A method for estimating the state of battery health in electric vehicles

Abstract

This invention discloses a method for estimating the state of health (SOH) of batteries in electric vehicles. The method includes: estimating the real-time health of a lithium battery based on a battery cycle mean stress-SOH cubic polynomial model, using real-time acquired battery mechanical stress signals. The battery cycle mean stress-SOH cubic polynomial model is obtained through the following steps: recording the lithium battery dimensions and pre-processed battery capacity; setting test conditions simulating real-world operating conditions; acquiring battery data in real-time during charge-discharge cycles under the test conditions; processing the real-time acquired data to calculate the battery capacity and health under different cycles, and converting the stress data into corresponding mechanical stress values; and fitting the relationship between battery stress and battery health to obtain the battery cycle mean stress-SOH cubic polynomial model. This invention provides data support for the optimized design of electric vehicle battery module structures and is applicable to various online health monitoring scenarios for electric vehicle lithium batteries.

Description

Technical Field

[0001] This invention relates to lithium battery health status monitoring technology and methods, specifically to a battery health status estimation method for electric vehicles. Background Technology

[0002] Lithium-ion batteries are a key component of electric vehicles, involved in energy storage. Their health status directly affects the vehicle's driving range, power, and safety and reliability. State of Health (SOH) is defined as the ratio of remaining capacity to new battery capacity, reflecting the extent of capacity decay. Accurately calculating the current SOH of the battery is crucial for optimal battery management and performance assurance.

[0003] Currently, common SOH estimation methods mainly fall into two categories: model-driven methods and data-driven methods. Model-driven methods typically use electrochemical models or equivalent circuit models, which not only require accurate acquisition of battery-related parameters, but also involve complex models and computational difficulties, making them unsuitable for online applications. Data-driven methods rely on various intelligent algorithms, requiring large amounts of training data and being highly dependent on battery aging trajectories, thus lacking adaptability to dynamic discharge and temperature fluctuations under real-world operating conditions.

[0004] In addition, traditional SOH estimation is generally based on electrical parameters such as voltage, current, and capacity. However, in real driving conditions, current fluctuations and temperature changes may distort these electrical parameters. Existing research has shown that mechanical stress is directly related to the lithium insertion and extraction process inside the battery. However, existing mechanical stress testing methods fail to consider the actual structure, dynamic power, and temperature control of electric vehicle battery modules, resulting in significant deviations between measurement results and actual operating conditions, making them difficult to directly use for battery SOH estimation.

[0005] Therefore, there is an urgent need for a SOH estimation method that can simulate real working conditions, is simple to measure, and has reliable accuracy, in order to solve the problems of adaptability and accuracy of existing technologies in real application scenarios. Summary of the Invention

[0006] To address at least one of the problems existing in the prior art, this invention provides a method for estimating the state of health (SOH) of lithium-ion batteries under real-world operating conditions based on mechanical stress measurement. This method can simulate the real-world operating conditions of electric vehicles, establish a correlation model between mechanical stress and battery health, and achieve high-precision, easily integrated SOH estimation.

[0007] To achieve the objective of this invention, the present invention provides a method for estimating the real-world health status of lithium-ion batteries based on mechanical stress measurement, comprising the following steps: (1) Construct a lithium-ion battery aging test device that simulates real working conditions: The device includes a constant temperature charge and discharge integrated machine, a multi-channel battery tester, an expansion force fixture, and a pressure sensor.

[0008] (2) Commercial LFP prismatic lithium batteries were selected as test samples. The batteries were pre-treated and their size and capacity after pre-treatment were recorded.

[0009] (3) Set test conditions simulating real working conditions: Control the ambient temperature inside the constant temperature charge-discharge machine to be constant at 45℃. The dynamic discharge curves A and B used during the cycle are designed with reference to the BEV cycle test content in IEC 62660-1:2018 standard. The maximum test power is set as follows. The calculation formula is as follows:

[0010] in, N It is the ratio of the vehicle's maximum power requirement to the battery energy; It refers to the electrical energy of a battery at room temperature.

[0011] The specific process of a single loop is as follows: Step 1: Use 1C constant current charging.

[0012] Step 2: Determine if the current is less than 0.05C. If so, proceed to the next step.

[0013] Step 3: Let it stand for 1 hour.

[0014] Step 4: Perform 16 discharge cycles under normal driving conditions (dynamic discharge curve A).

[0015] Step 5: Perform one simulated uphill discharge (dynamic discharge curve B).

[0016] Step 6: Perform 8 discharge cycles under normal driving conditions (dynamic discharge curve A).

[0017] Step 7: Use 1C constant current discharge.

[0018] Step 8: Check if the battery voltage has dropped to 2.5V. If so, proceed to the next step. Step 9: Let it stand for 1 hour.

[0019] (4) During the cycle, voltage and current data are collected by a multi-channel battery tester and surface pressure signals are collected by a pressure sensor. The data sampling frequency is synchronized with the electrical parameters and is recorded and stored in real time.

[0020] (5) Process the collected data: Define the instantaneous current measured during the test. I Units are A, test timet The unit is s. and Indicates the initial time and the end time, and the loop capacity. It is calculated by integrating the current over time, and the formula is as follows:

[0021] The battery state of health (SOH) is calculated by the ratio of the actual capacity to the reference capacity, using the following formula:

[0022] in, This represents the reference capacity, specifically the discharge capacity corresponding to the third cycle during the preprocessing process.

[0023] The stress on the battery is obtained by converting the pressure sensor readings, and the conversion formula is as follows:

[0024] The definitions and units of each parameter in the formula are as follows: This represents the stress exerted on the battery, measured in MPa. The reading is from the pressure sensor, in kg. The battery surface area is in mm. 2 .

[0025] (6) Model establishment: The relationship between battery stress and SOH is fitted to obtain a cubic polynomial model of battery cycle average stress-SOH.

[0026] (7) Based on the correlation model (i.e., the battery cycle average stress-SOH cubic polynomial model), the SOH is directly calculated by substituting the real-time battery surface pressure signal into the model, thereby realizing the SOH estimation under real working conditions.

[0027] The present invention also provides a battery health status estimation device for electric vehicles.

[0028] The present invention also provides a computer device.

[0029] The present invention also provides a computer-readable storage medium.

[0030] Compared with the prior art, the present invention has the following advantages: 1) Simulate real working conditions: Through constant temperature design and dynamic discharge simulation, the actual working environment of electric vehicle batteries is restored, and the estimation results are more in line with actual application scenarios; 2) Simple and reliable measurement: Mechanical stress measurement does not require complex modification and can be achieved by integrating small sensors. It has strong anti-electromagnetic interference capability and high data stability. 3) High estimation accuracy: the coefficient of determination for the correlation model is ≥0.99; 4) High engineering practicality: It can be directly integrated into the existing battery management system, with low computational burden, no need for a large amount of training data or complex parameter identification, and is suitable for commercial production needs. 5) Considering structural design: It reveals the stress increase caused by aging, provides data support for the structural design of battery modules, and improves the safety and durability of modules. Attached Figure Description

[0031] Figure 1 is a schematic diagram of the test device in an embodiment of the present invention.

[0032] Figure 2(a) is a schematic diagram of dynamic discharge curve A.

[0033] Figure 2(b) is a schematic diagram of dynamic discharge curve B.

[0034] Figure 3 is a flowchart of a single loop.

[0035] Figure 4 is a flowchart of the estimation method of the present invention.

[0036] Figure 5 shows the curve of the battery cycle average stress and the cubic polynomial fitting of SOH. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0038] This invention provides a lithium battery health state estimation method based on battery mechanical stress estimation, applicable to real-world operating conditions of electric vehicles, comprising the following steps: Step 1: Build a lithium-ion battery aging test device to simulate real working conditions.

[0039] In one embodiment, as shown in Figure 1, the device includes a constant-temperature charge-discharge integrated machine (such as the Guangdong Bell TX-007-27 dual-layer high and low temperature charge-discharge integrated machine), a multi-channel battery tester (such as the Shenghong BTS-5V300A16CH multi-channel battery tester), an expansion force clamp, and a pressure sensor mounted thereon (such as the Xinjingcheng XJC-P554-7T pressure sensor). The constant-temperature charge-discharge integrated machine supports control of ambient temperature, input voltage, and input current. The multi-channel battery tester is connected to an LFP square-shell lithium battery placed inside the constant-temperature charge-discharge integrated machine. The expansion force clamp, the pressure sensor mounted on the clamp, and the lithium battery are all housed inside the constant-temperature charge-discharge integrated machine. The multi-channel battery tester is connected to the lithium battery inside the integrated machine via cables to perform charge-discharge testing. It can measure parameters such as voltage, current, and charge / discharge amount.

[0040] A pressure sensor is mounted on an expansion force clamp. In one embodiment, the expansion force clamp comprises a multi-layer steel plate structure. Specifically, the first and fourth steel plates are fixed ends, connected by steel columns; the second and third steel plates can slide along the direction of the steel columns. An LFP prismatic lithium battery is placed between the third and fourth steel plates, with the large surface area of ​​the battery in direct contact with the lower surface of the third steel plate. The pressure sensor is bolted to the second steel plate, and its measuring end passes through the second steel plate and directly contacts the upper surface of the third steel plate, used to collect the vertical pressure generated when the battery expands. An internally threaded sleeve is fixedly installed in the center of the uppermost steel plate, and a pre-tightening screw is threadedly engaged with a nut seat, with the lower end of the pre-tightening screw connected to a pre-tightening pressure plate. By rotating the pre-tightening screw, the pre-tightening pressure plate and the second steel plate are pushed, thereby applying a pre-tightening force to the third steel plate and the battery through the pressure sensor. In this embodiment, the initial preload was set to 1 kgf, and the actual test result was 1.074 kgf, which is equivalent to 0.02433 MPa stress.

[0041] Step 2: Battery pretreatment: Select a commercial LFP prismatic lithium battery as the test sample and record the initial parameters: size 200.7×35.9×215.5 mm, nominal capacity 206Ah; pretreat the lithium battery and perform multiple charge-discharge cycles (e.g., 3 times), and take the discharge capacity of the last cycle as the initial capacity, which is 216.73Ah.

[0042] Step 3: Set test conditions to simulate real-world conditions: Maintain a constant ambient temperature and perform charge-discharge cycles on the battery using dynamic discharge under real driving conditions, including simulated discharge during normal driving and simulated discharge during hill climbing.

[0043] In one embodiment, the operating conditions are set as follows: the internal temperature of the constant-temperature charge-discharge integrated machine is set to 45°C to accelerate the battery aging process and simulate the actual temperature control conditions of an electric vehicle battery system; dynamic discharge curves A and B are used during the cycle, designed according to the BEV cycle test content in the IEC 62660-1:2018 standard. Dynamic discharge curve A simulates normal driving conditions (such as smooth driving on urban roads), while dynamic discharge curve B simulates high-load conditions such as climbing hills. Using two discharge curves can comprehensively simulate different power demand scenarios in actual electric vehicle operation, making the test conditions closer to real and complex driving conditions, thereby more accurately evaluating the aging characteristics of the battery under varying conditions. Figure 2(a) shows a schematic diagram of dynamic discharge curve A, and Figure 2(b) shows a schematic diagram of dynamic discharge curve B. The horizontal axis of the discharge curve represents time (unit: s), and the vertical axis represents the power at the corresponding moment and the maximum test power. The ratio. Maximum test power. The calculation formula is as follows:

[0044] in, N It is the ratio of the vehicle's maximum power requirement to the battery energy; It refers to the electrical energy of a lithium battery at room temperature.

[0045] In one embodiment, a 200kW commercial vehicle is equipped with 100 LF206 batteries. The maximum test power can be calculated. Multiply the power percentage-time curves shown in dynamic discharge curves A and B by... Obtain the actual power-time curve. Use the actual power to control the discharge during normal driving and hill climbing conditions in the cycle.

[0046] The process of a single charge-discharge cycle is as follows: Figure 3 As shown, the specific process is as follows: Step 1: Use 1C constant current charging.

[0047] Step 2: Determine if the current is less than 0.05C. If so, proceed to the next step.

[0048] Step 3: Let it stand for 1 hour.

[0049] Step 4: Perform 16 discharge cycles under normal driving conditions (using dynamic discharge curve A).

[0050] Step 5: Perform one simulated uphill discharge (using dynamic discharge curve B).

[0051] Step 6: Perform 8 discharges under normal driving conditions (using dynamic discharge curve A).

[0052] Step 7: Use 1C constant current discharge.

[0053] Step 8: Determine if the battery voltage has dropped to 2.5V. If yes, proceed to the next step.

[0054] Step 9: Let it stand for 1 hour.

[0055] Step 4: Data Acquisition: During the charge-discharge cycle, the voltage, current, pressure, and temperature data of the lithium battery are collected and recorded simultaneously, with a data sampling interval of 1 second.

[0056] In one embodiment, four temperature monitoring points are provided on the lithium battery: the positive electrode tab, the negative electrode tab, the side of the battery, and the ambient temperature.

[0057] Step 5: Process the collected voltage, current, pressure and temperature data, calculate the battery capacity and state of health (SOH) under different cycles, and convert the pressure data into the corresponding mechanical stress value using a formula.

[0058] This step includes: Define the instantaneous current measured during testing. (Unit: A) Test time (Unit: seconds) and The initial and final times represent the current and time, respectively. The cycle capacity is calculated by integrating the current over time. The calculation formula is:

[0059] The battery state of health (SOH) is calculated by the ratio of the actual capacity to the reference capacity, using the following formula:

[0060] in, The reference capacity is indicated. In one embodiment, the specific value of the reference capacity is the discharge capacity corresponding to the third cycle in the preprocessing process.

[0061] The stress on the battery is obtained by converting the pressure sensor readings, and the conversion formula is as follows:

[0062] In the formula: This represents the stress exerted on the battery, measured in MPa. The reading is from the pressure sensor, representing the pressure signal on the battery surface, in kg. The battery surface area is expressed in mm. 2 , The acceleration due to gravity is expressed in m / s², and is taken as 9.8 m / s². In one embodiment, the battery size is used for calculation. 43250.85mm 2 .

[0063] Step 6: Model Establishment: Fit the relationship between the mechanical stress value of the lithium battery and the state of health (SOH) to obtain a cubic polynomial model of the lithium battery cycle average stress-SOH.

[0064] The average mechanical stress data of each cycle battery is used to fit the SOH of the battery under the corresponding cycle to form a cubic polynomial model; this model does not require the introduction of additional parameters or other training samples.

[0065] In one embodiment, after removing outliers using MATLAB software, the fitted cubic polynomial model of the average stress and SOH for each cycle of the lithium battery is as follows.

[0066] The coefficient of determination is calculated as follows: R²= The value of 0.9991 indicates that the cubic polynomial model fits the data very well.

[0067] Step 7: Based on the lithium battery cycle mean stress-SOH cubic polynomial model, the battery surface pressure signal is collected in real time. The real-time state of health (SOH) of the lithium battery can be estimated.

[0068] The embodiments of the present invention can achieve rapid estimation of battery health status through a lithium battery cycle average stress-SOH cubic polynomial model.

[0069] In one embodiment, such as Figure 5 As shown, by comparing the original data with the fitted cubic polynomial model, the error between the estimated health state SOH and the actual measured value is ≤1.58%, indicating that the method described in this embodiment of the invention has high accuracy.

[0070] The method described in this invention is adapted to the actual operating scenarios of electric vehicles. Leveraging the convenience and stability of mechanical stress measurement, it significantly improves the accuracy and reliability of health status estimation by constructing a high-precision correlation model. This method requires no complex model construction or large amounts of training data and can be directly integrated into existing battery management systems, balancing engineering practicality and adaptability. It also provides data support for the optimized design of electric vehicle battery module structures and is suitable for online health status monitoring scenarios of various electric vehicle lithium batteries.

[0071] In one embodiment, a lithium battery health status estimation device for electric vehicles is provided to implement the method described in the foregoing embodiments. The device includes the following modules: The testing module is used to estimate the real-time health status of lithium batteries based on the battery cycle mean stress-SOH cubic polynomial model, using real-time acquired battery mechanical stress signals. The battery cycle mean stress-SOH cubic polynomial model is obtained through the following modules: The recording module is used to pre-process the lithium battery and record the size of the lithium battery and the capacity of the battery after pre-processing. The test condition setting module is used to set test conditions that simulate real working conditions; The data acquisition module is used to collect the voltage, current, and pressure data of the battery in real time during charge-discharge cycles under the test conditions. The data processing module is used to process real-time acquired data, calculate the battery's capacity and health status under different cycles, and convert pressure data into corresponding mechanical stress values. The fitting module is used to fit the relationship between battery stress and battery health status to obtain a cubic polynomial model of battery cycle average stress-SOH.

[0072] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the methods described in the foregoing embodiments.

[0073] In one embodiment, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program that, when executed by a processor, implements the methods described in the foregoing embodiments.

[0074] The device, equipment, and medium described herein have the same technical effects as those achieved by the methods described in the foregoing embodiments.

[0075] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined in this invention may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for estimating the state of health of batteries in electric vehicles, characterized in that, Based on the battery cycle mean stress-SOH cubic polynomial model, the real-time health status of the battery is estimated by real-time acquisition of battery mechanical stress signals. The battery cycle mean stress-SOH cubic polynomial model is obtained through the following steps: Build a battery aging test device that simulates real working conditions; The battery was pretreated, and its size and capacity after pretreatment were recorded. Set test conditions to simulate real working conditions; Under the test conditions, charge-discharge cycles were performed, and the battery's voltage, current, and pressure data were collected in real time. Process real-time acquired data, calculate battery capacity and health status under different cycles, and convert pressure data into corresponding mechanical stress values; By fitting the relationship between battery stress and battery health status, a cubic polynomial model of battery cycle average stress-SOH is obtained.

2. The battery health state estimation method for electric vehicles according to claim 1, characterized in that, The test conditions include: maintaining a constant ambient temperature and performing charge-discharge cycles on the battery using dynamic discharge under real driving conditions.

3. The battery health state estimation method for electric vehicles according to claim 2, characterized in that, The dynamic discharge includes simulated discharge during normal driving and simulated discharge during hill climbing.

4. The battery health state estimation method for electric vehicles according to claim 1, characterized in that, The stress on the battery is obtained by converting the readings from the pressure sensor. The conversion formula is as follows: In the formula, The stress on the battery, For the pressure sensor readings, It is the acceleration due to gravity. It represents the large surface area of ​​the battery.

5. The battery health state estimation method for electric vehicles according to claim 1, characterized in that, The State of Health (SOH) of a battery is calculated as the ratio of its actual capacity to its reference capacity, expressed as: In the formula, Indicates reference capacity. Indicates the loop capacity.

6. The battery health state estimation method for electric vehicles according to claim 5, characterized in that, The cycle capacity The value is obtained by integrating the current over time, and the calculation formula is as follows: This represents the instantaneous measured current. For testing time, and Indicates the initial time and the end time.

7. A method for estimating the state of battery health for electric vehicles according to any one of claims 1-6, characterized in that, The average mechanical stress data of the battery in each cycle is used to fit the SOH of the battery in the corresponding cycle, forming a cubic polynomial model of battery cycle average stress-SOH.

8. A battery health status estimation device for electric vehicles, characterized in that, For implementing the method according to any one of claims 1-7, the apparatus comprises the following modules: The testing module is used to estimate the real-time health status of the battery based on the battery cycle mean stress-SOH cubic polynomial model, using real-time acquired battery mechanical stress signals. The battery cycle mean stress-SOH cubic polynomial model is obtained through the following modules: The recording module is used to preprocess the battery and record its size and capacity after preprocessing. The test condition setting module is used to set test conditions that simulate real working conditions; The data acquisition module is used to collect the voltage, current, and pressure data of the battery in real time during charge-discharge cycles under the test conditions. The data processing module is used to process real-time acquired data, calculate the battery's capacity and health status under different cycles, and convert pressure data into corresponding mechanical stress values. The fitting module is used to fit the relationship between battery stress and battery health status to obtain a cubic polynomial model of battery cycle average stress-SOH.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method according to any one of claims 1-7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1-7.

Images