CN114660468A - Retired battery state of charge prediction method and device and electronic equipment - Google Patents
Retired battery state of charge prediction method and device and electronic equipment Download PDFInfo
- Publication number
- CN114660468A CN114660468A CN202210318734.9A CN202210318734A CN114660468A CN 114660468 A CN114660468 A CN 114660468A CN 202210318734 A CN202210318734 A CN 202210318734A CN 114660468 A CN114660468 A CN 114660468A
- Authority
- CN
- China
- Prior art keywords
- neural network
- retired battery
- network model
- characteristic parameters
- charge
- 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]
- G01R31/385—Arrangements for measuring battery or accumulator variables
- G01R31/387—Determining ampere-hour charge capacity or SoC
Abstract
The invention relates to a retired battery state of charge prediction method, a retired battery state of charge prediction device and electronic equipment, wherein the retired battery state of charge prediction method comprises the following steps: acquiring data of a plurality of retired batteries in different charge states, performing curve fitting by using the data, and obtaining characteristic parameters according to curve fitting results; constructing a neural network model, and training the neural network model according to the characteristic parameters to obtain a neural network model with complete training; and acquiring real-time characteristic parameters of the retired battery to be predicted, and acquiring the charge state of the retired battery to be predicted according to the real-time characteristic parameters and the neural network model with complete training. The method for predicting the charge state of the retired battery improves the accuracy of predicting the charge state of the retired battery.
Description
Technical Field
The invention relates to the technical field of retired batteries, in particular to a retired battery state of charge prediction method and device, electronic equipment and a computer readable storage medium.
Background
Batteries are widely applied to the fields of communication, power systems, transportation and the like as energy storage power sources, the requirements of the fields for the batteries are different, so that a large number of retired batteries can be generated, if the retired batteries cannot be recycled for two times or even multiple times, a large amount of energy can be wasted, and the charge state of the retired batteries is the basis of recycling and is used for judging whether recycling is proper or not.
The single individual detection aiming at the state of charge of the retired battery is unrealistic, and the workload is too large, so the existing common methods for predicting the state of charge of the retired battery are an ampere-hour integration method and the like, but the ampere-hour integration method is easily influenced by the current measurement precision and has low precision.
Disclosure of Invention
In view of the above, it is desirable to provide a method, an apparatus, an electronic device and a computer readable storage medium for predicting a state of charge of a retired battery, so as to solve the problem of inaccurate prediction of the state of charge of the retired battery in the prior art.
In order to solve the above problems, the present invention provides a retired battery state of charge prediction method, which includes:
acquiring data of a plurality of retired batteries in different charge states, performing curve fitting by using the data, and obtaining characteristic parameters according to curve fitting results;
constructing a neural network model, and training the neural network model according to the characteristic parameters to obtain a neural network model with complete training;
and acquiring real-time characteristic parameters of the retired battery to be predicted, and acquiring the state of charge of the retired battery to be predicted according to the real-time characteristic parameters and the neural network model which is trained completely.
Further, using the data for curve fitting, comprises:
the data was used for Nyquist curve fitting.
Further, obtaining the characteristic parameters according to the curve fitting result includes:
and obtaining initial characteristic parameters of a curve fitting result, and screening the initial characteristic parameters to obtain screened characteristic parameters.
Further, the screening of the initial characteristic parameters includes:
acquiring a corresponding retired battery state-of-charge value, and calculating the correlation between the initial characteristic parameter and the corresponding retired battery state-of-charge value by utilizing a Pearson correlation coefficient;
and when the absolute value of the correlation degree is smaller than a set threshold value, rejecting the initial characteristic parameters.
Further, constructing a neural network model, comprising:
and constructing a DBN-DNN neural network model.
Further, the DBN-DNN neural network model includes 1 input layer, 3 hidden layers and 1 output layer, the DBN-DNN neural network model is composed of 3 restricted boltzmann machine units, the restricted boltzmann machine units include two layers, the upper layer is a hidden layer, the lower layer is a display layer, and the hidden layer, i.e., the output layer, of the previous restricted boltzmann machine unit is used as the display layer, i.e., the input layer, of the next restricted boltzmann machine unit.
Further, training the neural network model according to the characteristic parameters comprises:
and taking the characteristic parameters as input of an input layer, and taking the corresponding retired battery state of charge value as output of an output layer to train the neural network model.
The invention also provides a retired battery state of charge prediction device, which comprises a data acquisition module, a model training module and a state prediction module;
the data acquisition module is used for acquiring data of a plurality of retired batteries in different charge states, performing curve fitting by using the data, and obtaining characteristic parameters according to curve fitting results;
the model training module is used for constructing a neural network model and training the neural network model according to the characteristic parameters to obtain a neural network model with complete training;
the state prediction module is used for acquiring real-time characteristic parameters of the retired battery to be predicted and obtaining the charge state of the retired battery to be predicted according to the real-time characteristic parameters and the neural network model which is trained completely.
The invention further provides an electronic device, which comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the retired battery state of charge prediction method according to any one of the above technical schemes is realized.
The invention also provides a computer readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the method for predicting the state of charge of the retired battery according to any one of the above technical solutions is implemented.
The beneficial effects of adopting the above embodiment are: according to the retired battery state of charge prediction method provided by the invention, curve fitting is carried out on data of the retired battery in different states of charge, characteristic parameters are obtained according to a curve fitting result, the constructed neural network model is trained by using the characteristic parameters to obtain a completely trained neural network model, and the accuracy of the retired battery state of charge prediction is improved by predicting the retired battery state of charge by using the completely trained neural network model.
Drawings
Fig. 1 is a schematic flowchart of an embodiment of a retired battery soc prediction method according to the present invention;
FIG. 2 is a schematic diagram of a DBN-DNN neural network model provided in an embodiment of the present invention;
fig. 3 is a block diagram illustrating an exemplary embodiment of a retired battery soc prediction apparatus according to the present invention;
fig. 4 is a block diagram of an embodiment of an electronic device according to the present invention.
Detailed Description
The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate preferred embodiments of the invention and together with the description, serve to explain the principles of the invention and not to limit the scope of the invention.
The invention provides a retired battery state of charge prediction method, a retired battery state of charge prediction device, electronic equipment and a computer readable storage medium, which are respectively described in detail below.
The embodiment of the invention provides a retired battery state of charge prediction method, which is a flow diagram, and as shown in fig. 1, the retired battery state of charge prediction method comprises the following steps:
s101, acquiring data of a plurality of retired batteries in different charge states, performing curve fitting by using the data, and obtaining characteristic parameters according to curve fitting results;
s102, constructing a neural network model, and training the neural network model according to the characteristic parameters to obtain a neural network model with complete training;
step S103, acquiring real-time characteristic parameters of the retired battery to be predicted, and obtaining the state of charge of the retired battery to be predicted according to the real-time characteristic parameters and the neural network model which is completely trained.
In a specific embodiment, step S101 includes selecting a plurality of retired batteries, adjusting the plurality of retired batteries to different SOC states at 0.05-1C, standing for a sufficient time to measure open-circuit voltages U of the plurality of retired batteries, measuring ac impedances of the retired batteries at different SOCs with an impedance tester at a temperature of (25 ± 2) ° C and a frequency of 0.05-100 KHz, importing measured impedance spectrum data into impedance spectrum analysis software, selecting a suitable equivalent circuit model to perform Nyquist curve fitting, and taking values of components and open-circuit voltages U in the fitted equivalent circuit as initial characteristic parameters.
As a preferred embodiment, using the data for curve fitting includes:
the data was used for Nyquist curve fitting.
As a preferred embodiment, the obtaining the characteristic parameters according to the curve fitting result includes:
and obtaining initial characteristic parameters of a curve fitting result, and screening the initial characteristic parameters to obtain screened characteristic parameters.
It should be noted that the workload can be greatly reduced by screening the initial characteristic parameters, and the screened characteristic parameters are more accurate, so that the state of charge of the retired battery is more accurate.
As a preferred embodiment, the screening of the initial characteristic parameters includes:
acquiring a corresponding state of charge value of the retired battery, and calculating the correlation degree of the initial characteristic parameter and the corresponding state of charge value of the retired battery by utilizing a Pearson correlation coefficient;
and when the absolute value of the correlation degree is smaller than a set threshold value, rejecting the initial characteristic parameters.
In a specific embodiment, the Pearson correlation coefficient is calculated byWherein Cov (X, Y) is covariance of initial characteristic parameter X and corresponding retired battery state of charge value Y, Var [ X [ ]]Variance of X, Var [ Y ]]The variance of Y, r is the correlation value, and the setting of the threshold is defined as the case may be, and is typically set to 98%.
As a preferred embodiment, constructing a neural network model includes:
and constructing a DBN-DNN neural network model.
In a specific embodiment, as shown in fig. 2, it can be seen from fig. 2 that the DBN-DNN neural network model includes a 1-layer input layer, a 3-layer hidden layer, and a 1-layer output layer, which includes 3 finite boltzmann machine units RBM1, RBM2, and RBM3, and forms a DBN structure, and the DBN-DNN neural network model is formed by taking an output of a DBN as an input of a BP neural network.
As a preferred embodiment, the DBN-DNN neural network model includes a 1-layer input layer, a 3-layer hidden layer and a 1-layer output layer, and the DBN-DNN neural network model is composed of 3 restricted boltzmann machine units, each restricted boltzmann machine unit includes two layers, an upper layer is a hidden layer, a lower layer is a display layer, and the hidden layer, i.e., the output layer, of the previous restricted boltzmann machine unit serves as the display layer, i.e., the input layer, of the next restricted boltzmann machine unit.
It should be noted that the DBN-DNN neural network model constructed in the present invention is a DNN structure constructed by using a deep belief network, and the limited boltzmann machine unit is pre-trained by greedy layer by layer, so that the training efficiency is greatly improved, the problem of local optimization is improved, the solving process is relatively simple, and the application is convenient.
As a preferred embodiment, training the neural network model according to the feature parameters includes:
and taking the characteristic parameters as input of an input layer, and taking the corresponding state of charge value of the retired battery as output of an output layer to train the neural network model.
The embodiment of the invention also provides a retired battery state of charge prediction device, which has a structural block diagram, as shown in fig. 3, and comprises a data acquisition module 301, a model training module 302 and a state prediction module 303;
the data acquisition module 301 is configured to acquire data of a plurality of retired batteries in different charge states, perform curve fitting using the data, and obtain characteristic parameters according to a curve fitting result;
the model training module 302 is configured to construct a neural network model, and train the neural network model according to the characteristic parameters to obtain a neural network model with complete training;
the state prediction module 303 is configured to obtain a real-time characteristic parameter of the retired battery to be predicted, and obtain a state of charge of the retired battery to be predicted according to the real-time characteristic parameter and the neural network model which is trained completely.
As shown in fig. 4, the invention further provides an electronic device, which may be a mobile terminal, a desktop computer, a notebook, a palm computer, a server, or other computing devices. The electronic device comprises a processor 403, a display 402 and a memory 401.
The storage 401 may be an internal storage unit of the computer device in some embodiments, such as a hard disk or a memory of the computer device. The memory 401 may also be an external storage device of the computer device in other embodiments, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), etc. provided on the computer device. Further, the memory 401 may also include both an internal storage unit and an external storage device of the computer device. The memory 401 is used for storing application software installed in the computer device and various data, such as program codes for installing the computer device. The memory 401 may also be used to temporarily store data that has been output or is to be output. In an embodiment, the memory 401 stores a retired battery soc prediction program 404, and the retired battery soc prediction program 404 may be executed by the processor 403, thereby implementing the retired battery soc prediction method according to the embodiments of the present invention.
The display 402 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch panel, or the like in some embodiments. The display 402 is used to display information at the computer device and to display a visual user interface. The components 401 and 403 of the computer device communicate with each other via a system bus.
In one embodiment, the following steps are implemented when processor 403 executes retired battery state of charge prediction program 404 in memory 401:
acquiring data of a plurality of retired batteries in different charge states, performing curve fitting by using the data, and obtaining characteristic parameters according to curve fitting results;
constructing a neural network model, and training the neural network model according to the characteristic parameters to obtain a neural network model with complete training;
and acquiring real-time characteristic parameters of the retired battery to be predicted, and acquiring the charge state of the retired battery to be predicted according to the real-time characteristic parameters and the neural network model with complete training.
The present embodiment also provides a computer readable storage medium having a retired battery state of charge prediction program stored thereon, which when executed by a processor, performs the following steps:
acquiring data of a plurality of retired batteries in different charge states, performing curve fitting by using the data, and obtaining characteristic parameters according to curve fitting results;
constructing a neural network model, and training the neural network model according to the characteristic parameters to obtain a neural network model with complete training;
and acquiring real-time characteristic parameters of the retired battery to be predicted, and acquiring the charge state of the retired battery to be predicted according to the real-time characteristic parameters and the neural network model with complete training.
According to the retired battery state of charge prediction method, the retired battery state of charge prediction device, the electronic equipment and the computer readable storage medium, curve fitting is conducted on data of the retired battery under different states of charge, the characteristic parameters are obtained according to curve fitting results, the built neural network model is trained through the characteristic parameters, the well-trained neural network model is obtained, and the accuracy of the retired battery state of charge prediction is improved by predicting the retired battery state of charge through the well-trained neural network model.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention.
Claims (10)
1. A retired battery state of charge prediction method is characterized by comprising the following steps:
acquiring data of a plurality of retired batteries in different charge states, performing curve fitting by using the data, and obtaining characteristic parameters according to curve fitting results;
constructing a neural network model, and training the neural network model according to the characteristic parameters to obtain a neural network model with complete training;
and acquiring real-time characteristic parameters of the retired battery to be predicted, and acquiring the state of charge of the retired battery to be predicted according to the real-time characteristic parameters and the neural network model which is trained completely.
2. The retired battery state of charge prediction method of claim 1, wherein using the data for curve fitting comprises:
the data was used for Nyquist curve fitting.
3. The retired battery state of charge prediction method of claim 1, wherein obtaining the characteristic parameters from the curve fitting results comprises:
and obtaining initial characteristic parameters of a curve fitting result, and screening the initial characteristic parameters to obtain screened characteristic parameters.
4. The retired battery state of charge prediction method of claim 3, wherein the screening of the initial characteristic parameters comprises:
acquiring a corresponding state of charge value of the retired battery, and calculating the correlation degree of the initial characteristic parameter and the corresponding state of charge value of the retired battery by utilizing a Pearson correlation coefficient;
and when the absolute value of the correlation degree is smaller than a set threshold value, rejecting the initial characteristic parameters.
5. The retired battery state of charge prediction method of claim 1, wherein constructing a neural network model comprises:
and constructing a DBN-DNN neural network model.
6. The retired battery state-of-charge prediction method of claim 5, wherein the DBN-DNN neural network model comprises 1 input layer, 3 hidden layers and 1 output layer, the DBN-DNN neural network model is composed of 3 restricted Boltzmann machine units, the restricted Boltzmann machine units comprise two layers, an upper layer is a hidden layer, a lower layer is a display layer, and a hidden layer or an output layer of a previous restricted Boltzmann machine unit is used as a display layer or an input layer of a next restricted Boltzmann machine unit.
7. The retired battery state of charge prediction method of claim 4, wherein training the neural network model according to the feature parameters comprises:
and taking the characteristic parameters as input of an input layer, and taking the corresponding state of charge value of the retired battery as output of an output layer to train the neural network model.
8. The retired battery state of charge prediction device is characterized by comprising a data acquisition module, a model training module and a state prediction module;
the data acquisition module is used for acquiring data of a plurality of retired batteries in different charge states, performing curve fitting by using the data, and obtaining characteristic parameters according to curve fitting results;
the model training module is used for constructing a neural network model and training the neural network model according to the characteristic parameters to obtain a neural network model with complete training;
the state prediction module is used for acquiring real-time characteristic parameters of the retired battery to be predicted and obtaining the charge state of the retired battery to be predicted according to the real-time characteristic parameters and the neural network model which is trained completely.
9. An electronic device comprising a memory and a processor, the memory having stored thereon a computer program that, when executed by the processor, implements a retired battery state of charge prediction method according to any of claims 1-7.
10. A computer-readable storage medium, having stored thereon a computer program which, when executed by a processor, implements a method of retired battery state of charge prediction according to any of claims 1-7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210318734.9A CN114660468A (en) | 2022-03-29 | 2022-03-29 | Retired battery state of charge prediction method and device and electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210318734.9A CN114660468A (en) | 2022-03-29 | 2022-03-29 | Retired battery state of charge prediction method and device and electronic equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114660468A true CN114660468A (en) | 2022-06-24 |
Family
ID=82032893
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210318734.9A Pending CN114660468A (en) | 2022-03-29 | 2022-03-29 | Retired battery state of charge prediction method and device and electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114660468A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115656834A (en) * | 2022-11-02 | 2023-01-31 | 武汉动力电池再生技术有限公司 | Battery capacity prediction method and device and electronic equipment |
-
2022
- 2022-03-29 CN CN202210318734.9A patent/CN114660468A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115656834A (en) * | 2022-11-02 | 2023-01-31 | 武汉动力电池再生技术有限公司 | Battery capacity prediction method and device and electronic equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107506868B (en) | Method and device for predicting short-time power load | |
WO2021208079A1 (en) | Method and apparatus for obtaining power battery life data, computer device, and medium | |
EP4113139A1 (en) | Lithium battery soc estimation method and apparatus, and computer-readable storage medium | |
WO2021004318A1 (en) | Resource data processing method and apparatus, computer device and storage medium | |
US20230385597A1 (en) | Multi-granularity perception integrated learning method, device, computer equipment and medium | |
CN107015900A (en) | A kind of service performance Forecasting Methodology of video website | |
CN112949721A (en) | Power equipment static data quality assessment method and system | |
CN114563716A (en) | Retired battery SOC estimation method and device based on electrochemical impedance spectrum | |
CN110990135B (en) | Spark job time prediction method and device based on deep migration learning | |
CN114660468A (en) | Retired battery state of charge prediction method and device and electronic equipment | |
WO2023050534A1 (en) | Method and device for predicting energy consumption of rail transit station apparatus, and apparatus and storage medium | |
CN112257958A (en) | Power saturation load prediction method and device | |
CN101206727A (en) | Data processing apparatus, data processing method data processing program and computer readable medium | |
CN111914101A (en) | Abnormal identification method and device for file association relationship and computer equipment | |
CN115656834A (en) | Battery capacity prediction method and device and electronic equipment | |
CN116125279A (en) | Method, device, equipment and storage medium for determining battery health state | |
Wu et al. | Short-term power load forecasting using least squares support vector machines (LS-SVM) | |
CN113095963A (en) | Real estate cost data processing method, real estate cost data processing device, computer equipment and storage medium | |
CN113468421A (en) | Product recommendation method, device, equipment and medium based on vector matching technology | |
CN116936103B (en) | User health prediction management method and system based on homodromous network | |
CN111414398B (en) | Data analysis model determining method, device and storage medium | |
CN114563717A (en) | Battery capacity sorting method and device, electronic equipment and storage medium | |
CN115469136A (en) | Non-invasive three-phase voltage measuring method based on electric field sensor array | |
CN117031295A (en) | Battery SOC estimation method, device and medium | |
CN117654928A (en) | Retired battery sorting method, device and equipment based on time distance sequence |
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 |