A method and system for calculating maximum available energy of a lithium-ion battery

Abstract

The application provides a lithium ion battery maximum available energy calculation method and system fusing working conditions, energy consumption and temperature, aiming at the actual available energy of the lithium ion battery, considering the influence of working conditions, heating energy consumption and temperature on the available energy, establishing a battery working condition prediction model based on traffic signal timing information, respectively simplifying modeling of key components according to the whole vehicle power system configuration to establish a whole vehicle energy consumption prediction model, establishing a temperature prediction model based on a multi-physical field coupling model reduction, and being capable of outputting the output energy prediction under a given working condition, capable of improving the available energy estimation precision of the lithium ion battery and being used for guiding target temperature calculation of a battery thermal management system.

Description

Technical Field

[0001] This invention relates to the field of lithium-ion battery thermal management technology, specifically to a method and system for calculating the maximum available energy of a lithium-ion battery by integrating operating conditions, energy consumption, and temperature. Background Technology

[0002] As the production, sales and ownership of new energy vehicles continue to increase, the consumer market's demands for new energy vehicles are constantly rising. As a key component of new energy vehicles, the performance and control of power batteries directly affect the user experience of new energy vehicle customers.

[0003] However, the state estimation and energy consumption prediction of vehicle-mounted power batteries often lack widely applicable methods due to the difficulty in balancing computational load and accuracy in their models. During low-temperature start-up, the energy that the battery can release increases with the increase of heating temperature within the target temperature range, but the energy consumption caused by heating also increases with the increase of the target temperature. Setting an inappropriate heating target temperature may lead to a decrease in the energy released by the system. However, traditional methods lack consideration of the heating target temperature, and only decide whether to stop heating by analyzing whether the battery's maximum temperature has reached a threshold. Traditional methods for calculating the remaining energy of the battery will also have large errors as a result, and cannot assess whether the battery can release more energy after heating. The method for calculating the maximum available energy is missing or nonexistent. Therefore, new solutions are needed to address the complex operating conditions of vehicle-mounted lithium-ion batteries and the lack of continuous adaptive correction methods. Summary of the Invention

[0004] To address the challenges of balancing computational load and accuracy in calculating the maximum available energy of lithium-ion batteries due to the complexity of battery operating conditions and the lack of continuous adaptive correction methods, this invention proposes a fusion-based method for calculating the maximum available energy of lithium-ion batteries, integrating operating conditions, energy consumption, and temperature. This method considers the impact of operating conditions, heating energy consumption, and temperature on the actual available energy of lithium-ion batteries, establishing separate prediction models for operating conditions, vehicle energy consumption, and temperature. It can output a predicted output energy under a given operating condition. This method improves the accuracy of lithium-ion battery available energy estimation and can guide the calculation of target temperatures for battery thermal management systems. This invention also relates to a fusion-based system for calculating the maximum available energy of lithium-ion batteries, integrating operating conditions, energy consumption, and temperature.

[0005] The technical solution of the present invention is as follows:

[0006] A method for calculating the maximum available energy of a lithium-ion battery that integrates operating conditions, energy consumption, and temperature, characterized by comprising the following steps;

[0007] S1. Establish a battery usage condition prediction model based on traffic signal timing information, and comprehensively optimize the battery usage condition prediction model by combining vehicle fuel economy, road traffic flow and vehicle acceleration and deceleration frequency. Use the model predictive control algorithm to plan future vehicle speed for the optimized battery usage condition prediction model.

[0008] S2. Simplify the modeling of key components according to the configuration of the vehicle power system to establish a vehicle energy consumption prediction model. The vehicle energy consumption prediction model is used to predict the energy consumption under future operating conditions under a given input condition.

[0009] S3. Establish a temperature prediction model based on the order reduction of a multi-physics coupling model. This includes constructing a multi-physics coupling model of the lithium-ion battery based on the lithium-ion battery mechanism, performing multiple simulations of the multi-physics coupling model under different input conditions, and then using the PSO-CNN-LSTM fusion algorithm to reduce the order of the multi-physics coupling model based on the simulation results. Finally, a temperature prediction model is established using a data-driven method to predict the temperature field distribution and future temperature rise of the lithium-ion battery under given operating conditions and heating conditions.

[0010] S4. Establish a maximum available energy prediction model, including predicting the battery terminal voltage, and planning the input current based on the future vehicle speed calculated according to the vehicle operating conditions predicted in step S1. Then, based on the input current, battery terminal voltage, the energy consumption of the vehicle predicted in step S2, the temperature field and future temperature rise during the lithium-ion battery discharge process predicted in step S3, establish a maximum available energy prediction model that integrates operating conditions, energy consumption and temperature. The maximum available energy of the lithium-ion battery is calculated iteratively through the future operating conditions obtained by the maximum available energy prediction model.

[0011] Preferably, in or after step S2 or S3, the heating loss of the lithium-ion battery is predicted based on the power of the heating film under given operating conditions and heating conditions; in step S4, a maximum available energy prediction model integrating operating conditions, energy consumption, and temperature is established based on the input current, battery terminal voltage, the energy consumption of the whole vehicle predicted in step S2, the temperature field and future temperature rise during the discharge process of the lithium-ion battery predicted in step S3, and the predicted heating loss of the lithium-ion battery.

[0012] Preferably, step S1 utilizes V2V and V2I communication to acquire vehicle position, speed, and traffic light status information in the road environment, establishes a traffic signal timing model to calculate the speed range of a single vehicle or vehicle queue when passing through a traffic light in a green light state, and then establishes a battery usage condition prediction model based on traffic signal timing information. This model is then comprehensively optimized by combining vehicle fuel economy, road traffic flow, and the number of vehicle accelerations and decelerations while meeting vehicle power requirements. The optimized battery usage condition prediction model is then used to plan future vehicle speeds through a model predictive control algorithm. Fuel economy includes vehicle energy consumption, electricity consumption, or carbon emissions; traffic flow includes traffic congestion time, waiting time, or low-speed operation time; the number of accelerations and decelerations includes average and minimum distances between multiple vehicles; and vehicle power includes vehicle power requirements.

[0013] Preferably, in step S2, based on the configuration of the vehicle powertrain system, the key components are simplified and modeled to establish a vehicle energy consumption prediction model. The vehicle energy consumption prediction model includes an engine model, a motor model, a lithium-ion battery model, a battery thermal management system model, a transmission model, and a vehicle dynamics model. Through the mutual coupling relationship between the models, the vehicle energy consumption prediction model can predict the energy consumption under future operating conditions under a given input operating condition.

[0014] Preferably, in step S3, the input to the electrochemical-thermal-mechanical multiphysics coupling model of the ion battery is a given power / current-time curve, ambient temperature, and initial battery state, and the output is the battery virtual cycle voltage output and temperature under this operating condition. The PSO-CNN-LSTM fusion algorithm simplifies the electrochemical-thermal-mechanical multiphysics coupling model based on the simulation results. Using the power / current-time curve, ambient temperature, and initial battery state from the simulation data as input and the battery temperature as output, a training set and a test set are established and divided into training, test, and validation sets according to a specific ratio. In the PSO-CNN-LSTM fusion algorithm, PSO optimizes the hyperparameters of CNN and LSTM, CNN performs pooling dimensionality reduction on the original features to extract key parameters, and LSTM uses the dimensionality-reduced features to predict battery temperature from time-series data. Finally, a temperature prediction model is established through a data-driven approach.

[0015] Preferably, in step S4, the battery terminal voltage is predicted based on a finite element model, a data-driven model, or a circuit model. The finite element model is a finite element model coupled with a P2D model, a thermal model, and an aging model based on electrochemical mechanisms. The driving model is a data-driven model with temperature, voltage, and current as inputs. The circuit model is a circuit model coupled with an equivalent circuit model, a heat generation model, and a heat transfer model.

[0016] A lithium-ion battery maximum available energy calculation system integrating operating conditions, energy consumption, and temperature is characterized by comprising an operating condition prediction model establishment module, a vehicle energy consumption prediction model establishment module, a temperature prediction model establishment module, and a maximum available energy prediction model establishment and calculation module, wherein the operating condition prediction model establishment module, the vehicle energy consumption prediction model establishment module, and the temperature prediction model establishment and calculation module are all connected to the maximum available energy prediction model establishment and calculation module.

[0017] The working condition prediction model building module establishes a battery usage working condition prediction model based on traffic signal timing information, and comprehensively optimizes the battery usage working condition prediction model by combining the vehicle's fuel economy, road traffic flow, and the number of vehicle acceleration and deceleration. The optimized battery usage working condition prediction model is then used to plan future vehicle speed through a model prediction control algorithm.

[0018] The vehicle energy consumption prediction model building module simplifies the modeling of key components according to the configuration of the vehicle power system to build a vehicle energy consumption prediction model. The vehicle energy consumption prediction model is used to predict the energy consumption under future operating conditions under a given input condition.

[0019] The temperature prediction model building module establishes a temperature prediction model based on a reduced-order multiphysics coupling model. This includes constructing an electrochemical-thermal-mechanical multiphysics coupling model of a lithium-ion battery based on the lithium-ion battery mechanism. By giving different input operating conditions, the electrochemical-thermal-mechanical multiphysics coupling model is simulated multiple times. Then, the PSO-CNN-LSTM fusion algorithm is used to reduce the order of the electrochemical-thermal-mechanical multiphysics coupling model based on the simulation results. Finally, a temperature prediction model is established through a data-driven method to predict the temperature field distribution and future temperature rise of the lithium-ion battery under given operating conditions and heating conditions.

[0020] The maximum available energy prediction model establishment and calculation module establishes a maximum available energy prediction model, including predicting the battery terminal voltage, and calculating the future vehicle speed based on the battery terminal voltage and the vehicle operating conditions predicted by the operating condition prediction model establishment module to obtain the input current. Then, based on the input current, battery terminal voltage, the vehicle energy consumption prediction model establishment module predicting the energy consumption of the whole vehicle, and the temperature prediction model establishment module predicting the temperature field during the lithium-ion battery discharge process and future temperature rise, a maximum available energy prediction model integrating operating conditions, energy consumption, and temperature is established. The maximum available energy of the lithium-ion battery is calculated iteratively through the future operating conditions obtained by the maximum available energy prediction model.

[0021] Preferably, it also includes a heating energy consumption calculation module, which is connected to the maximum available energy prediction model establishment calculation module. The heating energy consumption calculation module is used to calculate the heating loss of the lithium-ion battery based on the power of the heating film under given operating conditions and heating conditions. The maximum available energy prediction model establishment calculation module establishes a maximum available energy prediction model that integrates operating conditions, energy consumption, and temperature based on the input current, battery terminal voltage, the energy consumption of the whole vehicle predicted by the whole vehicle energy consumption prediction model establishment module, the temperature field and future temperature rise during the discharge process of the lithium-ion battery predicted by the temperature prediction model establishment module, and the heating loss of the lithium-ion battery predicted by the heating energy consumption calculation module.

[0022] Preferably, the vehicle energy consumption prediction model building module simplifies and models the key components of the vehicle powertrain system configuration to establish a vehicle energy consumption prediction model. The vehicle energy consumption prediction model includes an engine model, a motor model, a lithium-ion battery model, a battery thermal management system model, a transmission model, and a vehicle dynamics model. Through the mutual coupling relationship between the models, the vehicle energy consumption prediction model can predict the energy consumption under future operating conditions under a given input operating condition.

[0023] Preferably, the input of the electrochemical-thermal-mechanical multiphysics coupling model of the ion battery is a given power / current-time curve, ambient temperature, and initial battery state, and the output is the battery virtual cycle voltage output and temperature under the operating condition. The PSO-CNN-LSTM fusion algorithm is used to reduce the order of the electrochemical-thermal-mechanical multiphysics coupling model based on the simulation results. The power / current-time curve, ambient temperature, and initial battery state in the simulation data are used as inputs, and the battery temperature is used as the output to establish training and testing sets, which are divided into training, testing, and validation sets according to a specific ratio. In the PSO-CNN-LSTM fusion algorithm, PSO is used to optimize the hyperparameters of CNN and LSTM, CNN is used to perform pooling and dimensionality reduction on the original features to extract key parameters, and LSTM is used to predict the battery temperature based on the dimensionality-reduced features. Then, a temperature prediction model is established through a data-driven method.

[0024] The beneficial effects of this invention are as follows:

[0025] This invention proposes a method for calculating the maximum usable energy of lithium-ion batteries by integrating operating conditions, energy consumption, and temperature. This method deeply analyzes the factors affecting the actual usable energy of lithium-ion batteries, considering the impact of temperature, heating energy consumption, and operating conditions on usable energy. It establishes temperature prediction models, vehicle energy consumption prediction models, and operating condition prediction models, and uses the future operating conditions obtained from the maximum usable energy prediction model to iteratively calculate the maximum usable energy of the lithium-ion battery. This method can accurately predict the output energy under given conditions. The method decouples various factors affecting the energy used by lithium-ion batteries and establishes separate models for predictive analysis. Step S1 predicts the vehicle operating conditions and calculates the battery operating conditions and input current. Step S2 predicts the energy consumption of the entire vehicle. Step S3 predicts the temperature field and future temperature rise during the discharge process of the lithium-ion battery. Step S4 establishes the maximum usable energy prediction model integrating operating conditions, energy consumption, and temperature. The future operating conditions obtained from this model are used to iteratively calculate the maximum usable energy of the lithium-ion battery, achieving accurate prediction and simulation of the actual usable energy of the lithium-ion battery. This is beneficial for improving the estimation of the vehicle's driving range and enhancing the user experience.

[0026] This invention also relates to a system for calculating the maximum available energy of a lithium-ion battery by integrating operating conditions, energy consumption, and temperature. Corresponding to the aforementioned method for calculating the maximum available energy of a lithium-ion battery by integrating operating conditions, energy consumption, and temperature, this system can be understood as a system that implements the aforementioned method for calculating the maximum available energy of a lithium-ion battery by integrating operating conditions, energy consumption, and temperature. This system includes a module for establishing an operating condition prediction model, a module for establishing a vehicle energy consumption prediction model, a module for establishing a temperature prediction model, and a module for establishing and calculating the maximum available energy prediction model. These modules work collaboratively to establish future vehicle speed planning based on a traffic signal timing model, establish an energy consumption prediction model based on vehicle dynamics, and establish a temperature prediction model based on a reduced-order multiphysics coupling model. A reduced-order model of the electrochemical-thermal-mechanical multiphysics coupling model is established using a data-driven method, achieving accuracy comparable to the original model while significantly improving computational efficiency and reducing computational complexity. Finally, the maximum available energy prediction model is established, and the actual maximum available energy of the battery is calculated iteratively. This method comprehensively considers the influencing factors on the true available energy of lithium-ion batteries, improves the accuracy of lithium-ion battery available energy estimation, and can be used to guide the calculation of the target temperature of the battery thermal management system. Attached Figure Description

[0027] Figure 1 This is a flowchart of the method for calculating the maximum available energy of a lithium-ion battery according to the present invention.

[0028] Figure 2 This is the result of using a driving condition prediction model to plan future vehicle speeds.

[0029] Figure 3 This is a schematic diagram illustrating the principle of temperature prediction for lithium-ion batteries.

[0030] Figure 4 This is the result of the maximum available energy calculation. Detailed Implementation

[0031] The present invention will now be described with reference to the accompanying drawings.

[0032] This invention relates to a method for calculating the maximum available energy of a lithium-ion battery that integrates lifespan, energy consumption, and temperature. The flowchart of this method is shown below. Figure 1 As shown, it includes the following steps;

[0033] Step S1: Establish a battery usage condition prediction model

[0034] A battery usage condition prediction model based on traffic signal timing information is established. The model is then comprehensively optimized by combining vehicle fuel economy, road traffic flow, and the number of accelerations and decelerations. The optimized battery usage condition prediction model is then used to plan future vehicle speeds through a model predictive control algorithm.

[0035] The battery usage condition prediction model described in this embodiment refers to network information such as traffic light status information, distance of driving segments, and road speed limits obtained based on V2I and V2V communication. Traffic environment information, including traffic light timing (SPAT) information, can be obtained interactively through V2I communication. After acquiring this connected information based on V2V and V2I communication, a traffic signal timing model is established to calculate the speed range of a single vehicle or vehicle convoy when passing through a traffic light in a green light state. Then, a battery usage condition prediction model based on traffic signal timing information is established. Taking into account vehicle fuel economy, road traffic flow, and the number of vehicle accelerations and decelerations as objectives, the battery usage condition prediction model is comprehensively optimized while meeting vehicle power requirements. The optimized battery usage condition prediction model is then used to plan future vehicle speeds using a model predictive control algorithm to calculate the optimal vehicle speed, reducing the number of accelerations, decelerations, and stops, thereby improving vehicle fuel economy. Fuel economy includes vehicle energy consumption, electricity consumption, or carbon emissions; traffic flow includes traffic congestion time, waiting time, or low-speed operation time; acceleration and deceleration frequency includes average distance between multiple vehicles and minimum distance between vehicles; and vehicle power performance includes vehicle power requirements.

[0036] For multi-vehicle platoons, a longitudinal dynamics model can be established for any vehicle in the platoon as shown in formulas (1)(2)(3). In this case, a single vehicle is a special case of multi-vehicle platoons, where there is only the vehicle at the head of the platoon on the road, and it will not be discussed separately.

[0037] x i (t+1)=x i (t)+f i (xi (t),u i (t))Δt (1)

[0038]

[0039] x i (t)=[s i (t),v i (t)] T (3)

[0040] In the formula, i represents the i-th vehicle; x is the state variable of the longitudinal dynamics model; s is the current position of the vehicle, in meters; v is the velocity of the vehicle, in meters per second; and u is the current acceleration of the vehicle, in meters per second. 2 Δt is the time step of the upper-level control system; F f F is the rolling resistance of a vehicle, measured in N (N). w F is air resistance, measured in N (N). i F is the slope resistance, measured in N (N). j It is the acceleration resistance, measured in N.

[0041] The optimization objective of the upper-level control system is to reduce vehicle fuel consumption by rationally planning future vehicle speeds while satisfying system constraints. A fuel consumption model based on vehicle dynamics is selected as the energy consumption model for connected hybrid electric vehicles. The specific energy consumption calculation method is as follows:

[0042]

[0043]

[0044]

[0045] v imin ≤v i (t)≤v imax (7)

[0046] a imin ≤a i (t)≤a imax (8)

[0047] In the formula, η is the fuel consumption rate of the i-th vehicle; ieff H is the driving efficiency of the i-th vehicle; LHV It is the low calorific value of fuel; P idrive It is the effective driving power of the i-th vehicle; a i v is the acceleration of the i-th vehicle, which is also the control variable; i η is the speed of the i-th vehicle; β is a flag indicating whether the hybrid vehicle is in driving or braking mode; η irecv is the regenerative braking efficiency of the i-th vehicle; imin It is the minimum speed allowed under road driving conditions; v imax It is the maximum permissible speed under road driving conditions; aimin is the minimum permissible acceleration under road driving conditions; a imax It is the maximum allowable acceleration under road driving conditions.

[0048] Because common vehicle cruise control methods cannot enable hybrid vehicles to pass through traffic lights on a green light, in a connected environment, hybrid vehicles can pre-plan their speed based on network information obtained from V2I and V2V connections, effectively avoiding situations where the vehicle stops at a red light. This embodiment calculates the vehicle's target speed range based on traffic signal timing information obtained from SPAT.

[0049]

[0050]

[0051]

[0052] t c =t g +t r +t y (12)

[0053]

[0054] In the formula, v ilow It is the lower limit of the target speed of the i-th vehicle; v iup d is the upper limit of the target speed of the i-th vehicle; t is the total travel time of the vehicle; ia k is the distance of the i-th vehicle from the traffic light ahead; w It is the number of traffic light cycles, usually an integer, when k = k w t c At that time, k w Increase by 1; t c It is the duration of the traffic light cycle; t r It refers to the duration of the red light on a traffic light; t g It is the duration of the green light; t y It represents the duration of the yellow light; the mod function is the modulo function, which represents the total vehicle travel time t divided by the traffic light cycle t. c The remainder.

[0055] In single-vehicle traffic scenarios, the prediction of the optimal target speed for hybrid electric vehicles should primarily consider the vehicle's fuel economy, road traffic flow, and the number of accelerations and decelerations as objectives, employing a multi-objective comprehensive optimization. The specific optimization function is as follows:

[0056]

[0057] In the formula, ω1, ω2, and ω3 are the weight coefficients of different optimization terms in the objective function; T is the time window of model predictive control; s is the vehicle position; v is the vehicle speed; v target 'a' is the target vehicle speed; 'a' is the vehicle acceleration; 'Δt' is the simulation calculation step size. The final future vehicle speed planning result is as follows: Figure 2 As shown.

[0058] Step S2: Establish an energy consumption prediction model based on vehicle dynamics.

[0059] Based on the configuration of the vehicle powertrain system, key components are simplified and modeled to establish a vehicle energy consumption prediction model. The vehicle energy consumption prediction model is used to predict energy consumption under future operating conditions under a given input condition.

[0060] In this embodiment, based on the vehicle powertrain configuration (powertrain structure), key components are simplified and modeled to establish a vehicle energy consumption prediction model. Preferably, the vehicle energy consumption prediction model includes an engine model, a motor model, a lithium-ion battery model, a battery thermal management system model, a transmission model, and a vehicle dynamics model. Through the coupling relationship between the models, the vehicle energy consumption prediction model can predict energy consumption under future operating conditions under a given input condition. This embodiment only uses the vehicle dynamics model as an example to predict the total energy consumption and instantaneous energy consumption under future operating conditions under a given input condition.

[0061] In the vehicle dynamics model established in this embodiment, the vehicle's motion state is determined by traction force, driving resistance, etc., where driving resistance includes rolling resistance, air resistance, gradient resistance, and acceleration resistance. The driving balance equation can be obtained as follows:

[0062] F t =F f +F w +F i +F j (15)

[0063] In the formula, F t It is the traction force during vehicle movement, F f F is the frictional rolling resistance caused by tire deformation and wear. w It is the air resistance experienced by the vehicle, F i It is the slope resistance, F j These are acceleration resistance values, all in N.

[0064] The specific formulas for calculating the relevant resistance are as follows:

[0065] Rolling resistance

[0066] F f =mgfcosα (16)

[0067] In the formula, m is the effective mass of the vehicle, in kg; f is the rolling resistance coefficient; α is the road gradient; and g is the acceleration due to gravity, in m / s². 2 .

[0068] air resistance

[0069]

[0070] In the formula, ρ is the air density, with units of kg / m³. 3 C d A is the air resistance coefficient; A is the vehicle's frontal area, measured in meters. 2 u a This is the vehicle's speed, measured in m / s.

[0071] Slope resistance

[0072] F i =mgsinα (18)

[0073] Acceleration resistance

[0074] F j =δma (19)

[0075] In the formula, δ is the vehicle rotational mass conversion factor; a is the vehicle acceleration, in m / s². 2 .

[0076] Vehicle energy consumption Q based on vehicle dynamics calculation v The calculation formula is as follows:

[0077] Q v (t)=F t vΔt (20)

[0078] In the formula, F t Δt is the total traction force of the vehicle, v is the vehicle speed, and Δt is the time interval.

[0079] Step S3: Establish a temperature prediction model based on a reduced-order multiphysics coupling model.

[0080] An electrochemical-thermal-mechanical multiphysics coupling model (MPMM) for lithium-ion batteries was established based on the lithium-ion battery mechanism. The model's inputs included a given power-time or current-time curve, ambient temperature, and initial battery state. The model's outputs were the virtual cycle voltage output and temperature under the given operating conditions. Multiple simulations were performed on the MPMM under different input conditions, and the simulation data was stored. Then, a PSO-CNN-LSTM fusion algorithm was used to reduce the order of the MPMM model based on the simulation results, using the power-time or current-time curve, ambient temperature, and initial battery state from the simulation data. With the initial state as input and battery temperature as output, training and testing sets are established and divided into training, testing, and validation sets according to a ratio of 70%-15%-15%. In the PSO-CNN-LSTM fusion algorithm, PSO is used to optimize the hyperparameters of CNN and LSTM, CNN is used to perform pooling dimensionality reduction on the original features to extract key parameters, and LSTM is used to predict battery temperature from time-series data. Then, a reduced-order model of the electrochemical-thermal-mechanical multiphysics coupling model is established through a data-driven method to predict the temperature field distribution and future temperature rise of lithium-ion batteries under given operating conditions and heating conditions. Its accuracy is comparable to the original model, the computational efficiency is significantly improved, and the computational complexity is significantly reduced.

[0081] In this embodiment, an electrochemical-thermal-mechanical coupling model based on electrochemical mechanisms is employed, enabling high-precision simulation of the entire process of internal reaction principles and material changes in lithium-ion batteries. The traditional quasi-two-dimensional (P2D) model of lithium-ion batteries was proposed by Doyle and Newman in 1993, based on concentrated solution theory and electrode reaction kinetics. The P2D model divides the lithium-ion battery into one-dimensional and three-dimensional spaces. In the one-dimensional space, the lithium-ion battery is divided into three parts: the positive electrode, the negative electrode, and the separator. In the three-dimensional space, the positive and negative electrodes are constructed as three-dimensional spherical solids. Lithium ions move from the interior of the three-dimensional sphere to the surface, are extracted from the surface, and then, through the electrolyte and separator, reach the surface of the other electrode's three-dimensional sphere for embedding. In the one-dimensional model, the solid volume fraction, liquid volume fraction, porosity, etc., of the positive electrode, negative electrode, and separator are calculated, with the electrolyte filling the model. A mathematical model of mass conservation and charge conservation is established.

[0082] (1) Conservation of matter

[0083] The conservation of mass in lithium-ion batteries mainly concerns the diffusion and transfer of lithium ions in the electrolyte and electrode materials. Inside the spherical electrode material particles, lithium ions undergo solid-phase diffusion, which is described using Fick's second law.

[0084]

[0085] Where C s The solid phase concentration of the lithium-ion battery is denoted by r; the particle radius is denoted by D. s t represents the solid-phase diffusion coefficient of a lithium-ion battery; t represents time.

[0086] In an electrolyte, lithium ions move through diffusion and migration. The partial derivative of the electrolyte concentration and the liquid phase diffusion constant determine the diffusion rate of ions, while the liquid phase potential distribution affects the migration distribution of ions.

[0087] The total volume current density j is

[0088] j = j int +j SEI +j LP (twenty two)

[0089] Where, j SEI and j LP These represent the current densities for SEI formation and lithium plating, respectively. Therefore, the lithium ion concentration in the liquid phase can be expressed as...

[0090]

[0091] Among them, c e Lithium ion concentration at the electrode, mol / cm³ 3 ;t + It is the transfer coefficient; ε e It is the porosity of the porous electrode; D eff It is the equivalent lithium-ion diffusion coefficient of the electrolyte, in cm. 2 / s.

[0092] (2) Charge conservation

[0093] The conservation of charge means that during a chemical reaction, the current density at the same location inside the battery is equal to the sum of the current densities of the liquid and solid phases inside the battery. The conservation of charge in both the liquid and solid phases of lithium-ion batteries is calculated using Faraday's law.

[0094] Charge conservation primarily concerns the transfer of electrons and lithium ions in the P2D model. The solid-state potential distribution is explained by Ohm's law.

[0095]

[0096] Where, φ s V is the solid-state electronic potential; j is the sum of the current densities of all reactions, mA / cm². 2 ; χ is the electrode thickness, in meters; σ eff is the solid-phase equivalent conductivity, 1 / (Ω·cm). Wherein, Brug is the Bruggman factor.

[0097] The liquid phase potential distribution is measured using a reference electrode, which is simultaneously influenced by Ohm's law and the theory of concentrated solutions.

[0098]

[0099] R is the ideal gas constant; T is the temperature (K); f is the electrolyte activity coefficient; κ eff Electrode equivalent conductivity.

[0100] (3) Butler-Volmer equation

[0101] The Butler-Volmer equation is used to describe the electrode reaction kinetics within P2D. Electrode reactions at the electrolyte-solid electrode interface are described using the Butler-Volmer equation. The kinetics of lithium insertion and extraction are governed by the Butler-Volmer equation.

[0102]

[0103] i 0,int α is the exchange current density of the reaction. a,int and α c,int Here, η is the oxidation and reduction charge transfer coefficient, typically 0.5, R is the ideal gas constant, and η is the redox charge transfer coefficient. int To react overpotential

[0104] The exchange current density is calculated by the following formula.

[0105]

[0106] k int It is the initial reaction rate constant of the electrode; c e,ref This is the electrode reference lithium-ion concentration density, typically taken as 1 mol / m³. 3 ;c s,max It represents the maximum lithium-ion concentration in the electrode active material, in mol / m 3 ;c s,surf The concentration of lithium ions on the solid surface.

[0107] The overpotential of the reaction is shown in equation (28).

[0108]

[0109] Among them, U int V; R represents the reaction equilibrium potential controlled by the lithium-ion stoichiometry on the electrode material surface. film Ω represents the internal resistance formed on the surface of the electrode active material.

[0110] Consider two side reactions: lithium plating and SEI film thickening, namely SEI film growth and lithium-ion deposition occurring at the negative electrode. Assume both reactions occur at the negative electrode. A mixture of SEI film and lithium metal forms on the negative electrode surface. New SEI films continuously deposit on the negative electrode, consuming lithium ions and causing continuous electrode dissolution. Lithium deposition, or lithium plating, occurs under conditions of low-temperature charging, overcharging, and fast charging. In the simplified model, it is assumed that the deposited lithium metal will not be re-oxidized into SEI film or lithium ions. SEI film growth is determined by the electrochemical kinetics of the reaction and the electrolyte concentration at the negative electrode surface. SEI film formation follows the Tafel equation, a hallmark of irreversible kinetics.

[0111]

[0112] Where, j SEI The current density for the SEI film formation reaction; k SEI c is the reaction rate constant for SEI film formation. EC The concentration of EC on the negative electrode surface; α c,SEI =0.5 is the charge transfer coefficient; U SEI The equilibrium electromotive force for SEI formation is given. The lithium plating reaction follows the Butler-Volmer equation and occurs only when the overpotential of the reaction is negative, as shown in equation (30).

[0113]

[0114] Where j LP The current density for lithium plating; k LP α is the reaction rate constant. a,LP and α c,LP η is the charge transfer coefficient. LP This is the overpotential. In this study, α... a,LP =0.3, α c,LP =0.7. Due to the lithium deposition ratio relative to Li / Li + If the overpotential is 0, then the overpotential can be calculated using equation (31):

[0115]

[0116] The thickness of the lithium plating can be calculated using equation (32).

[0117]

[0118] Among them, M LP ρ is the molar mass of lithium metal. LP ρ represents the density of lithium metal.

[0119] The resistance of the mixture is determined by the sum of the two products and is calculated using equation (33).

[0120]

[0121] As the surface layer grows, the pores of the anode are occupied, resulting in a decrease in the electrolyte volume fraction, which is calculated using equation (34).

[0122] ε e =ε e,0 -aδ film (34)

[0123] Where, ε e,0 The initial electrode porosity.

[0124] Since many parameters in the electrochemical model are temperature-dependent, temperature corrections are required for these parameters. The relationship between parameters and temperature follows the Arrhenius equation, and the values ​​of these parameters can be calculated using equation (35).

[0125]

[0126] φ and φ ref The parameters E are at the current temperature and the reference temperature, respectively. a It is the corresponding useful work, T ref For reference temperature, 25°C is used in this embodiment.

[0127] The heat generation model of the electrochemical mechanism can be obtained through the following process, and the energy conservation formula of the battery is given.

[0128]

[0129] Where m is the battery mass, kg; C p q is the specific heat capacity, J / kg / K; q is the heat production rate; A is the surface area, m². 2 h is the thermal convection coefficient; T is the battery temperature, K; T amp The ambient temperature is in K.

[0130] Heat generation inside a battery follows the law of conservation of energy in the battery system. However, in electrochemical models, heat generation from different sources needs to be calculated separately. The mechanism of heat generation in a battery is very complex and mainly consists of the following three parts:

[0131]

[0132]

[0133]

[0134] In the formula: q0 is the ohmic heat, J; qp The heat generated by the overpotential is called polarization heat, J; q rev Irreversible heat; κ eff It is the diffusion conductivity, S / m; n is the lithium-ion battery charging coefficient; ΔS is the entropy change, calculated by the following formula:

[0135]

[0136] The overall heat generation model of the battery is shown in the following equation.

[0137]

[0138] This embodiment takes a certain 18650 NCM battery as an example. The battery capacity is 2Ah, the charging current is 10A, and the discharging current is 1A. COMSOL is used to model the above electrochemical mechanism model. With time-current as the input value, data such as the battery's output voltage and temperature change can be obtained.

[0139] This embodiment simulates battery temperature prediction information under various operating conditions, including NEDC (New European Driving Cycle), WLTC (Worldwide Harmonized Light Vehicle Test Cycle), and CLTC (China Light Vehicle Driving Cycle). A training set is constructed, and a PSO (Particle Swarm Optimization)-CNN (Convolutional Neural Network)-LSTM (Long Short-Term Memory Neural Network) fusion algorithm is used to simplify the complex electrochemical-thermal-mechanical multiphysics coupling model of lithium-ion batteries, achieving battery temperature prediction under limited computing power. Figure 3 The schematic diagram is shown. In this embodiment, CNN is used to perform pooling and dimensionality reduction on the original data features, LSTM predicts battery temperature based on the dimensionality-reduced features, and PSO is used to optimize the hyperparameters (learning rate, number of network layers, etc.) in the neural network model. The LSTM neural network architecture used in this embodiment is as follows: Figure 3 As shown.

[0140] Furthermore, in or after step S2 or S3, the heating loss of the lithium-ion battery is predicted based on the power of the heating film under given operating conditions and heating conditions.

[0141] Q h =P h t (42)

[0142] Step S4: Establish the maximum available energy prediction model

[0143] This embodiment uses a second-order equivalent circuit model to predict the battery terminal voltage U. L .

[0144]

[0145] Among them Uo The voltage drop caused by the ohmic internal resistance; I L R is the time-domain input current; o The internal resistance is ohmic; R is the concentration polarization voltage; p1 For concentration polarization resistance; C p1 For concentration polarization capacitors; R is the diffusion polarization voltage; p2 For diffusion polarization resistance; C p2 For diffusion polarization capacitance; U ocv U is the open-circuit voltage. L This is the terminal voltage.

[0146] Based on battery terminal voltage U L The input current I is obtained by calculating the future vehicle speed based on the predicted vehicle operating conditions in step S1. The calculation method for the input current I is as follows:

[0147]

[0148]

[0149] I = P c / U L (46)

[0150] In the formula, m is the vehicle's curb weight, f is the rolling resistance coefficient, g is the gravitational acceleration, α is the angle of attack, A is the frontal area, and C... d It is the air drag coefficient, v is the vehicle speed loss (future speed planning), η is the speed reduction coefficient. T It is the transmission efficiency, η E It is the electric traction efficiency, V c Q is the nominal voltage of a single cell. c It is the nominal capacity of a single battery cell, V b It is the nominal voltage of the battery system, Q. b It is the nominal capacity of the battery system, P b It is the system output voltage, P c It is the output power of a single unit, U L It refers to the voltage of a single battery cell, i.e., the terminal voltage.

[0151] Therefore, at each moment, based on the input current I (Formula 46) and the battery terminal voltage U... L (Formula 43) Step S2 predicts the energy consumption Q of the whole vehicle. v (Equation 20), Step S3 predicts the temperature field and future temperature rise (W) during the discharge process of a lithium-ion battery. e (affected by temperature), and the predicted heating loss Q of lithium-ion batteries. h(Formula 42) A comprehensive maximum available energy prediction model integrating operating conditions, energy consumption, and temperature is established. The maximum available energy (SOE) of the lithium-ion battery is calculated iteratively based on the future operating conditions obtained from the maximum available energy prediction model.

[0152] W e (T) is calculated using interpolation, specifically linear interpolation. The interpolation table is as follows:

[0153] energy 96 105 110 112 116

[0154]

[0155] Because the iterative process involves a large amount of computation, the SOE is not calculated at every moment in this embodiment. Instead, a time-scale approach is used, calculating the available energy every 500 seconds. The final available energy estimation result is as follows: Figure 4 As shown.

[0156] The overall iterative calculation process is as follows:

[0157] Initialization: After given SOE1, the vehicle speed v is calculated in step S1, the current I is calculated using formulas 44-46, and U is calculated using formula 43. L The temperature T is calculated in step S3, and the energy consumption Q is calculated using formula 20. v The heating loss Q is calculated using formula 42. h W is calculated using an interpolation table. e (T), SOE2 is calculated using formula 47, and then a similar iterative process is used to calculate SOE at the k-th second. k .

[0158] like Figure 4 The figure shows a comparison between the calculated and experimental values ​​of the usable energy of the lithium-ion battery. The solid line represents the actual usable energy value obtained through experimental measurement, while the dashed line represents the calculated usable energy value obtained using this invention. As can be seen from the figure, the usable energy calculated by this invention is close to the actual value, demonstrating good performance.

[0159] This invention also relates to a lithium-ion battery maximum available energy calculation system that integrates operating conditions, energy consumption, and temperature. Corresponding to the aforementioned lithium-ion battery maximum available energy calculation method that integrates operating conditions, energy consumption, and temperature, this system can be understood as a system that implements the aforementioned lithium-ion battery maximum available energy calculation method that integrates operating conditions, energy consumption, and temperature. The system includes an operating condition prediction model establishment module, a vehicle energy consumption prediction model establishment module, a temperature prediction model establishment module, and a maximum available energy prediction model establishment and calculation module. The operating condition prediction model establishment module, the vehicle energy consumption prediction model establishment module, and the temperature prediction model establishment module are all connected to the maximum available energy prediction model establishment and calculation module.

[0160] The battery usage prediction model building module establishes a battery usage prediction model based on traffic signal timing information. It then optimizes the model by considering vehicle fuel economy, road traffic flow, and the number of accelerations and decelerations. The optimized model is then used to plan future vehicle speeds via a model predictive control algorithm. The vehicle energy consumption prediction model building module simplifies and models key components according to the vehicle's powertrain configuration to establish a vehicle energy consumption prediction model. This model predicts energy consumption under given input conditions. The temperature prediction model building module establishes a temperature prediction model based on a reduced-order multiphysics coupling model. This includes constructing a lithium-ion battery electrochemical-thermal-mechanical multiphysics coupling model based on the lithium-ion battery mechanism. The model is simulated multiple times under different input conditions, and then PSO-CNN is used for simulation. The LSTM fusion algorithm simplifies the electrochemical-thermal-mechanical multiphysics coupling model based on simulation results, and then establishes a temperature prediction model through a data-driven method to predict the temperature field distribution and future temperature rise of the lithium-ion battery under given operating conditions and heating conditions. The maximum available energy prediction model establishment module establishes the maximum available energy prediction model, including predicting the battery terminal voltage. Based on the battery terminal voltage and the future vehicle speed calculated by the operating condition prediction module, the input current is obtained. Then, based on the input current, battery terminal voltage, the vehicle energy consumption prediction module predicting the energy consumption of the whole vehicle, and the temperature prediction model establishing module predicting the temperature field and future temperature rise during the lithium-ion battery discharge process, the operating condition-energy consumption-temperature fusion maximum available energy prediction model is established. The maximum available energy of the lithium-ion battery is calculated iteratively through the future operating conditions obtained by the maximum available energy prediction model.

[0161] Preferably, it also includes a heating energy consumption calculation module, which is connected to the maximum available energy prediction model establishment calculation module. The heating energy consumption calculation module is used to calculate the heating loss of the lithium-ion battery based on the power of the heating film under given operating conditions and heating conditions. The maximum available energy prediction model establishment calculation module establishes a maximum available energy prediction model that integrates operating conditions, energy consumption, and temperature based on the input current, battery terminal voltage, the energy consumption of the whole vehicle predicted by the whole vehicle energy consumption prediction model establishment module, the temperature field and future temperature rise during the discharge process of the lithium-ion battery predicted by the temperature prediction model establishment module, and the heating loss of the lithium-ion battery predicted by the heating energy consumption calculation module.

[0162] Preferably, the vehicle energy consumption prediction model building module simplifies and models the key components of the vehicle powertrain system configuration to establish a vehicle energy consumption prediction model. The vehicle energy consumption prediction model includes an engine model, a motor model, a lithium-ion battery model, a battery thermal management system model, a transmission model, and a vehicle dynamics model. Through the mutual coupling relationship between the models, the vehicle energy consumption prediction model can predict the energy consumption under future operating conditions under a given input operating condition.

[0163] Preferably, the input of the electrochemical-thermal-mechanical multiphysics coupling model of the ion battery is a given power / current-time curve, ambient temperature, and initial battery state, and the output is the battery virtual cycle voltage output and temperature under the operating condition. The PSO-CNN-LSTM fusion algorithm is used to reduce the order of the electrochemical-thermal-mechanical multiphysics coupling model based on the simulation results. The power / current-time curve, ambient temperature, and initial battery state in the simulation data are used as inputs, and the battery temperature is used as the output to establish training and testing sets, which are divided into training, testing, and validation sets according to a specific ratio. In the PSO-CNN-LSTM fusion algorithm, PSO is used to optimize the hyperparameters of CNN and LSTM, CNN is used to perform pooling and dimensionality reduction on the original features to extract key parameters, and LSTM is used to predict the battery temperature based on the dimensionality-reduced features. Then, a temperature prediction model is established through a data-driven method.

[0164] The present invention relates to a method and system for calculating the maximum available energy of a lithium-ion battery by integrating operating conditions, energy consumption, and temperature. It considers the impact of operating conditions, heating energy consumption, and temperature on available energy, establishes a future vehicle speed planning model based on a traffic signal timing model, establishes an energy consumption prediction model based on vehicle dynamics, and establishes a temperature prediction model based on a reduced-order multiphysics coupling model. A reduced-order model of the electrochemical-thermal-mechanical multiphysics coupling model is established through a data-driven method, achieving accuracy comparable to the original model while significantly improving computational efficiency and reducing computational complexity. Finally, a maximum available energy prediction model integrating operating conditions, energy consumption, and temperature is established, and the actual maximum available energy of the battery is calculated iteratively. This method comprehensively considers the factors influencing the true available energy of lithium-ion batteries, improving the accuracy of lithium-ion battery available energy estimation and guiding the calculation of target temperatures for battery thermal management systems.

[0165] It should be noted that the specific embodiments described above enable those skilled in the art to more fully understand the present invention, but do not limit the present invention in any way. Therefore, although the present invention has been described in detail with reference to the accompanying drawings and embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the present invention. In short, all technical solutions and improvements that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the present invention patent.

Claims

1. A method for calculating the maximum available energy of a lithium-ion battery by integrating operating conditions, energy consumption, and temperature, characterized in that, Includes the following steps; S1. Establish a battery usage condition prediction model based on traffic signal timing information, and comprehensively optimize the battery usage condition prediction model by combining vehicle fuel economy, road traffic flow and vehicle acceleration and deceleration frequency. Use the model predictive control algorithm to plan future vehicle speed for the optimized battery usage condition prediction model. S2. Simplify the modeling of key components according to the configuration of the vehicle power system to establish a vehicle energy consumption prediction model. The vehicle energy consumption prediction model is used to predict the energy consumption under future operating conditions under a given input condition. S3. Establish a temperature prediction model based on a reduced-order multiphysics coupling model, including constructing an electrochemical-thermal-mechanical multiphysics coupling model of a lithium-ion battery based on the lithium-ion battery mechanism. The input of the lithium-ion battery electrochemical-thermal-mechanical multiphysics coupling model is a given power / current-time curve, ambient temperature, and initial battery state. The output is the battery's virtual cycle voltage output and temperature under the given operating conditions. By giving different input operating conditions, the electrochemical-thermal-mechanical multiphysics coupling model is simulated multiple times. Then, the PSO-CNN-LSTM fusion algorithm is used to refine the electrochemical-thermal-mechanical multiphysics coupling model based on the simulation results. To simplify the process, the power / current-time curve, ambient temperature, and initial battery state from the simulation data are used as inputs, and battery temperature is used as the output. Training and test sets are established and divided into training, test, and validation sets according to a specific ratio. In the PSO-CNN-LSTM fusion algorithm, PSO is used to optimize the hyperparameters of CNN and LSTM. CNN is used to perform pooling and dimensionality reduction on the original features to extract key parameters. LSTM is used to predict battery temperature based on the dimensionality-reduced features. Then, a temperature prediction model is established through a data-driven method to predict the temperature field distribution and future temperature rise of lithium-ion batteries under given operating conditions and heating conditions. In or after step S2 or S3, the heating loss of the lithium-ion battery is also calculated based on the power of the heating film under given operating conditions and heating conditions. S4. Establish a maximum available energy prediction model, including predicting the battery terminal voltage, and planning the input current based on the future vehicle speed calculated according to the vehicle operating conditions predicted in step S1. Then, based on the input current, battery terminal voltage, the energy consumption of the vehicle predicted in step S2, the temperature field and future temperature rise predicted in the lithium-ion battery discharge process in step S3, and the predicted heating loss of the lithium-ion battery, establish a maximum available energy prediction model that integrates operating conditions, energy consumption, and temperature. The maximum available energy of the lithium-ion battery is calculated iteratively through the future operating conditions obtained by the maximum available energy prediction model.

2. The method for calculating the maximum available energy of a lithium-ion battery according to claim 1, characterized in that, Step S1 utilizes V2V and V2I communication to acquire vehicle position, speed, and traffic light status information in the road environment. A traffic signal timing model is established to calculate the speed range of a single vehicle or vehicle queue when passing through a traffic light in a green light state. This leads to the establishment of a battery usage condition prediction model based on traffic signal timing information. The model is then comprehensively optimized by considering vehicle fuel economy, road traffic flow, and the number of accelerations and decelerations while ensuring vehicle performance. The optimized battery usage condition prediction model is then used to plan future vehicle speeds through a model predictive control algorithm. Fuel economy includes vehicle energy consumption, electricity consumption, and carbon emissions; traffic flow includes traffic congestion time, waiting time, or low-speed operation time; acceleration and deceleration times include average and minimum distances between vehicles; and vehicle performance includes overall vehicle power requirements.

3. The method for calculating the maximum available energy of a lithium-ion battery according to claim 1, characterized in that, Step S2 involves simplifying and modeling key components of the vehicle's powertrain configuration to establish a vehicle energy consumption prediction model. This model includes an engine model, a motor model, a lithium-ion battery model, a battery thermal management system model, a transmission model, and a vehicle dynamics model. Through the coupling relationships between these models, the vehicle energy consumption prediction model can predict energy consumption under future operating conditions given input conditions.

4. The method for calculating the maximum available energy of a lithium-ion battery according to claim 1, characterized in that, In step S4, the battery terminal voltage is predicted based on a finite element model, a data-driven model, or a circuit model. The finite element model is a finite element model coupled with a P2D model, a thermal model, and an aging model based on electrochemical mechanisms. The driving model is a data-driven model with temperature, voltage, and current as inputs. The circuit model is a circuit model coupled with an equivalent circuit model, a heat generation model, and a heat transfer model.

5. A system for calculating the maximum available energy of a lithium-ion battery by integrating operating conditions, energy consumption, and temperature, for executing the method for calculating the maximum available energy of a lithium-ion battery by integrating operating conditions, energy consumption, and temperature as described in any one of claims 1 to 4, characterized in that, It includes a working condition prediction model establishment module, a vehicle energy consumption prediction model establishment module, a temperature prediction model establishment module, and a maximum available energy prediction model establishment and calculation module. The working condition prediction model establishment module, the vehicle energy consumption prediction model establishment module, and the temperature prediction model establishment module are all connected to the maximum available energy prediction model establishment and calculation module. The working condition prediction model building module establishes a battery usage working condition prediction model based on traffic signal timing information, and comprehensively optimizes the battery usage working condition prediction model by combining the vehicle's fuel economy, road traffic flow, and the number of vehicle acceleration and deceleration. The optimized battery usage working condition prediction model is then used to plan future vehicle speed through a model prediction control algorithm. The vehicle energy consumption prediction model building module simplifies the modeling of key components according to the configuration of the vehicle power system to build a vehicle energy consumption prediction model. The vehicle energy consumption prediction model is used to predict the energy consumption under future operating conditions under a given input condition. The temperature prediction model building module establishes a temperature prediction model based on a reduced-order multiphysics coupling model. This includes constructing an electrochemical-thermal-mechanical multiphysics coupling model of a lithium-ion battery based on the lithium-ion battery mechanism. By giving different input operating conditions, the electrochemical-thermal-mechanical multiphysics coupling model is simulated multiple times. Then, the PSO-CNN-LSTM fusion algorithm is used to reduce the order of the electrochemical-thermal-mechanical multiphysics coupling model based on the simulation results. Finally, a temperature prediction model is established through a data-driven method to predict the temperature field distribution and future temperature rise of the lithium-ion battery under given operating conditions and heating conditions. The maximum available energy prediction model establishment and calculation module establishes a maximum available energy prediction model, including predicting the battery terminal voltage, and calculating the future vehicle speed based on the battery terminal voltage and the vehicle operating conditions predicted by the operating condition prediction model establishment module to obtain the input current. Then, based on the input current, battery terminal voltage, the vehicle energy consumption prediction model establishment module predicting the energy consumption of the whole vehicle, and the temperature prediction model establishment module predicting the temperature field during the lithium-ion battery discharge process and future temperature rise, a maximum available energy prediction model integrating operating conditions, energy consumption, and temperature is established. The maximum available energy of the lithium-ion battery is calculated iteratively through the future operating conditions obtained by the maximum available energy prediction model.

6. The lithium-ion battery maximum available energy calculation system according to claim 5, characterized in that, It also includes a heating energy consumption calculation module, which is connected to the maximum available energy prediction model establishment calculation module. The heating energy consumption calculation module is used to calculate the heating loss of the lithium-ion battery based on the power of the heating film under given operating conditions and heating conditions. The maximum available energy prediction model establishment calculation module establishes a maximum available energy prediction model that integrates operating conditions, energy consumption, and temperature based on the input current, battery terminal voltage, the energy consumption of the whole vehicle predicted by the whole vehicle energy consumption prediction model establishment module, the temperature field and future temperature rise during the discharge process of the lithium-ion battery predicted by the temperature prediction model establishment module, and the heating loss of the lithium-ion battery predicted by the heating energy consumption calculation module.

7. The lithium-ion battery maximum available energy calculation system according to claim 5 or 6, characterized in that, The vehicle energy consumption prediction model building module establishes a vehicle energy consumption prediction model by simplifying and modeling the key components of the vehicle power system configuration. The vehicle energy consumption prediction model includes an engine model, a motor model, a lithium-ion battery model, a battery thermal management system model, a transmission model, and a vehicle dynamics model. Through the mutual coupling relationship between the models, the vehicle energy consumption prediction model can predict the energy consumption under future operating conditions under a given input operating condition.

8. The lithium-ion battery maximum available energy calculation system according to claim 5 or 6, characterized in that, The input to the electrochemical-thermal-mechanical multiphysics coupling model of the ion battery is a given power / current-time curve, ambient temperature, and initial state of the battery, and the output is the virtual cycle voltage output and temperature of the battery under the operating condition. A PSO-CNN-LSTM fusion algorithm is adopted to simplify the electrochemical-thermal-mechanical multiphysics coupling model based on the simulation results. The power / current-time curve, ambient temperature, and battery initial state from the simulation data are used as inputs, and battery temperature is used as the output. Training and test sets are established and divided into training, test, and validation sets according to a specific ratio. In the PSO-CNN-LSTM fusion algorithm, PSO is used to optimize the hyperparameters of CNN and LSTM. CNN is used to perform pooling and dimensionality reduction on the original features to extract key parameters. LSTM is used to predict battery temperature based on the dimensionality-reduced features. Finally, a temperature prediction model is established through a data-driven method.

Images