Multi-feature coupled energy storage thermal runaway monitoring and disposal method

Abstract

The invention discloses a multi-feature coupled energy storage thermal runaway monitoring and disposal method, which relates to the technical field of batteries, and comprises the following steps: approximately representing the internal temperature change gradient of a battery by using radial finite difference, and constructing a one-dimensional thermal model of the battery based on a thermal balance principle; based on the influence of battery aging on the actual capacity and ohmic internal resistance of the battery as evaluation indexes, constructing a double-index aging model of the battery; combining the one-dimensional thermal model and the double-index aging model of the battery to construct a multi-feature coupling model; selecting a particle swarm optimization algorithm to optimize model parameters of the multi-feature coupling model; designing a thermal runaway early warning method; when the monitoring system detects that the battery has a thermal runaway risk, linkage control of the fire extinguishing system is triggered immediately, and meanwhile, the electrical switch cuts off a battery circuit. Characteristic parameters required by thermal runaway early warning are calculated or estimated more accurately through the multi-characteristic coupling model, and the reliability of thermal runaway early warning is ensured.

Description

Technical Field

[0001] The present invention relates to the field of battery technology, and in particular to a multi-feature coupled energy storage thermal runaway monitoring and disposal method. Background Art

[0002] Thermal runaway of lithium-ion batteries is a phenomenon of internal exothermic chain reaction, which usually causes a sharp rise in the temperature inside the battery, instability of the internal structure of the battery or departure from normal working state. If allowed to spread, it will lead to complete failure of the battery. With the rapid development of electric vehicles, lithium-ion batteries have been widely used, and the number of thermal runaway accidents has also increased year by year. Therefore, the research on thermal runaway early warning methods has become an important topic.

[0003] At present, the thermal runaway warning methods all consider the detection of a single signal. In actual situations, it may be affected by noise, detection accuracy or other faults, so it is necessary to consider a warning strategy that combines multiple signals to ensure the reliability of thermal runaway warning. The effect of thermal runaway warning depends on the accuracy of the parameters and the methods used. Compared with independent equivalent circuit models, multi-scale coupling models can more accurately calculate or estimate the characteristic parameters required for thermal runaway warning. Summary of the invention

[0004] In order to solve the above technical problems, a multi-feature coupled energy storage thermal runaway monitoring and disposal method is provided. This technical solution solves the problems raised in the above background technology.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] A multi-feature coupled energy storage thermal runaway monitoring and disposal method, comprising:

[0007] The temperature gradient inside the battery is approximated using radial finite differences, and a one-dimensional thermal model of the battery is constructed based on the principle of thermal balance.

[0008] Based on the actual battery capacity and the impact of battery aging on ohmic internal resistance as evaluation indicators, a dual-index aging model for the battery is constructed;

[0009] Installing a gas sensor to monitor the concentration of characteristic gases in real time, wherein the characteristic gases include hydrogen and carbon monoxide;

[0010] Use environmental soundprint technology to monitor the sound characteristics of the battery when it is running;

[0011] Combined with the one-dimensional thermal model and dual-index aging model of the battery, the impact of heat and aging on the battery environmental parameters is analyzed, and a multi-feature coupling model is constructed. The environmental parameters include the concentration of characteristic gases and the sound characteristics during operation.

[0012] Measure the response of the multi-feature coupling model under special excitation, and select the particle swarm algorithm to optimize the model parameters of the multi-feature coupling model;

[0013] The thermal runaway warning is converted into an internal short circuit detection problem. Based on a multi-feature coupling model, a thermal runaway warning method with four characteristic parameters, namely battery voltage data, thermal signal, characteristic gas concentration and sound characteristics during operation, is designed. The battery cell data includes the actual battery capacity and ohmic internal resistance.

[0014] When the monitoring system detects that the battery is at risk of thermal runaway, it immediately triggers the linkage control of the fire protection system and the electrical switch cuts off the battery circuit.

[0015] Preferably, the method of using radial finite difference to approximate the temperature gradient inside the battery and constructing a one-dimensional thermal model of the battery based on the thermal balance principle specifically includes:

[0016] Obtain the ambient temperature and battery surface temperature, and use the thermal radiation formula to calculate the thermal radiation of the battery;

[0017] Use the thermal convection formula to calculate the convection heat dissipation of the battery;

[0018] Add the thermal radiation and convection heat dissipation of the battery and output it as the heat dissipation rate of the battery;

[0019] Obtain the thermal balance principle that the battery complies with;

[0020] The winding body and electrolyte inside the battery are considered to be uniformly distributed, and the radial temperature variation law of the battery is calculated based on the principle of thermal balance;

[0021] Starting from the radial center, the battery is divided into at least one node along the radial direction inside, and the temperature of each node is approximately represented by a finite difference;

[0022] Build a one-dimensional thermal model of the battery based on the temperature of each node;

[0023] The thermal radiation formula is:

[0024] ,

[0025] In the formula, is the thermal radiation of the battery, is the thermal emissivity, is the Stefan-Boltzmann constant, is the battery surface temperature, is the ambient temperature;

[0026] The heat convection formula is:

[0027] ,

[0028] In the formula, To dissipate heat from the battery through convection, is the convection coefficient;

[0029] The thermal balance principle is:

[0030] ,

[0031] In the formula, is the material density, is the specific heat capacity, is a function of temperature, For time, is the thermal conductivity, is the heat generation rate function, is the heat dissipation function of the battery, is the partial derivative of the temperature function with respect to time, is the vector differential operator;

[0032] The radial temperature variation rule of the battery is:

[0033] ,

[0034] In the formula, is the temperature function at a point y away from the center of the battery, is the volume of the battery.

[0035] Preferably, the dual-index aging model of the battery is constructed based on the actual battery capacity and the influence of the ohmic internal resistance on the battery aging as evaluation indicators, and specifically includes:

[0036] Use the Arrhenius formula to describe the battery capacity decay model;

[0037] The internal resistance growth is divided into three parts: the internal resistance growth caused by date, the internal resistance growth caused by cycle and the internal resistance growth affected by randomness;

[0038] Use the internal resistance growth formula to describe the ohmic internal resistance growth model of the battery;

[0039] The capacity attenuation model and the ohmic internal resistance growth model are combined and output as a dual-index aging model for the battery;

[0040] The Arrhenius formula is:

[0041] ,

[0042] In the formula, is the capacity loss rate, is the cycle charge and discharge rate, is the reaction activation energy, is the molar gas constant, is the average temperature inside the battery;

[0043] The internal resistance growth formula is:

[0044] ,

[0045] In the formula, is the battery internal resistance growth value, is the internal resistance growth affected by randomness, is the internal resistance growth coefficient caused by date, is the internal resistance growth coefficient caused by the cycle, is the number of cycles.

[0046] Preferably, the use of environmental soundprint technology to monitor the sound characteristics of the battery during operation specifically includes:

[0047] Use highly sensitive sound sensors deployed at the location of the battery pack;

[0048] Preprocess the collected sound signal to remove the noise signal in the sound signal to obtain the environmental voiceprint signal;

[0049] Use spectrum analysis to extract features from environmental voiceprint signals;

[0050] The process of decomposing an environmental soundprint signal into at least one cosine function basic signal superposition;

[0051] Outputting the coefficients of each cosine-type function basic signal as the sound characteristics when the battery is running;

[0052] The cosine type function is:

[0053] ,

[0054] In the formula, is a cosine function signal, is the amplitude of the signal, is the angular frequency of the signal, is the phase angle of the signal.

[0055] Preferably, the one-dimensional thermal model and the dual-index aging model of the battery are combined to analyze the influence of heat and aging on the battery environmental parameters, and the construction of the multi-feature coupling model specifically includes:

[0056] Obtain an equivalent circuit model of a lithium-ion battery;

[0057] Changes in battery environmental parameters are accompanied by changes in charge and discharge currents, which are input into the equivalent circuit model to calculate the real-time changes in terminal voltage and state of charge;

[0058] The equivalent circuit model transfers the internal resistance heat generation data into the one-dimensional thermal model;

[0059] The one-dimensional thermal model combines the ambient temperature to estimate the temperature inside the battery in real time, and calculates the changes in the parameters of the equivalent circuit model under the influence of thermal effects;

[0060] The dual-index aging model substitutes the calculation results of the equivalent circuit model and the one-dimensional thermal model, outputs the battery capacity decay rate, and returns the updated ohmic internal resistance to the equivalent circuit model and the one-dimensional thermal model;

[0061] The state of charge is:

[0062] ,

[0063] In the formula, is the state of charge, is the remaining capacity of the battery, is the actual capacity of the battery.

[0064] Preferably, the step of measuring the response of the multi-feature coupling model under special excitation and selecting a particle swarm algorithm to optimize the model parameters of the multi-feature coupling model specifically includes:

[0065] S101: Design at least one series of experiments of low rate cycling and constant capacity pulse discharge under constant temperature conditions;

[0066] S102: Obtaining an initial parameter group of a multi-feature coupling model;

[0067] S103: Setting key parameters required for the particle swarm algorithm, wherein the key parameters include an upper limit of the number of iterations, a minimum and maximum value of the inertia weight, and an initial value of the acceleration factor;

[0068] S104: calculating the inverse proportional function value of the data error generated by the model parameters of each multi-feature coupling model in the series of experiments, and recording it as the individual extreme value, and recording the maximum value of the individual extreme value as the group extreme value;

[0069] S105: Calculating the inertia coefficient according to the logistic regression equation;

[0070] S106: calculating the value function and the change speed of the model parameters of the multi-feature coupling model, and calculating the inverse proportional function value of the data error generated in the series of experiments according to the value function;

[0071] S107: Determine whether the inverse proportional function value of the data error generated in the series of experiments is higher than the individual extreme value. If so, update the individual extreme value and the corresponding value function. If not, do not output.

[0072] S108: judging whether the sum of the inverse proportional function values ​​of the data errors generated in the series of experiments is higher than the group extreme value, if so, updating the group extreme value and the corresponding value function, if not, no output;

[0073] S109: Determine whether the number of iterations reaches the upper limit of the number of iterations. If so, output the model parameters of the current multi-feature coupling model. If not, return to step S105.

[0074] The logistic regression equation is:

[0075]

[0076] In the formula, is the inertia coefficient, , are the maximum and minimum values ​​of the inertia weight, Q is the upper limit of the number of iterations, and q is the current number of iterations;

[0077] The value function and change rate of the model parameters of the multi-feature coupling model are:

[0078]

[0079] In the formula, , are the changing speed and value function of the model parameters of the i-th multi-feature coupling model at the q-th iteration, and j is the iteration parameter.

[0080] Preferably, the thermal runaway warning is converted into an internal short circuit detection problem, and based on a multi-feature coupling model, a thermal runaway warning method for designing four characteristic parameters of battery voltage data, thermal signal, characteristic gas concentration, and sound characteristics during operation specifically includes:

[0081] The changes of four characteristic parameters, namely, battery voltage data, thermal signal, characteristic gas concentration and sound characteristics during operation, are input into the multi-characteristic coupling model to obtain the threshold range of each parameter under safe conditions;

[0082] Determine whether the battery voltage data exceeds the first parameter threshold range, if so, output the proportion of the exceeding portion, if not, do not output;

[0083] Determine whether the thermal signal exceeds the second parameter threshold range, if so, output the proportion of the exceeding portion, if not, do not output;

[0084] Determine whether the characteristic gas concentration exceeds the third parameter threshold range, if so, output the proportion of the exceeding portion, if not, no output;

[0085] Determine whether the sound feature during operation exceeds the fourth parameter threshold range, if so, output the proportion of the exceeding portion, if not, do not output;

[0086] Use the warning signal level formula to calculate the current thermal runaway warning signal level;

[0087] The warning signal level formula is:

[0088] ,

[0089] In the formula, The warning signal level is , , , are the calculation weights of battery voltage data, thermal signal, characteristic gas concentration and sound characteristics during operation, , , , They are the proportion of excess data of battery voltage, thermal signal, characteristic gas concentration and sound characteristics during operation.

[0090] Preferably, when the monitoring system detects that the battery has a thermal runaway risk, the linkage control of the fire protection system is immediately triggered, and the electrical switch cuts off the battery circuit, specifically including:

[0091] Get the warning signal level;

[0092] Determine whether the warning signal level is greater than a first preset threshold value, if so, control the electrical switch to cut off the battery circuit, if not, dispatch relevant personnel to perform on-site maintenance;

[0093] Determine whether the warning signal level is greater than the second preset threshold. If so, trigger the linkage control of the fire protection system, start the fire extinguishing device, release the fire extinguishing agent, and start the smoke exhaust system. If not, no output is made.

[0094] Compared with the prior art, the present invention has the following beneficial effects:

[0095] It can accurately describe the mechanism of the battery. The more scales there are, the more comprehensively it can describe the state of the battery. A multi-feature coupling model suitable for the entire life cycle and a wide temperature range has been established. The multi-feature coupling model can more accurately calculate or estimate the characteristic parameters required for thermal runaway warning, ensure the reliability of thermal runaway warning, and achieve thermal runaway warning with high confidence. BRIEF DESCRIPTION OF THE DRAWINGS

[0096] Figure 1 It is a flow chart of the multi-feature coupled energy storage thermal runaway monitoring and disposal method of the present invention;

[0097] Figure 2 A flow chart of a method for constructing a one-dimensional thermal model of a battery according to the present invention;

[0098] Figure 3 A flow chart of a method for constructing a dual-index aging model for a battery according to the present invention;

[0099] Figure 4 This is a flow chart of the method for monitoring the sound characteristics of a battery during operation according to the present invention;

[0100] Figure 5 A flow chart of the method for constructing a multi-feature coupling model of the present invention;

[0101] Figure 6 A flow chart of a method for selecting a particle swarm algorithm to optimize model parameters of a multi-feature coupling model according to the present invention;

[0102] Figure 7 The flowchart of the thermal runaway early warning method of the present invention is as follows;

[0103] Figure 8 This is a flow chart of the linkage control method for triggering a fire protection system of the present invention. DETAILED DESCRIPTION

[0104] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0105] Reference Figure 1 As shown, a multi-feature coupled energy storage thermal runaway monitoring and disposal method includes:

[0106] The temperature gradient inside the battery is approximated using radial finite differences, and a one-dimensional thermal model of the battery is constructed based on the principle of thermal balance.

[0107] Based on the actual battery capacity and the impact of battery aging on ohmic internal resistance as evaluation indicators, a dual-index aging model for the battery is constructed;

[0108] Installing a gas sensor to monitor the concentration of characteristic gases in real time, wherein the characteristic gases include hydrogen and carbon monoxide;

[0109] Use environmental soundprint technology to monitor the sound characteristics of the battery when it is running;

[0110] Combined with the one-dimensional thermal model and dual-index aging model of the battery, the impact of heat and aging on the battery environmental parameters is analyzed, and a multi-feature coupling model is constructed. The environmental parameters include the concentration of characteristic gases and the sound characteristics during operation.

[0111] Measure the response of the multi-feature coupling model under special excitation, and select the particle swarm algorithm to optimize the model parameters of the multi-feature coupling model;

[0112] The thermal runaway warning is converted into an internal short circuit detection problem. Based on a multi-feature coupling model, a thermal runaway warning method with four characteristic parameters, namely battery voltage data, thermal signal, characteristic gas concentration and sound characteristics during operation, is designed. The battery cell data includes the actual battery capacity and ohmic internal resistance.

[0113] When the monitoring system detects that the battery is at risk of thermal runaway, it immediately triggers the linkage control of the fire protection system and the electrical switch cuts off the battery circuit.

[0114] As shown in Figure 2, the temperature gradient inside the battery is approximated using radial finite differences, and a one-dimensional thermal model of the battery is constructed based on the principle of thermal balance, specifically including:

[0115] Obtain the ambient temperature and battery surface temperature, and use the thermal radiation formula to calculate the thermal radiation of the battery;

[0116] Use the thermal convection formula to calculate the convection heat dissipation of the battery;

[0117] Add the thermal radiation and convection heat dissipation of the battery and output it as the heat dissipation rate of the battery;

[0118] Obtain the thermal balance principle that the battery complies with;

[0119] The winding body and electrolyte inside the battery are considered to be uniformly distributed, and the radial temperature variation law of the battery is calculated based on the principle of thermal balance;

[0120] Starting from the radial center, the battery is divided into at least one node along the radial direction inside, and the temperature of each node is approximately represented by a finite difference;

[0121] Build a one-dimensional thermal model of the battery based on the temperature of each node;

[0122] The thermal radiation formula is:

[0123] ,

[0124] In the formula, is the thermal radiation of the battery, is the thermal emissivity, is the Stefan-Boltzmann constant, is the battery surface temperature, is the ambient temperature;

[0125] The heat convection formula is:

[0126] ,

[0127] In the formula, To dissipate heat from the battery through convection, is the convection coefficient;

[0128] The thermal balance principle is:

[0129] ,

[0130] In the formula, is the material density, is the specific heat capacity, is a function of temperature, For time, is the thermal conductivity, is the heat generation rate function, is the heat dissipation function of the battery, is the partial derivative of the temperature function with respect to time, is the vector differential operator;

[0131] The radial temperature variation rule of the battery is:

[0132] ,

[0133] In the formula, is the temperature function at a point y away from the center of the battery, is the volume of the battery.

[0134] A system in any equilibrium state, in the absence of macroscopic work, relies on direct interaction between the system and the outside world to change the system state is called thermal contact (or heat exchange). When two thermodynamic systems are in thermal contact, the original equilibrium state of the system will generally change; after a sufficiently long time, the heat exchange stops; at this time, the two systems can be considered to be in thermal equilibrium. If the state does not change when the two systems are in thermal contact, it means that the two systems are in thermal equilibrium with each other. It can be considered that the two systems in thermal equilibrium have the same degree of hotness and coldness and the same temperature.

[0135] As shown in FIG3 , based on the actual battery capacity and the influence of the ohmic internal resistance on the battery aging as evaluation indicators, the dual-index aging model of the battery is constructed, which specifically includes:

[0136] Use the Arrhenius formula to describe the battery capacity decay model;

[0137] The internal resistance growth is divided into three parts: the internal resistance growth caused by date, the internal resistance growth caused by cycle and the internal resistance growth affected by randomness;

[0138] Use the internal resistance growth formula to describe the ohmic internal resistance growth model of the battery;

[0139] The capacity attenuation model and the ohmic internal resistance growth model are combined and output as a dual-index aging model for the battery;

[0140] The Arrhenius formula is:

[0141] ,

[0142] In the formula, is the capacity loss rate, is the cycle charge and discharge rate, is the reaction activation energy, is the molar gas constant, is the average temperature inside the battery;

[0143] The internal resistance growth formula is:

[0144] ,

[0145] In the formula, is the battery internal resistance growth value, is the internal resistance growth affected by randomness, is the internal resistance growth coefficient caused by date, is the internal resistance growth coefficient caused by the cycle, is the number of cycles.

[0146] The Arrhenius equation is generally applicable to situations where the temperature change range is not large. At this time, the reaction activation energy does not change much, and the Arrhenius equation has good applicability. If the temperature range is large, the Arrhenius equation will produce errors.

[0147] As shown in FIG4 , the sound characteristics of the battery during operation are monitored by using the environmental soundprint technology, including:

[0148] Use highly sensitive sound sensors deployed at the location of the battery pack;

[0149] Preprocess the collected sound signal to remove the noise signal in the sound signal to obtain the environmental voiceprint signal;

[0150] Use spectrum analysis to extract features from environmental voiceprint signals;

[0151] The process of decomposing an environmental soundprint signal into at least one cosine function basic signal superposition;

[0152] Outputting the coefficients of each cosine-type function basic signal as the sound characteristics when the battery is running;

[0153] The cosine type function is:

[0154] ,

[0155] In the formula, is a cosine function signal, is the amplitude of the signal, is the angular frequency of the signal, is the phase angle of the signal.

[0156] Spectral analysis is a technique that decomposes complex noise signals into simpler signals. Many physical signals can be expressed as the sum of many simple signals of different frequencies. Spectral analysis is the process of finding out the information of a signal at different frequencies, such as amplitude, power, intensity or phase.

[0157] As shown in Figure 5, the one-dimensional thermal model and dual-index aging model of the battery are combined to analyze the impact of heat and aging on the battery environmental parameters, and the construction of a multi-feature coupling model specifically includes:

[0158] Obtain an equivalent circuit model of a lithium-ion battery;

[0159] Changes in battery environmental parameters are accompanied by changes in charge and discharge currents, which are input into the equivalent circuit model to calculate the real-time changes in terminal voltage and state of charge;

[0160] The equivalent circuit model transfers the internal resistance heat generation data into the one-dimensional thermal model;

[0161] The one-dimensional thermal model combines the ambient temperature to estimate the temperature inside the battery in real time, and calculates the changes in the parameters of the equivalent circuit model under the influence of thermal effects;

[0162] The dual-index aging model substitutes the calculation results of the equivalent circuit model and the one-dimensional thermal model, outputs the battery capacity decay rate, and returns the updated ohmic internal resistance to the equivalent circuit model and the one-dimensional thermal model;

[0163] The state of charge is:

[0164] ,

[0165] In the formula, is the state of charge, is the remaining capacity of the battery, is the actual capacity of the battery.

[0166] The state of charge is the ratio of the remaining capacity of a battery after it has been used for a period of time or has been unused for a long time to its capacity in a fully charged state. It is usually expressed as a percentage, and its value range is 0~1. When SOC=0, it means that the battery is fully discharged, and when SOC=1, it means that the battery is fully charged.

[0167] As shown in FIG6 , the response of the multi-feature coupling model under special excitation is measured, and the particle swarm algorithm is selected to optimize the model parameters of the multi-feature coupling model, specifically including:

[0168] S101: Design at least one series of experiments of low rate cycling and constant capacity pulse discharge under constant temperature conditions;

[0169] S102: Obtaining an initial parameter group of a multi-feature coupling model;

[0170] S103: Setting key parameters required for the particle swarm algorithm, wherein the key parameters include an upper limit of the number of iterations, a minimum and maximum value of the inertia weight, and an initial value of the acceleration factor;

[0171] S104: calculating the inverse proportional function value of the data error generated by the model parameters of each multi-feature coupling model in the series of experiments, and recording it as the individual extreme value, and recording the maximum value of the individual extreme value as the group extreme value;

[0172] S105: Calculating the inertia coefficient according to the logistic regression equation;

[0173] S106: calculating the value function and the change speed of the model parameters of the multi-feature coupling model, and calculating the inverse proportional function value of the data error generated in the series of experiments according to the value function;

[0174] S107: Determine whether the inverse proportional function value of the data error generated in the series of experiments is higher than the individual extreme value. If so, update the individual extreme value and the corresponding value function. If not, do not output.

[0175] S108: judging whether the sum of the inverse proportional function values ​​of the data errors generated in the series of experiments is higher than the group extreme value, if so, updating the group extreme value and the corresponding value function, if not, no output;

[0176] S109: Determine whether the number of iterations reaches the upper limit of the number of iterations. If so, output the model parameters of the current multi-feature coupling model. If not, return to step S105.

[0177] The logistic regression equation is:

[0178]

[0179] In the formula, is the inertia coefficient, , are the maximum and minimum values ​​of the inertia weight, Q is the upper limit of the number of iterations, and q is the current number of iterations;

[0180] The value function and change rate of the model parameters of the multi-feature coupling model are:

[0181]

[0182] In the formula, , are the changing speed and value function of the model parameters of the i-th multi-feature coupling model at the q-th iteration, and j is the iteration parameter.

[0183] The idea of ​​particle swarm optimization algorithm originates from the simulation of the foraging process of bird flocks. The solution domain of the optimization problem is analogous to the flight space of birds, and the candidate solutions of the optimization problem are analogous to the birds in the flock. Each bird is abstracted as a massless and volumeless particle in the flight space. The optimal solution to the problem is equivalent to the food that the flock is looking for. Its core idea is to use the information sharing of individuals in the group to make the movement of the entire group evolve from disorder to order in the problem-solving space, so as to obtain a feasible solution to the problem.

[0184] As shown in Figure 7, the thermal runaway warning is converted into an internal short circuit detection problem. Based on the multi-feature coupling model, a thermal runaway warning method is designed with four characteristic parameters: battery voltage data, thermal signal, characteristic gas concentration, and sound characteristics during operation. Specifically, the method includes:

[0185] The changes of four characteristic parameters, namely, battery voltage data, thermal signal, characteristic gas concentration and sound characteristics during operation, are input into the multi-characteristic coupling model to obtain the threshold range of each parameter under safe conditions;

[0186] Determine whether the battery voltage data exceeds the first parameter threshold range, if so, output the proportion of the exceeding portion, if not, do not output;

[0187] Determine whether the thermal signal exceeds the second parameter threshold range, if so, output the proportion of the exceeding portion, if not, do not output;

[0188] Determine whether the characteristic gas concentration exceeds the third parameter threshold range, if so, output the proportion of the exceeding portion, if not, no output;

[0189] Determine whether the sound feature during operation exceeds the fourth parameter threshold range, if so, output the proportion of the exceeding portion, if not, do not output;

[0190] Use the warning signal level formula to calculate the current thermal runaway warning signal level;

[0191] The warning signal level formula is:

[0192] ,

[0193] In the formula, The warning signal level is , , , are the calculation weights of battery voltage data, thermal signal, characteristic gas concentration and sound characteristics during operation, , , , They are the proportion of excess data of battery voltage, thermal signal, characteristic gas concentration and sound characteristics during operation.

[0194] At present, thermal runaway warning methods all consider the detection of a single signal. In actual situations, it may be affected by noise, detection accuracy or other faults. Therefore, it is necessary to consider a warning strategy that combines multiple signals to ensure the reliability of thermal runaway warning.

[0195] As shown in FIG8 , when the monitoring system detects that the battery has a risk of thermal runaway, the linkage control of the fire protection system is immediately triggered, and the electrical switch cuts off the battery circuit. Specifically, the following steps are performed:

[0196] Get the warning signal level;

[0197] Determine whether the warning signal level is greater than a first preset threshold value, if so, control the electrical switch to cut off the battery circuit, if not, dispatch relevant personnel to perform on-site maintenance;

[0198] Determine whether the warning signal level is greater than the second preset threshold. If so, trigger the linkage control of the fire protection system, start the fire extinguishing device, release the fire extinguishing agent, and start the smoke exhaust system. If not, no output is made.

[0199] The data processing unit uses a multiple coupling algorithm to analyze and process the received data to determine whether the battery has a risk of thermal runaway. Based on the analysis results of the data processing unit, a decision is made whether to trigger the fire protection system linkage control and cut off the battery circuit.

[0200] Furthermore, the present solution also proposes a computer-readable storage medium on which a computer-readable program is stored. When the computer-readable program is called, the above-mentioned multi-feature coupled energy storage thermal runaway monitoring and disposal method is executed.

[0201] It is understandable that the storage medium may be a magnetic medium, such as a floppy disk, a hard disk, or a magnetic tape; an optical medium, such as a DVD; or a semiconductor medium, such as a solid state drive (SSD).

[0202] To sum up, the advantages of the present invention are: it can accurately describe the mechanism of the battery. The more scales there are, the more comprehensively it can describe the state of the battery. A multi-feature coupling model suitable for the entire life cycle and a wide temperature range is established. The multi-feature coupling model is used to more accurately calculate or estimate the characteristic parameters required for thermal runaway warning, thereby ensuring the reliability of thermal runaway warning and achieving thermal runaway warning with high confidence.

[0203] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A multi-feature coupled energy storage thermal runaway monitoring and disposal method, characterized in that: include: The temperature gradient inside the battery is approximated using radial finite differences, and a one-dimensional thermal model of the battery is constructed based on the principle of thermal balance. Based on the actual battery capacity and the impact of battery aging on ohmic internal resistance as evaluation indicators, a dual-index aging model for the battery is constructed; Installing a gas sensor to monitor the concentration of characteristic gases in real time, wherein the characteristic gases include hydrogen and carbon monoxide; Use environmental soundprint technology to monitor the sound characteristics of the battery when it is running; Combined with the one-dimensional thermal model and dual-index aging model of the battery, the impact of heat and aging on the battery environmental parameters is analyzed, and a multi-feature coupling model is constructed. The environmental parameters include the concentration of characteristic gases and the sound characteristics during operation. Measure the response of the multi-feature coupling model under special excitation, and select the particle swarm algorithm to optimize the model parameters of the multi-feature coupling model; The thermal runaway warning is converted into an internal short circuit detection problem. Based on a multi-feature coupling model, a thermal runaway warning method with four characteristic parameters, namely battery voltage data, thermal signal, characteristic gas concentration and sound characteristics during operation, is designed. The battery cell data includes the actual battery capacity and ohmic internal resistance. When the monitoring system detects that the battery is at risk of thermal runaway, it immediately triggers the linkage control of the fire protection system and the electrical switch cuts off the battery circuit.

2. The multi-feature coupled energy storage thermal runaway monitoring and disposal method according to claim 1, characterized in that: The method of using radial finite difference to approximate the temperature gradient inside the battery and constructing a one-dimensional thermal model of the battery based on the thermal balance principle specifically includes: Obtain the ambient temperature and battery surface temperature, and use the thermal radiation formula to calculate the thermal radiation of the battery; Use the thermal convection formula to calculate the convection heat dissipation of the battery; Add the thermal radiation and convection heat dissipation of the battery and output it as the heat dissipation rate of the battery; Obtain the thermal balance principle that the battery complies with; The winding body and electrolyte inside the battery are considered to be uniformly distributed, and the radial temperature variation law of the battery is calculated based on the principle of thermal balance; Starting from the radial center, the battery is divided into at least one node along the radial direction inside, and the temperature of each node is approximately represented by a finite difference; Build a one-dimensional thermal model of the battery based on the temperature of each node; The thermal radiation formula is: ， In the formula, is the thermal radiation of the battery, is the thermal emissivity, is the Stefan-Boltzmann constant, is the battery surface temperature, is the ambient temperature; The heat convection formula is: ， In the formula, To dissipate heat from the battery through convection, is the convection coefficient; The thermal balance principle is: ， In the formula, is the material density, is the specific heat capacity, is a function of temperature, For time, is the thermal conductivity, is the heat generation rate function, is the heat dissipation function of the battery, is the partial derivative of the temperature function with respect to time, is the vector differential operator; The radial temperature variation rule of the battery is: ， In the formula, is the temperature function at a point y away from the center of the battery, is the volume of the battery.

3. The multi-feature coupled energy storage thermal runaway monitoring and disposal method according to claim 2 is characterized in that: The dual-index aging model of the battery is constructed based on the actual battery capacity and the influence of the ohmic internal resistance on the battery aging as evaluation indicators, and specifically includes: Use the Arrhenius formula to describe the battery capacity decay model; The internal resistance growth is divided into three parts: the internal resistance growth caused by date, the internal resistance growth caused by cycle and the internal resistance growth affected by randomness; Use the internal resistance growth formula to describe the ohmic internal resistance growth model of the battery; The capacity attenuation model and the ohmic internal resistance growth model are combined and output as a dual-index aging model for the battery; The Arrhenius formula is: ， In the formula, is the capacity loss rate, is the cycle charge and discharge rate, is the reaction activation energy, is the molar gas constant, is the average temperature inside the battery; The internal resistance growth formula is: ， In the formula, is the battery internal resistance growth value, is the internal resistance growth affected by randomness, is the internal resistance growth coefficient caused by date, is the internal resistance growth coefficient caused by the cycle, is the number of cycles.

4. The multi-feature coupled energy storage thermal runaway monitoring and disposal method according to claim 3 is characterized in that: The use of environmental soundprint technology to monitor the sound characteristics of battery operation specifically includes: Use highly sensitive sound sensors deployed at the location of the battery pack; Preprocess the collected sound signal to remove the noise signal in the sound signal to obtain the environmental voiceprint signal; Use spectrum analysis to extract features from environmental voiceprint signals; The process of decomposing an environmental soundprint signal into at least one cosine function basic signal superposition; Outputting the coefficients of each cosine-type function basic signal as the sound characteristics when the battery is running; The cosine type function is: ， In the formula, is a cosine function signal, is the amplitude of the signal, is the angular frequency of the signal, is the phase angle of the signal.

5. The multi-feature coupled energy storage thermal runaway monitoring and disposal method according to claim 4 is characterized in that: The one-dimensional thermal model and dual-index aging model of the battery are combined to analyze the influence of heat and aging on the battery environmental parameters, and to construct a multi-feature coupling model, which specifically includes: Obtain an equivalent circuit model of a lithium-ion battery; Changes in battery environmental parameters are accompanied by changes in charge and discharge currents, which are input into the equivalent circuit model to calculate the real-time changes in terminal voltage and state of charge; The equivalent circuit model transfers the internal resistance heat generation data into the one-dimensional thermal model; The one-dimensional thermal model combines the ambient temperature to estimate the temperature inside the battery in real time, and calculates the changes in the parameters of the equivalent circuit model under the influence of thermal effects; The dual-index aging model substitutes the calculation results of the equivalent circuit model and the one-dimensional thermal model, outputs the battery capacity decay rate, and returns the updated ohmic internal resistance to the equivalent circuit model and the one-dimensional thermal model; The state of charge is: ， In the formula, is the state of charge, is the remaining capacity of the battery, is the actual capacity of the battery.

6. The multi-feature coupled energy storage thermal runaway monitoring and disposal method according to claim 5, characterized in that: The step of measuring the response of the multi-feature coupling model under special excitation and selecting the particle swarm algorithm to optimize the model parameters of the multi-feature coupling model specifically includes: S101: Design at least one series of experiments of low rate cycling and constant capacity pulse discharge under constant temperature conditions; S102: Obtaining an initial parameter group of a multi-feature coupling model; S103: Setting key parameters required for the particle swarm algorithm, wherein the key parameters include an upper limit of the number of iterations, a minimum and maximum value of the inertia weight, and an initial value of the acceleration factor; S104: calculating the inverse proportional function value of the data error generated by the model parameters of each multi-feature coupling model in the series of experiments, and recording it as the individual extreme value, and recording the maximum value of the individual extreme value as the group extreme value; S105: Calculating the inertia coefficient according to the logistic regression equation; S106: calculating the value function and the change speed of the model parameters of the multi-feature coupling model, and calculating the inverse proportional function value of the data error generated in the series of experiments according to the value function; S107: Determine whether the inverse proportional function value of the data error generated in the series of experiments is higher than the individual extreme value. If so, update the individual extreme value and the corresponding value function. If not, do not output. S108: judging whether the sum of the inverse proportional function values ​​of the data errors generated in the series of experiments is higher than the group extreme value, if so, updating the group extreme value and the corresponding value function, if not, no output; S109: Determine whether the number of iterations reaches the upper limit of the number of iterations. If so, output the model parameters of the current multi-feature coupling model. If not, return to step S105. The logistic regression equation is: In the formula, is the inertia coefficient, , are the maximum and minimum values ​​of the inertia weight, Q is the upper limit of the number of iterations, and q is the current number of iterations; The value function and change rate of the model parameters of the multi-feature coupling model are: In the formula, , are the changing speed and value function of the model parameters of the i-th multi-feature coupling model at the q-th iteration, and j is the iteration parameter.

7. The multi-feature coupled energy storage thermal runaway monitoring and disposal method according to claim 6, characterized in that: The thermal runaway warning method is designed to convert the thermal runaway warning into an internal short circuit detection problem based on a multi-feature coupling model, and to design four characteristic parameters of battery voltage data, thermal signal, characteristic gas concentration, and sound characteristics during operation. Specifically, the method includes: The changes of four characteristic parameters, namely, battery voltage data, thermal signal, characteristic gas concentration and sound characteristics during operation, are input into the multi-characteristic coupling model to obtain the threshold range of each parameter under safe conditions; Determine whether the battery voltage data exceeds the first parameter threshold range, if so, output the proportion of the exceeding portion, if not, do not output; Determine whether the thermal signal exceeds the second parameter threshold range, if so, output the proportion of the exceeding portion, if not, do not output; Determine whether the characteristic gas concentration exceeds the third parameter threshold range, if so, output the proportion of the exceeding portion, if not, no output; Determine whether the sound feature during operation exceeds the fourth parameter threshold range, if so, output the proportion of the exceeding portion, if not, do not output; Use the warning signal level formula to calculate the current thermal runaway warning signal level; The warning signal level formula is: ， In the formula, The warning signal level is , , , are the calculation weights of battery voltage data, thermal signal, characteristic gas concentration and sound characteristics during operation, , , , They are the proportion of the excess parts of battery voltage data, thermal signal, characteristic gas concentration and sound characteristics during operation.

8. The multi-feature coupled energy storage thermal runaway monitoring and disposal method according to claim 7, characterized in that: When the monitoring system detects that the battery has a risk of thermal runaway, the linkage control of the fire protection system is immediately triggered, and the electrical switch cuts off the battery circuit, specifically including: Get the warning signal level; Determine whether the warning signal level is greater than a first preset threshold value, if so, control the electrical switch to cut off the battery circuit, if not, dispatch relevant personnel to perform on-site maintenance; Determine whether the warning signal level is greater than the second preset threshold. If so, trigger the linkage control of the fire protection system, start the fire extinguishing device, release the fire extinguishing agent, and start the smoke exhaust system. If not, no output is made.

Images