Lithium ion battery system thermal runaway fire risk assessment method

Abstract

The invention relates to a lithium ion battery system thermal runaway fire risk assessment method, and belongs to the technical field of energy storage system safety assessment and risk management. According to the method, a multi-parameter coupled thermal runaway risk analysis model is constructed by comprehensively considering thermal response characteristics, gas release behaviors and spreading path parameters, and the dynamic evolution law in the thermal runaway process of the lithium ion battery system is simulated. According to the method, dynamic failure paths and risk levels of the battery system under different working conditions are quantified, so that thermal runaway risk assessment of the lithium ion battery system is realized, and technical support is provided for safety and reliability analysis of the battery system.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of energy storage system safety assessment and risk management, and particularly relates to a lithium ion battery system thermal runaway fire risk assessment method. BACKGROUND

[0002] Currently, lithium ion batteries have been widely used in electric vehicles, energy storage systems and new power systems due to their high energy density, long cycle life and other advantages. However, lithium ion batteries may experience thermal runaway under extreme conditions, leading to serious safety accidents. As the scale of lithium battery application expands and the capacity of the battery increases, thermal runaway risk assessment of lithium ion batteries effectively improves the reliability of operation and ensures the safe and stable operation of the system. Traditional lithium ion battery thermal runaway risk assessment methods mainly rely on experimental tests and simplified models, lack of control over the dynamic evolution process of thermal runaway, and cannot accurately predict the thermal runaway propagation path in complex systems. At the same time, the existing evaluation methods cannot achieve multi-parameter coupling, and cannot comprehensively consider the coupling of thermal response characteristics, gas release behavior and structure propagation, etc. Multi-dimensional factors, resulting in significant differences between the risk assessment results and the actual situation.

[0003] However, the existing technology still has the following deficiencies in the field of lithium ion battery thermal runaway risk assessment: first, most of the existing methods rely on experimental tests and static simplified models, lack of research on the dynamic evolution process of thermal runaway in complex battery systems, and it is difficult to accurately predict the thermal runaway propagation path; second, the coupling relationship between thermal response characteristics, gas release behavior and structure propagation, etc. Multi-dimensional factors are not systematically considered in the risk assessment process, resulting in a large deviation between the evaluation results and the actual situation; third, there is no risk level quantification and classification method suitable for different working conditions and system structures, which limits the promotion and universality of the evaluation model in engineering applications. SUMMARY

[0004] In order to solve the problems of the existing technology, the present application proposes a lithium ion battery system thermal runaway fire risk assessment method, which constructs a multi-parameter coupled thermal runaway risk analysis model by comprehensively considering the thermal response characteristics, gas release behavior and propagation path parameters, and simulates the dynamic evolution law in the thermal runaway process of lithium ion battery system. This method can quantify the dynamic failure path and risk level of the battery system under different working conditions, provide technical support for the safety and reliability analysis of the battery system, and has high model universality and practical engineering application value.

[0005] The present application is realized by the following technical scheme: a lithium ion battery system thermal runaway fire risk assessment method, the steps of which are: Step 1: analyze the typical evolution mechanism of lithium-ion battery system thermal runaway, identify the physical characteristics of the induction, initiation, expansion and disaster stages, and establish the evolution path of the thermal runaway process; Step 1.1: identify the physical characteristics of the induction, initiation, expansion and disaster stages, establish the evolution path of the thermal runaway process, extract the key physical parameters, and construct the path state vector X pa :

[0006] In the formula, t is the induction factor strength; is the temperature rise rate; D g is the gas diffusion coefficient; P s is the thermal runaway expansion probability; Q f is the fire heat release rate; P f is the total probability of system thermal runaway.

[0007] The induction factor strength t model:

[0008] In the formula, T is the battery temperature; T sei is the thermal runaway critical temperature; t is 1 when triggered, and 0 when not triggered; The model of the gas diffusion coefficient D g :

[0009] In the formula, D0 is the reference state diffusion coefficient; T0 is the reference temperature when D0 is calibrated; is the temperature index; p is the instantaneous pressure inside / outside the battery; p0 is the reference pressure when D0 is calibrated; β is the porosity; τ is the tortuosity; E D is the diffusion activation energy; R is the gas constant; The model of the thermal runaway expansion probability P s :

[0010] In the formula, k e is the equivalent thermal conductivity of the battery; A is the heat dissipation area; d is the heat propagation distance; ΔT is the temperature difference; The model of the fire heat release rate Q f :

[0011] In the formula, η is the combustion efficiency; Low heating value of gas i; Mass burning rate of gas i; The total probability P of system thermal runaway f Model:

[0012] Wherein, T is the instantaneous temperature of the battery; T f is the critical temperature of battery failure.

[0013] Step 2: According to the evolution path obtained in step 1, the key thermal response characteristics, gas release rate and structure propagation path of the battery system in the thermal runaway process are extracted, and a risk model of multi-parameter fusion of temperature rise rate, gas diffusion coefficient and health state is established; The multi-parameter fusion risk model is

[0014] Wherein, SOH is the health state of the battery; P s is the thermal runaway propagation path probability; is a function of quantifying risk level; F is a nonlinear risk weight mapping function.

[0015] Step 3: Construct a dynamic failure path evaluation model coupled with thermal diffusion equation and fire propagation probability model; The dynamic failure path evaluation model is:

[0016] Wherein, α is the thermal diffusion rate; is the Laplace operator; is the heat generation rate; is the failure path probability; is the occurrence probability of the i th failure mode; is the probability that n failure modes do not occur at the same time.

[0017] Step 4: Based on the dynamic failure path evaluation model, the dynamic failure path evaluation analysis is carried out, and the risk level evaluation model, risk mitigation cost model and thermal runaway probability model of the lithium ion battery system are constructed; Step 4-1: The dimensionless risk index R is obtained by weighted summation :

[0018] Wherein, , , , The weight coefficients of system failure path probability, temperature rise rate, gas diffusion and health state are respectively, wherein, ​is the maximum characteristic temperature; is the maximum gas diffusion coefficient; Step 4-2: Based on the risk index threshold, segmented output ,

[0019] wherein, is the demarcation value between low risk and medium risk, is the demarcation value between medium risk and high risk, when is 1, it is low risk, is 2, it is medium risk, and is 3, it is high risk; Step 4-3: Through weighted summation, the dimensionless risk mitigation cost at time t is obtained :

[0020] wherein, , , are the weight coefficients of thermal runaway probability, risk level response, and key parameter control, respectively; is the mitigation cost of the total probability of system thermal runaway; is the risk level mitigation cost; , , are the weight coefficients of temperature rise rate, gas diffusion coefficient, and health status, respectively; , , are the mitigation costs of temperature rise rate, gas diffusion coefficient, and battery health status, respectively.

[0021] Step 5: The dynamic failure path evaluation model obtained in step 3 and the risk level evaluation model obtained in step 4 are combined to construct a global risk minimization solving model, and the risk occurrence probability and corresponding index parameters are output;

[0022] wherein, R l (t) is the risk level mapping function; C m (t) is the risk mitigation cost function; P f (t) is the total probability of system thermal runaway under the specified working condition; , , are the target weight coefficients, wherein, ; After normalization processing, it is:

[0023] wherein, , , is the normalized target, apply linear normalization formula to each index, convert to dimensionless [0,1] interval value; is the global thermal runaway risk probability; is compared with the global thermal runaway risk threshold γ1, if <γ1, the risk probability and corresponding formula (1) index parameters are output; if >γ1, the parameters of formula (2)~(6) are optimized, and step 3 is recalculated.

[0024] Compared with the prior art, the beneficial effects of the present application are: the present application comprehensively considers key influencing factors such as thermal response characteristics, gas release behavior and system structure propagation path, and constructs a multi-parameter coupled thermal runaway risk modeling system to describe the dynamic evolution characteristics of the lithium ion battery system in the thermal runaway process; further, based on the thermal diffusion equation and the fire propagation probability function, a thermal runaway path evolution model and a risk level evaluation model are established; finally, taking the system thermal risk level, the risk mitigation cost, and the minimum and fastest total probability of thermal runaway under the specified working condition as the target, an evaluation method suitable for multiple working conditions and multiple structures is proposed, thereby realizing the thermal runaway fire risk of the lithium ion battery system. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 is a flow chart of lithium ion battery system thermal runaway risk evaluation; Figure 2 is a temperature distribution diagram of lithium ion battery system thermal runaway risk. DETAILED DESCRIPTION

[0026] The technical solutions in the embodiments of the present application will be described clearly and completely below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0027] The present application will be further described in detail below in combination with the drawings, and the specific process is shown in Figure 1 A lithium ion battery system thermal runaway fire risk evaluation method comprises the following steps: Step 1: analyze the typical evolution process of lithium ion battery system thermal runaway.

[0028] Identify the physical characteristics of the induction, initiation, expansion and disaster stages, and establish the evolution path of the thermal runaway process, specifically: Induction ( )→ Initiation ( )→ Expansion (​ , )→catastrophe , ) where, t is the induced factor intensity, such as overcharge, external short circuit, mechanical impact, etc. representing the index of the triggering factor; is the temperature rise rate; D g is the gas diffusion coefficient; P s is the thermal runaway propagation probability; Q f is the fire heat release; P f is the system catastrophic failure probability.

[0029] Extract the key physical parameters to form the path state vector X pa :

[0030] Establish a temperature threshold trigger model:

[0031] where T is the battery temperature; T sei is the thermal runaway critical temperature; t is 1 indicating triggering, and 0 indicating non-triggering.

[0032] Establish a gas diffusion coefficient model:

[0033] where D0 is the reference state diffusion coefficient; T0 is the reference temperature when D0 is calibrated; is the temperature index; p is the instantaneous pressure inside / outside the battery; p0 is the reference pressure when D0 is calibrated; β is the porosity; τ is the tortuosity; E D is the diffusion activation energy; R is the gas constant.

[0034] Establish a thermal runaway propagation probability model:

[0035] where k e is the equivalent thermal conductivity of the battery; A is the heat dissipation area; d is the heat propagation distance; ΔT is the temperature difference.

[0036] Establish a fire heat release model:

[0037] where η is the combustion efficiency; is the low heat value of gas i; is the mass combustion rate of gas i.

[0038] Establish a system catastrophic failure probability model:

[0039] Where T is the instantaneous temperature of the battery; T f is the critical temperature of battery failure.

[0040] Step 2: Extract the key thermal response characteristics of the battery system during the thermal runaway process and establish a multi-parameter fusion risk model, specifically:

[0041] Where SOH is the state of health of the battery; P s is the probability of thermal runaway propagation path; is a function of quantifying the risk level; F is a nonlinear risk weight mapping function.

[0042] Step 3: Build a dynamic failure path model, model the failure path probability coupling:

[0043] Where α is the thermal diffusivity; is the Laplacian operator; is the heat generation rate; is the failure path probability; is the probability of the occurrence of the i-th failure mode; is the probability of n failure modes not occurring simultaneously.

[0044] Step 4: Build a dynamic failure path probability model, consider different risk levels to establish a risk level assessment model, and a risk mitigation cost model.

[0045] Step 4-1: Through weighted summation, obtain the dimensionless risk index :

[0046] Where , , , are the weight coefficients of the system failure path probability, temperature rise rate, gas diffusion, and state of health, respectively, where ; is the maximum characteristic temperature; is the maximum value of the gas diffusion coefficient.

[0047] Step 4-2: Based on the risk index threshold, output , ,

[0048] wherein, R is the boundary value between low risk and medium risk, R is the boundary value between medium risk and high risk, when R = 1 is low risk, R = 2 is medium risk, and R = 3 is high risk.

[0049] Step 4-3: Obtain the dimensionless risk mitigation cost at time t by weighted summation

[0050] wherein, , , are the weight coefficients of thermal runaway probability, risk level response, and key parameter control, respectively; Cm(t) is the mitigation cost of the total probability of system thermal runaway; Cm(t) is the mitigation cost of the risk level; , , are the weight coefficients of temperature rise rate, gas diffusion coefficient, and health status, respectively; , , are the mitigation costs of temperature rise rate, gas diffusion coefficient, and battery health status, respectively.

[0051] For step 5, a minimum model is constructed based on the dynamic failure path model as shown in equation (8) constructed in step 3 and the risk level evaluation model formed in step 4, with the risk level, risk mitigation cost, and the total probability of thermal runaway under the specified working condition as the target, specifically:

[0052] wherein, Rl(t) is the risk level mapping function output by step 4; Cm(t) is the risk mitigation cost function; Pf(t) is the total probability of system thermal runaway under the specified working condition output by step 3; , , are the weight coefficients of each target, wherein, .

[0053] Apply a linear normalization formula to each index to convert it to a dimensionless value in the [0, 1] interval:

[0054] wherein, is , , ; is the minimum value of the index; is the maximum value of the index.​

[0055] and normalized as:

[0056] wherein, , , is the normalized target, and a linear normalization formula is applied to each index to convert it to a dimensionless [0, 1] interval value; , , is a weight coefficient, and the sum is 1; is the global thermal runaway risk probability.

[0057] is compared with the global thermal runaway risk threshold γ1, if <γ1, the risk occurrence probability and corresponding formula (1) index parameters are output.

[0058] According to steps 1-5, thermal runaway analysis and modeling are performed considering the thermal runaway evolution stage, key physical parameters and failure modes, and the risk index threshold, risk mitigation cost, total probability of thermal runaway under specified working conditions and main performance parameters affecting thermal runaway are output, ensuring that the thermal runaway process is traceable, the risk is quantifiable, and the probability of catastrophic failure of the system is minimized.

[0059] As shown in Figure 2 , the temperature field distribution simulation results of the lithium ion battery system during the thermal runaway process are given. As can be seen from Figure 2 , the temperature of the single battery rises rapidly under the action of the heat source, and is conducted to the adjacent battery along the surface of the battery and the connecting piece area, forming a clear high temperature concentration area and gradient distribution characteristics. The temperature difference between different battery cells clearly shows the heat diffusion and propagation path, where the battery near the heat source first appears a red high temperature area, and then the heat is diffused to the adjacent battery along the series connecting piece and side wall. This result directly reflects the spatial temperature distribution law of the battery system during the thermal runaway evolution process, provides reliable numerical support for dynamic failure path modeling and fire risk assessment, and also verifies the effectiveness of the proposed model in revealing the thermal runaway propagation path.​

Claims

1. A method of lithium-ion battery system thermal runaway fire risk assessment, characterized in that, The steps are: Step 1: analyze the typical evolution mechanism of lithium ion battery system thermal runaway, identify the physical characteristics of the induction, initiation, expansion and disaster stage, and establish the evolution path of the thermal runaway process; Step 2: according to the evolution path obtained in step 1, extract the key thermal response characteristics, gas release rate and structure spread path of the battery system in the thermal runaway process, and establish a risk model of multi-parameter fusion of temperature rise rate, gas diffusion coefficient and health state; Step 3: construct a dynamic failure path evaluation model coupled with thermal diffusion equation and fire propagation probability model; Step 4: based on the dynamic failure path evaluation model, perform dynamic failure path evaluation analysis, and construct a risk level evaluation model, a risk mitigation cost model and a thermal runaway probability model of the lithium ion battery system; Step 5: combine the dynamic failure path evaluation model obtained in step 3 with the risk level evaluation model obtained in step 4 to construct a global risk minimization solving model, and output the risk occurrence probability and corresponding index parameters.

2. The method of claim 1, wherein the method further comprises: The specific method in step 1 is: Step 1.1: Identify the physical features of induction, initiation, propagation and catastrophe stage, establish the evolution path of thermal runaway process, extract the key physical parameters, and construct the path state vector X pa : wherein t is the factor intensity; is the temperature rise rate; D g is the gas diffusion coefficient; P s is the thermal runaway propagation probability; Q f is the fire heat release; P f is the system thermal runaway total probability.

3. The method of claim 2, wherein the method further comprises: The strength of the eliciting factor t Model: where T is the battery temperature; T sei is the thermal runaway critical temperature; t 1 indicates triggered, and 0 indicates not triggered. The gas diffusion coefficient D g of the model: wherein D0 is the reference state diffusion coefficient; T0 is the reference temperature at which D0 is calibrated; is the temperature exponent; p is the instantaneous pressure inside / outside the cell; p0 is the reference pressure at which D0 is calibrated; β is the porosity; τ is the tortuosity; E D is the diffusion activation energy; R is the gas constant; The probability of thermal runaway propagation P s of the model: where k is the thermal conductivity of the battery; A is the heat dissipation area; d is the heat propagation distance; and ΔT is the temperature difference. e where k is the thermal conductivity of the battery; A is the heat dissipation area; d is the heat propagation distance; and ΔT is the temperature difference. The fire heat release rate Q f of the model: wherein η is the combustion efficiency; is the lower heating value of gas i; is the mass burn rate of gas i; The system thermal runaway total probability P f of the model: where T is the battery instantaneous temperature; T f is the battery failure threshold temperature.

4. The method of claim 1, wherein the method further comprises: In step 2, the multi-parameter fusion risk model is where SOH is the state of health of the battery; P s is the probability of the thermal runaway propagation path; is a function quantifying the risk level; F is a non-linear risk weight mapping function.

5. The method of claim 1, wherein the method further comprises: In step 3, the dynamic failure path evaluation model is: wherein a is the thermal diffusivity; is the Laplacian operator; is the heat generation rate; is the failure path probability; is the occurrence probability of the i-th failure mode; is the probability that none of the n failure modes occur simultaneously.

6. The method of claim 1, wherein the method further comprises: The specific method in step 4 is: Step 4-1 : Obtain the dimensionless risk index by weighted summation : In the formula, , , , are respectively the system failure path probability, the temperature rise rate, the gas diffusion, the health state weight coefficient, wherein, ; is the maximum characteristic temperature; is the maximum value of the gas diffusion coefficient; Step 4-2: Risk Index Based Threshold, piecewise output , wherein is the cut-off value between low and intermediate risk, is the cut-off value between intermediate and high risk, when is 1 for low risk, 2 for intermediate risk, and 3 for high risk. Step 4-3: Obtain the dimensionless risk mitigation cost at time t by weighted summation : In the formula, , , are weight coefficients of thermal runaway probability, risk level response, and key parameter control, respectively; is the mitigation cost of the total probability of system thermal runaway; is the risk level mitigation cost; , , are weight coefficients of temperature rise rate, gas diffusion coefficient, and health status, respectively; , , are mitigation costs of temperature rise rate, gas diffusion coefficient, and battery health status, respectively.

7. The method of claim 1, wherein the method further comprises: The specific method in step 5 is: wherein R l (t) is a risk level mapping function; C m (t) is a risk mitigation cost function; P f (t) is the total probability of thermal runaway of the system under the specified operating conditions; , , are target weight coefficients, wherein, ; The normalization processing is: In the formula, , , is the normalized target, and a linear normalization formula is applied to each index to convert it to a dimensionless [0, 1] interval value; is the global thermal runaway risk probability; Will Compared with the global thermal runaway risk threshold γ1, if If γ < 1, then output the probability of risk occurrence and the corresponding index parameters in formula (1); if >γ1, optimize the parameters of formulas (2) to (6), and recalculate step 3.

Images