Multi-level networking energy storage system reliability evaluation method and device based on electric-thermal-capacity coupling

Abstract

The invention provides an electricity-heat-capacity coupling-based reliability evaluation method and device for a multi-level networking energy storage system, and the method comprises the steps: constructing a battery electricity-heat-capacity coupling model which comprises three sub-models, namely an electrical sub-model, a thermal characteristic sub-model and a capacity degradation sub-model, and achieving the dynamic synchronous updating of current, temperature and capacity degradation through the closed-loop coupling of the three sub-models, key parameters are extracted; constructing an inter-battery thermal runaway spreading model, calculating heat transfer through a heat conduction network of a two-dimensional lumped convection model, and simulating a thermal runaway spreading path, rate and capacity decline trend in combination with a joint failure probability model based on a Copula function; the model is embedded into a Monte Carlo simulation framework, and reliability evaluation is carried out by adopting a'monomer-branch-system 'three-level structure. According to the method, the high-magnification working condition simulation precision is improved through dynamic closed-loop updating, physical transmission and probability driving are considered, safety control is enhanced, and a global view angle is provided for system reliability analysis through a multi-level system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of safety and reliability management of lithium-ion battery energy storage systems, in particular to a multi-level networked energy storage system reliability evaluation method and device based on electro-thermal-capacity coupling. BACKGROUND

[0002] Networked energy storage systems with autonomous voltage and frequency regulation capabilities are widely recognized as an important technical path for building new power systems because they can replace part of the traditional units to provide inertia support and voltage support functions. However, under dynamic operating conditions, the electro-thermal-aging effects of networked energy storage systems exhibit complex multi-time scale performance evolution characteristics. Not only do they face the problem of continuous capacity degradation, but they also have the risk of sudden thermal runaway, which is highly contagious within the battery pack and can lead to simultaneous failure of multiple cells, becoming a safety bottleneck for large-scale energy storage system operation. This poses higher requirements for dynamic modeling and system-level reliability evaluation.

[0003] Existing research mainly uses equivalent circuit models and simplified thermal networks to describe electro-thermal behavior, and capacity degradation is based on experimental data to fit a two-section empirical formula, or combined with online electrochemical impedance testing and machine learning algorithms to achieve health state estimation and life prediction. However, this type of research often treats thermal runaway risk and capacity degradation separately, making it difficult to reflect their interaction in actual applications. Research on thermal runaway also lacks a simultaneous depiction of thermal conduction chain spread and temperature field distribution. In terms of system-level reliability analysis, the multi-state system reliability theory provides a mathematical framework, but it is difficult to solve analytically due to the high dimensionality of the state space. Monte Carlo simulation is widely used in life prediction and risk assessment due to its flexibility and scalability, but it focuses on a single failure mode, and few studies have combined electro-thermal-aging coupling models with Monte Carlo for full-life-cycle reliability evaluation of networked battery energy storage systems (BESS).

[0004] In summary, current technology lacks a comprehensive method that can both finely depict electro-thermal-capacity coupling behavior and efficiently simulate the impact of thermal runaway propagation and long-term degradation on reliability in a system-level scenario. How to integrate improved ECM, thermal model, and aging model organically while ensuring computational efficiency, and embed them into the Monte Carlo simulation and scheduling optimization framework to achieve dynamic reliability evaluation of networked BESS under multiple operating conditions is a key technology that needs to be broken through. SUMMARY

[0005] The purpose of the present application is to provide a multi-level networked energy storage system reliability evaluation method and device based on electro-thermal-capacity coupling, to evaluate the reliability of networked energy storage systems under various high-rate operating conditions, analyze the coupling failure of thermal runaway and capacity degradation, and the chain spread of thermal runaway in battery packs, and make an evaluation of the battery energy storage system at the system level.

[0006] A multi-level networked energy storage system reliability evaluation method based on electro-thermal-capacity coupling, comprising the following steps:

[0007] Step 1: Constructing a battery electro-thermal-capacity coupling model

[0008] Based on the multi-physical field coupling failure behavior of networked energy storage systems under various high-rate charging and discharging operating conditions, a battery electro-thermal-capacity coupling model suitable for high-rate operating conditions is constructed; the model includes an electrical sub-model, a thermal characteristic sub-model, and a capacity degradation sub-model, which realizes dynamic synchronous updating of current, temperature, and capacity degradation through closed-loop coupling of the three, and extracts key parameters inside the battery, including temperature, state of charge SOC, and real-time capacity;

[0009] Step 2: Constructing a thermal runaway spread model between batteries

[0010] The thermal runaway spread model between batteries includes a heat conduction network and a joint failure probability model, wherein the heat conduction network is based on the geometric arrangement and thermal coupling characteristics of the cells in the battery pack, a two-dimensional lumped convection model is constructed to calculate the heat transfer amount between adjacent battery cells, and the heat flow path and boundary conditions are determined;

[0011] The joint failure probability model is constructed by introducing a Copula function, considering the thermal runaway trigger probability and capacity degradation probability of the battery cell as correlated random variables, and embedding the joint failure probability model into the heat conduction network as a probability driving mechanism for triggering thermal runaway spread;

[0012] Through the cooperation of the failure probability model and the heat conduction network W, the spread path, propagation rate, and capacity degradation evolution trend of thermal runaway in the battery pack under different operating conditions are simulated;

[0013] Step 3: Multi-level reliability evaluation based on Monte Carlo simulation

[0014] Embed the electro-thermal-capacity coupling model and the thermal runaway spread model into the Monte Carlo simulation framework, and use a three-level analysis structure of "cell level-branch level-system whole group level" for reliability evaluation.

[0015] Further, the step 3 uses a three-level analysis structure of "cell level-branch level-system whole group level" for reliability evaluation, specifically including:

[0016] Initial parameter sampling: For the initial state parameters of each battery cell in the system, a large number of independent samples are generated through a set probability distribution as the initial state of each simulation round, and each sample represents the possible state of the system in a simulation round, and the initial state parameters include SOC, cell temperature, capacity state and Copula correlation parameters;

[0017] Online simulation and index calculation: In each simulation round, the electrical response, temperature change and capacity degradation process of the battery cell are simulated based on the electro-thermal-capacity coupling model, and the thermal runaway probability and capacity failure probability are determined by combining the thermal runaway propagation model; whether the cell fails is determined by Bernoulli sampling, and the following is calculated:

[0018] Cell level: Based on the state of charge and temperature conditions, the cell failure probability is calculated;

[0019] Branch level: The minimum value of the reliability of each cell in the series branch is used to calculate the branch failure probability and unpowered quantity;

[0020] System group level: The survival / failure state of all branches is aggregated according to the parallel topology, the system level failure probability is calculated, and the reliability curve is drawn combined with the scheduling strategy and load model;

[0021] Result statistics: After repeating a large number of simulation rounds, all results are statistically analyzed to obtain the reliability evolution curve and system reliability evolution map in the whole life cycle of the system, and the reliability evolution curve includes power failure probability, reliability trend and risk distribution at different times.

[0022] Further, in the electro-thermal-capacity coupling model:

[0023] Electrical sub-model: The input is the external charging and discharging current, the initial state of charge SOC and the remaining available capacity, and the output is the battery terminal voltage, the updated SOC and the generated heat energy;

[0024] Thermal characteristic sub-model: The input is the heat energy output by the electrical sub-model, the environmental temperature and the core and shell temperature at the previous time, and the output is the updated internal temperature of the cell and the surface temperature of the shell;

[0025] Capacity degradation sub-model: The input is the SOC of the electrical sub-model, the temperature history of the thermal characteristic sub-model and the cumulative cycle number, and the output is the remaining available capacity;

[0026] The three sub-models realize dynamic synchronous update of current, temperature and capacity degradation through closed loop feedback.

[0027] Further, the joint failure probability model is based on the data of the electro-thermal-capacity coupling model, uses a Copula function to describe the correlation between the battery thermal runaway probability and the capacity degradation probability, and quantitatively evaluates the joint risk of the single battery simultaneously occurring thermal runaway and capacity failure under a given state of charge, temperature and cycle history.

[0028] Further, the thermal runaway propagation model is based on the relationship between the discretized heat conduction network and the thermal resistance between nodes, establishes a thermal runaway chain propagation mechanism, triggers a propagation event when the temperature of adjacent monomers exceeds a threshold, iteratively transmits the monomer-level joint failure rate to the entire battery cluster, and simulates the diffusion process of thermal runaway from a single point to multiple points.

[0029] Further, the electrical sub-model is based on an improved equivalent circuit model to calculate the terminal voltage, the thermal characteristic sub-model is coupled with the thermal energy output by the electrical sub-model through a simplified two-node thermal network and environmental heat exchange, the capacity degradation sub-model estimates the capacity degradation rate by integrating the effects of cycle aging and calendar aging, and the three form an electro-thermal-capacity linkage evolution mechanism.

[0030] Further, the battery thermal runaway probability is obtained by referring to a pre-constructed thermal runaway probability mapping table, the mapping table is based on the probability density function and the cumulative distribution function fitted from multiple sets of experimental data, and reflects the risk of battery monomers under different chemical systems, manufacturing processes and environmental conditions under each SOC-temperature condition.

[0031] Further, the two-dimensional lumped convection model of the heat conduction network regards each battery cell in the battery pack as a node, and heat transfer comes from the surrounding eight adjacent battery cells and fluid nodes, reflecting the thermal interaction characteristics between batteries at different spatial positions.

[0032] A multi-level network-based energy storage system reliability evaluation device based on electro-thermal-capacity coupling includes:

[0033] A battery electro-thermal-capacity coupling model construction module is used to construct a battery electro-thermal-capacity coupling model suitable for high-rate operation based on the multi-physical field coupling failure behavior of the network-based energy storage system under multiple high-rate charging and discharging conditions. The model includes an electrical sub-model, a thermal characteristic sub-model and a capacity degradation sub-model, which realize dynamic synchronous updating of current, temperature and capacity degradation through closed-loop coupling of the three, and extract key parameters inside the battery, including temperature, state of charge SOC and real-time capacity.

[0034] The battery inter-thermal runaway propagation model module is configured to construct a battery inter-thermal runaway propagation model including a heat conduction network and a joint failure probability model, wherein the heat conduction network is based on the geometric arrangement and thermal coupling characteristics of the battery cells in the battery pack, a two-dimensional lumped convection model is constructed to calculate the heat transfer amount between adjacent battery cells, and the heat flow path and boundary conditions are determined; the joint failure probability model is constructed by introducing a Copula function by regarding the thermal runaway triggering probability and the capacity degradation probability of the battery cells as correlated random variables, and the joint failure probability model is embedded in the heat conduction network as a probability driving mechanism for triggering thermal runaway propagation; the failure probability model and the heat conduction network W are matched to simulate the propagation path, propagation rate and capacity degradation evolution trend of thermal runaway in the battery pack under different working conditions.

[0035] The multi-level reliability evaluation module is configured to embed the electro-thermal-capacity coupling model and the thermal runaway propagation model into a Monte Carlo simulation framework, and perform reliability evaluation by using a three-level analysis structure of "cell level - branch level - system whole group level".

[0036] Further, the multi-level reliability evaluation module performs reliability evaluation by using a three-level analysis structure of "cell level - branch level - system whole group level", and specifically includes:

[0037] Initial parameter sampling: a large number of independent samples are generated by a set probability distribution for the initial state parameters of each battery cell in the system as the initial state of each simulation round, and each sample represents the possible state of the system in a simulation round, and the initial state parameters include SOC, cell temperature, capacity state and Copula correlation parameters;

[0038] Online simulation and index calculation: in each simulation round, the electrical response, temperature change and capacity degradation process of the battery cell are simulated based on the electro-thermal-capacity coupling model, and the thermal runaway probability and capacity failure probability are determined in combination with the thermal runaway propagation model; whether the cell fails is determined by Bernoulli sampling, and the following is calculated in turn:

[0039] Cell level: based on the state of charge and temperature conditions, the cell failure probability is counted;

[0040] Branch level: the minimum value of the reliability of each cell in the series branch is used to calculate the branch failure probability and the unsupplied power;

[0041] System whole group level: the survival / failure state of all branches is aggregated according to the parallel topology, the system level failure probability is counted, and the reliability curve is drawn in combination with the scheduling strategy and the load model;

[0042] Result statistics: after repeating a large number of simulation rounds, all results are statistically analyzed to obtain the reliability evolution curve and system reliability evolution atlas in the whole life cycle of the system, wherein the reliability evolution curve includes power failure probability, reliability trend and risk distribution at different times.

[0043] The present application has the following beneficial effects:

[0044] 1. Through the establishment of electro-thermal-capacity coupling simulation, dynamic closed-loop updating of current, temperature and capacity degradation is realized, more detailed single battery state information can be obtained, and the simulation accuracy of battery electrical and thermal response under high rate working conditions is significantly improved. 2. The joint failure probability model based on Copula is embedded in the heat propagation network, which can simultaneously simulate the triggering and spreading path of thermal runaway, and combined with the capacity degradation evolution, a reliability evaluation method considering both physical transmission and probability driving is provided, which improves the control ability of system safety under high risk operating conditions.

[0045] 3. A multi-level network type BESS evaluation system from single battery, module level to overall system is established, which can cooperatively depict failure propagation and performance degradation law at each level, and provide a global perspective for system level reliability analysis. BRIEF DESCRIPTION OF DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments of the present application will be briefly introduced as follows. Obviously, the drawings described below are only some embodiments of the present application, and other drawings can also be obtained according to these drawings without creative labor for those skilled in the art.

[0047] Figure 1 is a schematic diagram of the second-order equivalent circuit model adopted by the electrical sub-model in the electro-thermal-capacity coupling model of the embodiments of the present application;

[0048] Figure 2 is a schematic diagram of the simplified battery thermal circuit model adopted by the thermal characteristic sub-model in the electro-thermal-capacity coupling model of the embodiments of the present application;

[0049] Figure 3 is a schematic diagram of the two-dimensional lumped convection model established in the embodiments of the present application;

[0050] Figure 4 is a flowchart of a multi-level network energy storage system reliability evaluation method based on electro-thermal-capacity coupling according to an embodiment of the present application;

[0051] Figure 5 is a BESS reliability analysis schematic diagram according to an embodiment of the present application. DETAILED DESCRIPTION

[0052] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work belong to the scope of protection of the present application.

[0053] As shown in Figure 4 , the embodiments of the present application provide a reliability evaluation method for a multi-level network energy storage system based on electro-thermal-capacity coupling, comprising the following steps:

[0054] Step 1: Constructing an electro-thermal-capacity coupling model of a single battery, the electro-thermal-capacity coupling model comprising an electrical sub-model, a thermal characteristic sub-model, a capacity degradation model and other sub-models, and each sub-model is modeled in detail to accurately describe the electrical characteristics of the battery under various working conditions.

[0055] The electrical sub-model: as shown in Figure 1 , a second-order equivalent circuit model is adopted, which is composed of an open circuit voltage U oc , a series internal resistance R eq , a first RC branch and a second RC branch, the two branches being composed of R1, C1 and Rn, Cn respectively, and each RC data is obtained by experimental parameter fitting. The branch voltages of the first and second RC branches are U1 and U n , respectively, and the following can be obtained:

[0056]

[0057] t represents the time of the value, Δt represents the time interval between two sampling points, I represents the battery current, which is negative when discharging and positive when charging.

[0058]

[0059] U L represents the terminal voltage of the battery, f OCV (SOC, T) represents the function of the open circuit voltage of the battery and SOC and T, which is obtained through experiments.

[0060]

[0061] η I , η T represent the coulomb efficiency factor and the temperature efficiency factor, respectively, both of which can be obtained through experimental calibration, and Q aged represents the actual capacity of the battery under long-term operation.

[0062] Considering the polarization heat Qj Joule heat Q r and reaction heat Q p The battery heat source calculation equation of the second-order equivalent circuit model is as follows:

[0063]

[0064]

[0065]

[0066]

[0067] I is the battery current, T i represents the internal temperature of the battery, dU OCV / dT represents the entropy heat coefficient of the battery and T is the calvin temperature of the battery, H i is the total heat generated inside the battery.

[0068] The thermal characteristic sub-model can be described by the state space equation in the time domain as shown in the simplified battery thermal circuit model: Figure 2

[0069]

[0070] Where T ia , T sa represent the mutual differences between the internal temperature, surface temperature and ambient temperature of the battery respectively, Ri, Rs are the equivalent thermal resistances of the battery inside and surface respectively, Ci, Cs are the equivalent thermal capacities of the battery inside and surface respectively. The nodes T i and T s in the thermal characteristic model are discretized by the first-order forward difference method, and the following formula is obtained.

[0071]

[0072] The capacity degradation model quantifies the long-term influence of the discharge depth on the battery performance by the equivalent full cycle number method. For the i-th discharge interval depth DOD i , its rated cycle life is recorded as L i , which is provided by the manufacturer.

[0073] The discharge depth of the single battery in the k-th cycle is calculated. The DOD i of the single battery in each charge and discharge period is calculated by the change of the energy stored by the battery:

[0074]

[0075] C dis,i is the discharge capacity of this cycle, C​rated is the rated capacity of the battery. The equivalent cycles C eq is calculated. i When the discharge depth reaches DOD DODi at the end of a discharge cycle, one charge-discharge cycle is recorded in the corresponding C eq counter. To facilitate calculation, the cycle number of each DOD interval is normalized by its rated life, and the equivalent full cycle number C

[0076]

[0077] where C DODi records the total cycle number at discharge depth DOD i , L i is the cycle life of the battery under DOD i , L based is the cycle life of the battery at selected discharge depth DOD based , and C eq calculated represents the equivalent full charge-discharge cycle number that the single battery has consumed.

[0078] Further, to distinguish between normal and accelerated aging, C eq is decomposed into normal degradation part C nom and accelerated degradation part C eq,n according to the nominal discharge depth DOD eq,a .

[0079]

[0080] According to C eq,n and C eq,a , the capacity loss is estimated using a two-segment empirical model:

[0081]

[0082] where a i , b i , Z i are experimental fitting parameters, Q age,n , Q age,a represent the capacity loss of the battery due to normal and accelerated degradation, respectively, and Q0 is the initial capacity of the battery. The current available capacity Q aged of the battery is:

[0083]

[0084] where Q cal represents the calendar aging of the battery, which can be obtained based on the Eyring equation:

[0085]

[0086] where k B is the Boltzmann constant, A SOC is the pre-exponential factor, the activation energy Ecal a, SOC varies with SOC.

[0087] Step 2: Constructing the thermal runaway propagation model between cells. First, a two-dimensional lumped convection model is constructed, as shown in Figure 3 , which considers each battery cell Cjk in the battery pack as a node (assuming uniform battery surface properties) and surrounded by fluid nodes. Heat transfer includes both convective heat transfer between the battery surface and fluid nodes and convective heat transfer between adjacent fluid nodes. For example, for an internal battery Cjk, heat comes from both the surrounding eight adjacent battery cells and the surrounding fluid nodes. At the same time, in order to facilitate calculation, the influence of fluid nodes is removed, i.e., temperature propagation between each battery only considers a few adjacent batteries, and the simplified heat conduction equation is represented as (for a given internal battery node jk):

[0088]

[0089] T jk represents the temperature of the current battery C jk , dT jk / dt is the rate of change of temperature with time, K, p, C p are the thermal conductivity, density and specific heat capacity of the battery material, respectively, V is the volume of the lithium battery, A l , A c are the available areas of the battery directly or diagonally adjacent. In actual operation, for a battery in a thermal runaway state, the rapid temperature rise of the battery and a series of complex electrochemical reactions need to be considered. In order to facilitate calculation, it is assumed that when the battery triggers thermal runaway, its temperature will quickly rise to the critical temperature T crit , and at this temperature the battery enters a complete failure state.

[0090] Step 3: The main function of this step is to quantitatively evaluate the thermal runaway risk of each battery monomer under different temperature and SOC conditions based on the established two-dimensional lumped convection heat conduction model, and generate the corresponding thermal runaway probability mapping relationship for subsequent chain spread simulation and system reliability calculation. The specific implementation is to introduce the joint failure probability model based on the established two-dimensional lumped convection model: first, the survival function under two modes is constructed based on the thermal runaway probability curve and the capacity degradation failure probability curve of the battery respectively; second, the Copula function is used to couple the two edge survival functions to obtain the joint survival function; finally, the instantaneous failure rate of the battery at any time is derived from the joint survival function as the criterion for thermal runaway triggering and system reliability analysis. The specific operation is as follows.

[0091] For a battery energy storage system with N batteries inside, the set of electrical states of each monomer battery at time t can be obtained according to the coupling model SOC i = [SOCt 1, SOCt 2, …, SOCt N] and T = [Tt 1, Tt 2, …, Tt N], on which basis, for monomer battery thermal runaway, the thermal runaway probability is obtained by using the thermal runaway mapping function.

[0092]

[0093] p TR,t (i) is the thermal runaway probability of the i-th monomer battery at time t, f(SOC i (t), T i (t)) is the mapping relationship of battery SOC and temperature to thermal runaway probability.

[0094] In addition, define the capacity failure event of the monomer battery:

[0095]

[0096] Where Q th is the threshold value of the battery capacity failure taken, and since Q aged can be input as a determined value, the event can be taken as:

[0097]

[0098] Further, the cumulative survival function of a certain monomer battery i under two failure conditions can be obtained:

[0099]

[0100] Since the two types of failures are not completely independent, according to Sklar's theorem, the joint survival function of any multivariate distribution can be obtained by coupling its marginal survival function using Gumbel Copula:

[0101]

[0102] For the dependency strength parameters of the two failure modes on the i-th component, firstly, based on the sample failure time, the empirical Kendall's τ is calculated. i Then, using the theoretical relationship of Gumbel Copula... S was obtained. i Let be the joint survival function of the i-th component under the coupling of two failure modes. Furthermore, the instantaneous failure rate is derived from the joint survival function:

[0103]

[0104] λ i Let represent the instantaneous probability of battery failure at time t. From this, the battery reliability can be obtained as follows: represents the power supply reliability of a single battery cell at time t:

[0105]

[0106] In this way, the health state probability vector R of each individual battery cell can be obtained. i =[Rt 1,Rt 2,…,Rt N], based on this, the states of all individual cells and their corresponding probabilities can be aggregated to construct a system-level multi-state performance distribution model. Specifically, assume that the energy storage system consists of M series branches connected in parallel, and the j-th branch is numbered from the set S. j {1,2,3,4,…,N T The number in} is S j It consists of several batteries connected in series, and the rated output power of the branch is P. j .

[0107] Within the same branch, the failure of any single cell leads to the failure of the entire branch. Therefore, the reliability of the j-th branch can be defined as the minimum probability of health of all cells in that branch:

[0108]

[0109] Where RSj represents the reliability of the j-th branch, Rj i Let R be the normal operating probability of the i-th battery in this branch. In a parallel system, as long as at least one series branch survives, the system can continue to supply power. Then the reliability R of the entire system is... sysThe complement of the product of branch failure probabilities, as follows.

[0110]

[0111] where R sys is the reliability of the system, in the Monte Carlo simulation, based on the single reliability obtained by each round of sampling, real-time call the above series and parallel aggregation formula, dynamic calculation and statistics system level power failure probability and expected power supply.

[0112] Step 4: After completing the electric-thermal-capacity coupling model in step 1, the thermal runaway propagation model in step 2, and the single, branch and system level reliability aggregation and parameter calibration in step 3, integrate each model into a complete Monte Carlo simulation platform to dynamically evaluate the system failure risk and reliability with actual or random working condition sequences (including charging and discharging power curves, environmental temperature and load demand, etc.) as driving input. Specifically, each simulation round first calls the coupling model to calculate the SOC, terminal voltage, internal and surface temperature, and remaining capacity of all battery cells, then simulates the heat propagation between battery cells under the given working condition according to the thermal runaway propagation model, performs Bernoulli sampling on each cell based on the joint failure probability constructed by the Copula function to determine the failure occurrence, and updates the temperature field after triggering thermal runaway to the adjacent battery. In turn, the failure event and the power supply time are counted at the single cell level, the branch failure rate and the power supply are determined at the branch level with the weakest single reliability, and the system reliability is calculated at the system level according to the parallel system topology to aggregate all branch states. After the simulation is completed, all rounds of three-level indicators at each discrete time are statistically summarized to generate the power failure probability curve and system reliability evolution map throughout the life cycle, providing quantitative decision-making basis for capacity configuration, topology optimization and operation scheduling of network-type BESS. See the following sub-steps for details:

[0113] 1) First, the model performs data fitting to obtain the initial parameter set θ0={E 0,i ,R 0,i ,SOC 0,i ,T 0,i …}N i=1. At each discrete time step, the system scheduling decisions and external load curves are scheduled and output by scheduling the state vector of each battery cell {SOC i (t),T i (t),Q aged,i}N t=1.

[0114] 2) After accumulating the probabilities of each time step, the cumulative survival function of the single cell is derived and the reliability vector R(t)={R i(t)}N i=1. Further, according to the BESS electrical topology, the single cell level health evolution is mapped to the series branch and parallel system level, and a complete multi-state system performance description framework is constructed.

[0115] 3) Set the Monte Carlo simulation round N MC , for each round r = 1, 2, 3, …, N MC , and each time t k , the state evolution of each single battery is dynamically calculated with the external load curve and scheduling strategy as the driving signal. At each discrete time t k , according to the current health state probability, Bernoulli sampling is performed to update the health state H(r) i(t k ). At the same time, the heat diffusion path of the out-of-control single cell to the adjacent battery is tracked using a discrete heat conduction network, and the global temperature field is dynamically adjusted.

[0116] 4) After all N MC round simulations are completed, the system overall health state H(r) i(t k ) is statistically averaged to obtain the empirical reliability curve R sys (t) of the system at each time. The reliability evaluation process is shown in Figure 4 .

[0117] The following is further described in combination with a specific implementation method:

[0118] A 10x10 two-dimensional battery grid model is constructed in the MATLAB environment, and it is divided into four subarrays composed of 25 single batteries. The thermal runaway trigger threshold and its propagation behavior in each subarray are simulated to quantitatively evaluate the influence of single cell out-of-control on the overall safety and reliability of the battery pack. On the basis of measuring the battery cell performance degradation data using a battery detection device and setting the relevant parameters of the model, the reliability of the battery energy storage system of this structure is analyzed.

[0119] In the example, the daily charging start time is set to 5 o'clock and 17 o'clock, and the discharging start time is set to 8 o'clock and 20 o'clock. Each time the expected charging and discharging duration is 2h.

[0120] The simulation results are shown in Figure 5As shown, the system operation process can be divided into three typical stages, corresponding to the three key time points marked a, b, c in the figure: in the initial stage before the a point (0-600 seconds), the system temperature slowly rises, the performance index basically maintains at the nominal level, and no thermal runaway occurs; near the b point (about 700 seconds), part of the battery monomers trigger thermal runaway events due to temperature rise, the heat released locally spreads rapidly to the surrounding batteries along the spatial topological structure, resulting in a rapid decline trend of the system performance; after the c point (about 1000 seconds), the thermal runaway propagation range tends to be stable, the boundary between the runaway area and the healthy area gradually becomes clear, the system performance attenuation rate slows down significantly, and the whole enters a stable recession stage. This process reflects the sudden and phased influence characteristics of thermal runaway propagation on the reliability of the energy storage system.

[0121] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto, any change or replacement within the technical range disclosed by the present application can be easily thought by any person skilled in the art, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A reliability evaluation method for a multi-level grid-connected energy storage system based on electro-thermal-capacity coupling, characterized in that, Comprising the following steps: Step 1: Constructing battery electric-thermal-capacity coupling model Based on the multi-physical field coupling failure behavior of network-based energy storage system under multiple high-rate charging and discharging conditions, a battery electric-thermal-capacity coupling model suitable for high-rate conditions is constructed; the model includes an electrical sub-model, a thermal characteristic sub-model, and a capacity degradation sub-model, through the closed-loop coupling of the three to realize the dynamic synchronous update of current, temperature and capacity degradation, and extract the key parameters inside the battery, including temperature, state of charge SOC and real-time capacity; Step 2: Constructing inter-battery thermal runaway propagation model The inter-battery thermal runaway propagation model includes a heat conduction network and a joint failure probability model, wherein the heat conduction network is based on the geometric arrangement and thermal coupling characteristics of the battery monomers in the battery pack, a two-dimensional lumped convection model is constructed to calculate the heat transfer amount between adjacent battery monomers, and the heat flow path and boundary conditions are determined; The joint failure probability model is constructed by introducing the Copula function, considering the thermal runaway triggering probability and capacity degradation probability of the battery monomer as correlated random variables, and embedding the joint failure probability model into the heat conduction network as the probability driving mechanism for triggering thermal runaway propagation; Through the cooperation of the failure probability model and the heat conduction network W, the propagation path, propagation rate and capacity degradation evolution trend of thermal runaway in the battery pack under different conditions are simulated; Step 3: Multi-level reliability evaluation based on Monte Carlo simulation Embed the electric-thermal-capacity coupling model and the thermal runaway propagation model into the Monte Carlo simulation framework, and use a three-level analysis structure of "monomer level-branch level-system whole group level" for reliability evaluation.

2. The method of claim 1, wherein, The step 3 uses a three-level analysis structure of "monomer level-branch level-system whole group level" for reliability evaluation, specifically including: Initial parameter sampling: For the initial state parameters of each battery monomer in the system, a large number of independent samples are generated through a set probability distribution as the initial state of each simulation round, and each sample represents the possible state of the system in a simulation round, the initial state parameters include SOC, cell temperature, capacity state and Copula correlation parameters; Online simulation and index calculation: In each simulation round, the electrical response, temperature change and capacity degradation process of the battery monomer are simulated based on the electric-thermal-capacity coupling model, and the thermal runaway probability and capacity failure probability are determined combined with the thermal runaway propagation model; through Bernoulli sampling, it is judged whether the monomer fails, and in turn: Monomer level: Based on the state of charge and temperature conditions, the monomer failure probability is calculated; Branch level: Calculate the branch failure probability and unsupplied power based on the minimum reliability of each monomer in the series branch; System whole group level: Aggregate the survival / failure status of all branches according to the parallel topology, calculate the system-level failure probability, and draw the reliability curve combined with the scheduling strategy and load model; Result statistics: After repeating a large number of simulation rounds, statistically analyze all the results to obtain the reliability evolution curve and system reliability evolution map in the whole life cycle of the system, the reliability evolution curve includes power failure probability, reliability trend and risk distribution at different times.

3. The method of claim 1, wherein, In the electric-thermal-capacity coupling model: The electrical sub-model: the input is external charging and discharging current, initial state of charge (SOC) and remaining available capacity, and the output is battery terminal voltage, updated SOC and generated heat energy; The thermal characteristic sub-model: the input is heat energy output by the electrical sub-model, ambient temperature and core and shell temperature at the previous time, and the output is updated internal temperature of the battery cell and surface temperature of the shell; The capacity degradation sub-model: the input is SOC of the electrical sub-model, temperature history of the thermal characteristic sub-model and cumulative cycle number, and the output is remaining available capacity; The three sub-models realize dynamic synchronous update of current, temperature and capacity degradation through closed-loop feedback.

4. The method of claim 1, wherein, The joint failure probability model is based on data of the electrical-thermal-capacity coupling model, uses a Copula function to describe the correlation between battery thermal runaway probability and capacity degradation probability, and quantitatively evaluates the joint risk of simultaneous thermal runaway and capacity failure of a single battery under a given state of charge, temperature and cycle history.

5. The method of claim 1, wherein, The thermal runaway propagation model is based on the relationship between the discretized heat conduction network and the thermal resistance between nodes, establishes a thermal runaway chain propagation mechanism; when the temperature of adjacent single bodies exceeds the threshold, a propagation event is triggered, the joint failure rate of the single body level is iteratively transmitted to the entire battery cluster, and the diffusion process of thermal runaway from a single point to multiple points is simulated.

6. The method of claim 1, wherein, The electrical sub-model calculates the terminal voltage based on an improved equivalent circuit model, the thermal characteristic sub-model couples the heat energy output by the electrical sub-model and the environmental heat exchange through a simplified two-node thermal network, and the capacity degradation sub-model estimates the capacity degradation rate by comprehensively considering the effects of cycle aging and calendar aging, thereby forming an electrical-thermal-capacity linkage evolution mechanism.

7. The method of claim 1, wherein, The battery thermal runaway probability is obtained by referring to a pre-constructed thermal runaway probability mapping table, the mapping table is based on the probability density function and the cumulative distribution function fitted from multiple sets of experimental data, and reflects the risk of battery single body under different chemical systems, manufacturing processes and environmental conditions under each SOC-temperature condition.

8. The method of claim 1, wherein, The two-dimensional lumped convection model of the heat conduction network regards each battery cell in the battery pack as a node, and heat transfer comes from the surrounding eight adjacent battery cells and fluid nodes, reflecting the thermal interaction characteristics between batteries at different spatial positions.

9. A device for reliability evaluation of a multi-level grid-connected energy storage system based on electro-thermal-capacity coupling, characterized in that, The battery electrical-thermal-capacity coupling model construction module is used to construct a battery electrical-thermal-capacity coupling model suitable for high-rate operation conditions based on the multi-physical field coupling failure behavior of the network-based energy storage system under various high-rate charging and discharging conditions; the model includes an electrical sub-model, a thermal characteristic sub-model and a capacity degradation sub-model, and through the closed-loop coupling of the three sub-models, dynamic synchronous update of current, temperature and capacity degradation is realized, and internal key parameters of the battery are extracted, including temperature, state of charge (SOC) and real-time capacity; The battery thermal runaway propagation model construction module is used to construct a battery thermal runaway propagation model including a heat conduction network and a joint failure probability model, wherein the heat conduction network is based on the geometric arrangement and thermal coupling characteristics of the single bodies in the battery pack, a two-dimensional lumped convection model is constructed to calculate the heat transfer amount between adjacent battery single bodies, and the heat flow path and boundary conditions are determined; ​ The joint failure probability model is constructed by introducing a Copula function, taking the thermal runaway triggering probability and the capacity degradation probability of the battery cell as correlated random variables, and embedding the joint failure probability model into the heat conduction network as a probability driving mechanism for triggering the spread of thermal runaway; the failure probability model is combined with the heat conduction network W to simulate the spread path, propagation rate and capacity degradation evolution trend of thermal runaway in the battery pack under different working conditions; The multi-level reliability evaluation module is used for embedding the electro-thermal-capacity coupling model and the thermal runaway spread model into a Monte Carlo simulation framework, and performing reliability evaluation by adopting a three-level analysis structure of "cell level-branch level-system whole group level".

10. The apparatus of claim 9, wherein, The multi-level reliability evaluation module adopts a three-level analysis structure of "cell level-branch level-system whole group level" to perform reliability evaluation, specifically including: Initial parameter sampling: a large number of independent samples are generated by a set probability distribution for the initial state parameters of each battery cell in the system as the initial state of each simulation round, and each sample represents the possible state of the system in a simulation round, the initial state parameters including SOC, cell temperature, capacity state and Copula correlation parameters; Online simulation and index calculation: in each simulation round, the electrical response, temperature change and capacity degradation process of the battery cell are simulated based on the electro-thermal-capacity coupling model, and the thermal runaway probability and capacity failure probability are determined by combining the thermal runaway spread model; whether the cell fails is judged by Bernoulli sampling, and the following is calculated: Cell level: based on the state of charge and temperature conditions, the cell failure probability is calculated; Branch level: the minimum value of the reliability of each cell in the series branch is used to calculate the branch failure probability and the unsupplied power; System whole group level: the survival / failure state of all branches is aggregated according to the parallel topology, the system level failure probability is calculated, and the reliability curve is drawn by combining the scheduling strategy and the load model; Result statistics: after a large number of simulation rounds, all results are statistically analyzed to obtain the reliability evolution curve and system reliability evolution atlas in the whole life cycle of the system, and the reliability evolution curve includes the power failure probability, reliability trend and risk distribution at different times.

Images