New energy automobile power battery safety monitoring method
By combining LSTM and CNN neural networks, multi-angle monitoring and prediction of new energy vehicle power batteries is achieved, and the problem that traditional methods cannot accurately detect battery status is solved, which improves the safety and reliability of the battery, extends the life span and reduces the Maintenance cost.
Patent Information
- Application Number
- CN202411400260.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-09
- Publication Date
- 2024-12-06
AI Technical Summary
The traditional safety monitoring method of power batteries in new energy vehicles relies on traditional electrical parameters and safety indicators, and cannot fully and effectively detect the internal state of the battery, resulting in insufficient safety and reliability.
Using a method combining LSTM neural network and CNN neural network, the multi-angle monitoring and prediction of new energy vehicle power batteries is achieved by extracting and classifying historical battery operation data and monitoring data, and safe detection is carried out by fusing LSTM-CNN model.
It improves the safety and reliability of the battery, extends the battery life, reduces maintenance costs, and improves the competitiveness of new energy vehicles in the market.
Smart Images

Figure CN119087235A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of new energy vehicle batteries, and in particular to a new energy vehicle power battery safety monitoring method. Background Art
[0002] The safety of new energy vehicle power batteries has always been one of the hot topics in the development of new energy vehicles. Traditional power battery safety monitoring methods mainly rely on traditional electrical parameters and safety indicators, but these parameters and indicators do not accurately reflect the internal state of the battery, and therefore cannot fully and effectively detect the safety status of the battery. In recent years, deep learning technology has been widely used in the field of safety monitoring of new energy vehicle power batteries. Among them, the method combining LSTM neural network and CNN neural network has attracted much attention due to its good prediction performance and applicability.
[0003] LSTM neural network is a deep learning model that can process sequence data and is often used in speech recognition, natural language processing, time series prediction and other fields. In the safety monitoring of power batteries for new energy vehicles, LSTM neural network can use historical battery operation data, especially data such as cycle number, temperature and current, to predict the remaining battery life and safety.
[0004] The CNN neural network is a deep learning model for image and image recognition tasks, which usually uses convolutional layers for feature extraction. In the safety monitoring of new energy vehicle power batteries, the CNN neural network can form a three-dimensional image input model for monitoring data such as battery internal temperature, current and voltage, and perform feature extraction and classification to achieve more accurate battery safety monitoring.
[0005] The method combining LSTM neural network and CNN neural network can comprehensively monitor and predict the power batteries of new energy vehicles from multiple angles and improve the safety and reliability of batteries. Summary of the invention
[0006] The purpose of the present invention is to provide a new energy vehicle power battery safety monitoring method, which can achieve more accurate power battery status prediction through data drive, improve battery safety and reliability, and has broad application value in the design, manufacture and use of power batteries.
[0007] To achieve the above object, the present invention provides a new energy vehicle power battery safety monitoring method, comprising the following steps: S1. Collect power battery data and perform preprocessing to obtain preprocessed power battery data; S2, using the preprocessed power battery data as the input of the LSTM network and the CNN network; S3, LSTM network learns the historical time series data of power batteries, transforms each one-dimensional time series data into two-dimensional time series data for spatial domain modeling, and extracts the changing characteristic variables; S4, CNN network classifies and learns existing battery safety features; S5. Integrate the LSTM network and the CNN network to achieve safety detection of power batteries.
[0008] Preferably, in step S1, the power battery data includes single cell voltage, single cell temperature, vehicle speed, maximum temperature and minimum temperature.
[0009] Preferably, in step S1, preprocessing includes data cleaning, missing value processing, outlier processing, normalization and standardization.
[0010] Preferably, in step S4, the specific operations are: using a CNN network to extract features from a three-dimensional image, using the extracted feature data to train the CNN network, and optimizing the trained CNN network through a back propagation algorithm and a gradient descent method.
[0011] Preferably, in step S4 and step S5, the LSTM network and the CNN network are integrated, and the specific operations are: Integrate processed time series data and 3D images into a unified dataset; The LSTM network and the CNN network are combined by stacking, series connection or parallel connection to obtain the LSTM-CNN fusion model; The LSTM-CNN fusion model processes time series data and 3D image data simultaneously to achieve comprehensive analysis of multi-dimensional indicators; Multi-dimensional indicators include the battery's historical status, internal structure, and temperature distribution.
[0012] The present invention also provides an application of a new energy vehicle power battery safety monitoring method in power battery design.
[0013] Therefore, the present invention adopts the above-mentioned new energy vehicle power battery safety monitoring method, and its technical effects are as follows: (1) Improving battery safety and reliability: This technical solution improves the accuracy of prediction and monitoring of the power battery status of new energy vehicles, monitors the battery status in real time, reduces the risk of battery failure and accidents, and further improves the safety and reliability of power batteries; (2) Extending battery life: By real-time monitoring of the power battery status, the number of times the battery is charged and discharged is reduced, reducing battery wear and tear, thereby extending the battery life; (3) Reduce maintenance costs: This technical solution can help new energy vehicle manufacturers to perform regular maintenance and timely detect battery status, reduce unnecessary losses and costs, and reduce maintenance costs; (4) Increase the competitiveness of new energy vehicles in the market: By improving the safety and reliability of power batteries and extending battery life, the competitiveness of new energy vehicles in the market can be improved; The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS
[0014] Figure 1 The present invention is a flow chart of a method for safety monitoring of a power battery of a new energy vehicle. DETAILED DESCRIPTION
[0015] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.
[0016] Unless otherwise defined, technical or scientific terms used in the present invention shall have the common meanings understood by one having ordinary skills in the field to which the present invention belongs.
[0017] Embodiment 1
[0018] like Figure 1 FIG. 1 is a flow chart of a method for monitoring the safety of a power battery of a new energy vehicle according to the present invention, which specifically includes the following steps: S1. Collect power battery data, including cell voltage, cell temperature, vehicle speed, maximum temperature and minimum temperature, etc., and perform preprocessing (data cleaning, missing value processing, outlier processing, normalization and standardization) to obtain preprocessed power battery data; S2, using the preprocessed power battery data as the input of the LSTM network and the CNN network; S3. Based on the Keras deep learning framework, build an LSTM network. The LSTM network learns the historical time series data of the power battery, converts each one-dimensional time series data into two-dimensional time series data for spatial domain modeling, and extracts the changing feature variables at the same time; S4, CNN network classifies and learns existing battery safety features; The specific operation is: use the CNN network to extract features from the three-dimensional image, use the extracted feature data to train the CNN network, and optimize the trained CNN network through the back propagation algorithm and gradient descent method. The CNN model will learn how to map the input three-dimensional image data to the corresponding state classification label. After the training is completed, the CNN model is tested and verified to evaluate its performance in the power battery state classification task.
[0019] S5. Integrate the LSTM network and the CNN network to achieve safety detection of power batteries.
[0020] The LSTM network and CNN network are integrated. The specific operations are as follows: Integrate processed time series data and 3D images into a unified dataset; The LSTM network and the CNN network are combined by stacking, series connection or parallel connection to obtain the LSTM-CNN fusion model; The LSTM-CNN fusion model processes time series data and three-dimensional image data simultaneously to achieve comprehensive analysis of multi-dimensional indicators; including the battery's historical status, internal structure, temperature distribution and other information, enabling the model to have a more comprehensive understanding of the battery's operating conditions.
[0021] Applying the fusion model to real-time data streams enables real-time monitoring of the power battery status. Once the model detects an abnormality, an alarm can be issued to the relevant system or operator in a timely manner to ensure the safe operation of the system.
[0022] Therefore, the present invention adopts the above-mentioned new energy vehicle power battery safety monitoring method, realizes more accurate power battery status prediction through data drive, improves battery safety and reliability, and has broad application value in the design, manufacture and use of power batteries.
[0023] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.
Claims
1. A new energy vehicle power battery safety monitoring method, characterized in that: The following steps are involved: S1. Collect power battery data and perform preprocessing to obtain preprocessed power battery data; S2, using the preprocessed power battery data as the input of the LSTM network and the CNN network; S3, LSTM network learns the historical time series data of power batteries, transforms each one-dimensional time series data into two-dimensional time series data for spatial domain modeling, and extracts the changing characteristic variables; S4, CNN network classifies and learns existing battery safety features; S5. Integrate the LSTM network and the CNN network to achieve safety detection of power batteries.
2. A new energy vehicle power battery safety monitoring method according to claim 1, characterized in that: In step S1 , the power battery data includes single cell voltage, single cell temperature, vehicle speed, maximum temperature and minimum temperature.
3. A new energy vehicle power battery safety monitoring method according to claim 2, characterized in that: In step S1, preprocessing includes data cleaning, missing value processing, outlier processing, normalization and standardization.
4. A new energy vehicle power battery safety monitoring method according to claim 3, characterized in that: In step S4, the specific operations are: using the CNN network to extract features from the three-dimensional image, using the extracted feature data to train the CNN network, and optimizing the trained CNN network through the back propagation algorithm and the gradient descent method.
5. A new energy vehicle power battery safety monitoring method according to claim 4, characterized in that: In step S4 and step S5, the LSTM network and the CNN network are fused. The specific operations are as follows: Integrate processed time series data and 3D images into a unified dataset; The LSTM network and the CNN network are combined by stacking, series connection or parallel connection to obtain the LSTM-CNN fusion model; The LSTM-CNN fusion model processes time series data and 3D image data simultaneously to achieve comprehensive analysis of multi-dimensional indicators; Multi-dimensional indicators include the battery's historical status, internal structure, and temperature distribution.
