CN114639881B - Deep learning lithium ion battery thermal runaway early warning method - Google Patents
Deep learning lithium ion battery thermal runaway early warning method Download PDFInfo
- Publication number
- CN114639881B CN114639881B CN202210162536.8A CN202210162536A CN114639881B CN 114639881 B CN114639881 B CN 114639881B CN 202210162536 A CN202210162536 A CN 202210162536A CN 114639881 B CN114639881 B CN 114639881B
- Authority
- CN
- China
- Prior art keywords
- thermal runaway
- battery
- lithium ion
- data
- ion battery
- 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
- HBBGRARXTFLTSG-UHFFFAOYSA-N Lithium ion Chemical compound [Li+] HBBGRARXTFLTSG-UHFFFAOYSA-N 0.000 title claims abstract description 42
- 229910001416 lithium ion Inorganic materials 0.000 title claims abstract description 42
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000013135 deep learning Methods 0.000 title claims abstract description 17
- 239000013598 vector Substances 0.000 claims abstract description 44
- WHXSMMKQMYFTQS-UHFFFAOYSA-N Lithium Chemical compound [Li] WHXSMMKQMYFTQS-UHFFFAOYSA-N 0.000 claims abstract description 11
- 229910052744 lithium Inorganic materials 0.000 claims abstract description 11
- 230000032683 aging Effects 0.000 claims abstract description 10
- 230000004927 fusion Effects 0.000 claims abstract description 10
- 238000011156 evaluation Methods 0.000 claims abstract description 4
- 230000036541 health Effects 0.000 claims abstract description 4
- 239000011159 matrix material Substances 0.000 claims description 35
- 238000012549 training Methods 0.000 claims description 15
- 230000004913 activation Effects 0.000 claims description 10
- 230000006870 function Effects 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 9
- 238000013527 convolutional neural network Methods 0.000 claims description 8
- 238000012360 testing method Methods 0.000 claims description 6
- 230000015572 biosynthetic process Effects 0.000 claims description 4
- 238000012423 maintenance Methods 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 230000003213 activating effect Effects 0.000 claims 1
- 238000004146 energy storage Methods 0.000 abstract description 5
- 241000288105 Grus Species 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 238000012983 electrochemical energy storage Methods 0.000 description 1
- 230000017525 heat dissipation Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 239000002341 toxic gas Substances 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
- H01M10/00—Secondary cells; Manufacture thereof
- H01M10/42—Methods or arrangements for servicing or maintenance of secondary cells or secondary half-cells
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
- H01M10/00—Secondary cells; Manufacture thereof
- H01M10/42—Methods or arrangements for servicing or maintenance of secondary cells or secondary half-cells
- H01M10/48—Accumulators combined with arrangements for measuring, testing or indicating the condition of cells, e.g. the level or density of the electrolyte
Landscapes
- Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Theoretical Computer Science (AREA)
- Electrochemistry (AREA)
- Manufacturing & Machinery (AREA)
- Chemical & Material Sciences (AREA)
- General Chemical & Material Sciences (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Secondary Cells (AREA)
Abstract
The invention relates to the technical field of lithium batteries, in particular to a deep learning lithium ion battery thermal runaway early warning method, which is used for realizing high-precision battery thermal runaway early warning, improving the speed and precision of lithium ion battery thermal runaway early warning, helping to promote the intelligent progress of new energy storage safety and accelerating the application of a fourth industrial revolution in energy and energy storage industry by acquiring an original data set of a lithium battery, acquiring fusion characteristics of external environment data and text data, extracting and selecting related health indexes, acquiring time characteristic vectors of key characteristics, acquiring input of a graph convolution network, acquiring spatial characteristics of battery aging data, establishing a lithium ion battery thermal runaway early warning model and safety early warning evaluation steps of a battery.
Description
Technical Field
The invention relates to the technical field of lithium batteries, in particular to a thermal runaway early warning method for a deep learning lithium ion battery.
Background
The lithium ion battery has the advantages of high working voltage, high energy density, long cycle life and the like, is increasingly used as an energy storage carrier, is widely applied to various electronic consumer products, and is also widely applied to the fields of electric automobiles, electrochemical energy storage and the like at present. However, with the large-scale application of lithium ion batteries, the problem of system safety attracts great attention. Through statistical investigation of accidents, it is found that thermal runaway of the lithium ion battery caused by self chemical reaction or external influence becomes a main cause of occurrence of safety problems. However, the lithium ion battery not only has safety problems due to abuse (such as overcharge and overheat), manufacturing defects (such as internal short circuit and package damage) and the like, but also has uneven temperature distribution of the battery itself due to heat dissipation problems in the normal use process, and local overhigh temperature and the like, which can cause internal short circuit of the battery and further generate thermal runaway. After thermal runaway of a lithium battery, the lithium battery has the characteristics of high diffusion speed, high flame intensity, large production of a large amount of toxic gas and the like, and serious personal and property losses can be caused, so that early warning of the thermal runaway is particularly important under the condition that the thermal runaway cannot be completely avoided, and a method capable of accurately reflecting the temperature distribution of the battery and a measure for early warning of the thermal runaway of the lithium ion battery based on the method are needed.
Disclosure of Invention
In order to solve the problems in the background art, the invention provides a thermal runaway early warning method for a deep learning lithium ion battery.
A thermal runaway early warning method for a deep learning lithium ion battery is characterized by comprising the following steps:
s1, acquiring an original data set of a lithium battery: collecting battery data of a lithium battery, and dividing the battery data into external environment data, text data and battery aging data;
s2, acquiring fusion characteristics of external environment data and text data: splicing the external environment data E and the text data M, and recording the spliced external environment data E and the text data M as external features O= [ E, M]∈R 2k Taking the external feature O as the input of a fully-connected network, and obtaining the fusion feature O' E R of the external environment and the text by nonlinear transformation k ;
S3, extracting and selecting relevant health indexes: calculating a correlation coefficient rho of the voltage, current and temperature correlation characteristics and the battery capacity by using a correlation coefficient matrix shown in the formula 1, and screening key characteristics by designing a threshold;
wherein X is a characteristic value, Y is a target value, mu X Sum mu Y Is an average value of
S4, obtaining a time feature vector of the key feature: inputting each key feature into a gating circulation unit GRU, and obtaining a time feature vector of the key feature through a fully connected network, and marking the time feature vector as H E R k ;
S5, acquiring input of a graph convolution network: extracting local features of key features by adopting convolutional neural network CNN to obtain feature matrix X epsilon R n×k N is the number of key features, and k is the dimension of the time feature vector; calculating Euclidean distance between feature vectors of key features, and calculating to obtain a topology network by using a formula shown in formula 2, wherein the topology network forms an adjacent matrix A epsilon R n×n ;
Wherein n is the number of key features, dist (X, Y) is the Euclidean distance of the two vectors of X and Y
S6, acquiring spatial characteristics of battery aging data: inputting the adjacent matrix A and the feature matrix X into a graph convolution network G of 3 to obtain feature vectorsInputting the feature vector Z into a fully connected network to obtain a spatial feature Z' E R k ;
H (l+1) =σ(AH (l) W (l) ) (3)
Wherein A is E R n×n Is a contiguous matrix of which the number of cells is,is the output of the first layer, H when l=0 (l) =X,Is the weight matrix of the first layer, sigma is the activation function
S7, establishing a thermal runaway early warning model of the lithium ion battery: the feature O' E R will be fused k Feature vector H ε R k And spatial features Z' ∈R k Splicing to obtain integral features vector f= [ O ] I H Z']∈R 3k Inputting the integral feature vector F into a multi-layer full-connection network MLP for nonlinear mapping;
s8, safety precaution evaluation of the battery: the output information phi of the multi-layer full-connection network MLP is input into an activation function ReLU, and the output information of the activation function ReLU is used for safety pre-warning of the battery.
Preferably, in the step S1, the external environment data is time-series data, the text data is maintenance records, and the battery aging data includes battery voltage, current, temperature and battery capacity.
Preferably, in S2, one-hot is usedCode means for converting text data into matrix form M 1 ,M 2 ,...,M n ],M n A one-hot encoding vector representing an nth word; scaling the time series data to between 0 and 1 by min-max normalization, extracting potential features of external environment information by LSTM, and marking the environment features as E R k 。
Preferably, in S3, if ρ > threshold is a key feature, ρ < threshold is filtered out.
Preferably, in the step S5, the topology network forming process specifically includes setting a threshold of formula 2, comparing the euclidean distance with the threshold, and connecting the euclidean distance when the euclidean distance is greater than the threshold, and not connecting the euclidean distance when the euclidean distance is less than the threshold until the topology network is formed.
Preferably, in the step S5, the condition for forming the adjacency matrix by the topology network is that the connection with the edge is 1, and the connection without the edge is 0.
Preferably, the feature vector in S6The formation process is specifically that the adjacent matrices A and H in formula 3 (l) The adjacent node information connection of a certain node is gathered in the node by multiplication, in order to avoid the loss of the node information, an adjacent matrix A is added with a unit matrix I to form a graph convolution network G of a formula (4), an L-layer network is assumed, and the finally obtained feature vector is
Wherein,is->Degree matrix of (2)
Preferably, the lithium ion battery thermal runaway early warning model structure in S7 includes fusion characteristics of external environment data and text data obtained by 2 layers of LSTM and 2 layers of fully connected networks, time characteristic vectors obtained by 2 layers of GRUs and 2 layers of fully connected networks, and spatial characteristics obtained by 1 layer of CNN, 2 layers of graph convolution networks and 2 layers of fully connected networks.
Preferably, training and verifying the lithium ion battery thermal runaway early warning model of S7: and (3) taking 70% of the original data set obtained in the step (S1) as a model training set and 30% as a model test set, training the lithium ion battery thermal runaway early warning model by using the training set, determining model parameters, and inputting the test set into the determined lithium ion battery thermal runaway early warning model to early warn the battery thermal runaway.
Preferably, the output data of the activation function ReLU in S8 is at [ t ] i ,t j ]In between the two,
when t i <t<t j The system starts to alarm;
when t > t j Indicating a serious danger to the battery.
The beneficial effects are that: compared with the prior art, the method for warning the thermal runaway of the lithium ion battery for deep learning expands various data acquisition sources of the lithium ion battery, adopts a fully-connected network, a graph convolution network and a convolution neural network to extract external features, time features and space features respectively, and finally splices and establishes a warning model for the thermal runaway of the lithium ion battery, thereby realizing high-precision warning for the thermal runaway of the battery, improving the speed and precision of the warning for the thermal runaway of the lithium ion battery, being beneficial to promoting the intelligent progress of energy storage safety of new energy sources and accelerating the application of the fourth industrial revolution in energy and energy storage industries.
Drawings
FIG. 1 is a flow chart of a model training method of the present invention;
fig. 2 is a topological network formation process.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in detail with reference to specific embodiments. It should be further noted that, in order to avoid obscuring the present invention due to unnecessary details, only structures and/or processing steps closely related to aspects of the present invention are shown in the specific embodiments, and other details not greatly related to the present invention are omitted.
In addition, it should be further noted that 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.
Referring to fig. 1, the method for early warning thermal runaway of a deep learning lithium ion battery provided by the invention comprises the following steps:
s1, acquiring an original data set of a lithium battery: collecting battery data of a lithium battery, and dividing the battery data into external environment data, text data and battery aging data; the external environment data are time series data, the text data are maintenance records, and the battery aging data comprise battery voltage, current, temperature and battery capacity;
s2, acquiring fusion characteristics of external environment data and text data: external environment data E, scaling the data to between 0 and 1 by adopting min-max normalization, extracting potential characteristics of external environment information by utilizing LSTM, and marking the environment characteristics as E R k The method comprises the steps of carrying out a first treatment on the surface of the Converting maintenance record into matrix form [ M ] by using one-hot coding mode 1 ,M 2 ,...,M n ],M n One-hot encoding vector representing nth word, and also extracting potential characteristics of text information by using LSTM, wherein the text characteristics are marked as M epsilon R k Splicing the external environment data E and the text data M, and recording the spliced external environment data E and the text data M as external features O= [ E, M]∈R 2k Taking the external feature O as the input of a fully-connected network, and obtaining the fusion feature O' E R of the external environment and the text by nonlinear transformation k ;
S3, extracting and selecting relevant health indexes: calculating a correlation coefficient rho of the correlation characteristics of the voltage, the current and the temperature and the battery capacity by using a correlation coefficient matrix shown in the formula 1, and screening out the correlation coefficient rho by designing a threshold value threshold, wherein if rho is more than threshold, the correlation coefficient rho is the key characteristic, and if rho is less than threshold;
wherein X is a characteristic value, Y is a target value, mu X Sum mu Y Is an average value of
S4, obtaining a time feature vector of the key feature: inputting each key feature into a gating circulation unit GRU, and obtaining a time feature vector of the key feature through a fully connected network, and marking the time feature vector as H E R k ;
S5, acquiring input of a graph convolution network: extracting local features of key features by adopting convolutional neural network CNN to obtain feature matrix X epsilon R n×k N is the number of key features, and k is the dimension of the time feature vector; calculating Euclidean distance between feature vectors of key features, and calculating to obtain a topology network by using a formula shown in formula 2, wherein the topology network forms an adjacent matrix A epsilon R n×n The method comprises the steps of carrying out a first treatment on the surface of the The topology network forming process specifically includes setting a threshold of formula 2, comparing the Euclidean distance with the threshold, and connecting when the Euclidean distance is larger than the threshold, and not connecting when the Euclidean distance is smaller than the threshold until the topology network is formed (see fig. 2); the condition of the topology network forming the adjacency matrix is that the connection with the edge is 1, and the connection without the edge is 0;
wherein n is the number of key features, dist (X, Y) is the Euclidean distance of the two vectors of X and Y
S6, acquiring spatial characteristics of battery aging data: inputting the adjacent matrix A and the feature matrix X into a graph convolution network G of 3 to obtain feature vectorsInputting the characteristic vector Z into a fully-connected network to obtain a space featureSign Z' ∈R k ;
H (l+1) =σ(AH (l) W (l) ) (3)
Wherein A is E R n×n Is a contiguous matrix of which the number of cells is,is the output of the first layer, H when l=0 (l) =X,Is the weight matrix of the first layer, sigma is the activation function
S7, establishing a thermal runaway early warning model of the lithium ion battery: the feature O' E R will be fused k Feature vector H ε R k And spatial features Z' ∈R k Splicing to obtain integral features vector f= [ O ] I H Z']∈R 3k Inputting the integral feature vector F into a multi-layer full-connection network MLP for nonlinear mapping to form a lithium ion battery thermal runaway early warning model, wherein the model comprises 2 layers of LSTM and 2 layers of full-connection networks to obtain fusion features of external environment data and text data, 2 layers of GRUs and 2 layers of full-connection networks to obtain time feature vectors, and 1 layer of CNN, 2 layers of graph convolution networks and 2 layers of full-connection networks to obtain space features;
wherein the feature vectorThe formation process is specifically that the adjacent matrices A and H in formula 3 (l) The adjacent node information connection of a certain node is gathered in the node by multiplication, in order to avoid the information loss of the node, an adjacent matrix A is added with a unit matrix I to form a graph rolling network G of a formula (4), and the characteristic vector obtained finally is given by assuming an L-layer network>
Wherein,is->Degree matrix of (2)
Training and verifying a thermal runaway early warning model of the lithium ion battery: taking 70% of the original data set obtained in the step S1 as a model training set and 30% as a model testing set, training the thermal runaway early-warning model of the lithium ion battery by using the training set, determining model parameters, inputting the testing set into the determined thermal runaway early-warning model of the lithium ion battery for training and optimizing until the model converges, and determining the thermal runaway early-warning model;
s8, safety precaution evaluation of the battery: inputting an activation function ReLU into the thermal runaway early warning model, wherein the output information of the activation function ReLU is [ t ] i ,t j ]In between the two,
when t i <t<t j The system starts to alarm;
when t > t j Indicating that the battery is severely dangerous for safety precautions of the battery.
Finally, it is noted that the above embodiments are only for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications and equivalents may be made thereto without departing from the spirit and scope of the technical solution of the present invention, which is intended to be covered by the scope of the claims of the present invention.
Claims (9)
1. A deep learning lithium ion battery thermal runaway early warning method is characterized by comprising the following steps:
s1, acquiring an original data set of a lithium battery: collecting battery data of a lithium battery, and dividing the battery data into external environment data, text data and battery aging data;
s2, acquiring fusion characteristics of external environment data and text data: the external environment data E and the text data M are spliced,is marked as an external featureExternal features->As input of fully connected network, nonlinear transformation gets fusion feature of external environment and text +.>;
S3, extracting and selecting relevant health indexes: calculating the correlation coefficient of the voltage, current and temperature correlation characteristics and the battery capacity by using the correlation coefficient matrix shown in the formula 1By designing threshold->Screening key characteristics,
(1)
wherein,is a characteristic value->Is a target value, & lt & gt>And->Is the average value;
s4, obtaining a time feature vector of the key feature: inputting each key feature into a gate control circulation unit GRU, obtaining a time feature vector of the key feature through a fully connected network, and marking the time feature vector as;
S5, obtaining input of a graph convolution network: using convolutional neural networksExtracting local features of key features to obtain a feature matrix +.>N is the number of key features, and k is the dimension of the time feature vector; calculating Euclidean distance between feature vectors of key features, and calculating to obtain a topology network by using a formula shown in formula 2, wherein the topology network forms an adjacency matrix +.>,
(2)
Wherein n is the number of key features,dist(X,Y)is thatXAndYeuclidean distance of the two vectors;
s6, acquiring spatial characteristics of battery aging data: inputting the adjacent matrix A and the feature matrix X into a graph convolution network G of 3 to obtain feature vectorsInputting the feature vector Z into a fully connected network to obtain the spatial feature +.>,
(3)
Wherein,is an adjacency matrix->Is->Layer output, when->Then->,Is->Weight matrix of layer,/>Is an activation function;
s7, establishing a thermal runaway early warning model of the lithium ion battery: will fuse featuresFeature vector->And spatial featuresSplicing to obtain an overall feature vector->The whole feature vector +.>Input multi-layer full-connection networkPerforming nonlinear mapping;
s8, safety precaution evaluation of the battery: connecting multiple layers of networkOutput information phi of (a) input activation function +.>Activating function->The output information of the battery is used for safety precaution of the battery; the output data t of the activation function is [ ti, tj ]]In between the two,
when ti is less than t and less than tj, the system starts to alarm;
when t > tj, the battery is severely dangerous.
2. The deep learning lithium ion battery thermal runaway early warning method according to claim 1, wherein in the step S1, the external environment data is time series data, the text data is maintenance records, and the battery aging data comprises battery voltage, current, temperature and battery capacity.
3. The deep learning thermal runaway warning method for lithium ion batteries according to claim 1, wherein in the step S2, the method is characterized in thatThe coding mode converts the text data into a matrix form +.>,/>Indicate->Individual words +.>Encoding the vector; by->Normalization scales time series data between 0 and 1, with +.>Extracting potential characteristics of external environment information, wherein the environment characteristics are marked as +.>。
4. The deep learning thermal runaway warning method for lithium ion batteries according to claim 1, wherein in said step S3, ifThen is a key feature->Then screening out.
5. The method for warning thermal runaway of a deep learning lithium ion battery according to claim 1, wherein in the step S5, a threshold value of formula 2 is set, the euclidean distance is compared with the threshold value, and when the euclidean distance is greater than the threshold value, the euclidean distance is not required to be connected until a topology network is formed.
6. The method for warning thermal runaway of a deep learning lithium ion battery according to claim 1, wherein the condition for forming the adjacent matrix by the topology network in S5 is that the connection with the edge is 1, and the connection without the edge is 0.
7. A deep learning lithium ion battery thermal runaway pre-heater as defined in claim 1The alert method is characterized in that the feature vector in the S6The formation process is specifically, the adjacency matrix +.>And->Multiplying to gather adjacent node information connection of a node to the node, and let adjacent matrix +_in order to avoid losing the node information>Plus an identity matrix->Form a graph roll-up network G of formula (4), assuming +.>Layer network, the final feature vector is +.>,
(4)
Wherein,,/>is->Is a degree matrix of (2).
8. The deep learning lithium ion battery thermal runaway early warning method according to claim 1, wherein the method comprises the following steps: the lithium ion battery thermal runaway early warning model structure in the S7 comprises fusion characteristics of external environment data and text data obtained by a layer 2 LSTM and a layer 2 full-connection network, time characteristic vectors obtained by a layer 2 GRU and a layer 2 full-connection network, and space characteristics obtained by a layer 1 CNN, a layer 2 graph convolution network and a layer 2 full-connection network.
9. The deep learning lithium ion battery thermal runaway early warning method according to claim 1, wherein the method comprises the following steps: training and verifying the lithium ion battery thermal runaway early warning model of S7: and (3) taking 70% of the original data set obtained in the step (S1) as a model training set and 30% as a model test set, training the thermal runaway early-warning model of the lithium ion battery by using the training set, determining model parameters, inputting the test set into the determined thermal runaway early-warning model of the lithium ion battery for training and optimizing until the model converges, and determining the thermal runaway early-warning model.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210162536.8A CN114639881B (en) | 2022-02-22 | 2022-02-22 | Deep learning lithium ion battery thermal runaway early warning method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210162536.8A CN114639881B (en) | 2022-02-22 | 2022-02-22 | Deep learning lithium ion battery thermal runaway early warning method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114639881A CN114639881A (en) | 2022-06-17 |
CN114639881B true CN114639881B (en) | 2024-04-12 |
Family
ID=81945842
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210162536.8A Active CN114639881B (en) | 2022-02-22 | 2022-02-22 | Deep learning lithium ion battery thermal runaway early warning method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114639881B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114882021B (en) * | 2022-07-07 | 2022-09-09 | 江苏中清先进电池制造有限公司 | Efficient processing method and system for battery lithium film |
CN115688592B (en) * | 2022-11-09 | 2023-05-09 | 福建德尔科技股份有限公司 | Rectification control system and method for preparing electronic-grade carbon tetrafluoride |
CN116315189B (en) * | 2023-05-25 | 2023-08-11 | 澄瑞电力科技(上海)股份公司 | Data fusion-based battery Bao Re out-of-control prediction method and system |
CN116359748B (en) * | 2023-05-31 | 2023-08-01 | 苏州精控能源科技有限公司 | Thermal runaway simulation test method and device for cylindrical lithium battery |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107229269A (en) * | 2017-05-26 | 2017-10-03 | 重庆工商大学 | A kind of wind-driven generator wheel-box method for diagnosing faults of depth belief network |
CN112949164A (en) * | 2021-01-27 | 2021-06-11 | 上海电机学院 | Lithium battery health state prediction method |
CN113609791A (en) * | 2021-10-11 | 2021-11-05 | 武汉云侦科技有限公司 | Active safety monitoring and early warning method and system for lithium ion battery energy storage power station |
WO2022034983A1 (en) * | 2020-08-14 | 2022-02-17 | 주식회사 한국파워셀 | Battery cell defectiveness and fire early diagnosis method and apparatus based on neural network |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11131713B2 (en) * | 2018-02-21 | 2021-09-28 | Nec Corporation | Deep learning approach for battery aging model |
US20200011932A1 (en) * | 2018-07-05 | 2020-01-09 | Nec Laboratories America, Inc. | Battery capacity fading model using deep learning |
-
2022
- 2022-02-22 CN CN202210162536.8A patent/CN114639881B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107229269A (en) * | 2017-05-26 | 2017-10-03 | 重庆工商大学 | A kind of wind-driven generator wheel-box method for diagnosing faults of depth belief network |
WO2022034983A1 (en) * | 2020-08-14 | 2022-02-17 | 주식회사 한국파워셀 | Battery cell defectiveness and fire early diagnosis method and apparatus based on neural network |
CN112949164A (en) * | 2021-01-27 | 2021-06-11 | 上海电机学院 | Lithium battery health state prediction method |
CN113609791A (en) * | 2021-10-11 | 2021-11-05 | 武汉云侦科技有限公司 | Active safety monitoring and early warning method and system for lithium ion battery energy storage power station |
Non-Patent Citations (1)
Title |
---|
基于卷积自编码神经网络的锂离子电池健康状况评估方法研究;侯瑞磊;范秋华;;计算机测量与控制;20200825(第08期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN114639881A (en) | 2022-06-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN114639881B (en) | Deep learning lithium ion battery thermal runaway early warning method | |
CN113343633A (en) | Thermal runaway fault classification and risk prediction method and system for power lithium battery | |
CN116154900B (en) | Active safety three-stage prevention and control system and method for battery energy storage power station | |
Zhang et al. | State‐of‐charge estimation of lithium‐ion battery pack based on improved RBF neural networks | |
CN115436821B (en) | Method and device for identifying abnormal heat generation of battery system | |
CN115214424A (en) | Temperature determination method and device of battery thermal management system and electronic equipment | |
Liang et al. | Evaluation of battery modules state for electric vehicle using artificial neural network and experimental validation | |
He et al. | Voltage abnormality-based fault diagnosis for batteries in electric buses with a self-adapting update model | |
Xing et al. | Lithium-ion battery state-of-charge estimation based on a dual extended Kalman filter and BPNN correction | |
CN116125306A (en) | Power battery thermal runaway diagnosis method and system based on hybrid prediction framework | |
Xu et al. | State-of-charge estimation and health prognosis for lithium-ion batteries based on temperature-compensated Bi-LSTM network and integrated attention mechanism | |
CN113253125B (en) | Information fusion-based lithium iron phosphate battery thermal runaway monitoring method and system | |
CN116315189B (en) | Data fusion-based battery Bao Re out-of-control prediction method and system | |
Xiao et al. | Battery state of health prediction based on voltage intervals, BP neural network and genetic algorithm | |
US20240106232A1 (en) | Thermal-runaway warning method, system, and terminal for power station | |
CN115659790B (en) | Real-time temperature detection method for power battery pack | |
Wang et al. | An inconsistency fault diagnosis method for lithium-ion cells in the battery pack based on piecewise dimensionality reduction and outlier identification | |
CN116227650B (en) | Lithium battery temperature distribution prediction model construction method based on orthogonal enhancement type local maintenance projection algorithm | |
Sarmah et al. | Numerical and experimental investigation of state of health of Li-ion battery | |
CN114924192A (en) | Parallel battery pack safety early warning method based on neural network | |
CN113721151B (en) | Battery capacity estimation model and method based on double-tower deep learning network | |
CN117805616A (en) | Abnormal cell identification method for lithium ion energy storage system | |
Li et al. | Remaining useful life prediction of lithium battery based on ACNN-Mogrifier LSTM-MMD | |
Li et al. | Hybrid data-driven approach for predicting the remaining useful life of lithium-ion batteries | |
Zhang et al. | Lithium-ion battery state of health estimation based on improved deep extreme learning machine |
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 |