A thermal runaway propagation prediction method and prediction system

Abstract

The present invention discloses a method and system for predicting the spread of thermal runaway, the method comprising the following steps: a maximum temperature rise rate, an operating temperature and a maximum surface temperature are preset in a control device; the control device receives the battery surface temperature fed back by a second temperature sensor; the control device receives the ambient temperature fed back by a first temperature sensor and a thermal cloud map fed back by a thermal imager; the control device obtains the real-time voltage value, real-time current value and real-time state of charge of a battery pack; the control device obtains a predicted temperature value according to the ambient temperature, the thermal cloud map, the battery surface temperature, the real-time voltage value, the real-time current value and the real-time state of charge; the control device calculates the temperature rise rate according to the predicted temperature value; when the temperature rise rate is greater than the preset maximum temperature rise rate, the control device outputs a control instruction indicating the presence of a thermal runaway problem; the prediction method disclosed in the present application can realize the quantitative prediction of the spread of thermal runaway, improve the safety of lithium batteries during operation, and avoid accidents.

Description

Technical Field

[0001] The present invention relates to the field of battery safety technology, and in particular to a thermal runaway propagation prediction method and prediction system. Background Art

[0002] The thermal runaway problem of lithium-ion batteries is a key issue restricting the industrial development of the electric vehicle industry. Lithium batteries may experience thermal runaway under extreme working conditions. Thermal runaway of a single cell may cause heat spread, leading to safety problems in the module, battery box and even the entire vehicle where the thermal runaway battery is located. Therefore, the heat spread problem caused by thermal runaway must be studied and evaluated.

[0003] Existing thermal runaway propagation prediction methods are generally based on experiments or models, but both prediction methods have the problem of large errors in prediction results and cannot quantitatively estimate the thermal runaway propagation of lithium batteries.

[0004] It can be seen that the existing technology still needs to be improved and enhanced. Summary of the invention

[0005] In view of the above-mentioned deficiencies in the prior art, an object of the present invention is to provide a method for predicting the spread of thermal runaway, which can achieve quantitative prediction of the spread of thermal runaway and improve the safety of lithium batteries during operation.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A method for predicting the spread of thermal runaway is implemented by a thermal runaway spread prediction system, wherein the thermal runaway spread prediction system comprises a control device and a first temperature sensor, a thermal imager and a plurality of second temperature sensors electrically connected to the control device respectively; the prediction method comprises the steps of:

[0008] S100, the maximum temperature rise rate, operating temperature and maximum surface temperature are preset in the control device;

[0009] S200, the control device receives the battery surface temperature fed back by the second temperature sensor;

[0010] S300, the control device receives the ambient temperature fed back by the first temperature sensor and the thermal cloud map fed back by the thermal imager;

[0011] S400, the control device obtains the real-time voltage value, real-time current value and real-time state of charge of the battery pack;

[0012] S500, the control device obtains a predicted temperature value according to the ambient temperature, the thermal cloud map, the battery surface temperature, the real-time voltage value, the real-time current value and the real-time state of charge;

[0013] S600, the control device calculates the temperature rise rate according to the predicted temperature value;

[0014] S700: When the temperature rise rate is greater than a preset maximum temperature rise rate, the control device outputs a control instruction indicating that a thermal runaway problem exists.

[0015] In the method for predicting the spread of thermal runaway, the following steps are further included between step S200 and step S300:

[0016] S210: When the battery surface temperature is ≥ the preset operating temperature, execute steps S300 to S700.

[0017] In the method for predicting the spread of thermal runaway, step S500 specifically includes the following steps:

[0018] S501, the control device normalizes the ambient temperature, the thermal cloud map, the battery surface temperature, the real-time voltage value, the real-time current value and the real-time state of charge;

[0019] S502, the control device inputs the normalized ambient temperature, battery surface temperature, real-time voltage value, real-time current value and real-time state of charge into the long short-term memory neural network to obtain a first feature;

[0020] S503, the control device inputs the normalized thermal cloud map into the three-dimensional convolutional neural network to obtain the second feature;

[0021] S504: The control device inputs the first feature and the second feature into a temperature prediction network to obtain a predicted temperature value.

[0022] In the prediction method for the spread of thermal runaway, the long short-term memory neural network includes a first LSTM layer, a second LSTM layer, a third LSTM layer, a first fully connected layer, and a second fully connected layer connected in sequence; the three-dimensional convolutional neural network includes a first three-dimensional convolutional layer, a second three-dimensional convolutional layer, a third three-dimensional convolutional layer, a fourth three-dimensional convolutional layer, a third fully connected layer, and a fourth fully connected layer connected in sequence; the temperature prediction network includes a feature fusion layer, a fifth fully connected layer, and a sixth fully connected layer connected in sequence.

[0023] In the method for predicting the spread of thermal runaway, in step S600, the temperature rise rate is calculated as follows:

[0024]

[0025] Among them, ΔT is the temperature rise rate; T(k) represents the predicted temperature value at the current moment, in degrees Celsius; T(k-1) represents the predicted temperature value at the previous moment, in degrees Celsius; t represents the time between two moments, in seconds.

[0026] In the method for predicting the spread of thermal runaway, the step S200 further includes the following steps:

[0027] S220: When the battery surface temperature is greater than or equal to a preset maximum surface temperature, the control device outputs a control instruction indicating that a thermal runaway problem exists.

[0028] The present invention also correspondingly provides a thermal runaway propagation prediction system, comprising a control device and a first temperature sensor, a thermal imager and a plurality of second temperature sensors electrically connected to the control device respectively; the first temperature sensor is used to obtain the ambient temperature of the environment in which the battery pack is located, the thermal imager is used to obtain the thermal cloud map of the battery pack, and the plurality of second temperature sensors are used to obtain the surface temperature of a single battery in the battery pack; the control device is electrically connected to the positive and negative poles of the battery pack respectively.

[0029] The thermal runaway propagation prediction system further includes an alarm system, which is electrically connected to the control device and is used to display alarm information output by the control device.

[0030] In the thermal runaway propagation prediction system, the alarm system includes a display screen, an indicator light and a broadcasting device electrically connected to the control device respectively; the display screen is used to display the thermal cloud map, predicted temperature value and temperature rise rate of the battery pack; the broadcasting device and the indicator light are used to output sound and light alarm signals respectively.

[0031] Beneficial effects:

[0032] The present invention provides a method for predicting the spread of thermal runaway, which can obtain the predicted temperature value of a single battery in a battery pack and calculate the temperature rise rate, intuitively obtain the temperature change of the lithium battery during operation, and realize quantitative prediction of the spread of thermal runaway, thereby realizing early warning of thermal runaway problems, reducing the degree of harm of accidents or avoiding accidents, and improving the safety of lithium batteries during operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 A first logic flow chart of the prediction method provided by the present invention;

[0034] Figure 2 A second logic flow chart of the prediction method provided by the present invention;

[0035] Figure 3 A logic flow chart of an embodiment of step S500 provided by the present invention;

[0036] Figure 4 This is a schematic diagram of the structure of the prediction system provided by the present invention.

[0037] Explanation of the symbols of the main components: 1-control device, 2-first temperature sensor, 3-thermal imager, 4-second temperature sensor, 5-alarm system, 51-display screen, 52-indicator light, 53-broadcasting device. DETAILED DESCRIPTION

[0038] The present invention provides a thermal runaway propagation prediction method and prediction system. In order to make the purpose, technical solution and effect of the present invention clearer and more specific, the present invention is further described in detail with reference to the accompanying drawings and examples.

[0039] In the description of the present invention, it should be understood that the terms "installation", "connection" and the like should be understood in a broad sense, and a person skilled in the art can understand the specific meanings of the above terms in the present invention according to specific circumstances.

[0040] See also Figures 1 to 3 The present invention provides a method for predicting the spread of thermal runaway, which is implemented by a thermal runaway spread prediction system. The thermal runaway spread prediction system includes a control device and a first temperature sensor, a thermal imager and a plurality of second temperature sensors electrically connected to the control device respectively; the prediction method includes the steps of:

[0041] S100. The control device is preset with a maximum temperature rise rate, an operating temperature and a maximum surface temperature. In one embodiment, the maximum temperature rise rate is 1°C, the operating temperature is 60°C, and the maximum surface temperature is 300°C. The set values ​​of the maximum temperature rise rate, the operating temperature and the maximum surface temperature are related to the working environment of the battery pack and the brand of the battery pack.

[0042] S200, the control device receives the battery surface temperature fed back by the second temperature sensor; the battery surface temperature is the surface temperature of a single battery in the battery pack, and a plurality of the second temperature sensors correspond one-to-one to the single batteries in the battery pack.

[0043] S300, the control device receives the ambient temperature fed back by the first temperature sensor and the thermal cloud map fed back by the thermal imager; the first temperature sensor detects the ambient temperature of the environment in which the battery pack is located, and the thermal imager obtains the thermal cloud map of the battery pack.

[0044] S400, the control device obtains the real-time voltage value, real-time current value and real-time state of charge of the battery pack; the control device is electrically connected to the positive and negative electrodes of the battery pack respectively to obtain the real-time current value, real-time voltage value and real-time state of charge of the battery pack.

[0045] S500, the control device obtains a predicted temperature value according to the ambient temperature, the thermal cloud map, the battery surface temperature, the real-time voltage value, the real-time current value and the real-time state of charge; the predicted temperature value includes several predicted temperature values, and the several predicted temperature values ​​correspond one to one to a single battery in the battery pack.

[0046] S600, the control device calculates a temperature rise rate according to the predicted temperature value; the temperature rise rate includes a plurality of temperature rise rates, and the plurality of temperature rise rates correspond one to one to individual batteries in the battery pack.

[0047] S700: When the temperature rise rate is greater than a preset maximum temperature rise rate, the control device outputs a control instruction indicating that a thermal runaway problem exists.

[0048] The method for predicting the spread of thermal runaway disclosed in the present invention can obtain the predicted temperature value of a single battery in a battery pack and calculate the temperature rise rate of the single battery pack, so as to intuitively obtain the temperature change of each battery when the lithium battery pack is working, and realize quantitative prediction of the spread of thermal runaway, thereby realizing early warning of thermal runaway problems, reducing the degree of harm of accidents or avoiding accidents, and improving the safety of lithium batteries when working.

[0049] For further information, see Figure 2 , the step between step S200 and step S300 also includes the following steps:

[0050] S210. When the battery surface temperature is ≥ the preset operating temperature, execute steps S300 to S700. Specifically, when the surface temperature of any single battery in the battery pack is greater than or equal to the preset operating temperature, the control device starts to predict the temperature value and calculate the temperature difference rate, that is, the control device executes steps S300 to S700; when the surface temperatures of all single batteries in the battery pack are lower than the operating temperature, the first temperature sensor and the thermal imager stop working, and the control device stops obtaining the predicted temperature value and calculating the temperature rise rate, so as to reduce the working energy consumption and working temperature of the prediction system.

[0051] For further information, see Figure 3 , the step S500 specifically includes the steps of:

[0052] S501, the control device normalizes the ambient temperature, the thermal cloud map, the battery surface temperature, the real-time voltage value, the real-time current value and the real-time state of charge;

[0053] S502, the control device inputs the normalized ambient temperature, battery surface temperature, real-time voltage value, real-time current value and real-time state of charge into the long short-term memory neural network to obtain a first feature;

[0054] S503, the control device inputs the normalized thermal cloud map into the three-dimensional convolutional neural network to obtain the second feature;

[0055] S504: The control device inputs the first feature and the second feature into a temperature prediction network to obtain a predicted temperature value.

[0056] Furthermore, the long short-term memory neural network includes a first LSTM layer, a second LSTM layer, a third LSTM layer, a first fully connected layer and a second fully connected layer connected in sequence; the three-dimensional convolutional neural network includes a first three-dimensional convolutional layer, a second three-dimensional convolutional layer, a third three-dimensional convolutional layer, a fourth three-dimensional convolutional layer, a third fully connected layer and a fourth fully connected layer connected in sequence; the temperature prediction network includes a feature fusion layer, a fifth fully connected layer and a sixth fully connected layer connected in sequence; the first feature and the second feature are feature fused in a splicing manner in the feature fusion layer, and the feature fusion results are input into the fifth fully connected layer and the sixth fully connected layer in sequence, and the sixth fully connected layer outputs the predicted temperature value; in one embodiment, the control device uses a mean square error loss function to update the long short-term memory neural network and the three-dimensional convolutional neural network.

[0057] Furthermore, in step S600, the temperature rise rate is calculated as follows:

[0058]

[0059] Among them, ΔT is the temperature rise rate; T(k) represents the predicted temperature value at the current moment, in degrees Celsius; T(k-1) represents the predicted temperature value at the previous moment, in degrees Celsius; t represents the time between two moments, in seconds.

[0060] When the temperature rise rate of any single battery in the battery pack is greater than or equal to the preset maximum temperature rise rate, the control device marks the battery as a thermal runaway battery, and records the moment when the temperature rise rate exceeds the maximum temperature rise rate as the moment of battery thermal runaway; the control device can also calculate the sum of the number of batteries with thermal runaway problems in the battery pack based on the obtained several temperature rise rates, and realize statistics and early warning of the spread of thermal runaway, which is convenient for researchers to conduct subsequent research on the spread of thermal runaway of the battery pack and provide data support for the improvement of the battery pack.

[0061] For further information, see Figure 2 , the step S200 further includes the following steps:

[0062] S220. When the battery surface temperature is greater than or equal to the preset maximum surface temperature, the control device outputs a control instruction that a thermal runaway problem exists; that is, the determination conditions for whether a thermal runaway problem exists include two conditions, which improves the accuracy and reliability of the prediction results output by the prediction method.

[0063] See also Figure 4The present invention also provides a thermal runaway propagation prediction system accordingly, comprising a control device 1 and a first temperature sensor 2, a thermal imager 3 and a plurality of second temperature sensors 4 electrically connected to the control device 1 respectively; the first temperature sensor 2 is used to obtain the ambient temperature of the environment where the battery pack is located, the thermal imager is used to obtain the thermal cloud map of the battery pack, and the plurality of second temperature sensors 4 are used to obtain the surface temperature of a single battery in the battery pack; the control device 1 is electrically connected to the positive and negative electrodes of the battery pack respectively.

[0064] For further information, see Figure 4 The thermal runaway propagation prediction system also includes an alarm system 5, which is electrically connected to the control device 1. The alarm system 5 is used to display the alarm information output by the control device 1, so that the user can quickly learn about the safety status of the battery and avoid accidents.

[0065] For further information, see Figure 4 The alarm system 5 includes a display screen 51, an indicator light 52 and a broadcasting device 53 which are electrically connected to the control device 1 respectively; the display screen 51 is used to display the thermal cloud map, predicted temperature value and temperature rise rate of the battery pack; the broadcasting device 53 and the indicator light 52 are used to output sound and light alarm signals respectively; in one embodiment, the display screen 51, the indicator light 52 and the broadcasting device 53 are arranged on the center console of the car; the broadcasting device 53 can be a speaker or a sound box.

[0066] It is understandable that those skilled in the art can make equivalent substitutions or changes based on the technical solution and inventive concept of the present invention, and all these changes or substitutions should fall within the protection scope of the present invention.

Claims

1. A method for predicting the spread of thermal runaway, characterized in that: The method is implemented by a thermal runaway propagation prediction system, which includes a control device and a first temperature sensor, a thermal imager, and a plurality of second temperature sensors electrically connected to the control device respectively; the prediction method includes the steps of: S100, the maximum temperature rise rate, operating temperature and maximum surface temperature are preset in the control device; S200, the control device receives the battery surface temperature fed back by the second temperature sensor, wherein the battery surface temperature is the surface temperature of a single battery in the battery pack, and a plurality of the second temperature sensors correspond one to one to the single batteries in the battery pack; S300, the control device receives the ambient temperature fed back by the first temperature sensor and the thermal cloud map fed back by the thermal imager, wherein the first temperature sensor detects the ambient temperature of the environment where the battery pack is located, and the thermal imager obtains the thermal cloud map of the battery pack; S400, the control device obtains the real-time voltage value, real-time current value and real-time state of charge of the battery pack; S500, the control device obtains a predicted temperature value according to the ambient temperature, the thermal cloud map, the battery surface temperature, the real-time voltage value, the real-time current value and the real-time state of charge, wherein the predicted temperature value includes a plurality of predicted temperature values, and the plurality of predicted temperature values ​​correspond one to one to a single battery in the battery pack; Wherein, the step S500 specifically includes the steps of: S501, the control device normalizes the ambient temperature, the thermal cloud map, the battery surface temperature, the real-time voltage value, the real-time current value and the real-time state of charge; S502, the control device inputs the normalized ambient temperature, battery surface temperature, real-time voltage value, real-time current value and real-time state of charge into the long short-term memory neural network to obtain a first feature; S503, the control device inputs the normalized thermal cloud map into the three-dimensional convolutional neural network to obtain the second feature; S504, the control device inputs the first feature and the second feature into the temperature prediction network to obtain a predicted temperature value; S600, the control device calculates a temperature rise rate according to the predicted temperature value, wherein the temperature rise rate includes a plurality of temperature rise rates, and the plurality of temperature rise rates correspond one to one to a single battery in the battery pack; S700: When the temperature rise rate is greater than a preset maximum temperature rise rate, the control device outputs a control instruction indicating that a thermal runaway problem exists; The steps between step S200 and step S300 also include: S210, when the battery surface temperature is ≥ the preset operating temperature, execute steps S300 to S700. Specifically, when the surface temperature of any single battery in the battery pack is greater than or equal to the preset operating temperature, the control device starts to predict the temperature value and calculate the temperature rise rate, that is, the control device executes steps S300 to S700; when the surface temperatures of all single batteries in the battery pack are lower than the operating temperature, the first temperature sensor and the thermal imager stop working, and the control device stops obtaining the predicted temperature value and calculating the temperature rise rate; S220: When the battery surface temperature is greater than or equal to a preset maximum surface temperature, the control device outputs a control instruction indicating that a thermal runaway problem exists.

2. A method for predicting the spread of thermal runaway according to claim 1, characterized in that: The long short-term memory neural network includes a first LSTM layer, a second LSTM layer, a third LSTM layer, a first fully connected layer and a second fully connected layer connected in sequence; the three-dimensional convolutional neural network includes a first three-dimensional convolutional layer, a second three-dimensional convolutional layer, a third three-dimensional convolutional layer, a fourth three-dimensional convolutional layer, a third fully connected layer and a fourth fully connected layer connected in sequence; the temperature prediction network includes a feature fusion layer, a fifth fully connected layer and a sixth fully connected layer connected in sequence.

3. A method for predicting the spread of thermal runaway according to claim 1, characterized in that: In step S600, the temperature rise rate is calculated as follows: Among them, ΔT is the temperature rise rate; T(k) represents the predicted temperature value at the current moment, in degrees Celsius; T(k-1) represents the predicted temperature value at the previous moment, in degrees Celsius; t represents the time between two moments, in seconds.

4. A thermal runaway propagation prediction system, characterized in that: The thermal runaway propagation prediction system is used to implement the prediction method as described in any one of claims 1 to 3. The prediction system specifically includes a control device and a first temperature sensor, a thermal imager and a plurality of second temperature sensors electrically connected to the control device respectively; the first temperature sensor is used to obtain the ambient temperature of the environment in which the battery pack is located, the thermal imager is used to obtain the thermal cloud map of the battery pack, and the plurality of second temperature sensors are used to obtain the surface temperature of a single battery in the battery pack; the control device is electrically connected to the positive and negative poles of the battery pack respectively.

5. A thermal runaway propagation prediction system according to claim 4, characterized in that: It also includes an alarm system, which is electrically connected to the control device and is used to display the alarm information output by the control device.

6. A thermal runaway propagation prediction system according to claim 5, characterized in that: The alarm system includes a display screen, an indicator light and a broadcasting device which are electrically connected to the control device respectively; the display screen is used to display the thermal cloud map, predicted temperature value and temperature rise rate of the battery pack; the broadcasting device and the indicator light are used to output sound and light alarm signals respectively.

Images