CN109283469A - Battery management system failure prediction method, device and readable storage medium storing program for executing - Google Patents
Battery management system failure prediction method, device and readable storage medium storing program for executing Download PDFInfo
- Publication number
- CN109283469A CN109283469A CN201811109007.1A CN201811109007A CN109283469A CN 109283469 A CN109283469 A CN 109283469A CN 201811109007 A CN201811109007 A CN 201811109007A CN 109283469 A CN109283469 A CN 109283469A
- Authority
- CN
- China
- Prior art keywords
- layer
- management system
- node
- hidden layer
- output
- 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
Landscapes
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
Abstract
The invention discloses a kind of battery management system failure prediction method, device and readable storage medium storing program for executing.A kind of battery management system failure prediction method, include the following steps: the operating parameter that the battery pack different times of running are extracted from batteries management system, wherein, the operating parameter includes total voltage, monomer voltage, electric current, temperature and the vibration signal of battery pack;The operating parameter is subjected to feature extraction and normalization according to the output demand of neural network model to form the input quantity of neural network model;The input quantity is inputted into neural network, the neural network judges whether the operating status of batteries management system breaks down and export judging result;In conjunction with the characteristics of battery management system failure and the advantage of neural network, failure symptom is more accurate with fault signature to be predicted to battery management system failure using failure prediction method neural network based, can achieve better effect.
Description
Technical field
The present invention relates to battery management system field more particularly to a kind of battery management system failure prediction methods, device
And readable storage medium storing program for executing.
Background technique
Battery management system (Battery Management System) is the tie between battery and user, the system
Main application is secondary cell.Battery management system (hereinafter referred to as BMS system) is mainly for present in individual cells
Energy storage is few, the service life is short, the problems such as estimating is difficult to using completeness and battery capacity provides comprehensive solution, purpose
It is the real-time precise information of battery correlation performance parameters in order to obtain and it is monitored, reaches the mesh for improving battery utilization rate
's.BMS system mainly includes three functions: be first accurate estimation battery pack state-of-charge (State of Charge, with
Lower abbreviation SOC), i.e., the remaining capacity of battery, guarantee SOC value maintain in the reasonable scope, to prevent from overcharging and make battery with over-discharge
At damage.SOC value is indicated with the ratio of remaining capacity and fully charged state electricity.Followed by the charge and discharge process of battery
Population parameter (including every piece of battery terminal voltage, temperature, electric current, total voltage etc.) dynamic monitoring is carried out to battery pack, establishes each piece
The historical archives of battery and battery pack;It is finally to make each battery in battery pack all in balanced consistent using balancing technique
State, to reach the maximized purpose of battery utilization rate.Wherein accurately estimation can be carried out to SOC to determine to battery dynamic
The correct evaluation capacity of parameter, the behavior directly determine the accuracy of the important indicators such as over-charging of battery, over-discharge, harmony, are BMS
Can system effectively improve the most important factor of battery utilization rate.
With the development of the world today, the application of new energy especially new energy battery worldwide becomes one
A hot issue, effect of the battery management system (BMS) in new energy battery applications is also outstanding day by day, and battery management system can
To carry out charge and discharge control, heat management, balanced management to battery pack, the state of battery can also be fed back into user, in new energy
Irreplaceable role is played in source battery application.But since new energy battery pack has a large amount of series-parallel use, in system
Various failures can occur in operational process influences system normal operation, and battery management system failure predication can be prejudged with help system
Whether will appear all kinds of failures in operation, provide a kind of operational support mechanism for battery management system, assists user's prevention
The system failure being likely to occur.
Neural network is formed by connecting by the processing unit largely interconnected, it is based on modern neuro biology and cognition
The research achievement that science is applied in field of information processing.It has large-scale parallel simulation process, continuous time dynamics and net
The features such as network overall situation acts on, has very strong adaptive learning and nonlinear fitting ability, makes signal processing closer to the mankind
Thinking activities.The fault signature of battery management system has similitude, has stronger mould between failure symptom and fault signature
Paste property, fault signature is interweaved.
Summary of the invention
Purpose of this disclosure is to provide a kind of battery management system failure prediction method, device and readable storage medium storing program for executing, knots
The advantage of the characteristics of closing battery management system failure and neural network, using failure prediction method neural network based to electricity
Pond management system failure is predicted that failure symptom is more accurate with fault signature, can achieve better effect.
In order to achieve the above objectives, the embodiment of the present disclosure in a first aspect, providing a kind of battery management system failure predication side
Method includes the following steps:
The operating parameter of the battery pack different times of running is extracted from batteries management system, wherein the operating parameter
Total voltage, monomer voltage, electric current, temperature and vibration signal including battery pack;
The operating parameter is subjected to feature extraction and normalization according to the output demand of neural network model to be formed
The input quantity of neural network model;
By the input quantity input neural network, the neural network judge batteries management system operating status whether
It breaks down and exports judging result;
Wherein, the learning process of the neural network includes the following steps:
Build artificial neural network
Artificial neural network is built using single hidden layer mode, the artificial neural network is by input layer, hidden layer and output
Layer is constituted, and input layer is made of several input nodes, and the quantity of input node is equal to the number for needing acquisition parameter, if hidden layer by
Dry implicit node is constituted, and output layer is only made of two output nodes;
Using total voltage, monomer voltage, electric current, temperature and the vibration signal of the different times of running of battery pack as input layer
Input node;
The corresponding weight of each node, the hidden layer threshold value of input layer is randomly generated, obtains input layer reflecting to hidden layer
Penetrate rule;
The slave input layer mapping received according to each node that the mapping ruler of input layer to hidden layer extrapolates hidden layer
The information come;
It is used as output layer output node whether by battery management system failure, each node that hidden layer is randomly generated is corresponding
Weight, battery management system quasi- operating status break down threshold values obtain hidden layer to output layer mapping ruler;
According to the mapping ruler of hidden layer to output layer, the map information that output layer output node receives is extrapolated;
Carry out application condition
Above-mentioned output result is compared with the standard operating status of battery management system, judges whether comparison result is small
In allowing worst error, when comparison result, which is greater than, allows worst error, calibrates and each node for updating hidden layer is corresponding
Weight, pond management system quasi- operating status break down threshold values, calibrate and update the corresponding power of each node of input layer
Value, hidden layer threshold value, re-execute each node that the mapping ruler according to input layer to hidden layer extrapolates hidden layer
The information and subsequent step that the slave input layer mapping received comes;
If preservation finally enters the corresponding power of each node of layer input layer into hidden layer allowing within the scope of worst error
The corresponding weight of each node and output layer threshold value of value, hidden layer hidden layer into output layer, obtain battery management system
The mapping ruler of quasi- operating status and battery pack dynamic parameter;
Study terminates.
The second aspect of the embodiment of the present disclosure provides a kind of battery management system fault prediction device, comprising:
Data acquisition module is configured as extracting the operation ginseng of the battery pack different times of running from batteries management system
Number, wherein the operating parameter includes total voltage, monomer voltage, electric current, temperature and the vibration signal of battery pack;
Characteristic extracting module is configured as the operating parameter carrying out feature according to the output demand of neural network model
It extracts and normalizes to form the input quantity of neural network model;And
Neural network module, is configured as receiving and judges battery set management system from the input quantity of the characteristic extracting module
Whether the operating status of system breaks down and exports judging result;
Wherein the neural network module executes following learning process:
Build artificial neural network
Artificial neural network is built using single hidden layer mode, the artificial neural network is by input layer, hidden layer and output
Layer is constituted, and input layer is made of several input nodes, and the quantity of input node is equal to the number for needing acquisition parameter, if hidden layer by
Dry implicit node is constituted, and output layer is only made of two output nodes;
Using total voltage, monomer voltage, electric current, temperature and the vibration signal of the different times of running of battery pack as input layer
Input node;
The corresponding weight of each node, the hidden layer threshold value of input layer is randomly generated, obtains input layer reflecting to hidden layer
Penetrate rule;
The slave input layer mapping received according to each node that the mapping ruler of input layer to hidden layer extrapolates hidden layer
The information come;
It is used as output layer output node whether by battery management system failure, each node that hidden layer is randomly generated is corresponding
Weight, battery management system quasi- operating status break down threshold values obtain hidden layer to output layer mapping ruler;
According to the mapping ruler of hidden layer to output layer, the map information that output layer output node receives is extrapolated;
Carry out application condition
Above-mentioned output result is compared with the standard operating status of battery management system, judges whether comparison result is small
In allowing worst error, when comparison result, which is greater than, allows worst error, calibrates and each node for updating hidden layer is corresponding
Weight, pond management system quasi- operating status break down threshold values, calibrate and update the corresponding power of each node of input layer
Value, hidden layer threshold value, re-execute each node that the mapping ruler according to input layer to hidden layer extrapolates hidden layer
The information and subsequent step that the slave input layer mapping received comes;
If preservation finally enters the corresponding power of each node of layer input layer into hidden layer allowing within the scope of worst error
The corresponding weight of each node and output layer threshold value of value, hidden layer hidden layer into output layer, obtain battery management system
The mapping ruler of quasi- operating status and battery pack dynamic parameter;
Study terminates.
Optionally, pass through CAN bus, Zigbee transmission line between characteristic extracting module described in the data acquisition module
Or wifi transmits data.
The third aspect of the embodiment of the present disclosure provides a kind of computer readable storage medium, is stored thereon with computer journey
The step of sequence, which realizes any one of above-mentioned first aspect the method when being executed by processor.
The third aspect of the embodiment of the present disclosure provides a kind of device, comprising:
Computer readable storage medium described in the above-mentioned third aspect;And
One or more processor, for executing the program in the computer readable storage medium.
By adopting the above technical scheme, following technical effect can at least be reached:
It is pre- using failure neural network based in conjunction with the characteristics of battery management system failure and the advantage of neural network
Survey method predicts that failure symptom is more accurate with fault signature to battery management system failure, can achieve more preferable characterization
Whether effect, forecasting system will break down.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is neural network topological diagram;
Fig. 2 is the schematic diagram of one embodiment of the embodiment of the present invention;
Fig. 3 is the schematic diagram of another embodiment of the embodiment of the present invention.
Specific embodiment
In the following detailed description, many details are proposed, in order to complete understanding of the present invention.But
It will be apparent to those skilled in the art that bright can be the case where not needing some details in these details
Lower implementation.Below to the description of embodiment just for the sake of provided by showing example of the invention to it is of the invention preferably
Understand.
Below in conjunction with attached drawing, the technical solution of the embodiment of the present invention is described.
Disclosure implementation provides a kind of battery management system failure prediction method, includes the following steps:
The operating parameter of the battery pack different time of running is extracted from batteries management system, wherein operating parameter includes
Total voltage, monomer voltage, electric current, temperature and the vibration signal of battery pack;
Operating parameter is subjected to feature extraction and normalization according to the output demand of neural network model to form nerve
The input quantity of network model;
Input quantity is inputted into neural network, neural network judges whether the operating status of batteries management system breaks down
And export judging result;
Wherein, the learning process of neural network includes the following steps:
Build artificial neural network
Artificial neural network is built using single hidden layer mode, the artificial neural network is by input layer, hidden layer and output
Layer is constituted, and input layer is made of several input nodes, and the quantity of input node is equal to the number for needing acquisition parameter, if hidden layer by
Dry implicit node is constituted, and output layer is only made of two output nodes;
Using total voltage, monomer voltage, electric current, temperature and the vibration signal of the different times of running of battery pack as input layer
Input node;
The corresponding weight of each node, the hidden layer threshold value of input layer is randomly generated, obtains input layer reflecting to hidden layer
Penetrate rule;
The slave input layer mapping received according to each node that the mapping ruler of input layer to hidden layer extrapolates hidden layer
The information come;
It is used as output layer output node whether by battery management system failure, each node that hidden layer is randomly generated is corresponding
Weight, battery management system quasi- operating status break down threshold values obtain hidden layer to output layer mapping ruler;
According to the mapping ruler of hidden layer to output layer, the map information that output layer output node receives is extrapolated;
Carry out application condition
Above-mentioned output result is compared with the standard operating status of battery management system, judges whether comparison result is small
In allowing worst error, when comparison result, which is greater than, allows worst error, calibrates and each node for updating hidden layer is corresponding
Weight, pond management system quasi- operating status break down threshold values, calibrate and update the corresponding power of each node of input layer
Value, hidden layer threshold value, re-execute and are received according to each node that the mapping ruler of input layer to hidden layer extrapolates hidden layer
Slave input layer mapping come information and subsequent step;
If preservation finally enters the corresponding power of each node of layer input layer into hidden layer allowing within the scope of worst error
The corresponding weight of each node and output layer threshold value of value, hidden layer hidden layer into output layer, obtain battery management system
The mapping ruler of quasi- operating status and battery pack dynamic parameter;
Study terminates.
Wherein the algorithm of neural network is as follows:
Neural network is a kind of typical multilayer neural network, network Basic Topological as shown in Figure 1: it include three
Layer, is input layer, hidden layer and output layer respectively, hidden layer can be made of multistage, and the neuron on every layer is known as node or list
Member, they are interconnected by amendable weight.Algorithm is made of the positive transmitting of information and backpropagation two parts of error.Just
Into communication process, input information, which is successively calculated from input layer through hidden layer, is transmitted to output layer, and the state of each layer of neuron is only
Under the influence of one layer of neuron state.If not obtaining desired output in output layer, the error change of output layer is calculated
Value, then turns to backpropagation, error signal is modified each layer neuron along original connecting path anti-pass back by network
Weight until reach expectation target.Before using neural network prediction failure, need to be trained it, it is assumed that input layer
Node i (i=1,2 ... input value T n)iEqual to output valve Xi, output valve is passed to hidden layer;For hidden node j (j=1,
2 ... input value I p)jWith output valve OjIt can be calculated respectively by following:
Wherein, ωjiFor the weight between hidden node j and input layer i, θjFor the biasing of node j, f sigmoid
Function, expression formula are as follows:If output node layer k (k=1,2 ... input and output m) are IkWith
yk, calculation formula is respectively as follows:
Wherein, ωkjFor the weight between output node layer k and hiding node layer j, θkFor the biasing of node k.
For given training sample (xp1,xp2,...xpn), p is number of samples p=(1,2 ... P), then neural network
Mean square error between trained and training objective output valve can be written as:
Wherein, p is sample number, tplTarget for first of output unit of p-th of sample exports as a result, yplIt is p-th
The network operations result of first of output unit of sample.The process of network training includes the forward calculation and error of network internal
Backpropagation, purpose is exactly to keep network output error minimum by adjusting network internal connection weight.For multilayer feedforward
Connection weight is adjusted using algorithm between input layer and hidden layer, between hidden layer and output layer in network.
The training sample that the present invention uses comes from standard knowledge well known in the art, when system known to one group is run
Quarter, electric current, monomer voltage, total voltage, temperature and vibration signal are defeated as training sample with corresponding relationship whether breaking down
Enter, according to the method described above with algorithm training neural network, the neural network after the completion of training be can be used to battery management system
Carry out failure predication.In order to preferably illustrate training and use of the present invention to neural network, such as training sample shown in following table
This:
Serial number | Moment | Electric current | Monomer voltage | Total voltage | Temperature | Vibration | Consequent malfunction |
1 | 11:01 | 0.0025 | 0.0030 | 0.0315 | 0.0008 | 0.0000 | It is no |
2 | 11:30 | 0.0021 | 0.0027 | 0.0330 | 0.0000 | 0.0005 | It is |
3 | 12:15 | 0.0018 | 0.0027 | 0.0326 | 0.0009 | 0.0001 | It is no |
4 | 12:25 | 0.0021 | 0.0032 | 0.0318 | 0.0010 | 0.0002 | It is no |
5 | 13:00 | 0.0020 | 0.0035 | 0.0322 | 0.0015 | 0.0001 | It is |
... | ... | ... | ... | ... | ... | ... | ... |
50 | 20:07 | 0.0031 | 0.0025 | 0.0325 | 0.0009 | 0.0000 | It is |
After the completion of training, using the identical sample of another group of dimension as test sample, the training quality of neural network is examined.
Neural network after trained is tested using such as following table test sample:
As shown in Fig. 2, the embodiment of the present disclosure also provides a kind of battery management system fault prediction device, comprising:
Data acquisition module is configured as extracting the operation ginseng of the battery pack different times of running from batteries management system
Number, wherein operating parameter includes total voltage, monomer voltage, electric current, temperature and the vibration signal of battery pack;
Characteristic extracting module is configured as operating parameter carrying out feature extraction according to the output demand of neural network model
And it normalizes to form the input quantity of neural network model;And
Neural network module, is configured as receiving the input quantity from characteristic extracting module and judges batteries management system
Whether operating status breaks down and exports judging result;
Wherein neural network module executes following learning process:
Build artificial neural network
Artificial neural network is built using single hidden layer mode, the artificial neural network is by input layer, hidden layer and output
Layer is constituted, and input layer is made of several input nodes, and the quantity of input node is equal to the number for needing acquisition parameter, if hidden layer by
Dry implicit node is constituted, and output layer is only made of two output nodes;
Using total voltage, monomer voltage, electric current, temperature and the vibration signal of the different times of running of battery pack as input layer
Input node;
The corresponding weight of each node, the hidden layer threshold value of input layer is randomly generated, obtains input layer reflecting to hidden layer
Penetrate rule;
The slave input layer mapping received according to each node that the mapping ruler of input layer to hidden layer extrapolates hidden layer
The information come;
It is used as output layer output node whether by battery management system failure, each node that hidden layer is randomly generated is corresponding
Weight, battery management system quasi- operating status break down threshold values obtain hidden layer to output layer mapping ruler;
According to the mapping ruler of hidden layer to output layer, the map information that output layer output node receives is extrapolated;
Carry out application condition
Above-mentioned output result is compared with the standard operating status of battery management system, judges whether comparison result is small
In allowing worst error, when comparison result, which is greater than, allows worst error, calibrates and each node for updating hidden layer is corresponding
Weight, pond management system quasi- operating status break down threshold values, calibrate and update the corresponding power of each node of input layer
Value, hidden layer threshold value, re-execute and are received according to each node that the mapping ruler of input layer to hidden layer extrapolates hidden layer
Slave input layer mapping come information and subsequent step;
If preservation finally enters the corresponding power of each node of layer input layer into hidden layer allowing within the scope of worst error
The corresponding weight of each node and output layer threshold value of value, hidden layer hidden layer into output layer, obtain battery management system
The mapping ruler of quasi- operating status and battery pack dynamic parameter;
Study terminates.
Wherein, pass through CAN bus, Zigbee transmission line or wifi transmission between data acquisition module characteristic extracting module
Data.
The disclosure also provides a kind of computer readable storage medium, is stored thereon with computer program, and the program is processed
A kind of method and step of battery management system failure prediction method described in any of the above-described alternative embodiment is realized when device executes.
The embodiment of the present disclosure also provides a kind of device of battery management system failure predication, comprising:
Above-mentioned computer readable storage medium;And
One or more processor, for executing the program in the computer readable storage medium.
Fig. 3 is a kind of block diagram of the device 400 of battery management system failure predication shown according to an exemplary embodiment.
As shown in figure 3, the device 400 may include: processor 401, memory 402, multimedia component 403, input/output (I/O)
Interface 404 and communication component 405.
Wherein, processor 401 is used to control the integrated operation of the device 400, to complete a kind of above-mentioned battery management system
All or part of the steps in system failure prediction method.Memory 402 is for storing various types of data to support in the dress
400 operation is set, these data for example may include the finger of any application or method for operating on the device 400
It enables.The memory 402 can realize by any kind of volatibility or non-volatile memory device or their combination, such as
Static random access memory (Static Random Access Memory, abbreviation SRAM), electrically erasable is read-only to be deposited
Reservoir (Electrically Erasable Programmable Read-Only Memory, abbreviation EEPROM), it is erasable can
Program read-only memory (Erasable Programmable Read-Only Memory, abbreviation EPROM), may be programmed read-only deposit
Reservoir (Programmable Read-Only Memory, abbreviation PROM), and read-only memory (Read-Only Memory, referred to as
ROM), magnetic memory, flash memory, disk or CD.Multimedia component 403 may include screen and audio component.Wherein
Screen for example can be touch screen, and audio component is used for output and/or input audio signal.For example, audio component may include
One microphone, microphone is for receiving external audio signal.The received audio signal can be further stored in storage
Device 402 is sent by communication component 405.Audio component further includes at least one loudspeaker, is used for output audio signal.I/O
Interface 404 provides interface between processor 401 and other interface modules, other above-mentioned interface modules can be keyboard, mouse,
Button etc..These buttons can be virtual push button or entity button.Communication component 405 is used for the device 400 and other equipment
Between carry out wired or wireless communication.Wireless communication, such as Wi-Fi, bluetooth, near-field communication (Near Field
Communication, abbreviation NFC), 2G, 3G or 4G or they one or more of combination, therefore corresponding communication
Component 405 may include: Wi-Fi module, bluetooth module, NFC module.
In one exemplary embodiment, device 400 can be by one or more application specific integrated circuit
(Application Specific Integrated Circuit, abbreviation ASIC), digital signal processor (Digital
Signal Processor, abbreviation DSP), digital signal processing appts (Digital Signal Processing Device,
Abbreviation DSPD), programmable logic device (Programmable Logic Device, abbreviation PLD), field programmable gate array
(Field Programmable Gate Array, abbreviation FPGA), controller, microcontroller, microprocessor or other electronics member
Part realization, the method for executing a kind of above-mentioned battery management system failure predication.
In a further exemplary embodiment, a kind of computer readable storage medium including program instruction, example are additionally provided
It such as include the memory 402 of program instruction, above procedure instruction can be executed above-mentioned to complete by the processor 401 of device 400
A kind of battery management system failure prediction method.
It is pre- using failure neural network based in conjunction with the characteristics of battery management system failure and the advantage of neural network
Survey method predicts that failure symptom is more accurate with fault signature to battery management system failure, can achieve more preferable characterization
Whether effect, forecasting system will break down.
The preferred embodiment of the disclosure is described in detail in conjunction with attached drawing above, still, the disclosure is not limited to above-mentioned reality
The detail in mode is applied, in the range of the technology design of the disclosure, a variety of letters can be carried out to the technical solution of the disclosure
Monotropic type, these simple variants belong to the protection scope of the disclosure.
It is further to note that specific technical features described in the above specific embodiments, in not lance
In the case where shield, it can be combined in any appropriate way.In order to avoid unnecessary repetition, the disclosure to it is various can
No further explanation will be given for the combination of energy.
In addition, any combination can also be carried out between a variety of different embodiments of the disclosure, as long as it is without prejudice to originally
Disclosed thought equally should be considered as disclosure disclosure of that.
Claims (5)
1. a kind of battery management system failure prediction method, which comprises the steps of:
The operating parameter of the battery pack different time of running is extracted from batteries management system, wherein the operating parameter includes
Total voltage, monomer voltage, electric current, temperature and the vibration signal of battery pack;
The operating parameter is subjected to feature extraction and normalization according to the output demand of neural network model to form nerve
The input quantity of network model;
The input quantity is inputted into neural network, the neural network judges whether the operating status of batteries management system occurs
Failure simultaneously exports judging result;
Wherein, the learning process of the neural network includes the following steps:
Build artificial neural network
Artificial neural network is built using single hidden layer mode, the artificial neural network is by input layer, hidden layer and output layer structure
At input layer is made of several input nodes, and the quantity of input node is equal to the number for needing acquisition parameter, and hidden layer is by several hidden
It is constituted containing node, output layer is only made of two output nodes;
Using total voltage, monomer voltage, electric current, temperature and the vibration signal of the different times of running of battery pack as the defeated of input layer
Ingress;
The corresponding weight of each node, the hidden layer threshold value of input layer is randomly generated, the mapping for obtaining input layer to hidden layer is advised
Then;
According to the slave input layer mapping that each node that the mapping ruler of input layer to hidden layer extrapolates hidden layer receives come
Information;
It is used as output layer output node whether by battery management system failure, the corresponding power of each node of hidden layer is randomly generated
The threshold values that value, the quasi- operating status of battery management system break down obtains hidden layer to the mapping ruler of output layer;
According to the mapping ruler of hidden layer to output layer, the map information that output layer output node receives is extrapolated;
Carry out application condition
Above-mentioned output result is compared with the standard operating status of battery management system, it is fair to judge whether comparison result is less than
Perhaps worst error, when comparison result be greater than allow worst error when, calibrate and update hidden layer the corresponding weight of each node,
The corresponding weight of each node, hidden of input layer is calibrated and updated to the threshold values that the quasi- operating status of pond management system breaks down
Threshold value containing layer re-executes the mapping ruler according to input layer to hidden layer and extrapolates what each node of hidden layer received
The information and subsequent step come from input layer mapping;
If allow worst error within the scope of, preservation finally enter the corresponding weight of each node of layer input layer into hidden layer,
The corresponding weight of each node and output layer threshold value of hidden layer hidden layer into output layer obtain the quasi- fortune of battery management system
The mapping ruler of row state and battery pack dynamic parameter;
Study terminates.
2. a kind of battery management system fault prediction device characterized by comprising
Data acquisition module is configured as extracting the operating parameter of the battery pack different times of running from batteries management system,
Wherein, the operating parameter includes total voltage, monomer voltage, electric current, temperature and the vibration signal of battery pack;
Characteristic extracting module is configured as the operating parameter carrying out feature extraction according to the output demand of neural network model
And it normalizes to form the input quantity of neural network model;And
Neural network module, is configured as receiving and judges batteries management system from the input quantity of the characteristic extracting module
Whether operating status breaks down and exports judging result;
Wherein the neural network module executes following learning process:
Build artificial neural network
Artificial neural network is built using single hidden layer mode, the artificial neural network is by input layer, hidden layer and output layer structure
At input layer is made of several input nodes, and the quantity of input node is equal to the number for needing acquisition parameter, and hidden layer is by several hidden
It is constituted containing node, output layer is only made of two output nodes;
Using total voltage, monomer voltage, electric current, temperature and the vibration signal of the different times of running of battery pack as the defeated of input layer
Ingress;
The corresponding weight of each node, the hidden layer threshold value of input layer is randomly generated, the mapping for obtaining input layer to hidden layer is advised
Then;
According to the slave input layer mapping that each node that the mapping ruler of input layer to hidden layer extrapolates hidden layer receives come
Information;
It is used as output layer output node whether by battery management system failure, the corresponding power of each node of hidden layer is randomly generated
The threshold values that value, the quasi- operating status of battery management system break down obtains hidden layer to the mapping ruler of output layer;
According to the mapping ruler of hidden layer to output layer, the map information that output layer output node receives is extrapolated;
Carry out application condition
Above-mentioned output result is compared with the standard operating status of battery management system, it is fair to judge whether comparison result is less than
Perhaps worst error, when comparison result be greater than allow worst error when, calibrate and update hidden layer the corresponding weight of each node,
The corresponding weight of each node, hidden of input layer is calibrated and updated to the threshold values that the quasi- operating status of pond management system breaks down
Threshold value containing layer re-executes the mapping ruler according to input layer to hidden layer and extrapolates what each node of hidden layer received
The information and subsequent step come from input layer mapping;
If allow worst error within the scope of, preservation finally enter the corresponding weight of each node of layer input layer into hidden layer,
The corresponding weight of each node and output layer threshold value of hidden layer hidden layer into output layer obtain the quasi- fortune of battery management system
The mapping ruler of row state and battery pack dynamic parameter;
Study terminates.
3. a kind of battery management system fault prediction device according to claim 2, which is characterized in that the data acquisition
Data are transmitted by CAN bus, Zigbee transmission line or wifi between characteristic extracting module described in module.
4. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is held by processor
The step of claim 1 the method is realized when row.
5. a kind of device characterized by comprising
Memory is stored thereon with computer program;And
Processor, for executing the computer program in the memory, to realize the step of claim 1 the method
Suddenly.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811109007.1A CN109283469A (en) | 2018-09-21 | 2018-09-21 | Battery management system failure prediction method, device and readable storage medium storing program for executing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811109007.1A CN109283469A (en) | 2018-09-21 | 2018-09-21 | Battery management system failure prediction method, device and readable storage medium storing program for executing |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109283469A true CN109283469A (en) | 2019-01-29 |
Family
ID=65181474
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811109007.1A Pending CN109283469A (en) | 2018-09-21 | 2018-09-21 | Battery management system failure prediction method, device and readable storage medium storing program for executing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109283469A (en) |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110163263A (en) * | 2019-04-30 | 2019-08-23 | 首钢京唐钢铁联合有限责任公司 | Fault identification method and device |
CN111142026A (en) * | 2019-12-31 | 2020-05-12 | 联想(北京)有限公司 | Data processing method and device and electronic equipment |
CN111967570A (en) * | 2019-07-01 | 2020-11-20 | 嘉兴砥脊科技有限公司 | Implementation method, device and machine equipment of mysterious neural network system |
CN112396156A (en) * | 2019-08-12 | 2021-02-23 | 美光科技公司 | Predictive maintenance of automotive batteries |
CN112710956A (en) * | 2020-12-17 | 2021-04-27 | 四川虹微技术有限公司 | Battery management system fault detection system and method based on expert system |
CN113139328A (en) * | 2020-01-17 | 2021-07-20 | 丰田自动车株式会社 | Internal combustion engine state determination device, internal combustion engine state determination system, and data analysis device |
CN113255764A (en) * | 2021-05-21 | 2021-08-13 | 池测(上海)数据科技有限公司 | Method, system and device for detecting electrochemical energy storage system fault by using machine learning |
CN113547919A (en) * | 2021-08-26 | 2021-10-26 | 武汉海亿新能源科技有限公司 | Remote fault monitoring method and system for fuel cell vehicle |
CN113848494A (en) * | 2021-09-18 | 2021-12-28 | 北京经纬恒润科技股份有限公司 | Online monitoring method for power battery temperature and vehicle-mounted T-BOX |
CN115061049A (en) * | 2022-08-08 | 2022-09-16 | 山东卓朗检测股份有限公司 | Method and system for rapidly detecting UPS battery fault of data center |
CN115332649A (en) * | 2022-07-05 | 2022-11-11 | 厦门宇电自动化科技有限公司 | Battery temperature control management method, device, equipment and readable storage medium |
CN117289145A (en) * | 2023-11-27 | 2023-12-26 | 宁德时代新能源科技股份有限公司 | Fault analysis method, data acquisition method, device, equipment, system and medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105429230A (en) * | 2015-12-22 | 2016-03-23 | 重庆西南集成电路设计有限责任公司 | Artificial neural network algorithm-based battery management system and method |
CN107422266A (en) * | 2017-03-15 | 2017-12-01 | 中国电力科学研究院 | A kind of method for diagnosing faults and device of high capacity cell energy-storage system |
US20180143257A1 (en) * | 2016-11-21 | 2018-05-24 | Battelle Energy Alliance, Llc | Systems and methods for estimation and prediction of battery health and performance |
CN108363016A (en) * | 2018-02-22 | 2018-08-03 | 上海理工大学 | Battery micro-short circuit quantitative Diagnosis method based on artificial neural network |
-
2018
- 2018-09-21 CN CN201811109007.1A patent/CN109283469A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105429230A (en) * | 2015-12-22 | 2016-03-23 | 重庆西南集成电路设计有限责任公司 | Artificial neural network algorithm-based battery management system and method |
US20180143257A1 (en) * | 2016-11-21 | 2018-05-24 | Battelle Energy Alliance, Llc | Systems and methods for estimation and prediction of battery health and performance |
CN107422266A (en) * | 2017-03-15 | 2017-12-01 | 中国电力科学研究院 | A kind of method for diagnosing faults and device of high capacity cell energy-storage system |
CN108363016A (en) * | 2018-02-22 | 2018-08-03 | 上海理工大学 | Battery micro-short circuit quantitative Diagnosis method based on artificial neural network |
Non-Patent Citations (1)
Title |
---|
古昂: "基于RBF神经网络的动力电池故障诊断***研究", 《电源技术》 * |
Cited By (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110163263A (en) * | 2019-04-30 | 2019-08-23 | 首钢京唐钢铁联合有限责任公司 | Fault identification method and device |
CN111967570A (en) * | 2019-07-01 | 2020-11-20 | 嘉兴砥脊科技有限公司 | Implementation method, device and machine equipment of mysterious neural network system |
CN111967570B (en) * | 2019-07-01 | 2024-04-05 | 北京砥脊科技有限公司 | Implementation method, device and machine equipment of visual neural network system |
CN112396156A (en) * | 2019-08-12 | 2021-02-23 | 美光科技公司 | Predictive maintenance of automotive batteries |
CN111142026A (en) * | 2019-12-31 | 2020-05-12 | 联想(北京)有限公司 | Data processing method and device and electronic equipment |
CN113139328A (en) * | 2020-01-17 | 2021-07-20 | 丰田自动车株式会社 | Internal combustion engine state determination device, internal combustion engine state determination system, and data analysis device |
CN113139328B (en) * | 2020-01-17 | 2024-02-20 | 丰田自动车株式会社 | Internal combustion engine state determination device, internal combustion engine state determination system, and data analysis device |
CN112710956B (en) * | 2020-12-17 | 2023-08-04 | 四川虹微技术有限公司 | Expert system-based battery management system fault detection system and method |
CN112710956A (en) * | 2020-12-17 | 2021-04-27 | 四川虹微技术有限公司 | Battery management system fault detection system and method based on expert system |
CN113255764A (en) * | 2021-05-21 | 2021-08-13 | 池测(上海)数据科技有限公司 | Method, system and device for detecting electrochemical energy storage system fault by using machine learning |
CN113547919A (en) * | 2021-08-26 | 2021-10-26 | 武汉海亿新能源科技有限公司 | Remote fault monitoring method and system for fuel cell vehicle |
CN113848494A (en) * | 2021-09-18 | 2021-12-28 | 北京经纬恒润科技股份有限公司 | Online monitoring method for power battery temperature and vehicle-mounted T-BOX |
CN113848494B (en) * | 2021-09-18 | 2024-01-26 | 北京经纬恒润科技股份有限公司 | On-line monitoring method for temperature of power battery and vehicle-mounted T-BOX |
CN115332649A (en) * | 2022-07-05 | 2022-11-11 | 厦门宇电自动化科技有限公司 | Battery temperature control management method, device, equipment and readable storage medium |
CN115061049A (en) * | 2022-08-08 | 2022-09-16 | 山东卓朗检测股份有限公司 | Method and system for rapidly detecting UPS battery fault of data center |
CN115061049B (en) * | 2022-08-08 | 2022-11-01 | 山东卓朗检测股份有限公司 | Method and system for rapidly detecting UPS battery fault of data center |
CN117289145A (en) * | 2023-11-27 | 2023-12-26 | 宁德时代新能源科技股份有限公司 | Fault analysis method, data acquisition method, device, equipment, system and medium |
CN117289145B (en) * | 2023-11-27 | 2024-04-19 | 宁德时代新能源科技股份有限公司 | Fault analysis method, data acquisition method, device, equipment, system and medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109283469A (en) | Battery management system failure prediction method, device and readable storage medium storing program for executing | |
Hannan et al. | Neural network approach for estimating state of charge of lithium-ion battery using backtracking search algorithm | |
EP3770619B1 (en) | Degradation estimating device and degradation estimating method | |
Zhang et al. | Intelligent state of charge estimation of battery pack based on particle swarm optimization algorithm improved radical basis function neural network | |
CN110146817A (en) | The diagnostic method of lithium battery failure | |
EP4141460A1 (en) | Battery capacity measuring device and method, and battery control system comprising battery capacity measuring device | |
Wang et al. | A method based on improved ant lion optimization and support vector regression for remaining useful life estimation of lithium‐ion batteries | |
Zhang et al. | State of health estimation of lithium-ion batteries based on modified flower pollination algorithm-temporal convolutional network | |
Li et al. | A recurrent neural network with long short-term memory for state of charge estimation of lithium-ion batteries | |
Zafar et al. | Hybrid deep learning model for efficient state of charge estimation of Li-ion batteries in electric vehicles | |
Li et al. | Lithium-ion battery remaining useful life prognostics using data-driven deep learning algorithm | |
WO2024000756A1 (en) | Electrochemical model parameter identification method based on deep learning and heuristic algorithm | |
KR20240015240A (en) | Method and System for Estimating Online Charging and Health Status of Lithium Batteries Based on Neural Network Model Banks | |
JP2023534823A (en) | Battery capacity estimation device and method | |
Chen et al. | State of health estimation for lithium-ion battery based on long short term memory networks | |
Sangwan et al. | Estimation of battery parameters of the equivalent circuit models using meta-heuristic techniques | |
Chen et al. | State of health estimation of lithium-ion batteries based on equivalent circuit model and data-driven method | |
Gu et al. | A beetle antennae search optimized recurrent extreme learning machine for battery state of charge estimation | |
CN117445735B (en) | Cloud-edge cooperative charging early warning method, equipment and medium | |
Xue et al. | Battery degradation modelling and prediction with combination of machine learning and semi-empirical methods | |
Savargaonkar et al. | A novel neural network with gaussian process feedback for modeling the state-of-charge of battery cells | |
Zhu et al. | A noise‐immune model identification method for lithium‐ion battery using two‐swarm cooperative particle swarm optimization algorithm based on adaptive dynamic sliding window | |
Yüksek et al. | A novel state of health estimation approach based on polynomial model for lithium-ion batteries | |
Li et al. | The state of charge estimation of lithium-ion battery based on battery capacity | |
Ganjkhani et al. | Multi-Source Data Aggregation and Real-time Anomaly Classification and Localization in Power Distribution Systems |
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: 20190129 |