Method for evaluating online health state of lithium ion battery of electric vehicle through data fusion model

Abstract

The invention provides a method for evaluating the online health state of an electric vehicle lithium ion battery through a data fusion model, and relates to the technical field of lithium ion battery health state evaluation. A plurality of groups of voltage, discharge current and internal resistance data are collected under a specific working condition, respective average values are obtained through calculation, and then a voltage health factor, a current health factor and an internal resistance health factor are constructed by combining rated values with the average values and maximum and minimum values. On the basis of the health factors and variance data, on one hand, a comprehensive health index is obtained through a weighted average algorithm, a preset threshold value comparison result is assigned, on the other hand, a comprehensive scoring formula is constructed according to variance and threshold value comparison result assignment, and the lithium ion battery health state is comprehensively judged according to different result assignments; and outputting normal, sub-health and fault states.

Description

Technical Field

[0001] The present invention relates to the technical field of lithium-ion battery health status assessment, and in particular to an online health status assessment method for lithium-ion batteries of electric vehicles through a data fusion model. Background Art

[0002] As the global automotive industry accelerates its transition to electrification, the electric vehicle market continues to expand rapidly. Lithium-ion batteries, with their superior characteristics such as high voltage, high specific energy, and long cycle life, have become the mainstream power source of choice for electric vehicles. They not only provide strong power but also largely determine the performance of electric vehicles, such as range and acceleration, significantly driving their widespread adoption and application worldwide. Lithium-ion battery safety is crucial in electric vehicles. Poor battery health can lead to serious safety issues such as overheating, short circuits, and even fire and explosion. Accurately assessing the health of lithium-ion batteries can promptly identify potential safety hazards and enable appropriate measures, such as adjusting charging strategies, limiting battery power output, or issuing timely alarms, thereby ensuring the safe operation of electric vehicles and protecting the lives and property of drivers and passengers.

[0003] At present, in the prior art with publication number CN115236512A, a method for determining the state of a lithium-ion battery by calculating its SOH (state of health) is usually performed by extracting features under ideal operating conditions for calculating the SOH through statistical methods, thereby calculating the SOH of the lithium-ion battery to obtain the state of the lithium-ion battery. The huge difference between the actual operating conditions and the ideal operating conditions makes it difficult for the evaluation model established based on the ideal operating condition data to accurately reflect the health status of the lithium-ion battery in actual usage scenarios, and the prediction accuracy is low.

[0004] This method constructs a model by collecting data under speed conditions of 20-30 km / h. This speed range is more common in daily electric vehicle driving and is closer to actual usage. Compared with high-speed driving (such as exceeding 100 km / h), the instantaneous high-power output demand faced by lithium-ion batteries is smaller, thereby reducing the impact of interference factors such as changes in the internal material structure of the lithium-ion battery caused by high current shocks on data accuracy. This relatively stable operating condition ensures that the collected lithium-ion battery voltage, current, and internal resistance data can more accurately reflect the performance of the lithium-ion battery under normal operating conditions, better conform to the parameter characteristics in actual use, and help improve the practicality and reliability of the evaluation results.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0006] The purpose of the present invention is to provide an online health status assessment method for lithium-ion batteries of electric vehicles through a data fusion model to solve the problems raised in the above background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A method for online health status assessment of lithium-ion batteries of electric vehicles using a data fusion model includes the following steps:

[0009] Step 1: Preset a speed range as a monitoring range. When the vehicle is in a driving state and enters the monitoring range for the first time, continuously acquire characteristic data of the electric vehicle at the same time interval until the vehicle leaves the monitoring range for the first time. Construct a characteristic data table based on the data acquired within this range. The characteristic data includes the voltage, discharge current, and internal resistance of the lithium-ion battery.

[0010] Step 2: Extract the characteristic data table to obtain the rated voltage, rated discharge current, and rated internal resistance data of the lithium-ion battery, record the maximum and minimum voltage values, the maximum value data of the discharge current and the internal resistance, and calculate the average value of the voltage, discharge current, and internal resistance data respectively;

[0011] Step 3: Build the voltage health factor using the rated voltage, average voltage, maximum voltage, and minimum voltage. Build the current health factor using the rated discharge current, average discharge current, and minimum discharge current. Build the internal resistance health factor using the rated internal resistance, maximum internal resistance, and average internal resistance.

[0012] Step 4: Based on the voltage health factor, current health factor, and internal resistance health factor, a weighted average algorithm is used to obtain a comprehensive health index, and the variance of the discharge current, voltage, and internal resistance are calculated respectively;

[0013] Step 5: Set the health index threshold, set the variance threshold of the discharge current, voltage, and internal resistance respectively, compare the comprehensive health index with the health index threshold, and compare the variance of the discharge current, voltage, and internal resistance with the corresponding variance threshold respectively, and comprehensively judge the health status of the lithium-ion battery based on the comparison results.

[0014] Furthermore, for an operating period within a preset speed range, recording characteristic data of the lithium-ion battery during the period includes the following steps:

[0015] Set the speed range to be monitored to: u fw ∈[20, 30],u fwrepresents the real-time speed of the electric vehicle to be evaluated. The time interval is: Δt = 2s. After the electric vehicle starts and begins to travel, when the vehicle speed reaches 20km / h for the first time, the voltage, discharge current and internal resistance data are collected at 2s intervals. Data collection stops when the vehicle speed drops below 20km / h or exceeds 30km / h.

[0016] Furthermore, the method for calculating the average values of the voltage, discharge current and internal resistance data is as follows:

[0017] Extract the voltage, discharge current, and internal resistance data at each time interval from the characteristic data table and calculate the average voltage:

[0018]

[0019] Where V avg It is expressed as the average voltage, n is the number of time intervals, that is, the number of times the lithium-ion battery characteristic data is recorded, V i It is represented by the voltage value collected at the i-th time interval, i = 1, 2, 3, ..., n, where n is a positive integer ≥ 10. That is, if the data collection duration is less than 10s (n<10), the collected data will not be used in the health status assessment calculation. The assessment calculation can only be performed after the vehicle enters the monitoring speed range again and the collection duration reaches 10s or more.

[0020] Calculate the average discharge current:

[0021]

[0022] Where, I Avg Expressed as the average discharge current, I i It is represented by the discharge current value collected at the i-th time interval;

[0023] Calculate the average internal resistance:

[0024]

[0025] Where R avg Expressed as the average internal resistance, R i It is represented by the internal resistance value collected at the i-th time interval.

[0026] Furthermore, the method for constructing the voltage health factor, current health factor and internal resistance health factor is:

[0027] The voltage health factor is constructed by the rated voltage, average voltage, maximum voltage and minimum voltage:

[0028]

[0029] Where V ratedExpressed as the rated voltage of the lithium-ion battery, K v1 , K v2 is the weight coefficient, and K v1 >K v2 >0;

[0030] The current health factor is constructed by the rated discharge current, the average discharge current and the minimum discharge current:

[0031]

[0032] Where, I rated Expressed as the rated discharge current of the lithium-ion battery, K I1 , K I2 is the weight coefficient, and K I1 >K I2 >0,I min Expressed as the minimum discharge current;

[0033] The internal resistance health factor is constructed by the rated internal resistance, maximum internal resistance and average internal resistance:

[0034]

[0035] Where R rated Expressed as the rated internal resistance of the lithium-ion battery, K R1 , K R2 is the weight coefficient, and K R1 >K R2 >0, R max Expressed as the maximum internal resistance.

[0036] Furthermore, the weighted average algorithm is used to obtain the comprehensive health index:

[0037] Assign corresponding weights to the voltage health factor, current health factor, and internal resistance health factor: ω V 、ω I and ω R , and ω V >ω I =ω R ,ω V +ω I +ω R =1, and use the weighted average algorithm to construct the comprehensive health index formula:

[0038] H=ω V ·F v +ω I ·F I +ω R ·F R .

[0039] Furthermore, the methods for calculating the variance of discharge current, voltage, and internal resistance are as follows:

[0040] Construct the voltage variance formula:

[0041]

[0042] Where, Expressed as voltage variance;

[0043] Construct the discharge current variance formula:

[0044]

[0045] Where, Expressed as discharge current variance;

[0046] Construct the internal resistance variance formula:

[0047]

[0048] Where, Expressed as internal resistance variance.

[0049] Furthermore, the method for determining the health status of a lithium-ion battery is:

[0050] The threshold of the health index of the good state is preset as H1, the threshold of the health index of the critical health state is preset as H2, and H1>H2>0, and the voltage variance threshold is T V , the discharge current variance threshold is T I , the internal resistance variance threshold is T R ;

[0051] When H≥H1, the value is assigned to 3; when H1>H>H2, the value is assigned to 2; when H≤H2, the value is assigned to 1;

[0052] For the variance of each feature data, if the variance is less than the threshold, it is assigned a value of 1; if the variance is greater than or equal to the threshold, it is assigned a value of 0, that is:

[0053] when Assign a value of 0, when Assign a value of 1;

[0054] when Assign a value of 0, when Assign a value of 1;

[0055] when Assign a value of 0, when Assign a value of 1;

[0056] Construct a comprehensive scoring formula:

[0057] S total =Hscore +SV score +SI score +SR score

[0058] Where S total Expressed as a comprehensive score, H score Expressed as a health index score, SV score Expressed as the score of voltage variance, SI score Expressed as the score of the discharge current variance, SR score It is expressed as the score of internal resistance variance;

[0059] When the comprehensive score is 6, the health status of the lithium-ion battery is judged to be normal; when the comprehensive score is 5, the health status of the lithium-ion battery is judged to be sub-healthy; when the comprehensive score is less than or equal to 4, the health status of the lithium-ion battery is judged to be in a fault state.

[0060] Compared with the prior art, the present invention has the following beneficial effects:

[0061] The present invention pre-sets speed ranges as monitoring intervals to collect lithium-ion battery characteristic data. This allows the collection of voltage, discharge current, and internal resistance while the electric vehicle's operating state is relatively stable, avoiding interference with data collection caused by frequent vehicle starts and stops, sudden acceleration and braking, and extreme operating conditions such as high-speed driving. The rated parameters of the lithium-ion battery are obtained, and the voltage health factor, current health factor, and internal resistance health factor are constructed based on the collected characteristic data. This quantitative assessment of lithium-ion battery performance is conducted from multiple perspectives, avoiding the one-sidedness caused by relying solely on a single parameter assessment. Compared with existing technologies that rely on only a single or a few lithium-ion battery parameters for evaluation, this multi-dimensional factor construction method can more comprehensively characterize the health status of lithium-ion batteries.

[0062] A weighted average algorithm is then used to calculate a comprehensive health index (CHI), and corresponding health index thresholds are set as a key basis for determining the health status of lithium-ion batteries. Simultaneously, the variance values of voltage, discharge current, and internal resistance are calculated. Comparisons between these variances and the set variance thresholds are used, and these, along with the CHI threshold determination, form a multi-dimensional assessment system. By evaluating both the CHI and data variance, a more comprehensive, accurate, and integrated assessment of the health status of lithium-ion batteries is achieved, ultimately outputting the battery's health status. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 Schematic diagram of the overall method flow of the present invention;

[0064] Figure 2 This is a graph showing changes in the health index of the present invention;

[0065] Figure 3 This is a voltage variance monitoring diagram of the present invention;

[0066] Figure 4 This is a discharge current variance monitoring diagram of the present invention;

[0067] Figure 5 This is the internal resistance variance monitoring diagram of the present invention;

[0068] Figure 6 This is the comprehensive scoring monitoring chart of the present invention. DETAILED DESCRIPTION

[0069] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to specific embodiments.

[0070] It should be noted that, unless otherwise defined, the technical or scientific terms used in the present invention should have the usual meanings understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0071] Example:

[0072] See also Figures 1 to 6 , the present invention provides a technical solution:

[0073] A method for online health status assessment of lithium-ion batteries of electric vehicles using a data fusion model includes the following steps:

[0074] Step 1: Preset a speed range as a monitoring range. When the vehicle is in a driving state, continuously obtain the real-time speed information of the electric vehicle at the same time interval. For the operating period within the pre-set speed range, record the characteristic data of the lithium-ion battery during the period and construct a characteristic data table. The characteristic data includes voltage, discharge current and internal resistance.

[0075] When electric vehicles are driving, speed is one of the key factors affecting the working state of lithium-ion batteries. When the speed is in a higher range, such as exceeding 30km / h, the wind resistance increases sharply. The lithium-ion battery needs to continuously provide greater power to overcome the resistance, which causes a significant increase in the discharge current of the lithium-ion battery. At the same time, the voltage will also fluctuate greatly due to factors such as internal resistance voltage division, making it difficult to accurately reflect the subtle changes in the electrochemical performance of the lithium-ion battery. Therefore, the speed range is set to u fw =20km / h-30km / h is used as the monitoring range. The overall energy consumption of the vehicle is relatively stable, and external interference such as wind resistance is at a low level. The working status of the lithium-ion battery mainly depends on its own electrochemical properties, battery cell consistency, internal thermal management and other internal factors, which more accurately captures the performance of the lithium-ion battery under relatively stable working conditions.

[0076] The interval for acquiring real-time speed information for electric vehicles is set to 2 seconds. From the perspective of the lithium-ion battery's own electrochemistry, this 2-second interval can also reflect certain stage changes in the lithium-ion battery's internal chemical reactions. After the electric vehicle starts and drives, the vehicle control system acquires real-time speed information every 2 seconds. When the vehicle speed is within the monitoring range, it records the lithium-ion battery's voltage, discharge current, and internal resistance data. For example, when the speed is 22 km / h, 24 km / h, 28 km / h, and 35 km / h, the lithium-ion battery's voltage, discharge current, and internal resistance data at 22 km / h, 24 km / h, and 28 km / h are recorded and constructed into a characteristic data table. However, when the speed reaches 35 km / h, this speed exceeds the set range, and the system will not select the corresponding characteristic data for recording. Table 1 shows the voltage, discharge current, and internal resistance values of a lithium-ion battery used in a certain electric vehicle over time. By performing trend analysis on this data, it is possible to identify performance changes of lithium-ion batteries during use. For example, as time goes by, the health status of lithium-ion batteries can be monitored by checking whether the voltage continues to drop, whether the discharge current is stable, whether the internal resistance increases, etc.

[0077]

[0078]

[0079] Table 1 Parameter monitoring table

[0080] Step 2: Extract the characteristic data table to obtain the rated voltage, rated discharge current, and rated internal resistance data of the lithium-ion battery, record the maximum and minimum voltage values, the maximum value data of the discharge current and the internal resistance, and calculate the average value of the voltage, discharge current, and internal resistance data respectively;

[0081] Get the rated voltage V of the lithium-ion batteryracted , Rated discharge current I rated And the rated internal resistance R rated The data serves as the basic parameter for the subsequent construction of health factors, thereby measuring the degree to which the lithium-ion battery deviates from the ideal state during operation.

[0082] Then, the voltage, discharge current, and internal resistance data for each time interval are extracted from the characteristic data table, and their average values are calculated. This is because lithium-ion batteries are affected by various accidental factors during actual operation, resulting in local fluctuations in the data. Calculating the average value can effectively smooth out these local fluctuations, allowing the obtained data to better reflect the basic performance of lithium-ion batteries.

[0083] The formula for calculating the average voltage is:

[0084]

[0085] The formula for calculating the average discharge current is:

[0086]

[0087] The formula for calculating the average internal resistance is:

[0088]

[0089] If the average value is calculated when the number of data collection times is too small, the local fluctuations caused by these accidental factors will cause a large error in the result, resulting in the calculated average value being unable to truly and effectively reflect the basic performance of the lithium-ion battery. Therefore, when the number of collection times reaches at least 10 times, the average values of voltage, discharge current and internal resistance are calculated, and the health factor is subsequently constructed, which is closer to the actual performance level of the lithium-ion battery under this operating condition. That is, if the data collection duration is less than 10s (n<10), the collected data will not be used in the health status assessment calculation. The assessment calculation can only be performed after the vehicle enters the monitoring speed range again and the collection time reaches 10s or more.

[0090] Step 3: Build the voltage health factor using the rated voltage, average voltage, maximum voltage, and minimum voltage. Build the current health factor using the rated discharge current, average discharge current, and minimum discharge current. Build the internal resistance health factor using the rated internal resistance, maximum internal resistance, and average internal resistance.

[0091] The voltage health factor is constructed by the rated voltage, average voltage, maximum voltage and minimum voltage:

[0092]

[0093] Where, F v Represents the voltage health factor, V ratedExpressed as the rated voltage of the lithium-ion battery, K v1 , K v2 It is a weight coefficient used to adjust the degree of voltage influence. The specific value should be based on data analysis and actual needs. The voltage health factor can comprehensively evaluate the health of lithium-ion battery voltage. Through multi-dimensional voltage characteristic analysis, it can quantify the performance status related to lithium-ion battery voltage, intuitively display the degree of deviation between the average voltage of lithium-ion battery and the rated voltage, and judge whether the overall voltage performance is normal. This item is used to evaluate the overall performance level of lithium-ion battery voltage. It can intuitively show the degree of deviation of the average voltage of the lithium-ion battery under normal working conditions from the rated voltage. If this ratio is close to 1, it means that the average voltage of the lithium-ion battery is basically in the normal range. The higher the value of the voltage health factor, the lower the voltage average value will be. On the contrary, when the performance of the lithium-ion battery deteriorates, the voltage health factor value will decrease. It indicates the voltage fluctuation range, reflecting the long-term deviation degree. If the voltage fluctuation is too large, that is, the larger the value, it means that the chemical reaction inside the lithium-ion battery is unstable or there are some problems with the lithium-ion battery structure. This item can include the voltage fluctuation range in the consideration range of the health factor. The larger the value, The smaller the value of this item is, the smaller the value of the voltage health factor is. This item takes into account the deviation between the minimum voltage and the voltage value at each time interval, reflecting short-term fluctuations. The larger the value, the more serious the deviation. The smaller this value is, the smaller the value of the voltage health factor is. Voltage is a direct reflection of the state of the lithium-ion battery, but short-term fluctuations are less harmful than long-term deviations. Therefore, the voltage that reflects short-term fluctuations is not as good as the voltage that reflects long-term fluctuations. This item can set K v1 =0.3, K v2 =0.2;

[0094] The current health factor is constructed by the rated discharge current, the average discharge current and the minimum discharge current:

[0095]

[0096] Where, F I Represents the current health factor, I rated Expressed as the rated discharge current of the lithium-ion battery, K I1 , K I2 is the weight coefficient, which is used to adjust the degree of influence of current. The specific value should be based on data analysis and actual needs. minExpressed as the minimum discharge current, the current health factor can comprehensively evaluate the health of the lithium-ion battery at the discharge current level. Through multi-dimensional current characteristic analysis, it quantifies the performance status related to the lithium-ion battery current, intuitively displays the deviation between the average discharge current of the lithium-ion battery and the rated discharge current, and judges whether the overall current performance is normal. This item is used to evaluate the overall performance level of the lithium-ion battery discharge current. It can intuitively show the degree of deviation of the average discharge current of the lithium-ion battery under normal working conditions from the rated discharge current. The closer the ratio is to 1, the closer the average discharge current is to the rated current, and the healthier the overall performance of the lithium-ion battery current is. This item represents the evaluation of discharge capacity. Due to the aging of lithium-ion batteries, changes in internal structure and other reasons, the output capacity of lithium-ion batteries under low-load conditions may be limited, and lithium-ion batteries may find it difficult to release sufficient current. When the discharge capacity decreases, the minimum discharge current decreases, causing this item value to increase. As this item decreases, the value of the current health factor will also decrease. In this item, It represents the sum of the absolute values of the deviations between the discharge current value collected at each time interval and the average value, which can reflect the current fluctuation. When the current fluctuation is large, the value of this item will increase. The value of this item will decrease accordingly, reflecting that the more severe the current fluctuation, the less ideal the health of the lithium-ion battery. The discharge current directly affects the power output capacity of the lithium-ion battery. When the current is insufficient, it means that the capacity of the lithium-ion battery is declining. The current stability affects the cycle life of the lithium-ion battery, but the harm is less than continuous low current discharge. Therefore, the current stability is reflected. This item has a smaller weight, so you can set K I1 =0.4, K I2 =0.3;

[0097] The internal resistance health factor is constructed by the rated internal resistance, maximum internal resistance and average internal resistance:

[0098]

[0099] Where, F R Represents the internal resistance health factor, R rated Expressed as the rated internal resistance of the lithium-ion battery, K R1 , K R2 is the weight coefficient, which is used to adjust the influence of internal resistance. The specific value should be based on data analysis and actual needs. max Expressed as the maximum internal resistance, the internal resistance health factor is used to quantify the health status of the internal resistance of the lithium-ion battery, reflecting the degree of influence of the internal resistance characteristics on the performance of the lithium-ion battery. The larger the value, the better the performance of the lithium-ion battery. This item is used to measure the overall change of the internal resistance of the lithium-ion battery. It can intuitively show the degree of change of the average internal resistance relative to the rated internal resistance. With the increase of usage time, the increase of charge and discharge times, or the physical and chemical changes inside the lithium-ion battery, the internal resistance usually increases gradually. If this ratio is close to 1, it means that the average internal resistance of the lithium-ion battery is basically in the normal range. On the contrary, if the ratio is significantly less than 1, it means that the internal resistance of the lithium-ion battery has increased significantly, and local high internal resistance areas may appear inside the lithium-ion battery. When the value of this item increases, it can reflect the high internal resistance of the lithium-ion battery, that is, R max Increase, This item can reflect the degree of dispersion of the internal resistance, that is, the fluctuation of the internal resistance. When the internal resistance fluctuates greatly, the value of this item will increase, indicating that the stability of the internal structure or chemical reaction of the lithium-ion battery is poor. The decrease in the value of this item leads to a decrease in the value of the internal resistance health factor. The long-term increase in internal resistance will lead to capacity decline, reflecting the aging of the lithium-ion battery. It is a key signal for fault warning. It is common in the imbalance of cells in series lithium-ion battery packs. When the fluctuation is large, although it affects the performance, it is not an immediate fault. Therefore The weight of this item is small, so we can set K R1 =0.5, K R2 =0.3.

[0100] Step 4: Based on the voltage health factor, current health factor, and internal resistance health factor, a weighted average algorithm is used to obtain a comprehensive health index, and the variance of the discharge current, voltage, and internal resistance are calculated respectively;

[0101] Different lithium-ion battery health factors have different importance in reflecting the health status of lithium-ion batteries. By reasonably allocating weights, the role of each health factor in the comprehensive assessment can be balanced according to actual needs, so that the comprehensive health index can more accurately reflect the true health status of lithium-ion batteries in specific application scenarios.

[0102] According to the characteristics of lithium-ion batteries, the voltage health factor F v , current health factor F I and internal resistance health factor F R Determine the weight ω respectively V 、ω I and ω R , and ω V >ω i =ω R ,ω V +ω I +ω R=1. After research and experimental verification, for lithium-ion batteries, the voltage health factor is relatively more important and directly affects the operation of the electrical system. Therefore, based on the fact that voltage has a more direct impact on the battery status, the voltage health factor is given a higher weight and the weight is assigned to ω. V =0.4, the current is directly related to the power output of the lithium-ion battery, so the current health factor weight ω I =0.3, internal resistance has an important impact on the energy efficiency of lithium-ion batteries, so the internal resistance health factor weight ω R =0.3;

[0103] According to the weighted average algorithm, the comprehensive health index formula is constructed:

[0104] H=ω V ·F v +ω I ·F I +ω R ·F R

[0105] Voltage, current, and internal resistance are the core parameters of battery status. The closer the voltage, current, and internal resistance are to the rated values, the higher the scores of each health factor. By combining various health factors, the relative importance of each health factor and their respective actual values are taken into account. The health index formula avoids the one-sided evaluation that may be caused by a single health factor by weighted integration of various health factors, and realizes the assessment of the overall health status of lithium-ion batteries from the perspectives of discharge current, voltage, and internal resistance. The larger the value, the better the health status of the lithium-ion battery. The weight of each health factor can be adjusted according to the application scenario. As the current health factor, voltage health factor, and internal resistance health factor increase, the comprehensive health index increases, showing a positive correlation.

[0106] Table 2 shows the voltage of a lithium-ion battery for an electric vehicle at rated voltage V rated =3.7, rated discharge current I rated =50, rated internal resistance R rated=0.02, 40 sets of health factor and comprehensive health index data for lithium-ion batteries were randomly sampled in chronological order and summarized into a table. The voltage health factor reflects the health of the lithium-ion battery's voltage performance. The higher the value of this factor, the closer the lithium-ion battery's voltage state is to its rated value, which generally reflects the lithium-ion battery's good charging and discharging capabilities. The current health factor indicates the health of the lithium-ion battery's current output during discharge. A higher value indicates that the lithium-ion battery can stably output current within the normal range, reflecting the lithium-ion battery's discharge capacity. The internal resistance health factor reflects the health of the lithium-ion battery's internal resistance. The lower the internal resistance, the less energy loss the lithium-ion battery has and the better the overall performance. Therefore, the higher the value of this factor, the better. The comprehensive health index is calculated by weighted average voltage, discharge current, and internal resistance health factor. It represents the overall health of the lithium-ion battery. A higher value indicates better health of the lithium-ion battery and is one of the important data used to judge the health status of lithium-ion batteries.

[0107]

[0108]

[0109]

[0110] Table 2 Comprehensive health index data table

[0111] In practical applications, when lithium-ion batteries are operating normally, the voltage changes relatively smoothly, the fluctuation range of discharge current is usually larger than that of voltage, and the change of internal resistance of lithium-ion batteries is relatively slow. The variance formula can measure the degree of dispersion or fluctuation of voltage, current and internal resistance over a period of time.

[0112] Construct the voltage variance formula:

[0113]

[0114] Construct the discharge current variance formula:

[0115]

[0116] Construct the internal resistance variance formula:

[0117]

[0118] T V represents the voltage variance threshold, T I represents the discharge current variance threshold, T RIt represents the internal resistance variance threshold. In actual applications, when lithium-ion batteries are in normal working condition, the voltage changes are relatively stable, the fluctuation range of discharge current is usually larger than that of voltage, and the change of internal resistance of lithium-ion batteries will be relatively slow. Then, through a large number of experiments, performance tests and analysis of actual application scenarios of specific types of lithium-ion batteries, the voltage, current and internal resistance data changes of lithium-ion batteries in normal operation and when various fault signs appear are collected, and the variance threshold is set. Experiments have found that the variance of its voltage in normal working condition is usually less than a certain value, which can be set as the voltage variance threshold T. V For example, the voltage variance threshold T is preset V =0.003, discharge current variance threshold T I =6 and internal resistance variance threshold T R =0.0000035, when When the voltage fluctuation is abnormal, When the current fluctuation is abnormal, When the internal resistance fluctuation is abnormal, it is judged that the voltage variance formula is abnormal. It can quantify the degree of data fluctuation and can intuitively compare the voltage stability of lithium-ion batteries at different times. Table 3 shows the variance of lithium-ion batteries in 40 groups of voltage, discharge current and internal resistance. By comparing with the set thresholds, it can reflect whether the voltage, discharge current and internal resistance of the lithium-ion battery are normal or not.

[0119]

[0120]

[0121]

[0122] Table 3 Voltage, discharge current and internal resistance variance

[0123] Step 5: Set the health index threshold, set the variance thresholds of the discharge current, voltage, and internal resistance respectively, compare the comprehensive health index with the health index threshold, and compare the variances of the discharge current, voltage, and internal resistance with the corresponding variance thresholds respectively, and comprehensively judge the health status of the lithium-ion battery based on the comparison results;

[0124] The health index threshold includes a good health index threshold and a critical health index threshold. For example, the good health index threshold is pre-set to H1=0.85, and the critical health index threshold is pre-set to H2=0.8. These two thresholds are based on a large number of experiments and verifications on lithium-ion batteries of a certain brand of electric vehicles. Different lithium-ion batteries have different performance and parameters due to their different materials, structures and application scenarios. Therefore, it is necessary to determine the appropriate health index threshold for the specific lithium-ion battery type to accurately evaluate its status.

[0125] When H≥H1, the value is assigned to 3; when H1>H>H2, the value is assigned to 2; when H≤H2, the value is assigned to 1;

[0126] like Figure 2 As shown, the main significance of these data is to achieve real-time monitoring of the health status of lithium-ion batteries by collecting health indexes. By regularly recording these health factors, it can not only assist in judging the health status of lithium-ion batteries, but also promptly discover the degradation trend of lithium-ion batteries.

[0127] The two horizontal lines, H1 and H2, represent the health thresholds. The comprehensive health index shows a clear downward trend between sample numbers 0 and 20, dropping from a high value close to 1 to below 0.8. This may indicate that the lithium-ion battery has experienced significant degradation during this period, perhaps due to frequent charging and discharging or other external factors affecting the performance of the lithium-ion battery, requiring maintenance. Between sample numbers 20 and 35, the comprehensive health index fluctuates, indicating that the lithium-ion battery can still recover to a higher health level at certain moments, perhaps due to an improvement in the state of charge or a change in load conditions.

[0128] For the variance of each feature data, if the variance is less than the threshold, it is assigned a value of 1; if the variance is greater than or equal to the threshold, it is assigned a value of 0, that is:

[0129] when Assign a value of 0, when Assign a value of 1;

[0130] when Assign a value of 0, when Assign a value of 1;

[0131] when Assign a value of 0, when Assign a value of 1;

[0132] like Figure 3-Figure 5 As shown, between sample numbers 0 and 20, the voltage variance gradually increases, suggesting that the lithium-ion battery may have entered the degradation stage and its health status needs to be paid attention to. Between sample numbers 0 and 20, the discharge current variance is always at a low level, indicating that the discharge performance of the lithium-ion battery is relatively stable, but it is on an upward trend and requires timely attention. If the threshold of the discharge current variance is exceeded, it will affect the normal operation of the equipment. Between sample numbers 0 and 20, the internal resistance variance is relatively low, and the internal resistance of the lithium-ion battery shows good stability, but the continuous increase has certain signs of aging or loss. Starting around sample number 35, the variances rise rapidly and begin to exceed the set thresholds. This may be due to defects in the manufacturing process, such as uneven materials and poor welding, which lead to drastic fluctuations in the performance of the lithium-ion battery.

[0133] Construct a comprehensive scoring formula:

[0134] S total =H score +SV score +SI score +SR score

[0135] Where S total Expressed as a comprehensive score, H score Expressed as a health index score, SV score Expressed as the score of voltage variance, SI score Expressed as the score of the discharge current variance, SR score It is expressed as the score of internal resistance variance;

[0136] When the comprehensive score is 6, that is, the comprehensive health index is greater than or equal to the good health index threshold H1, and the variance of each feature data is less than the corresponding variance threshold, it is judged that the health status of the lithium-ion battery is normal;

[0137] When the comprehensive score is 5, that is, the comprehensive health index is greater than or equal to the health index threshold H1, only one feature data variance is greater than the corresponding threshold, or the comprehensive health index satisfies H1>H>H2, and the variance of each feature data is less than the corresponding variance threshold, it is judged that the health status of the lithium-ion battery is in a sub-healthy state;

[0138] When the comprehensive score is less than or equal to 4, that is, the comprehensive health index is greater than or equal to the good health index threshold H1, the variance of at least two feature data is greater than the corresponding variance threshold; when the comprehensive health index satisfies H1>H>H2, the variance of at least one feature data is greater than the corresponding threshold; when the comprehensive health index satisfies H≤H2, no matter how many feature data variances are greater than the corresponding threshold, the health status of the lithium-ion battery is judged to be in a fault state.

[0139] The Comprehensive Health Index (CHI) is a crucial metric in lithium-ion battery health assessment. It measures the overall health of a lithium-ion battery by taking a weighted average of the voltage health factor, current health factor, and internal resistance health factor. However, the CHI does have limitations in reflecting data fluctuations. When data show an abnormal increase, and the extent of the increase is not sufficiently reflected in the CHI, the variance can reveal this anomaly by showing a greater degree of dispersion.

[0140] Taking the voltage health factor as an example, its calculation formula is Although the voltage fluctuation range and voltage deviation at individual moments are involved, these fluctuation information may be masked after comprehensive calculation with the average voltage and subsequent weighted average calculation of the comprehensive health index, especially for some abnormal conditions with small fluctuations but persistent existence.

[0141] By using both the comprehensive health index and variance as evaluation criteria, the health status of lithium-ion batteries can be comprehensively assessed from different perspectives. The comprehensive health index focuses on the comprehensive evaluation of overall performance, while the variance focuses on the assessment of data fluctuations. This allows for a more detailed and accurate determination of whether a lithium-ion battery is in a normal, sub-healthy, or faulty state.

[0142] like Figure 6 As shown, for the sample numbers with a comprehensive score of 6, it indicates that the sample's health index, variance index, etc. are in good condition, the lithium-ion battery is in the best working condition, and can be used safely and effectively. For the sample numbers with a comprehensive score of 5, this means that the sample has minor problems in some aspects, which may affect the performance and life of the lithium-ion battery. For the sample numbers with a comprehensive score of 4 and below, this indicates that the sample's health has seriously deteriorated, which may lead to lithium-ion battery failure or safety hazards. By regularly evaluating the health status of lithium-ion batteries, the decline in lithium-ion battery performance or potential safety hazards can be discovered in time. The lithium-ion batteries in numbers 19 and 20 were monitored to be in a faulty state. After maintenance, they returned to normal and sub-healthy states. Failures occurred after a period of time, which was related to their own lithium-ion battery performance. Long-term use may have caused irreversible damage. It is recommended to replace the lithium-ion battery.

[0143] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0144] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software depends on the specific application and design constraints of the technical solution.

[0145] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, and may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment as needed.

[0146] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. A method for online health status assessment of electric vehicle lithium-ion batteries using a data fusion model, characterized in that: The specific steps include: Step 1: Preset a speed range as a monitoring range. When the vehicle is in a driving state and enters the monitoring range for the first time, continuously acquire characteristic data of the electric vehicle at the same time interval until the vehicle leaves the monitoring range for the first time. Construct a characteristic data table based on the data acquired within this range. The characteristic data includes the voltage, discharge current, and internal resistance of the lithium-ion battery. Step 2: Extract the characteristic data table to obtain the rated voltage, rated discharge current, and rated internal resistance data of the lithium-ion battery, record the maximum and minimum voltage values, the maximum value data of the discharge current and the internal resistance, and calculate the average value of the voltage, discharge current, and internal resistance data respectively; Step 3: Build the voltage health factor using the rated voltage, average voltage, maximum voltage, and minimum voltage. Build the current health factor using the rated discharge current, average discharge current, and minimum discharge current. Build the internal resistance health factor using the rated internal resistance, maximum internal resistance, and average internal resistance. Step 4: Based on the voltage health factor, current health factor, and internal resistance health factor, a weighted average algorithm is used to obtain a comprehensive health index, and the variance of the discharge current, voltage, and internal resistance are calculated respectively; Step 5: Set the health index threshold, set the variance threshold of the discharge current, voltage, and internal resistance respectively, compare the comprehensive health index with the health index threshold, and compare the variance of the discharge current, voltage, and internal resistance with the corresponding variance threshold respectively, and comprehensively judge the health status of the lithium-ion battery based on the comparison results.

2. The method for online health status assessment of electric vehicle lithium-ion batteries using a data fusion model according to claim 1, characterized in that: For an operation period within a preset speed range, recording characteristic data of the lithium-ion battery during the period includes the following steps: Set the speed range to be monitored to: u fw ∈[20, 30],u fw represents the real-time speed of the electric vehicle to be evaluated. The time interval is: Δt = 2s. After the electric vehicle starts and begins to travel, when the vehicle speed reaches 20km / h for the first time, the voltage, discharge current and internal resistance data are collected at 2s intervals. Data collection stops when the vehicle speed drops below 20km / h or exceeds 30km / h.

3. The method for online health status assessment of electric vehicle lithium-ion batteries using a data fusion model according to claim 1, characterized in that: The method for calculating the average value of voltage, discharge current and internal resistance data is: Extract the voltage, discharge current, and internal resistance data at each time interval from the characteristic data table and calculate the average voltage: Where V avg It is expressed as the average voltage, n is the number of time intervals, that is, the number of times the lithium-ion battery characteristic data is recorded, V i It is represented by the voltage value collected at the i-th time interval, i = 1, 2, 3, ..., n, where n is a positive integer ≥ 10. That is, if the data collection duration is less than 10s (n<10), the collected data will not be used in the health status assessment calculation. The assessment calculation can only be performed after the vehicle enters the monitoring speed range again and the collection duration reaches 10s or more. Calculate the average discharge current: Where, I avg Expressed as the average discharge current, I i It is represented by the discharge current value collected at the i-th time interval; Calculate the average internal resistance: Where R avg Expressed as the average internal resistance, R i It is represented by the internal resistance value collected at the i-th time interval.

4. The method for online health status assessment of electric vehicle lithium-ion batteries using a data fusion model according to claim 1, characterized in that: The method for constructing voltage health factor, current health factor and internal resistance health factor is: The voltage health factor is constructed by the rated voltage, average voltage, maximum voltage and minimum voltage: Where V rated Expressed as the rated voltage of the lithium-ion battery, K v1 , K v2 is the weight coefficient, and K v1 >K v2 >0; The current health factor is constructed by the rated discharge current, the average discharge current and the minimum discharge current: Where, I rated Expressed as the rated discharge current of the lithium-ion battery, K I1 , K I2 is the weight coefficient, and K I1 >K I2 >0,I min Expressed as the minimum discharge current; The internal resistance health factor is constructed by the rated internal resistance, maximum internal resistance and average internal resistance: Where R rated Expressed as the rated internal resistance of the lithium-ion battery, K R1 , K R2 is the weight coefficient, and K R1 >K R2 >0, R max Expressed as the maximum internal resistance.

5. The method for online health status assessment of electric vehicle lithium-ion batteries using a data fusion model according to claim 1, characterized in that: The method of using the weighted average algorithm to obtain the comprehensive health index is: Assign corresponding weights to the voltage health factor, current health factor, and internal resistance health factor: ω V 、ω I and ω R , and ω V >ω I =ω R ,ω V +ω I +ω R =1, and use the weighted average algorithm to construct the comprehensive health index formula: H=ω V ·F v +oh I ·F I +oh R ·F R 。 6. The method for online health status assessment of electric vehicle lithium-ion batteries using a data fusion model according to claim 1, characterized in that: The method for calculating the variance of discharge current, voltage and internal resistance is: Construct the voltage variance formula: Where, Expressed as voltage variance; Construct the discharge current variance formula: Where, Expressed as discharge current variance; Construct the internal resistance variance formula: Where, Expressed as internal resistance variance.

7. The method for online health status assessment of electric vehicle lithium-ion batteries using a data fusion model according to claim 1, characterized in that: The method to judge the health status of lithium-ion batteries is: The threshold of the health index of the good state is preset as H1, the threshold of the health index of the critical health state is preset as H2, and H1>H2>0, and the voltage variance threshold is T V , the discharge current variance threshold is T I , the internal resistance variance threshold is T R ; When H≥H1, the value is assigned to 3; when H1>H>H2, the value is assigned to 2; when H≤H2, the value is assigned to 1; For the variance of each feature data, if the variance is less than the threshold, it is assigned a value of 1; if the variance is greater than or equal to the threshold, it is assigned a value of 0, that is: when Assign a value of 0, when Assign a value of 1; when Assign a value of 0, when Assign a value of 1; when Assign a value of 0, when Assign a value of 1; Construct a comprehensive scoring formula: S total =H score +SV score +SI score +SR score Where S total Expressed as a comprehensive score, H score Expressed as a health index score, SV score Expressed as the score of voltage variance, SI score Expressed as the score of the discharge current variance, SR score It is expressed as the score of internal resistance variance; When the comprehensive score is 6, the health status of the lithium-ion battery is judged to be normal; when the comprehensive score is 5, the health status of the lithium-ion battery is judged to be sub-healthy; when the comprehensive score is less than or equal to 4, the health status of the lithium-ion battery is judged to be in a fault state.

Images