CN116683061A - Power battery thermal runaway prediction and suppression integrated system, method and storage medium - Google Patents
Power battery thermal runaway prediction and suppression integrated system, method and storage medium Download PDFInfo
- Publication number
- CN116683061A CN116683061A CN202310967238.0A CN202310967238A CN116683061A CN 116683061 A CN116683061 A CN 116683061A CN 202310967238 A CN202310967238 A CN 202310967238A CN 116683061 A CN116683061 A CN 116683061A
- Authority
- CN
- China
- Prior art keywords
- data
- module
- thermal runaway
- prediction
- acquisition module
- 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.)
- Granted
Links
- 230000001629 suppression Effects 0.000 title claims abstract description 70
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000003860 storage Methods 0.000 title claims abstract description 10
- 238000001816 cooling Methods 0.000 claims abstract description 68
- 238000012545 processing Methods 0.000 claims abstract description 44
- 238000011217 control strategy Methods 0.000 claims abstract description 43
- 230000005764 inhibitory process Effects 0.000 claims abstract description 9
- 238000012549 training Methods 0.000 claims description 34
- 238000005070 sampling Methods 0.000 claims description 20
- 238000012360 testing method Methods 0.000 claims description 19
- 238000011156 evaluation Methods 0.000 claims description 18
- 230000002159 abnormal effect Effects 0.000 claims description 17
- 238000013528 artificial neural network Methods 0.000 claims description 14
- 230000008569 process Effects 0.000 claims description 14
- 238000005457 optimization Methods 0.000 claims description 13
- 230000008859 change Effects 0.000 claims description 10
- 238000010606 normalization Methods 0.000 claims description 10
- 238000006243 chemical reaction Methods 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 5
- 230000003111 delayed effect Effects 0.000 claims description 5
- 230000005856 abnormality Effects 0.000 claims description 4
- 238000004140 cleaning Methods 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 2
- 230000001934 delay Effects 0.000 abstract description 2
- 230000015654 memory Effects 0.000 description 18
- 230000006870 function Effects 0.000 description 13
- 238000004364 calculation method Methods 0.000 description 6
- 238000002790 cross-validation Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000036541 health Effects 0.000 description 3
- 230000007787 long-term memory Effects 0.000 description 3
- 230000007774 longterm Effects 0.000 description 3
- 230000006403 short-term memory Effects 0.000 description 3
- 125000004122 cyclic group Chemical group 0.000 description 2
- 238000004146 energy storage Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000004913 activation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 238000012546 transfer Methods 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- H01M10/4285—Testing apparatus
-
- 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
- H01M10/60—Heating or cooling; Temperature control
- H01M10/61—Types of temperature control
- H01M10/613—Cooling or keeping cold
-
- 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
- H01M10/60—Heating or cooling; Temperature control
- H01M10/62—Heating or cooling; Temperature control specially adapted for specific applications
- H01M10/625—Vehicles
-
- 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
- H01M10/60—Heating or cooling; Temperature control
- H01M10/63—Control systems
- H01M10/633—Control systems characterised by algorithms, flow charts, software details or the like
-
- 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
- H01M10/60—Heating or cooling; Temperature control
- H01M10/65—Means for temperature control structurally associated with the cells
- H01M10/656—Means for temperature control structurally associated with the cells characterised by the type of heat-exchange fluid
- H01M10/6561—Gases
- H01M10/6563—Gases with forced flow, e.g. by blowers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/08—Thermal analysis or thermal optimisation
Landscapes
- Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Chemical & Material Sciences (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Electrochemistry (AREA)
- General Chemical & Material Sciences (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Geometry (AREA)
- Biomedical Technology (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Automation & Control Theory (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Hardware Design (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Secondary Cells (AREA)
Abstract
The invention belongs to the technical field of batteries, and particularly relates to a power battery thermal runaway prediction and inhibition integrated system and method. The system comprises a first acquisition module, a second acquisition module, a third acquisition module, a fourth acquisition module, a data processing module, a prediction module, a control module, an alarm module and a suppression module; the prediction module predicts by using a trained prediction model through the acquired acquisition data to obtain prediction result information of whether the battery module has a thermal runaway trend; the control module optimizes the acquired acquisition data by utilizing a trained control strategy to obtain the optimized parameters of the cooling system. The invention further provides a storage medium for executing the method. The invention not only can predict the thermal runaway risk of the power battery in advance, but also can adjust the working parameters of the cooling system according to the corresponding prediction result, so that the power battery works in a normal temperature area or delays the occurrence time of the thermal runaway, and the safety and the reliability are effectively improved.
Description
Technical Field
The invention relates to the technical field of batteries, in particular to a power battery thermal runaway prediction and suppression integrated system, a method and a storage medium.
Background
The demand of power batteries in modern society is increasing, including electric vehicles, renewable energy storage, and other fields. However, because the power battery is large in size and high in energy storage density, if the temperature of the power battery cannot be effectively controlled, the power battery can be caused to generate thermal runaway, serious potential safety hazards are caused, and even personnel safety is threatened, so that property loss is caused. Traditional power cell temperature control methods rely primarily on temperature sensors and protection devices to monitor and maintain cell temperature within safe ranges. However, the above method is implemented in the case that the power battery has thermal runaway, and can only passively respond to temperature changes, and lacks accurate prediction and active suppression capability of thermal runaway. Therefore, finding an efficient and reliable integrated system for predicting and suppressing thermal runaway of a power battery is an urgent need in the battery technology field.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a power battery thermal runaway prediction and suppression integrated system, a method and a storage medium, and aims to provide an innovative technical solution to realize accurate prediction and advanced suppression of the thermal runaway of the power battery.
According to the invention, by cooperatively using the prediction model based on the neural network and the control strategy based on the support vector regression, the thermal runaway risk of the power battery can be predicted in advance, and the working parameters of the cooling system can be adjusted according to the corresponding prediction result, so that the power battery works in a normal temperature area or the occurrence time of the thermal runaway is delayed, and the safety and reliability of the new energy automobile are effectively improved.
The invention provides a power battery thermal runaway prediction and suppression integrated system, which comprises a battery module and a cooling system, and further comprises:
the first acquisition module is used for acquiring battery minimum single voltage data, battery current data, battery minimum single temperature data, battery module SOH data and battery module SOC data of the battery module in real time;
the second acquisition module is used for acquiring vehicle environment temperature data and battery module temperature data of the battery module in real time;
the third acquisition module is used for acquiring the output power data of the battery module in real time;
the fourth acquisition module is used for acquiring the rotation speed data of the compressor, the air inlet quantity data of the evaporator and the air inlet quantity data of the condenser of the cooling system in real time;
The data processing module is used for respectively processing the acquired data of each acquisition module;
the prediction module is used for obtaining the acquired data of the processed first acquisition module and the processed second acquisition module, and predicting the acquired data through a trained prediction model to obtain the prediction result information of whether the battery module has a thermal runaway trend;
the control module is used for acquiring the acquired data of the processed second acquisition module, the processed third acquisition module and the processed fourth acquisition module, receiving the predicted result information without the thermal runaway trend sent by the predicting module, and optimizing the predicted result information through a trained control strategy to obtain optimized parameters of the cooling system;
the warning module receives the predicted result information with the thermal runaway trend sent by the predicting module, and after the predicted result information is matched with the built-in thermal runaway grade rule, the safety time and the corresponding thermal runaway grade are displayed, and a warning prompt instruction, a suppression instruction and a warning instruction of the corresponding grade are executed at the same time;
and the suppression module executes the suppression instruction sent by the alarm module.
Preferably, the data processing module performs processing operations of data cleaning, feature selection, feature conversion, data scaling and data encoding on the collected data of each collecting module.
Preferably, the method for constructing the trained prediction model in the prediction module specifically includes:
s11: acquiring historical working data of the battery module, wherein the historical working data comprises data acquired by a first acquisition module and a second acquisition module;
s12: normalizing the historical working data in the step S11;
s13: dividing the historical working data subjected to normalization processing in the step S12 into a training set and a testing set according to a proportion;
s14: inputting the training set data into a prediction model based on a neural network for training to obtain an initial prediction model;
s15: inputting the test set data into the initial prediction model for evaluation, and adjusting model parameters according to the evaluation result to obtain a trained prediction model, wherein an evaluation method in the prediction model adopts an average relative error (MRE) to evaluate the prediction result of each cross verification test data;
s16: and inputting the trained prediction model into a prediction module for predicting the thermal runaway trend of the battery module in the actual running process of the vehicle.
Preferably, the prediction result information includes three types of battery data, where the three types of battery data are respectively: temperature data including, but not limited to, temperature rise, temperature difference, temperature, voltage data including, but not limited to, voltage rate of change, voltage difference, voltage, current data including, but not limited to, current rate of change, current difference, current; in addition, the predicted result information also comprises the occurrence time of thermal runaway prediction.
Preferably, the built-in thermal runaway level is determined based on a number condition of abnormal battery data in the prediction result information and a occurrence time condition of thermal runaway prediction, and the determination condition is specifically:
condition one, judge on the basis of the quantity of each type battery abnormal data in the predicted result information:
first level: when the number of abnormal data in any one/any two/all types of data is three or more, the data is the first grade;
second level: when the number of abnormal data in any one type/any two types of data is two, the abnormal data is the second level;
third level: when the number of data with abnormality in any type of data is one, the data is the third grade;
condition two, judge according to the predicted occurrence time of thermal runaway:
first level: the predicted occurrence time of thermal runaway is within one minute including one minute;
second level: the predicted occurrence time of thermal runaway is from one minute to more than five minutes including five minutes;
third level: the predicted occurrence time of thermal runaway is more than five minutes to ten minutes including ten minutes;
thus, the rule for determining the final thermal runaway level is as follows:
if any one of the thermal runaway judging conditions is a first grade or both of the two thermal runaway judging conditions are a second grade, defining the final thermal runaway grade as a grade one, and executing an alarm prompt instruction, a suppression instruction and an alarm instruction;
If the combination of the thermal runaway judging conditions is the second grade and the third grade, defining the final thermal runaway grade as a grade two, and executing an alarm prompt instruction and a suppression instruction;
if the two thermal runaway judging conditions are the third grade, the final thermal runaway grade is defined as a third grade, and an alarm prompt instruction is executed.
Preferably, the method for constructing the trained control strategy in the control module specifically includes:
s21: acquiring historical working data of the battery module and the cooling system, wherein the historical working data comprises data acquired by a second acquisition module, a third acquisition module and a fourth acquisition module;
s22: normalizing the data acquired by the second acquisition module, the third acquisition module and the fourth acquisition module in the step S21;
s23: dividing the data acquired by the second acquisition module, the third acquisition module and the fourth acquisition module after normalization processing in the step S22 into a training set and a testing set according to a proportion;
s24: inputting the training set data into a control strategy based on support vector regression for training to obtain a preliminary control strategy;
s25: inputting the test set data into the preliminary control strategy for evaluation, and adjusting model parameters according to an evaluation result to obtain a trained control strategy, wherein a regression square correlation coefficient is adopted in an evaluation method in the control strategy;
S26: and inputting the trained control strategy into the control module for optimizing the working parameters of the cooling system in the actual running process of the vehicle.
Preferably, the predicted result information, which is sent by the prediction module and has no thermal runaway trend, and is received by the control module is the battery module temperature predicted in the last sampling period, and the optimization parameter is the optimal operation condition solution of the cooling system, which is obtained after the control strategy of the control module is solved, according to the battery module temperature, the current working state of the battery module and the current working state of the cooling system predicted in the last sampling period sent by the prediction module; and the suppression module is internally preset with a suppression parameter for executing a suppression instruction of the alarm module, wherein the suppression parameter is a maximum power working parameter of the cooling system.
Preferably, the cooling system includes: the system comprises an electric compressor, a condenser, an evaporator, an expansion valve, a cooling pipeline, a cooling air channel, a central processing unit and an independent power supply, wherein the central processing unit is used for receiving and executing the suppression parameters or the optimization parameters, and the independent power supply is used for supplying power to the cooling system when the prediction module predicts the thermal runaway risk.
The invention provides a power battery thermal runaway prediction and inhibition integrated method, which is applied to a power battery thermal runaway prediction and inhibition integrated system and comprises the following steps:
s31: after the vehicle is started, the first acquisition module, the second acquisition module, the third acquisition module and the fourth acquisition module acquire real-time driving condition data according to the sampling frequency of 5 seconds; meanwhile, the acquired real-time data is transmitted to a data processing module connected with the acquisition module;
s32: the data processing module respectively processes the acquired data of each acquisition module;
s33: the prediction module acquires the acquired data of the processed first acquisition module and the processed second acquisition module, predicts the acquired data through a trained prediction model, and acquires the prediction result information of whether the battery module has a thermal runaway trend or not;
s34: if the output result of the prediction model has no thermal runaway risk, returning to the step S31, and sending the prediction result information without thermal runaway trend to the control module; if the output result of the prediction model has a thermal runaway risk, the prediction result information with a thermal runaway trend is sent to the alarm module, and meanwhile, the sampling frequency of each acquisition module is changed to be once in 1 second, so that the thermal runaway prediction is more accurately carried out;
S35: when the control module receives the predicted result information without thermal runaway trend sent by the prediction module, acquiring the processed acquired data of the second acquisition module, the third acquisition module and the fourth acquisition module synchronously, optimizing the predicted battery module temperature in the last sampling period in the predicted result information and the acquired data of the second to fourth acquisition modules through a trained control strategy, acquiring optimized parameters aiming at a cooling system, and sending the optimized parameters to a central processor of the cooling system for execution so as to maintain the power module to work in a normal temperature interval;
s36: when the warning module receives the predicted result information with the thermal runaway trend sent by the predicting module, the warning module displays the safety time and the corresponding level after matching with the built-in thermal runaway level rule, and simultaneously executes warning prompt instructions, inhibition instructions and warning instructions of the corresponding level;
s37: when the suppression module receives the suppression instruction of the alarm module, the suppression module calls a preset suppression parameter and transmits the suppression parameter to the central processing unit of the cooling system for execution, so that the cooling system runs the maximum power working parameter under the thermal runaway suppression working condition, the occurrence time of the thermal runaway is further delayed, and larger loss is avoided.
The invention also provides a storage medium, wherein the storage medium is stored with a computer program, and when the computer program is executed by a processor, the integrated method of predicting and suppressing the thermal runaway of the power battery can be realized.
Compared with the prior art, the invention has the following beneficial effects:
1. the invention provides a power battery thermal runaway prediction and inhibition integrated system and a method, which can accurately predict and inhibit the occurrence of thermal runaway of a power battery in advance, reduce the occurrence probability of the thermal runaway and obviously reduce potential loss.
2. According to the invention, training data adopted by the prediction model come from a real vehicle, the influence of seasons and battery health states on battery thermal runaway is considered in the data, the accuracy of a prediction result is effectively improved, and the probability of false alarm is greatly reduced.
3. According to the invention, through the cooperative use of the prediction model based on the neural network and the control strategy based on the support vector regression, not only the thermal runaway risk of the power battery can be predicted in advance, but also the working parameters of the cooling system can be adjusted according to the corresponding prediction result, so that the power battery works in a normal temperature interval or the occurrence time of the thermal runaway is delayed. Compared with the prior art, the integrated system has obvious differences and advantages, can effectively predict and inhibit thermal runaway of the power battery, furthest protects the safety and reliability of the battery system, and provides an important progress technology for the industry.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.
FIG. 1 is a schematic block diagram of an integrated power cell thermal runaway prediction and suppression system provided by the present invention;
FIG. 2 is a schematic diagram of the workflow of the integrated power cell thermal runaway prediction and suppression system provided by the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In the description of the present invention, it should be noted that the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying any particular importance, and furthermore, unless explicitly specified and limited otherwise, the terms "mounted," "connected," "coupled," and "connected" are to be construed broadly, and may be, for example, fixedly coupled, detachably coupled, or integrally coupled; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention will be understood in specific cases by those of ordinary skill in the art.
As shown in fig. 1, the present invention provides an integrated power cell thermal runaway prediction and suppression system, comprising a battery module and a cooling system, further comprising:
the first acquisition module is used for acquiring the battery minimum cell voltage data, the battery current data, the battery minimum cell temperature data, the battery module SOH data and the battery module SOC data of the battery module in real time.
And the second acquisition module is used for acquiring the vehicle environment temperature data and the battery module temperature data of the battery module in real time.
And the third acquisition module is used for acquiring the output power data of the battery module in real time.
And the fourth acquisition module is used for acquiring the rotation speed data of the compressor, the air inlet quantity data of the evaporator and the air inlet quantity data of the condenser of the cooling system in real time.
In the embodiment, a first acquisition module, a second acquisition module and a third acquisition module are connected with a power supply module, and work state data of the power supply module are acquired in real time; the fourth acquisition module is connected with the cooling system and acquires working state data of the cooling system in real time.
In order to comprehensively monitor the power battery and further make accurate judgment, in this embodiment, a plurality of acquisition modules are provided, and different acquisition modules can monitor one or more of temperature, voltage, current, SOH, SOC, evaporator air intake, condenser air intake and the like to acquire acquisition data.
It should be noted that, because the battery module temperature data is the target to be optimized in the control strategy of the present invention, and the battery module output power data and the vehicle environment temperature data also affect the temperature of the battery module, the present invention adds the battery module temperature data, the battery module output power data and the vehicle environment temperature data in the control strategy, so as to more accurately optimize the working parameters of the cooling system.
The data processing module is used for respectively processing the acquired data of each acquisition module.
Preferably, the data processing module performs processing operations of data cleaning, feature selection, feature conversion, data scaling and data encoding on the collected data of each collecting module.
In this embodiment, the data cleansing, feature selection and feature conversion are to ensure that the input data has the same format and distribution as the training data so that the model can process the data correctly; data scaling is to scale features of input data so that they have similar dimensions; the data coding is to code the data so that it is suitable for the requirements of the model.
The prediction module acquires the acquired data of the processed first acquisition module and the second acquisition module, predicts the acquired data through a trained prediction model, and acquires the prediction result information of whether the battery module has a thermal runaway trend.
In this embodiment, the method for constructing the trained prediction model in the prediction module specifically includes:
s11: and acquiring historical working data of the battery module, wherein the historical working data comprises data acquired by a first acquisition module and a second acquisition module.
It should be noted that, the historical operating data of the battery module in the step S11 includes: the collected data in the first collection module is "battery minimum cell voltage data, battery current data, battery minimum cell temperature data, battery module SOH data and battery module SOC data", and the collected data in the second collection module is "vehicle environment temperature data, battery module temperature data".
Preferably, when training a prediction model based on a neural network, the historical working data of the battery module is obtained by years when the historical working data is obtained in consideration of the influence of seasons and battery health states on thermal runaway of the battery.
It should be noted that, in this embodiment, the data collected by the first collecting module and the second collecting module used in training the prediction model are driving condition data of a real automobile for three years obtained in the new energy big data center, where the condition data considers the influence of seasons and battery health states on thermal runaway of the battery, so that the prediction result is more accurate, and the probability of false alarm is reduced.
Because the structure, the model and the series-parallel connection mode of the battery cells of different battery packs are different, the time span of the historical working data in the embodiment can be adjusted according to actual requirements.
S12: and (3) carrying out normalization processing on the historical working data in the step S11, wherein a normalization formula is as follows:
，
wherein,,representing the original value +_>Representing the normalized value.
S13: and dividing the historical working data subjected to normalization processing in the step S12 into a training set and a testing set according to a proportion.
It should be noted that, in this embodiment, in order to ensure the stability and reproducibility of the training model, 10 times of cross-validation was performed, and the training set and the test set respectively account for 80% and 20% of the whole sample.
S14: and inputting the training set data into a prediction model based on a neural network for training to obtain an initial prediction model.
It should be noted that the long-term and short-term memory neural network is a special cyclic neural network, and aims to solve the difficulty encountered by the traditional cyclic neural network in processing the long-sequence problem. The conventional recurrent neural network is prone to the problem of gradient disappearance or gradient explosion when processing long sequences, so that long-term dependency is difficult to learn effectively. To overcome these problems, long and short term memory neural networks incorporate input gates, forget gates, and output gate mechanisms that can selectively screen and store information in candidate memory cells, and finally store and transfer long term information through the candidate memory cells, thereby enabling efficient processing of long sequences and capture of long term dependencies.
In the present embodiment, the current time is inputAnd previous time implicitHidden state->Input into long-short-term memory neural network by adopting +.>The full connection layer processing of the function is activated to calculate the values of an input gate, a forget gate and an output gate, and the calculation formula is as follows:
，
wherein,,for inputting door parameters +.>、/>For inputting the gate weight parameter, +.>Biasing parameters for the input gate; />For forgetting door parameters, < >>、/>For forgetting gate weight parameter, < >>Bias parameters for forgetting doors; />For outputting door parameters +.>、For outputting the gate weight parameter, +.>To output the gate bias parameters.
Further, candidate memory cellsBy using +.>The fully connected layer of the activation function is processed so as to provide a candidate memory cell +.>The cell value is limited to (-1, 1), and the calculation formula is:
，
wherein,,the candidate memory cell weight parameter; />Bias parameters for candidate memory cells.
Further, the memory cellThe input gate control is used to select memory list>The new data quantity of the element is kept from the old memory cell by means of forgetting gate control>The design is adopted to relieve the gradient vanishing problem and better capture the long-distance dependence in the sequence, and the memory unit formula is as follows:
。
Further, the hidden state is calculated by the output gate and the memory unit，/>Firstly by->Function processing, ensure->The value is within the interval (-1, 1), and the calculation formula is:
。
further, a full connection layer is added after the hidden state for connectingMapping into the required data.
It should be noted that the above-listed related formulas are only some of the calculation contents in the long-short term memory neural network. Through the related formulas and other formulas which are not listed, a prediction model based on the long-term and short-term memory neural network can be constructed.
S15: inputting the test set data into the initial prediction model for evaluation, and adjusting model parameters according to the evaluation result to obtain a trained prediction model, wherein the evaluation method in the prediction model adopts an average relative error MRE to evaluate the prediction result of each cross validation test data, the average relative error formula is as follows,
，
where MRE represents the average relative error, n represents the number of samples,representing the true value +_>Representing the predicted value.
It should be noted that the smaller the value of the MRE, the smaller the relative error between the predicted and actual values, the better the performance of the model, and the final MRE for the complete training and testing in this example is the average MRE of 10 cross-validation.
S16: and inputting the trained prediction model into a prediction module for predicting the thermal runaway trend of the battery module in the actual running process of the vehicle.
In this embodiment, the thermal runaway trend is determined based on the thermal runaway failure samples in the historical operating data of the first and second acquisition modules and the preset thermal runaway failure anomaly data threshold.
The control module acquires the acquired data of the processed second acquisition module, the processed third acquisition module and the processed fourth acquisition module, receives the predicted result information without thermal runaway trend sent by the prediction module, and optimizes the predicted result information through a trained control strategy to obtain the optimized working parameters of the cooling system.
In this embodiment, the method for constructing the trained control policy in the control module specifically includes:
s21: and acquiring historical working data of the battery module and the cooling system, wherein the historical working data comprises data acquired by a second acquisition module, a third acquisition module and a fourth acquisition module.
It should be noted that, the historical operation data of the battery module and the cooling system in step S21 includes: the collected data in the second collecting module are vehicle environment temperature data and battery module temperature data, the collected data in the third collecting module is battery module output power data, and the collected data in the fourth collecting module are compressor rotating speed data, evaporator air inlet data and condenser air inlet data.
It should be noted that the data collected by the second, third and fourth collection modules used in training the support vector regression based control strategy are obtained on a thermal management system model on a simulation software platform. Specifically, by changing input parameters such as output power of a power battery, rotation speed of a compressor, air inlet of an evaporator, air inlet of a condenser, vehicle environment temperature and the like, a battery module temperature of 25-35 ℃ is taken as a target, and a training data sample of a control strategy is obtained.
S22: and (3) carrying out normalization processing on the data acquired by the second acquisition module, the third acquisition module and the fourth acquisition module in the step S21, wherein the normalization formula is as follows:
，
wherein,,representing the original value +_>Representing the normalized value.
S23: and dividing the data acquired by the second acquisition module, the third acquisition module and the fourth acquisition module after normalization processing in the step S22 into a training set and a testing set according to the proportion.
It should be noted that, in this embodiment, in order to ensure the stability and reproducibility of the training model, 10 times of cross-validation was performed, and the training set and the test set respectively account for 80% and 20% of the whole sample.
S24: and inputting the training set data into a control strategy based on support vector regression for training to obtain a preliminary control strategy.
It should be noted that unlike conventional regression methods, support vector regression translates the regression problem into finding an optimal hyperplane in the feature space to place the sample points as close to the hyperplane as possible. Support vector regression has better performance in dealing with nonlinear relationships, noise and outliers than traditional linear regression methods. Support vector regression can be adapted to different types of data sets by selecting the appropriate kernel function and adjusting the boundary parameters.
Further, in the present embodiment, n training samples are provided，/>Is an output value, wherein the m-dimensional vector +.>Representing the ith training sample. Let the linear regression function be->. Wherein (1)>And a nonlinear mapping function, w and b are substitution parameters, and T is a transposed symbol.
In the training process, in order to ensure the existence of the support vector and reduce the influence of abnormal data on the model, a insensitive loss function is introducedAnd relaxation variable->And->Correcting constraint conditions of the functions and optimization targets by adopting the following formula, wherein the optimization targets are the following minimization target functions, and the optimization targets are specific:
，
Wherein,,is an empirical value when in fact +.>Output +.>The current loss is calculated only when the error is greater than the tolerance, allowing for the presence of a certain error,/->Smaller values of (c) indicate smaller requirements of the regression equation for errors. C is penalty factor, defining error greater than +.>The larger the value the cost to be paid for the sample points of (a), the less permissible the regression equation is to allow the existence of error points. By solving this optimization problem, the optimal substitution parameters and relaxation variables can be found +.>。
To simplify the calculation and introduce a kernel function to complete nonlinear conversion, the above formula is converted into solutionIs shown in the following formula:
，
wherein,,a variable representing the lagrangian multiplier.
Further, the radial basis function formula is as follows:
，
wherein,,variance of radial basis function, +.>And->Representing the eigenvector of the input sample.
Similarly, it should be noted that the above-listed related formulas are only some of the calculation contents in the support vector regression model, and the control strategy based on support vector regression can be constructed by using the above-listed related formulas and the other listed formulas.
S25: inputting the test set data into the preliminary control strategy for evaluation, and adjusting model parameters according to an evaluation result to obtain a trained control strategy, wherein the evaluation method in the control strategy adopts a regression square correlation coefficient ，The following formula is shown:
，
wherein n is the number of samples,for real data +.>For fitting data, +.>Is the average of the real data.
It should be noted that it is also possible to provide,the closer the value of 1 is to the 1, the higher the accuracy of the model fitting, and the more accurate the control strategy.
S26: and inputting the trained control strategy into the control module for optimizing the working parameters of the cooling system in the actual running process of the vehicle.
It should be noted that: in this example, the control module receives the predicted result information of no thermal runaway trend sent by the prediction module as the battery module temperature data predicted in the last sampling period.
The battery module temperature data predicted in the last sampling period sent by the prediction module specifically refers to: when the trained control strategy calls the collected data of the second to fourth collecting modules, the predicting module sends the predicted battery module temperature data in the sampling period which is the latest time of the sampling. The prediction model can predict ten minutes at most, for example, when the control module calls the collected data of the second to fourth collecting modules from 8 minutes and 30 seconds, the "predicted battery module temperature in the last sampling period sent by the prediction module" is the predicted battery module temperature data in the last sampling period before 8 minutes and 30 seconds. The predicted battery module temperature data is a predicted result, and can be set directly in the prediction module without additional data processing.
And after the control module receives the prediction result information without thermal runaway trend, the current working state data of the battery module and the current working state data of the cooling system, which are sent by the prediction module, the optimal operation condition solution of the cooling system, namely the optimal parameters of the cooling system, is obtained after the control strategy of the control module is solved. The current working state data of the battery module are acquired data of the second acquisition module and the third acquisition module which are acquired in real time when a trained control strategy is used. The current working state data of the cooling system is acquired data of a fourth acquisition module acquired in real time when a trained control strategy is used.
In this example, the collected data of the second to fourth collecting modules are the original data of the vehicle collected by the collecting modules, and the data can be used by the data processing module.
Therefore, the optimization parameter in this embodiment is a solution of the optimal operation condition of the cooling system, which combines the current operating state of the battery module, the current operating state of the cooling system, and the predicted temperature solution of the battery module in the last sampling period.
The battery module may be maintained to operate at a suitable temperature interval while the cooling system performs the optimization parameters. In this embodiment, the optimum operating temperature range of the battery module is set to 25-35 ℃ ±2℃.
The warning module receives the predicted result information with the thermal runaway trend sent by the predicting module, and displays the safety time and the corresponding thermal runaway grade after being matched with the built-in thermal runaway grade rule, and simultaneously executes warning prompt instructions, inhibition instructions and warning instructions of the corresponding grade.
Preferably, the prediction result information includes three types of battery data, where the three types of battery data are respectively: temperature data including, but not limited to, temperature rise, temperature difference, temperature, voltage data including, but not limited to, voltage rate of change, voltage difference, voltage, current data including, but not limited to, current rate of change, current difference, current; in addition, the predicted result information also comprises the occurrence time of thermal runaway prediction.
In this example, the thermal runaway level built in the alarm module is determined based on the abnormal battery data condition in the prediction result information and the occurrence time condition of thermal runaway prediction, and the determination conditions are specifically:
condition one, judge on the basis of the quantity of each type battery abnormal data in the predicted result information:
first level: when the number of abnormal data in any one/any two/all types of data is three or more, the data is the first grade;
Second level: when the number of abnormal data in any one type/any two types of data is two, the abnormal data is the second level;
third level: when the number of data with abnormality in any type of data is one, the data is the third grade;
condition two, judge according to the predicted occurrence time of thermal runaway:
first level: the predicted occurrence time of thermal runaway is within one minute including one minute;
second level: the predicted occurrence time of thermal runaway is from one minute to more than five minutes including five minutes;
third level: the predicted occurrence time of thermal runaway is more than five minutes to ten minutes including ten minutes.
Thus, the rule for determining the final thermal runaway level is as follows:
if any one of the thermal runaway determination conditions is a first level, or both of the thermal runaway determination conditions are a second level, the final thermal runaway level is defined as level one, and the warning prompt instruction, the suppression instruction and the warning instruction are executed.
For example, the first, second, and third levels of the first and second conditions are denoted by numerals 1, 2, and 3, respectively, and when the two thermal runaway determination conditions are combined to 11, 12, 13, 21, 31, and 22, the final thermal runaway level is defined as level one.
If the combination of the thermal runaway determination conditions is the second level and the third level, the final thermal runaway level is defined as level two, and the warning prompt instruction and the suppression instruction are executed.
If the two thermal runaway judging conditions are the third grade, the final thermal runaway grade is defined as a third grade, and an alarm prompt instruction is executed.
Meanwhile, in the present embodiment, the determination rule of the abnormal parameters of each type of battery data in the prediction result information is as follows:
temperature data determination rule: the temperature rise rate of the minimum serial units is more than or equal to 10 ℃/5s; the temperature difference between the maximum temperature and the minimum temperature of the minimum serial units is more than or equal to 20 ℃, and the duration time is more than or equal to 5s; the minimum serial unit temperature is more than or equal to 60 ℃, and the duration is more than or equal to 5s.
Voltage data determination rule: voltage data determination rule: the minimum series unit voltage is less than 1.0V, which represents undervoltage, and the duration is more than or equal to 5s; the minimum series unit voltage is within the range of 3.65V-3.9V, which indicates overvoltage, and the duration is more than or equal to 5s; the voltage difference between the maximum voltage and the minimum voltage of the minimum series unit is more than or equal to 0.5V, and the duration is more than or equal to 5s; the voltage change rate of the minimum series unit is more than or equal to 25 percent.
Current data determination rules: because the training samples used in the embodiment are large enough, the current change rate, the current difference, the current and the normal current change rate, the current difference and the current different from the historical data according to the current law in the training process of the prediction model are judged, the current change rate, the current difference and the current different from the thermal runaway law in the historical data are judged to be abnormal, and the different threshold range is set in combination with the actual demand and the historical data sample size. In this embodiment, the current difference is the current difference between the minimum parallel units of the battery modules.
In the embodiment, a trend of a current in a historical data sample, which is several minutes before thermal runaway occurs, is used as a criterion for determining the thermal runaway of the current, and a trend of a real-time current is compared with a trend of a current, which is several minutes before thermal runaway occurs in the historical data, in actual prediction, so as to determine whether the real-time current is a normal current, and in addition, a comparison time for determining whether the current is a normal current trend can be set according to actual requirements.
It should be noted that, because there are differences among various components such as the structure and model of different battery packs and the serial-parallel connection mode of the battery cells, the threshold value in the embodiment can be adjusted according to actual requirements.
The thermal runaway level judgment rule of the embodiment combines various data in temperature, voltage and current and occurrence time of thermal runaway prediction, can improve judgment accuracy, reduce false alarm risk and improve safety performance of the battery pack.
The suppression module is preset with suppression parameters for executing the suppression instruction of the alarm module, wherein the suppression parameters are maximum power working parameters of the cooling system.
The cooling system in this example implements the suppression parameter to enable the cooling system to operate at a maximum power operating parameter under thermal runaway suppression conditions to delay the time that thermal runaway occurs.
It should be noted that, the suppression parameters corresponding to different working conditions and the prediction result are consistent, namely, the maximum power working parameter; the main function of the control strategy is to optimize the working parameters of the cooling system according to the prediction result information of the prediction module, so that the power battery works in a proper temperature range. However, the suppression parameter values in the present embodiment are set for the present embodiment, and may be set according to the needs in actual use.
The cooling system of the present invention includes: the system comprises an electric compressor, a condenser, an evaporator, an expansion valve, a cooling pipeline, a cooling air channel, a central processing unit and an independent power supply, wherein the central processing unit is used for receiving and executing the suppression parameters or the optimization parameters, and the independent power supply is used for supplying power to the cooling system when the prediction module predicts the thermal runaway risk.
Further, a fan is arranged at the end part of the cooling air duct, and a central processing unit of the cooling system sends a control instruction to the fan to drive the fan to rotate and circulate cooling air so as to cool the battery, so that the power supply module works in a proper temperature range. It should be noted that, in this embodiment, the rotation speed of the fan is constant, and does not participate in the optimization process of the control module in this embodiment; the adjusting objects of the inhibition parameters or the optimization parameters are only the rotation speed of the compressor, the air inlet quantity of the condenser and the air inlet quantity of the evaporator.
It should be understood that the schematic block diagram of the integrated power cell thermal runaway prediction and suppression system shown in fig. 1 is only illustrative, and the number of the illustrated modules does not limit the scope of the present invention.
The invention not only can predict the thermal runaway risk of the power battery in advance, but also can adjust the working parameters of the cooling system according to the corresponding prediction result, so that the power battery works in a normal temperature interval or delays the occurrence time of the thermal runaway.
It should be noted that, when the prediction model and the control strategy are actually used, the acquired data of each acquisition module are acquired in real time in the running process of the vehicle.
Referring to fig. 2, the invention provides a power battery thermal runaway prediction and suppression integration method, which is applied to a power battery thermal runaway prediction and suppression integration system, and comprises the following steps:
s31: after the vehicle is started, the first acquisition module, the second acquisition module, the third acquisition module and the fourth acquisition module acquire real-time driving condition data according to the sampling frequency of 5 seconds; meanwhile, the acquired real-time data is transmitted to a data processing module connected with the acquisition module;
S32: the data processing module respectively processes the acquired data of each acquisition module;
s33: the testing module acquires the processed acquisition data of the first acquisition module and the second acquisition module, predicts the acquired acquisition data through a trained prediction model, and acquires the prediction result information of whether the battery module has a thermal runaway trend;
s34: if the output result of the prediction model has no thermal runaway risk, returning to the step S31, and sending the prediction result information without thermal runaway trend to the control module; if the output result of the prediction model has a thermal runaway risk, the prediction result information with a thermal runaway trend is sent to the alarm module, and meanwhile, the sampling frequency of each acquisition module is changed to be once in 1 second, so that the thermal runaway prediction is more accurately carried out;
s35: when the control module receives the predicted result information without thermal runaway trend sent by the prediction module, acquiring the processed acquired data of the second acquisition module, the third acquisition module and the fourth acquisition module synchronously, optimizing the predicted battery module temperature in the last sampling period in the predicted result information and the acquired data of the second to fourth acquisition modules through a trained control strategy, acquiring optimized parameters aiming at a cooling system, and sending the optimized parameters to a central processor of the cooling system for execution so as to maintain the power module to work in a normal temperature interval;
S36: when the warning module receives the predicted result information with the thermal runaway trend sent by the predicting module, the warning module displays the safety time and the corresponding level after matching with the built-in thermal runaway level rule, and simultaneously executes warning prompt instructions, inhibition instructions and warning instructions of the corresponding level;
s37: when the suppression module receives the suppression instruction of the alarm module, the suppression module calls a preset suppression parameter and transmits the suppression parameter to the central processing unit of the cooling system for execution, so that the cooling system runs the maximum power working parameter under the thermal runaway suppression working condition, the occurrence time of the thermal runaway is further delayed, and larger loss is avoided.
It will be appreciated that for the various embodiments described above, first, a series of combinations of acts are described for simplicity of illustration, but it will be understood by those skilled in the art that the application is not limited by the order of acts described, as some steps of the application may take on other orders. Second, the embodiments described in the specification are all preferred embodiments, and the modules and operations involved are not necessarily essential to the application.
Finally, the application also provides a storage medium, wherein the storage medium is stored with a computer program, and when the computer program is executed by a processor, the integrated method of predicting and suppressing the thermal runaway of the power battery can be realized.
In addition, each functional unit in the embodiment of the present application may be integrated in one unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated units described above, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable memory. It will be appreciated that the technical solution of the present application, or a part or all of the technical solution contributing to the prior art, may be embodied in the form of a software product stored in a memory, comprising several instructions for causing a computer device, such as a personal computer, a server or a network device, to perform all or part of the steps of the above-mentioned method of the embodiments of the present application. And the aforementioned memory includes: various media capable of storing program codes, such as a U disk, a read-only memory, a random access memory, a mobile hard disk, a magnetic disk or an optical disk.
Those of ordinary skill in the art will appreciate that all or a portion of the steps in the various methods of the above embodiments may be performed by hardware associated with program instructions, and that the program may be stored in a computer readable memory, which may include: u disk, ROM, RAM, magnetic disk or optical disk, etc.
Finally, it should be noted that: the foregoing description is only illustrative of the preferred embodiments of the present invention, and although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications may be made to the embodiments described, or equivalents may be substituted for elements thereof, and any modifications, equivalents, improvements or changes may be made without departing from the spirit and principles of the present invention.
Claims (10)
1. The integrated system for predicting and suppressing the thermal runaway of the power battery comprises a battery module and a cooling system, and is characterized by further comprising:
the first acquisition module is used for acquiring battery minimum single voltage data, battery current data, battery minimum single temperature data, battery module SOH data and battery module SOC data of the battery module in real time;
The second acquisition module is used for acquiring vehicle environment temperature data and battery module temperature data of the battery module in real time;
the third acquisition module is used for acquiring the output power data of the battery module in real time;
the fourth acquisition module is used for acquiring the rotation speed data of the compressor, the air inlet quantity data of the evaporator and the air inlet quantity data of the condenser of the cooling system in real time;
the data processing module is used for respectively processing the acquired data of each acquisition module;
the prediction module is used for obtaining the acquired data of the processed first acquisition module and the processed second acquisition module, and predicting the acquired data through a trained prediction model to obtain the prediction result information of whether the battery module has a thermal runaway trend;
the control module is used for acquiring the acquired data of the processed second acquisition module, the processed third acquisition module and the processed fourth acquisition module, receiving the predicted result information without the thermal runaway trend sent by the predicting module, and optimizing the predicted result information through a trained control strategy to obtain optimized parameters of the cooling system;
the warning module receives the predicted result information with the thermal runaway trend sent by the predicting module, and after the predicted result information is matched with the built-in thermal runaway grade rule, the safety time and the corresponding thermal runaway grade are displayed, and a warning prompt instruction, a suppression instruction and a warning instruction of the corresponding grade are executed at the same time;
And the suppression module executes the suppression instruction sent by the alarm module.
2. The integrated power cell thermal runaway prediction and suppression system according to claim 1, wherein the data processing module performs processing operations of data cleaning, feature selection, feature conversion, data scaling and data encoding on the collected data of each collection module, respectively.
3. The integrated system for predicting and suppressing thermal runaway of a power battery according to claim 2, wherein the method for constructing a trained prediction model in the prediction module specifically comprises:
s11: acquiring historical working data of the battery module, wherein the historical working data comprises data acquired by a first acquisition module and a second acquisition module;
s12: normalizing the historical working data in the step S11;
s13: dividing the historical working data subjected to normalization processing in the step S12 into a training set and a testing set according to a proportion;
s14: inputting the training set data into a prediction model based on a neural network for training to obtain an initial prediction model;
s15: inputting the test set data into the initial prediction model for evaluation, and adjusting model parameters according to the evaluation result to obtain a trained prediction model, wherein an evaluation method in the prediction model adopts an average relative error (MRE) to evaluate the prediction result of each cross verification test data;
S16: and inputting the trained prediction model into a prediction module for predicting the thermal runaway trend of the battery module in the actual running process of the vehicle.
4. The integrated power cell thermal runaway prediction and suppression system according to claim 3, wherein the prediction result information includes three types of battery data, wherein the three types of battery data are respectively: temperature data including, but not limited to, temperature rise, temperature difference, temperature, voltage data including, but not limited to, voltage rate of change, voltage difference, voltage, current data including, but not limited to, current rate of change, current difference, current; in addition, the predicted result information also comprises the occurrence time of thermal runaway prediction.
5. The integrated power cell thermal runaway prediction and suppression system according to claim 4, wherein the built-in thermal runaway level is determined based on a number condition of cell abnormality data in prediction result information and a occurrence time condition of thermal runaway prediction, and the determination condition is specifically:
condition one, judge on the basis of the quantity of each type battery abnormal data in the predicted result information:
first level: when the number of abnormal data in any one/any two/all types of data is three or more, the data is the first grade;
Second level: when the number of abnormal data in any one type/any two types of data is two, the abnormal data is the second level;
third level: when the number of data with abnormality in any type of data is one, the data is the third grade;
condition two, judge according to the predicted occurrence time of thermal runaway:
first level: the predicted occurrence time of thermal runaway is within one minute including one minute;
second level: the predicted occurrence time of thermal runaway is from one minute to more than five minutes including five minutes;
third level: the predicted occurrence time of thermal runaway is more than five minutes to ten minutes including ten minutes;
thus, the rule for determining the final thermal runaway level is as follows:
if any one of the thermal runaway judging conditions is a first grade or both of the two thermal runaway judging conditions are a second grade, defining the final thermal runaway grade as a grade one, and executing an alarm prompt instruction, a suppression instruction and an alarm instruction;
if the combination of the thermal runaway judging conditions is the second grade and the third grade, defining the final thermal runaway grade as a grade two, and executing an alarm prompt instruction and a suppression instruction;
if the two thermal runaway judging conditions are the third grade, the final thermal runaway grade is defined as a third grade, and an alarm prompt instruction is executed.
6. The integrated power cell thermal runaway prediction and suppression system of claim 5, wherein the method for constructing the trained control strategy in the control module specifically comprises:
s21: acquiring historical working data of the battery module and the cooling system, wherein the historical working data comprises data acquired by a second acquisition module, a third acquisition module and a fourth acquisition module;
s22: normalizing the data acquired by the second acquisition module, the third acquisition module and the fourth acquisition module in the step S21;
s23: dividing the data acquired by the second acquisition module, the third acquisition module and the fourth acquisition module after normalization processing in the step S22 into a training set and a testing set according to a proportion;
s24: inputting the training set data into a control strategy based on support vector regression for training to obtain a preliminary control strategy;
s25: inputting the test set data into the preliminary control strategy for evaluation, and adjusting model parameters according to an evaluation result to obtain a trained control strategy, wherein a regression square correlation coefficient is adopted in an evaluation method in the control strategy;
s26: and inputting the trained control strategy into the control module for optimizing the working parameters of the cooling system in the actual running process of the vehicle.
7. The integrated system for thermal runaway prediction and suppression of a power battery according to claim 6, wherein the predicted result information of no thermal runaway trend sent by the prediction module and received by the control module is a battery module temperature predicted in a last sampling period, and the optimization parameter is a solution of an optimal operation condition of the cooling system obtained after solving by a control strategy of the control module according to the battery module temperature predicted in the last sampling period, the current operation state of the battery module and the current operation state of the cooling system sent by the prediction module;
and the suppression module is internally preset with a suppression parameter for executing a suppression instruction of the alarm module, wherein the suppression parameter is a maximum power working parameter of the cooling system.
8. The integrated power cell thermal runaway prediction and suppression system of claim 7, wherein the cooling system comprises: the system comprises an electric compressor, a condenser, an evaporator, an expansion valve, a cooling pipeline, a cooling air channel, a central processing unit and an independent power supply, wherein the central processing unit is used for receiving and executing the suppression parameters or the optimization parameters, and the independent power supply is used for supplying power to the cooling system when the prediction module predicts the thermal runaway risk.
9. The integrated method for predicting and suppressing the thermal runaway of the power battery, which is applied to the integrated system for predicting and suppressing the thermal runaway of the power battery according to any one of claims 1 to 8, is characterized by comprising the following steps:
s31: after the vehicle is started, the first acquisition module, the second acquisition module, the third acquisition module and the fourth acquisition module acquire real-time driving condition data according to the sampling frequency of 5 seconds; meanwhile, the acquired real-time data is transmitted to a data processing module connected with the acquisition module;
s32: the data processing module respectively processes the acquired data of each acquisition module;
s33: the prediction module acquires the acquired data of the processed first acquisition module and the processed second acquisition module, predicts the acquired data through a trained prediction model, and acquires the prediction result information of whether the battery module has a thermal runaway trend or not;
s34: if the output result of the prediction model has no thermal runaway risk, returning to the step S31, and sending the prediction result information without thermal runaway trend to the control module; if the output result of the prediction model has a thermal runaway risk, the prediction result information with a thermal runaway trend is sent to the alarm module, and meanwhile, the sampling frequency of each acquisition module is changed to be once in 1 second, so that the thermal runaway prediction is more accurately carried out;
S35: when the control module receives the predicted result information without thermal runaway trend sent by the prediction module, acquiring the processed acquired data of the second acquisition module, the third acquisition module and the fourth acquisition module synchronously, optimizing the predicted battery module temperature in the last sampling period in the predicted result information and the acquired data of the second to fourth acquisition modules through a trained control strategy, acquiring optimized parameters aiming at a cooling system, and sending the optimized parameters to a central processor of the cooling system for execution so as to maintain the power module to work in a normal temperature interval;
s36: when the warning module receives the predicted result information with the thermal runaway trend sent by the predicting module, the warning module displays the safety time and the corresponding level after matching with the built-in thermal runaway level rule, and simultaneously executes warning prompt instructions, inhibition instructions and warning instructions of the corresponding level;
s37: when the suppression module receives the suppression instruction of the alarm module, the suppression module calls a preset suppression parameter and transmits the suppression parameter to the central processing unit of the cooling system for execution, so that the cooling system runs the maximum power working parameter under the thermal runaway suppression working condition, the occurrence time of the thermal runaway is further delayed, and larger loss is avoided.
10. A storage medium having a computer program stored thereon, which when executed by a processor, implements the integrated power cell thermal runaway prediction and suppression method of claim 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310967238.0A CN116683061B (en) | 2023-08-03 | 2023-08-03 | Power battery thermal runaway prediction and suppression integrated system, method and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310967238.0A CN116683061B (en) | 2023-08-03 | 2023-08-03 | Power battery thermal runaway prediction and suppression integrated system, method and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116683061A true CN116683061A (en) | 2023-09-01 |
| CN116683061B CN116683061B (en) | 2023-09-29 |
Family
ID=87785886
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310967238.0A Active CN116683061B (en) | 2023-08-03 | 2023-08-03 | Power battery thermal runaway prediction and suppression integrated system, method and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116683061B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117159953A (en) * | 2023-11-03 | 2023-12-05 | 安徽中科中涣智能装备股份有限公司 | Fire monitoring, prevention and control system and method for battery box |
| CN117346418A (en) * | 2023-12-05 | 2024-01-05 | 珠海科创储能科技有限公司 | Control method and device of water chilling unit, storage medium and electronic equipment |
| CN118144571A (en) * | 2024-05-11 | 2024-06-07 | 太原科技大学 | Multi-model fusion power battery thermal runaway early warning system and method |
| CN118277873A (en) * | 2024-06-03 | 2024-07-02 | 深圳康普盾科技股份有限公司 | Monitoring data analysis method and system of battery management system |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104573854A (en) * | 2014-12-23 | 2015-04-29 | 国家电网公司 | Iron steel electricity consumption forecasting method and device |
| CN113437371A (en) * | 2021-05-19 | 2021-09-24 | 湖南大学 | Early warning system and early warning method for thermal runaway of lithium ion battery of new energy automobile |
| CN113506924A (en) * | 2021-06-17 | 2021-10-15 | 重庆金康动力新能源有限公司 | Thermal runaway early warning method and system for battery pack |
| CN114066095A (en) * | 2021-11-26 | 2022-02-18 | 华能新能源股份有限公司山西分公司 | Method for predicting wind power generation by using artificial neural network and support vector regression model |
| CN114696010A (en) * | 2020-12-31 | 2022-07-01 | 宁德时代新能源科技股份有限公司 | Thermal runaway processing method, device and system and storage medium |
| CN114889433A (en) * | 2022-04-29 | 2022-08-12 | 奇瑞新能源汽车股份有限公司 | Thermal runaway alarm system and method for battery of electric vehicle |
| US11443569B1 (en) * | 2021-10-30 | 2022-09-13 | Beta Air, Llc | Systems and methods for battery management for a vehicle |
| CN115267555A (en) * | 2022-08-08 | 2022-11-01 | 上海乐驾智慧能源科技有限公司 | Battery SOH (State of health) evaluation system of energy storage system based on battery multipoint temperature measurement |
| US20230023641A1 (en) * | 2021-07-22 | 2023-01-26 | X Development Llc | Automated detection of chemical component of moving object |
| CN116263878A (en) * | 2022-10-09 | 2023-06-16 | 北京理工大学 | Lithium ion battery thermal runaway risk prediction method and device |
-
2023
- 2023-08-03 CN CN202310967238.0A patent/CN116683061B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104573854A (en) * | 2014-12-23 | 2015-04-29 | 国家电网公司 | Iron steel electricity consumption forecasting method and device |
| CN114696010A (en) * | 2020-12-31 | 2022-07-01 | 宁德时代新能源科技股份有限公司 | Thermal runaway processing method, device and system and storage medium |
| CN113437371A (en) * | 2021-05-19 | 2021-09-24 | 湖南大学 | Early warning system and early warning method for thermal runaway of lithium ion battery of new energy automobile |
| CN113506924A (en) * | 2021-06-17 | 2021-10-15 | 重庆金康动力新能源有限公司 | Thermal runaway early warning method and system for battery pack |
| US20230023641A1 (en) * | 2021-07-22 | 2023-01-26 | X Development Llc | Automated detection of chemical component of moving object |
| US11443569B1 (en) * | 2021-10-30 | 2022-09-13 | Beta Air, Llc | Systems and methods for battery management for a vehicle |
| CN114066095A (en) * | 2021-11-26 | 2022-02-18 | 华能新能源股份有限公司山西分公司 | Method for predicting wind power generation by using artificial neural network and support vector regression model |
| CN114889433A (en) * | 2022-04-29 | 2022-08-12 | 奇瑞新能源汽车股份有限公司 | Thermal runaway alarm system and method for battery of electric vehicle |
| CN115267555A (en) * | 2022-08-08 | 2022-11-01 | 上海乐驾智慧能源科技有限公司 | Battery SOH (State of health) evaluation system of energy storage system based on battery multipoint temperature measurement |
| CN116263878A (en) * | 2022-10-09 | 2023-06-16 | 北京理工大学 | Lithium ion battery thermal runaway risk prediction method and device |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117159953A (en) * | 2023-11-03 | 2023-12-05 | 安徽中科中涣智能装备股份有限公司 | Fire monitoring, prevention and control system and method for battery box |
| CN117159953B (en) * | 2023-11-03 | 2024-03-15 | 安徽中科中涣智能装备股份有限公司 | Fire monitoring, prevention and control system and method for battery box |
| CN117346418A (en) * | 2023-12-05 | 2024-01-05 | 珠海科创储能科技有限公司 | Control method and device of water chilling unit, storage medium and electronic equipment |
| CN117346418B (en) * | 2023-12-05 | 2024-03-22 | 珠海科创储能科技有限公司 | Control method and device of water chilling unit, storage medium and electronic equipment |
| CN118144571A (en) * | 2024-05-11 | 2024-06-07 | 太原科技大学 | Multi-model fusion power battery thermal runaway early warning system and method |
| CN118144571B (en) * | 2024-05-11 | 2024-07-02 | 太原科技大学 | Multi-model fusion power battery thermal runaway early warning system and method |
| CN118277873A (en) * | 2024-06-03 | 2024-07-02 | 深圳康普盾科技股份有限公司 | Monitoring data analysis method and system of battery management system |
| CN118277873B (en) * | 2024-06-03 | 2024-10-11 | 深圳康普盾科技股份有限公司 | Monitoring data analysis method and system of battery management system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116683061B (en) | 2023-09-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116683061B (en) | Power battery thermal runaway prediction and suppression integrated system, method and storage medium | |
| Wang et al. | A hybrid prognostics approach for estimating remaining useful life of rolling element bearings | |
| Li et al. | Battery life estimation based on cloud data for electric vehicles | |
| Xue et al. | Fault diagnosis and abnormality detection of lithium-ion battery packs based on statistical distribution | |
| Li et al. | Big data driven lithium-ion battery modeling method based on SDAE-ELM algorithm and data pre-processing technology | |
| Zhou et al. | A novel health indicator for on-line lithium-ion batteries remaining useful life prediction | |
| Wu et al. | Health prognosis with optimized feature selection for lithium-ion battery in electric vehicle applications | |
| Ma et al. | State of health estimation and remaining useful life prediction for lithium-ion batteries by improved particle swarm optimization-back propagation neural network | |
| CN113447828B (en) | Lithium battery temperature estimation method and system based on Bayesian neural network | |
| Wang et al. | Multiple indicators-based health diagnostics and prognostics for energy storage technologies using fuzzy comprehensive evaluation and improved multivariate grey model | |
| Zhao et al. | Model-based fault diagnosis methods for systems with stochastic process–a survey | |
| Liu et al. | An improved method of state of health prediction for lithium batteries considering different temperature | |
| CN115718258B (en) | Battery fault detection | |
| He et al. | SOH estimation for lithium-ion batteries: an improved GPR optimization method based on the developed feature extraction | |
| Wei et al. | A multi-scale learning approach for remaining useful life prediction of lithium-ion batteries based on variational mode decomposition and Monte Carlo sampling | |
| CN113359048A (en) | Indirect prediction method for remaining service life of lithium ion battery | |
| Wang et al. | A hybrid method for online cycle life prediction of lithium‐ion batteries | |
| Jeng et al. | Evaluation of cell inconsistency in lithium-ion battery pack using the autoencoder network model | |
| Hussein et al. | A review of battery state of charge estimation and management systems: Models and future prospective | |
| CN117445755A (en) | Cloud computing-based remote monitoring system for batteries of electric vehicle | |
| Wang et al. | Research on interactive multi‐model fault diagnosis method of Li‐ion battery based on noise suppression | |
| CN116736141A (en) | Lithium battery energy storage safety management system and method | |
| CN116505040A (en) | Energy storage management and diagnosis system and method for iron-chromium flow battery | |
| Cui et al. | Long-sequence voltage series forecasting for internal short circuit early detection of lithium-ion batteries | |
| Fang et al. | Early micro-short circuit fault diagnosis of lithium battery pack based on Pearson correlation coefficient and KPCA |
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 |