the time of day,
to representkThe second SOC estimation value at the time +1,
to representkOf time of dayThe second SOC estimation value is set to be,
to representkThe current at the moment of time is,
which represents the rated capacity of the lithium battery,
to representk+1 time andkthe time interval between the moments.
In a second aspect, an embodiment of the present invention further provides an online estimation system for a state of charge of a lithium battery of an electric vehicle, including: the system comprises a vehicle-mounted computing device and a cloud battery monitoring platform; wherein, the first and the second end of the pipe are connected with each other,
the in-vehicle computing device includes:
the acquisition module is used for acquiring battery data of the electric automobile at the current moment, including current, voltage and temperature of the lithium battery;
the uploading module is used for uploading the battery data at the current moment to a cloud battery monitoring platform so as to call the cloud battery monitoring platform to carry out SOC estimation and obtain a first SOC estimation value at the next moment;
the estimation module is used for carrying out SOC estimation through the vehicle end estimation model based on the current at the current moment to obtain a second SOC estimation value at the next moment;
the fusion module is used for taking the obtained first SOC estimation value as a measurement value and the second SOC estimation value as an observation value, and fusing through a Kalman filter to obtain a lithium battery SOC estimation value at the next moment;
the cloud battery monitoring platform is used for responding to the call of the uploading module, carrying out SOC estimation through a stored cloud estimation model based on the battery data at the current moment and the historical moment, obtaining a first SOC estimation value at the next moment, and transmitting the first SOC estimation value back to the uploading module.
In a third aspect, an embodiment of the present invention further provides an electronic device, which includes a memory and a processor, where the memory stores a computer program, and when the processor executes the computer program, the online estimation method for the state of charge of the lithium battery of the electric vehicle according to any embodiment of the present specification is implemented.
In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored, and when the computer program is executed in a computer, the computer is caused to execute the method for estimating a state of charge of a lithium battery of an electric vehicle on line according to any embodiment of the present specification.
The invention provides an electric vehicle lithium battery state of charge online estimation method, a system, electronic equipment and a storage medium, wherein based on battery data of the electric vehicle at the current moment, a first SOC estimation value at the next moment is obtained through cloud computing, a second SOC estimation value at the next moment is obtained through vehicle-side computing, and then fusion is carried out through a Kalman filter to finally obtain the SOC estimation value of the lithium battery at the next moment; according to the method, a vehicle cloud cooperation mode is adopted, the lithium battery state of charge estimation results obtained by two different modes of vehicle-end calculation and cloud calculation are combined, the problems of dependency and limitation caused by the fact that a single method is used for state of charge estimation are solved, the anti-interference capacity is higher, the vehicle end does not need to bear the calculation pressure of estimation by multiple methods independently, the cloud end can perform more complex and accurate calculation, the calculation accuracy is effectively improved, the calculation speed is higher, and the real-time requirement can be met.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 is a flowchart of an online estimation method for a state of charge of a lithium battery of an electric vehicle according to an embodiment of the present invention;
fig. 2 is a hardware architecture diagram of an electronic device according to an embodiment of the present invention;
fig. 3 is a structural diagram of an in-vehicle computing device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer and more complete, the technical solutions in the embodiments of the present invention will be described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention, and based on the embodiments of the present invention, all other embodiments obtained by a person of ordinary skill in the art without creative efforts belong to the scope of the present invention.
As previously mentioned, accurate state of charge estimation for lithium batteries is often difficult due to factors such as temperature, unknown noise, and uncertain outliers. In the prior art, the state of charge estimation of a lithium battery is mainly divided into an ampere-hour integration method, an open-circuit voltage method, a model-based method and a data driving method. The ampere-hour integration method has a disadvantage that an estimation result is easily affected by an initial SOC deviation and an accumulated error as an open-loop method. The open circuit voltage method has a disadvantage that the corresponding relationship between the open circuit voltage and the SOC is affected by the temperature and the battery aging, which results in a significant estimation error. A disadvantage of model-based methods is that building accurate battery models is often complicated and results in increased computational complexity. The data-driven method has the disadvantages that the estimation precision is highly dependent on the data quality, the estimation error is large under the complex working condition, and the complex network parameters and the huge calculation cost can cause the method to be difficult to be used for vehicles. At present, the estimation accuracy is limited due to factors such as calculation power of a vehicle-mounted BMS (battery management system), and the like, when a single method is used for estimating the state of charge of a lithium battery, the estimation accuracy is limited, and when a high-accuracy method is used for estimating, the real-time performance is poor, so that the safety monitoring requirements of the lithium battery and an electric vehicle are difficult to meet. In view of this, the invention provides a vehicle cloud collaborative online estimation method for battery SOC, so as to realize fast and high-precision SOC estimation.
Specific implementations of the above concepts are described below.
Referring to fig. 1, an embodiment of the present invention provides an online estimation method for a state of charge of a lithium battery of an electric vehicle, where the method includes:
102, uploading the battery data at the current moment to a cloud battery monitoring platform so as to realize that:
based on the battery data at the current moment and the historical moment, carrying out SOC estimation through a stored cloud estimation model to obtain a first SOC estimation value at the next moment, and returning the first SOC estimation value;
104, carrying out SOC estimation through a vehicle end estimation model based on the current at the current moment to obtain a second SOC estimation value at the next moment; the vehicle-end estimation model is constructed by adopting an ampere-hour integral method;
and 106, fusing by using the obtained first SOC estimation value as a measurement value and the second SOC estimation value as an observation value through a Kalman filter to obtain the SOC estimation value of the lithium battery at the next moment.
The method comprises the steps of utilizing a vehicle end estimation model to achieve vehicle end calculation, carrying out SOC estimation on a lithium battery of the electric vehicle, utilizing a cloud battery monitoring platform to carry out cloud end calculation, carrying out SOC estimation on the lithium battery of the electric vehicle through the stored cloud end estimation model at the cloud end, finally using a Kalman filtering technology, taking an SOC estimation result of the vehicle end calculation as an observation value and an SOC estimation result of the cloud end calculation as a measurement value, and obtaining optimal estimation as an SOC estimation value of the lithium battery at the next moment through fusion calculation. According to the invention, a vehicle-cloud cooperation mode is adopted, the lithium battery state-of-charge estimation results obtained by two different modes of vehicle-end calculation and cloud-end calculation are combined, so that the problems of dependence and limitation caused by the fact that a single method is used for state-of-charge estimation are solved, the anti-interference capability is stronger, and due to the utilization of the cloud-end calculation results, the vehicle end does not need to independently bear the calculation pressure of estimation by using multiple methods, the cloud end can perform more complex and accurate calculation, the calculation precision is effectively improved, the calculation speed is higher, and the real-time requirement can be met.
Alternatively, for step 100, the battery data of the electric vehicle at the current moment may be collected from a vehicle-mounted Battery Management System (BMS).
Optionally, for step 102, the cloud estimation model is a trained deep learning model.
The deep learning model has a complex structure and high calculation consumption, but the obtained SOC estimation value has high precision. The cloud computing adopts a deep learning model, so that the estimation result obtained by final fusion is more accurate.
Further, the deep learning model may adopt a CNN model, an RNN model, an LSTM model or a GRU model, and is more preferably a one-dimensional convolutional neural network model in the CNN model, i.e., a 1D-CNN model.
The deep learning model can realize more accurate estimation results. Considering that the SOC estimation of the lithium battery belongs to the time sequence problem, the training speed of the CNN model is higher, and meanwhile, the convolution operation can avoid the defects of a recursive model (such as an LSTM model), such as gradient disappearance, gradient explosion, lack of memory reservation and the like, therefore, the 1D-CNN model is preferably adopted, the 1D-CNN model applies convolution operation on a time axis through a one-dimensional kernel (also called time convolution), and further the time sequence problem is solved, and a network of the 1D-CNN model is constructed by stacking a plurality of time convolution units.
Further, the deep learning model is a 1D-CNN model, each layer of the network hidden layer in the 1D-CNN model includes a BN (Batch Normalization) layer for Normalization, and a gradient descent calculation method adopted in the 1D-CNN model is an adaptive moment estimation method.
In the above embodiment, the 1D-CNN model maintains the same distribution of inputs of each layer of neural network through a batch normalization technique, thereby accelerating convergence, preventing gradient disappearance and gradient explosion, and suppressing overfitting to a certain extent; in the model training process, the adaptive moment estimation method (namely the Adam method) can dynamically adjust the learning rate of each parameter through the first moment estimation and the second moment estimation of the gradient, and simultaneously introduce deviation correction to ensure that the learning rate has a determined range in each iteration, so that the parameters are stable, the convergence speed and the accuracy are obviously improved, and the updating of the learning rate and the deviation correction can be introduced through the adaptive moment estimation method.
Further, the 1D-CNN model includes an input layer, four time convolution units, a pooling layer, and an output layer, which are sequentially connected, wherein each time convolution unit is composed of a convolution layer, a BN layer, and a hash layer, which are sequentially connected, the widths of the kernels of the four time convolution units are 7, 5, 3, and 1, the numbers of the kernels are 16, 32, 16, and 1, respectively, the activation function of the convolution layer is a hash function, and the activation function of the output layer is a Relu function.
Optionally, the deep learning model is trained by:
battery data and corresponding lithium cell SOC of a plurality of different moments under two kind at least operating modes of electric automobile are obtained respectively, reach on uploading to high in the clouds battery monitor platform, in order to utilize high in the clouds battery monitor platform realizes:
constructing a deep learning model;
preprocessing the acquired data;
obtaining a training sample set and a testing sample set based on the preprocessed data; the training sample set comprises battery data and corresponding lithium battery SOC at a plurality of different moments under one working condition, and the testing sample set comprises battery data and corresponding lithium battery SOC at a plurality of different moments under other working conditions;
and training the constructed deep learning model based on the obtained training sample set and the test sample set until the deep learning model converges.
Adopt above-mentioned embodiment can train the degree of deep learning model in the high in the clouds battery monitoring platform, a data for the training, namely, the battery data and corresponding lithium cell SOC of a plurality of different moments of electric automobile, can come from on-vehicle battery management system collection, the producer provides and multiple approaches such as public data set, but should include battery data and corresponding lithium cell SOC under two kinds of operating modes at least, wherein every operating mode all corresponds the data that have a plurality of different moments, so that make the degree of deep learning model after the training possess the generalization, can be applicable to various complicated operating modes. Taking an A123 18650 lithium iron phosphate battery as an example, data for training can include three working conditions, the three working conditions are dynamic stress test, US06 driving plan and federal city driving plan respectively, a plurality of different time data corresponding to the dynamic stress test working conditions are selected to form a training sample set, and the rest working conditions, namely a plurality of different time data corresponding to the US06 driving plan and the federal city driving plan working conditions form a testing sample set, so that the generalization of the network of the deep learning model can be increased.
Further, the preprocessing the acquired data includes:
and performing abnormal data removing cleaning and fixed interval sampling on the battery data at different moments and the corresponding lithium battery SOC.
In the embodiment, through removing abnormal data and washing, can screen the obvious unusual data, the fixed interval sampling then is favorable to reducing the noise that the data acquisition brought, has solved the on-vehicle BMS in practical application and has gathered the estimation precision decline that data anomaly and fluctuation lead to, poor stability scheduling problem.
Further, the training the constructed deep learning model based on the obtained training sample set and the test sample set includes:
constructing an input matrix based on the obtained training sample set;
based on the input matrix, obtaining input data by enabling a preset sliding window to move in the input matrix according to a preset step length in a time dimension;
training the deep learning model based on the input data;
and testing the deep learning model based on the test sample set.
In the embodiment, the data is constructed into the input matrix, and the sliding window with a certain size moves in the input matrix according to a certain step length in the time dimension to obtain the input data with the preset format, so that local connection and weight sharing are realized, and the model training efficiency is improved. It is understood that the input data of the deep learning model during estimation and the input data during training should keep a corresponding format.
Optionally, for step 104, the SOC estimation is performed through a vehicle end estimation model, and the following expression is adopted:
wherein the content of the first and second substances,kthe time is the current time, and the time is,
to representkThe second SOC estimation value at the +1 time (i.e. the next time),
to representkThe second SOC estimation value at the time of day,
to representkThe current at the moment of time is,
which represents the rated capacity of the lithium battery,
to representk+1 time andkthe time interval between the moments.
By adopting the embodiment, the SOC estimation of the battery, namely the second SOC estimation value, can be obtained by integrating the current with respect to time and combining the initial SOC, an ampere-hour integration method is adopted for vehicle-end calculation, the model structure is simple, the calculation consumption is low, the speed is high, and the real-time performance of estimation can be improved.
In step 106, kalman filtering is performed by taking the second SOC estimation value calculated by the vehicle end as an observation value and taking the first SOC estimation value calculated by the cloud end as a measurement value, so that vehicle-cloud cooperation is realized, and a more accurate estimation result can be obtained. The Kalman filter integrates the vehicle end and the cloud computing result, the problem that the result precision is low due to the fact that a single method is used for estimating the state of charge of the lithium battery is solved, meanwhile, more computing pressure is not caused to the vehicle end, the estimation result can be obtained quickly, and the precision and the real-time performance are both considered.
Optionally, the online estimation method for the state of charge of the lithium battery of the electric vehicle further includes: and outputting the SOC estimated value of the lithium battery at the next moment.
The embodiment of the invention also provides an online estimation system for the state of charge of the lithium battery of the electric automobile, which comprises the following steps: the system comprises a vehicle-mounted computing device and a cloud battery monitoring platform;
the vehicle-mounted computing device comprises an acquisition module 301, an uploading module 302, an estimation module 303 and a fusion module 304; the acquisition module 301 is configured to acquire battery data of the electric vehicle at the current moment, including current, voltage and temperature of the lithium battery; the uploading module 302 is configured to upload the battery data at the current moment to a cloud battery monitoring platform, so as to call the cloud battery monitoring platform to perform SOC estimation, and obtain a first SOC estimation value at the next moment; the estimation module 303 is configured to perform SOC estimation through a vehicle-end estimation model based on the current at the current moment to obtain a second SOC estimation value at the next moment; the fusion module 304 is configured to fuse the obtained first SOC estimation value as a measurement value and the obtained second SOC estimation value as an observation value through a kalman filter to obtain a next-time SOC estimation value of the lithium battery;
the cloud battery monitoring platform stores a cloud estimation model, is used for responding to the call of the uploading module 302, performs SOC estimation on the basis of the battery data at the current moment and the historical moment through the stored cloud estimation model to obtain a first SOC estimation value at the next moment, and transmits the first SOC estimation value back to the uploading module 302.
As shown in fig. 2 and 3, an embodiment of the present invention provides an in-vehicle computing device. The apparatus may be implemented by software, or by hardware, or by a combination of hardware and software. From a hardware aspect, as shown in fig. 2, for a hardware architecture diagram of an electronic device in which an in-vehicle computing device is located according to an embodiment of the present invention, in addition to the processor, the memory, the network interface, and the nonvolatile memory shown in fig. 2, the electronic device in which the device is located may also include other hardware, such as a forwarding chip responsible for processing a packet. Taking a software implementation as an example, as shown in fig. 3, as a logical device, a CPU of the electronic device reads a corresponding computer program in the non-volatile memory into the memory for running.
In an embodiment of the present invention, the obtaining module 301 may be configured to perform step 100 in the foregoing method embodiment, the uploading module 302 may be configured to perform step 102 in the foregoing method embodiment, the estimating module 303 may be configured to perform step 104 in the foregoing method embodiment, and the fusing module 304 may be configured to perform step 106 in the foregoing method embodiment.
It is to be understood that the illustrated structure of the embodiment of the present invention does not specifically limit the in-vehicle computing device. In other embodiments of the invention, the in-vehicle computing device may include more or fewer components than illustrated, or combine certain components, or split certain components, or arrange different components. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.
Because the content of information interaction, execution process, and the like among the modules in the device is based on the same concept as the method embodiment of the present invention, specific content can be referred to the description in the method embodiment of the present invention, and is not described herein again.
Optionally, in the online estimation system for the state of charge of the lithium battery of the electric vehicle, the cloud estimation model stored by the cloud battery monitoring platform is a deep learning model after training. The deep learning model can adopt a CNN model, an RNN model, an LSTM model or a GRU model, and is more preferably a 1D-CNN model.
Further, the deep learning model is trained by:
constructing a deep learning model;
acquiring battery data and corresponding lithium battery SOC of the electric automobile at a plurality of different moments under at least two working conditions, and preprocessing the battery data and the corresponding lithium battery SOC;
obtaining a training sample set and a testing sample set based on the preprocessed data; the training sample set comprises battery data and corresponding lithium battery SOCs at a plurality of different moments under one working condition, and the test sample set comprises battery data and corresponding lithium battery SOCs at a plurality of different moments under other working conditions;
and training the constructed deep learning model based on the obtained training sample set and the test sample set until the deep learning model converges.
The deep learning model training process in the embodiment is preferably completed by the cloud-side battery monitoring platform, the cloud-side battery monitoring platform is stronger in calculation capacity, and the calculation amount of a vehicle end can be reduced through vehicle-cloud cooperation.
The embodiment of the invention also provides electronic equipment which comprises a memory and a processor, wherein the memory is stored with a computer program, and when the processor executes the computer program, the online estimation method for the state of charge of the lithium battery of the electric vehicle in any embodiment of the invention is realized.
The embodiment of the invention also provides a computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the processor is enabled to execute the method for estimating the state of charge of the lithium battery of the electric vehicle on line in any embodiment of the invention.
Specifically, a system or an apparatus equipped with a storage medium on which software program codes that realize the functions of any of the above-described embodiments are stored may be provided, and a computer (or a CPU or MPU) of the system or the apparatus is caused to read out and execute the program codes stored in the storage medium.
In this case, the program code itself read from the storage medium can realize the functions of any of the above-described embodiments, and thus the program code and the storage medium storing the program code constitute a part of the present invention.
Examples of the storage medium for supplying the program code include a floppy disk, a hard disk, a magneto-optical disk, an optical disk (e.g., CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD + RW), a magnetic tape, a nonvolatile memory card, and a ROM. Alternatively, the program code may be downloaded from a server computer via a communications network.
Further, it should be clear that the functions of any one of the above-described embodiments may be implemented not only by executing the program code read out by the computer, but also by causing an operating system or the like operating on the computer to perform a part or all of the actual operations based on instructions of the program code.
Further, it is to be understood that the program code read out from the storage medium is written to a memory provided in an expansion board inserted into the computer or to a memory provided in an expansion module connected to the computer, and then causes a CPU or the like mounted on the expansion board or the expansion module to perform part or all of the actual operations based on instructions of the program code, thereby realizing the functions of any of the above-described embodiments.
In summary, the invention provides an online estimation method, a system, an electronic device and a storage medium for a state of charge of a lithium battery of an electric vehicle, in which, in a vehicle cloud cooperation manner, a cloud estimation model with a complex cloud arrangement structure, large calculation consumption and high calculation accuracy is arranged at a vehicle end, and a vehicle end estimation model with a simple vehicle end, small calculation consumption and low calculation accuracy is arranged at the vehicle end, and a cloud calculation result (i.e., a first SOC estimation value) and a vehicle end calculation result (i.e., a second SOC estimation value) are respectively obtained, and then a kalman filter is used for fusion to obtain a more accurate estimation result, so that the online estimation of the battery SOC in the vehicle cloud cooperation is realized, and the problem that in practical application, when a single method is used for estimation, the estimation accuracy is limited and the real-time performance is poor due to the use of the high accuracy model is solved.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising a" \8230; "does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
Those of ordinary skill in the art will understand that: all or part of the steps for realizing the method embodiments can be completed by hardware related to program instructions, the program can be stored in a computer readable storage medium, and the program executes the steps comprising the method embodiments when executed; and the aforementioned storage medium includes: ROM, RAM, magnetic or optical disks, etc. that can store program codes.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, and not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (8)
1. An online estimation method for the state of charge of a lithium battery of an electric vehicle is characterized by comprising the following steps:
acquiring battery data of the electric automobile at the current moment, wherein the battery data comprises current, voltage and temperature of a lithium battery;
uploading the current battery data to a cloud battery monitoring platform to utilize the cloud battery monitoring platform to realize: based on the battery data at the current moment and the historical moment, carrying out SOC estimation through a stored cloud estimation model to obtain a first SOC estimation value at the next moment, and returning the first SOC estimation value;
based on the current at the current moment, carrying out SOC estimation through a vehicle end estimation model to obtain a second SOC estimation value at the next moment; the vehicle-end estimation model is constructed by adopting an ampere-hour integral method;
fusing through a Kalman filter by taking the obtained first SOC estimation value as a measurement value and the second SOC estimation value as an observation value to obtain a lithium battery SOC estimation value at the next moment;
the cloud estimation model is a trained deep learning model; the deep learning model is trained by the following method:
battery data and corresponding lithium cell SOC at a plurality of different moments under two kinds of operating modes of electric automobile are obtained respectively, upload to high in the clouds battery monitoring platform, in order to utilize high in the clouds battery monitoring platform realizes:
constructing a deep learning model;
preprocessing the acquired data;
obtaining a training sample set and a testing sample set based on the preprocessed data; the training sample set comprises battery data and corresponding lithium battery SOCs at a plurality of different moments under one working condition, and the test sample set comprises battery data and corresponding lithium battery SOCs at a plurality of different moments under other working conditions;
and training the constructed deep learning model based on the obtained training sample set and the test sample set until the deep learning model converges.
2. The online estimation method for the state of charge of the lithium battery of the electric vehicle according to claim 1, characterized in that: the deep learning model is a CNN model, an RNN model, an LSTM model or a GRU model.
3. The online estimation method for the state of charge of the lithium battery of the electric vehicle according to claim 2, characterized in that: the deep learning model is a 1D-CNN model;
each layer of the network hidden layer in the 1D-CNN model comprises a BN layer for normalization, and the gradient descent calculation method adopted in the 1D-CNN model is an adaptive moment estimation method.
4. The online estimation method for the state of charge of the lithium battery of the electric vehicle according to claim 3, wherein the preprocessing the acquired data comprises:
and carrying out abnormal data cleaning and fixed interval sampling on each battery data and the corresponding lithium battery SOC.
5. The online estimation method for the state of charge of the lithium battery of the electric automobile according to claim 1, wherein the SOC estimation is performed through a vehicle-end estimation model, and the following expression is adopted:
let the current time bekAt the time of day, the user may,
to representkThe second SOC estimation value at the time +1,
to representkThe second SOC estimation value at the time of day,
to representkThe current of the moment of time is,
which represents the rated capacity of the lithium battery,
representk+1 time andkthe time interval between the moments.
6. An electric vehicle lithium battery state of charge on-line estimation system is characterized by comprising: the system comprises a vehicle-mounted computing device and a cloud battery monitoring platform; wherein, the first and the second end of the pipe are connected with each other,
the in-vehicle computing device includes:
the acquisition module is used for acquiring battery data of the electric automobile at the current moment, including current, voltage and temperature of the lithium battery;
the uploading module is used for uploading the battery data at the current moment to a cloud battery monitoring platform so as to call the cloud battery monitoring platform to carry out SOC estimation and obtain a first SOC estimation value at the next moment;
the estimation module is used for carrying out SOC estimation through the vehicle end estimation model based on the current at the current moment to obtain a second SOC estimation value at the next moment;
the fusion module is used for taking the obtained first SOC estimation value as a measurement value and the second SOC estimation value as an observation value, and fusing through a Kalman filter to obtain a lithium battery SOC estimation value at the next moment;
the cloud battery monitoring platform is used for responding to the call of the uploading module, carrying out SOC estimation through a stored cloud estimation model based on the battery data at the current moment and the historical moment to obtain a first SOC estimation value at the next moment, and transmitting the first SOC estimation value back to the uploading module;
the cloud estimation model is a trained deep learning model; the deep learning model is trained by the following method:
battery data and corresponding lithium cell SOC at a plurality of different moments under two kinds of operating modes of electric automobile are obtained respectively, upload to high in the clouds battery monitoring platform, in order to utilize high in the clouds battery monitoring platform realizes:
constructing a deep learning model;
preprocessing the acquired data;
obtaining a training sample set and a testing sample set based on the preprocessed data; the training sample set comprises battery data and corresponding lithium battery SOC at a plurality of different moments under one working condition, and the testing sample set comprises battery data and corresponding lithium battery SOC at a plurality of different moments under other working conditions;
and training the constructed deep learning model based on the obtained training sample set and the test sample set until the deep learning model converges.
7. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method for online estimation of state of charge of a lithium battery of an electric vehicle according to any one of claims 1 to 5.
8. A storage medium having a computer program stored thereon, wherein when the computer program is executed in a computer, the computer program is used to make the computer execute the online estimation method for the state of charge of a lithium battery of an electric vehicle according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211352798.7A CN115407217B (en) | 2022-11-01 | 2022-11-01 | Online estimation method and system for state of charge of lithium battery of electric vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211352798.7A CN115407217B (en) | 2022-11-01 | 2022-11-01 | Online estimation method and system for state of charge of lithium battery of electric vehicle |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115407217A CN115407217A (en) | 2022-11-29 |
| CN115407217B true CN115407217B (en) | 2023-01-10 |
Family
ID=84169070
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211352798.7A Active CN115407217B (en) | 2022-11-01 | 2022-11-01 | Online estimation method and system for state of charge of lithium battery of electric vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115407217B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116338460B (en) * | 2023-04-11 | 2024-04-09 | 宁波禾旭汽车科技有限公司 | New energy automobile battery allowance identification system based on multi-parameter analysis |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111262896A (en) * | 2018-11-30 | 2020-06-09 | 联合汽车电子有限公司 | Network-connected automobile battery management system |
| CN111781529B (en) * | 2020-07-14 | 2023-06-27 | 上海理工大学 | Battery pack monomer capacity estimation method and device based on cloud data of electric automobile |
| KR102572652B1 (en) * | 2020-09-11 | 2023-08-31 | 삼성에스디아이 주식회사 | Method for estimating state of charge of battery |
| CN111999657B (en) * | 2020-10-29 | 2021-01-29 | 北京航空航天大学 | Method for evaluating driving mileage of lithium ion battery of electric vehicle in residual life |
| CN113109717B (en) * | 2021-03-27 | 2022-02-18 | 浙江大学 | Lithium battery state of charge estimation method based on characteristic curve optimization |
| CN113671381B (en) * | 2021-08-30 | 2022-09-06 | 武汉理工大学 | Lithium ion power battery estimation method based on time convolution network |
| CN113960476B (en) * | 2021-09-23 | 2023-10-17 | 北京理工大学 | Power battery fault monitoring method and system based on information physical fusion technology |
| CN113625175B (en) * | 2021-10-11 | 2022-03-25 | 北京理工大学深圳汽车研究院(电动车辆国家工程实验室深圳研究院) | SOC estimation method and system based on cloud big data platform |
-
2022
- 2022-11-01 CN CN202211352798.7A patent/CN115407217B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN115407217A (en) | 2022-11-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhao et al. | Elman neural network using ant colony optimization algorithm for estimating of state of charge of lithium-ion battery | |
| CN112433169B (en) | Cloud power battery health degree evaluation system and method | |
| CN115407211B (en) | Online prediction method and system for health state of lithium battery of electric vehicle | |
| CN109100655B (en) | Data processing method and device for power battery | |
| CN112068004A (en) | Method and device for determining battery abnormity and battery charging remaining time | |
| CN116826933B (en) | Knowledge-graph-based hybrid energy storage battery power supply backstepping control method and system | |
| CN115407217B (en) | Online estimation method and system for state of charge of lithium battery of electric vehicle | |
| CN116609676B (en) | Method and system for monitoring state of hybrid energy storage battery based on big data processing | |
| CN114290960B (en) | Method and device for acquiring battery health of power battery and vehicle | |
| Wu et al. | SOC prediction method based on battery pack aging and consistency deviation of thermoelectric characteristics | |
| CN115512777B (en) | Electrochemical model parameter identification method and system based on capacity change rate | |
| CN112083346B (en) | LSTM-based parallel battery pack internal current distribution estimation method and system | |
| CN112924866A (en) | Capacity conservation rate detection method, capacity conservation rate detection device, vehicle and storage medium | |
| CN115932634A (en) | Method, device, equipment and storage medium for evaluating health state of battery | |
| WO2021088216A1 (en) | Method and apparatus for detecting data on state of charge of battery pack, medium, terminal and vehicle | |
| CN117129879B (en) | Threshold adjustment method and training method of battery state of health prediction model | |
| CN113406496B (en) | Battery capacity prediction method, system, device and medium based on model migration | |
| CN117054892B (en) | Evaluation method, device and management method for battery state of energy storage power station | |
| WO2024060539A1 (en) | Annual carbon emission amount estimation method and device for power battery | |
| CN117074955A (en) | Cloud-end correction OCV-based lithium battery state joint estimation method | |
| CN116298927B (en) | Battery capacity estimation method and device, computer equipment and storage medium | |
| CN116224127A (en) | Battery state of health estimation method based on big data analysis | |
| CN113625183B (en) | Battery pack life prediction method and battery pack simulation system | |
| CN115759434A (en) | Method, system, computer device and storage medium for predicting battery capacity in real time | |
| CN112213644A (en) | Battery state of charge estimation method and battery management system |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |