A power battery risk identification and traceability system based on extreme voltage difference

Abstract

The present invention relates to the technical field of power batteries, and discloses a power battery risk identification and tracing system based on extreme voltage, comprising an acquisition module, a preprocessing module, an element extraction module, a risk quantification module, a risk identification module and a risk tracing module; the acquisition module is used to acquire historical operating data of the power battery as initial data; the preprocessing module is used to preprocess the initial data and obtain basic data; the element extraction module is used to extract risk elements; the risk quantification module is used to quantify risk elements and form quantitative features p; the risk identification module is used to identify the safety status based on the quantitative features and obtain the absolute risk probability; the risk tracing module is used to prepare a risk element image according to a tracing strategy, and determine the risk source based on feature changes in the safety element image.

Description

Technical Field

[0001] The present invention relates to the technical field of power batteries, and in particular to a power battery risk identification and tracing system based on extreme voltage difference. Background Art

[0002] With the advent of the electric vehicle era, electric vehicle safety has become a major concern. As the core component of new energy electric vehicles, power batteries face an increasingly prominent need for accurate quantitative estimation of their safety status and structural protection. Therefore, accurately identifying potential risks in power batteries, accurately quantifying and estimating their safety status, and accurately diagnosing power battery failures are crucial to improving the durability, safety, and reliability of new energy vehicles.

[0003] However, existing risk assessment or safety status assessment methods for new energy electric vehicle power battery safety often rely on professionals to score and conduct qualitative analysis based on their personal experience and subjective judgment, rarely involving quantitative analysis. This results in high labor costs and low analytical accuracy and reliability. Furthermore, with the continuous updating and optimization of new energy vehicles, their operating environments are becoming increasingly complex, generating multi-dimensional operational data that is simultaneously redundant, heterogeneous, and strongly coupled. This significantly hinders data analysis, resulting in low assessment accuracy and an inability to accurately capture fault anomalies.

[0004] Moreover, due to the limitations of the progress in battery mechanism research and the low feasibility of actual battery detection operations, the fault diagnosis of power batteries has also been limited by deficiencies caused by such objective factors. The accuracy and feasibility of abnormal detection and fault determination of power batteries are also unsatisfactory, so that the potential risks of power batteries cannot be captured in a timely manner and the source of risk cannot be accurately determined in a timely manner. Summary of the Invention

[0005] The present invention aims to provide a power battery risk identification and tracing system based on extreme voltage difference, which can accurately identify and quantify the power battery risks, accurately determine the power battery status and accurately trace the fault source.

[0006] The basic solution provided by the present invention is: a power battery risk identification and tracing system based on extreme voltage difference, including an acquisition module, a preprocessing module, an element extraction module, a risk quantification module, a risk identification module and a risk tracing module;

[0007] The acquisition module is used to acquire historical operating data of the power battery as initial data; the preprocessing module is used to preprocess the initial data and obtain basic data;

[0008] The element extraction module is used to extract risk factors, and when extracting, it calculates the extreme voltage based on the basic data and converts the extreme voltage into risk factors according to the conversion strategy. The risk factors include the first-level risk factors and secondary risk factors Among them, V aa is the extreme difference voltage, V av is the speed of change of extreme voltage; α is the signal amplification factor;

[0009] The risk quantification module is used to quantify risk factors and form a quantitative feature p, where the quantitative feature p is a quantitative feature value of the risk factor in the interval [0,1];

[0010] The risk identification module is used to identify the safety status based on quantitative characteristics and obtain the absolute risk probability;

[0011] The risk tracing module is used to prepare a risk factor image according to the tracing strategy, and determine the risk source based on the feature changes in the security factor image.

[0012] The working principle and advantages of the present invention are: selecting the extreme voltage parameter item and the extreme voltage change rate parameter item as risk factors, and quantifying them into abnormal risk characteristics of extreme pressure difference stability and extreme pressure difference change rate stability (i.e., quantitative characteristics p), so as to realize accurate judgment of the comprehensive status of power batteries and new energy vehicles and identification of risk sources (i.e., identification of abnormal fault causes).

[0013] In particular, the feature extraction module in this solution can effectively extract parameter features such as extreme pressure difference as risk factors based on the mechanism of the power battery, and then quantitatively describe the risk status of the power battery from a specific dimension (the extreme pressure difference dimension). Specifically, cell voltage is an important parameter that describes the safety status of the battery during vehicle operation. Many battery failures will cause fluctuations in cell voltage. However, considering the physical characteristics of the battery system itself, as well as factors such as sensor design and acquisition accuracy, information coupling, redundancy, and errors are inevitable between different signals, making it difficult to accurately extract and quantify features. This solution specifically selects the extreme voltage and the rate of change of the extreme pressure difference as features for extraction and quantification. The extreme voltage is the difference between the maximum and minimum values of all single cell voltages at any time, which can effectively express the extreme conditions in the single cell. The quantitative description of this extreme condition can accurately quantify the degree of difference in the status of the power battery during service. Compared with the single cell voltage, it is less susceptible to coupling, redundancy, and error caused by external factors such as the battery system, working conditions, and environment. It can further accurately evaluate the safety status of the power battery and the vehicle, accurately identify and quantify the risks of the power battery, and then accurately determine the status of the power battery and accurately trace the source of the fault. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a schematic diagram of the system structure of an embodiment of a power battery risk identification and tracing system based on extreme voltage difference of the present invention;

[0015] Figure 2 This is a first risk factor image of an embodiment of a power battery risk identification and tracing system based on extreme voltage difference according to the present invention;

[0016] Figure 3 This is a second risk factor image of an embodiment of a power battery risk identification and tracing system based on extreme voltage difference according to the present invention;

[0017] Figure 4 This is a third risk factor image of an embodiment of a power battery risk identification and tracing system based on extreme voltage difference according to the present invention;

[0018] Figure 5 This is the fourth risk factor image of an embodiment of a power battery risk identification and tracing system based on extreme voltage difference of the present invention;

[0019] Figure 6 This is the fifth risk factor image of an embodiment of a power battery risk identification and tracing system based on extreme voltage difference of the present invention;

[0020] Figure 7 This is the sixth risk factor image of an embodiment of a power battery risk identification and traceability system based on extreme voltage difference of the present invention;

[0021] Figure 8 This is the seventh risk factor image of an embodiment of a power battery risk identification and tracing system based on extreme voltage difference of the present invention;

[0022] Figure 9 This is an eighth risk factor image of an embodiment of a power battery risk identification and tracing system based on extreme voltage difference according to the present invention;

[0023] Figure 10 This is the ninth risk factor image of an embodiment of a power battery risk identification and tracing system based on extreme voltage difference of the present invention. DETAILED DESCRIPTION

[0024] The following is a further detailed description through specific implementation methods:

[0025] The embodiment is basically as shown in the attached Figure 1 Shown: A power battery risk identification and tracing system based on extreme voltage difference, including an acquisition module, a preprocessing module, an element extraction module, a risk quantification module, a risk identification module and a risk tracing module;

[0026] The acquisition module is used to acquire historical operating data of the power battery as initial data. Specifically, in this embodiment, corresponding battery signal data is obtained by parsing the message log of the power battery system that complies with the GB32960 protocol as the initial data of the power battery.

[0027] The preprocessing module is used to preprocess the initial data and obtain basic data.

[0028] In the preprocessing module, the preprocessing includes: (1) limiting the data signal boundary value; that is, the data of the voltage and current signal data exceeding the first preset threshold in the initial data is treated as abnormal data and removed. In this embodiment, the first specified threshold is set to [2.5, 4.25], which can fully clean the abnormal data. (2) Identifying and marking interference pulses; that is, if the difference between the current frame voltage data and the previous frame exceeds the second preset threshold, the frame data is marked. The second specified threshold is set to 3s d ;s d is the standard deviation of the voltage change rate. (3) Identify and mark time discontinuities; that is, if the difference between the timestamp data of the current frame and the previous frame exceeds the third preset threshold, mark the frame data. The third specified threshold is set to 120s. (4) Perform mean filtering on the initial data; and when performing mean filtering, specify a time window and calculate the mean of each characteristic signal in the time series in the specified time window.

[0029] The element extraction module is used to extract risk factors, and when extracting, it calculates the extreme voltage based on the basic data and converts the extreme voltage into risk factors according to the conversion strategy. The risk factors include the first-level risk factors and secondary risk factors Among them, V aa is the extreme difference voltage, V av is the speed of change of extreme voltage; α is the signal amplification factor.

[0030] Specifically, when calculating the extreme differential pressure, the voltage signal data is first extracted from the basic data; and the voltage signal data is recorded as V:

[0031]

[0032] Where, Represents the vector composed of the voltage data of all cells of the power battery at the nth sampling point.

[0033] Then calculate the maximum value of all cells at any moment, recorded as V max :

[0034]

[0035] Calculate the minimum value of all cells at any time, recorded as V min :

[0036]

[0037] Then we get the extreme voltage V aa :

[0038]

[0039] The conversion strategy includes: performing signal amplification and nonlinear conversion on the extreme voltage to obtain the first-level risk factor

[0040] Perform speed filtering on the extreme voltage difference to obtain the extreme voltage change speed; perform signal amplification and nonlinear conversion on the extreme voltage change speed to obtain the secondary risk factor

[0041] Specifically, when calculating the extreme voltage variation speed, the obtained extreme voltage variation speed V av :

[0042]

[0043] in, Indicates the speed of change of the extreme voltage at the i-th sampling point; specifically, It means that for the i-th sampling point, the sliding window length is specified as l, and the difference between the range voltages corresponding to the l / 2 sampling points before and after it is taken as the range change ΔV aa , and then take the ratio of this value to the time span Δt of the specified time window as the extreme voltage change rate corresponding to sampling point i. At the same time, for the first 1 / 2 sampling points and the last 1 / 2 sampling points, a valid padding operation is performed and their speed values are set to 0.

[0044] The risk quantification module is used to quantify risk factors and form a quantitative feature p, which is a quantitative feature value of the risk factor in the interval [0,1]. The quantitative feature p = 1-λ; where λ = E 2 (Sf) / E(Sf 2 ), 0≤λ≤1.

[0045] The risk identification module is used to identify safe states based on quantitative features and obtain absolute risk probabilities. Specifically, the closer λ is to 1, the smaller p is, indicating that the fluctuation of the extreme voltage characteristic on a time scale is smaller and the battery state is safer. Specifically, when performing safe state identification, the quantitative features on the time scale are discretely integrated to obtain Sp = ∑p; Sp is a monotonically increasing curve, and the slope value z of this curve is taken as the benchmark for safe state identification. The amplitude of z obtained by identification is used as the absolute risk probability corresponding to the consistent safety feature.

[0046] The risk tracing module is used to prepare a risk factor image according to the tracing strategy, and determine the risk source based on the feature changes in the security factor image.

[0047] The traceability strategy includes selecting the time point at which the absolute risk probability exceeds a specified threshold as the risk time point; collecting the cell voltage, current, and SOC values in the local time before and after the risk time point to create a risk factor image; and determining the risk source based on the fluctuations in the corresponding numerical curves (including the cell voltage curve, current curve, and SOC value curve) of each cell in the risk factor image. These risk sources include sampling anomalies, self-discharge anomalies, connection anomalies, consistency anomalies, internal short circuit anomalies, module replacement anomalies, and maintenance anomalies.

[0048] Specifically, when the traceability strategy is in operation, it includes the following risk source identification situations:

[0049] (1) When tracing back to the acquired risk point based on the first-level risk factors

[0050] As attached Figure 2 As shown, in the risk factor image, the fluctuation of each numerical curve is manifested as follows: before and during the charging process No. 1, the single cell voltage of a single battery cell is continuously lower than the single cell voltage of other battery cells, and there is a situation of poor consistency (that is, it is determined that the risk source is a consistency anomaly). The single cell voltage of the same battery cell returns to normal in the early stage of the charging process No. 2. In the middle and late stages of the discharge process, the voltage of the same battery cell drops rapidly in the medium and high SOC stages, and returns to normal in the charging process No. 3. It is determined that there is a balancing maintenance operation between the charging process No. 1 and the charging process No. 2; there is a pack replacement maintenance operation between the charging process No. 2 and the charging process No. 3. In this embodiment, the maintenance anomaly corresponds to the historical maintenance of the power battery. This solution can simultaneously realize the determination of the type of maintenance operation and whether the maintenance is effective, and can provide an effective data reference for power battery fault maintenance.

[0051] As attached Figure 3As shown in the risk factor image, the fluctuation of each numerical curve is as follows: the single cell voltage numerical curve of a single battery cell drops out of the group in the initial period, and the numerical abnormality intensifies in the middle period until the voltage drops sharply and the voltage difference exceeds 1000mV. It is then determined that the battery cell has an internal short circuit abnormality, and the internal short circuit abnormality is caused by self-discharge.

[0052] As attached Figure 4 As shown in the risk factor image, the fluctuation of each numerical curve is manifested as follows: there are two battery cells whose voltages always shift in opposite directions, and the electrical signals of the two battery cells are connected, that is, compared with other normal battery cells, the voltages of these two battery cells are obviously high and low, and there is no high-charge and low-discharge situation in a single battery cell, then it is determined that the power battery has an abnormal sampling fault.

[0053] As attached Figure 5 As shown in the risk factor image, the fluctuation of each numerical curve is manifested as follows: when a single battery cell (abnormal battery cell) suddenly experiences abnormal voltage fluctuation during vehicle driving, it is determined that there is a sampling abnormality fault, and the sampling abnormality is mostly caused by interference with the sampling chip. When the interference is eliminated, the abnormality will also disappear.

[0054] (2) When tracing back to the acquired risk point based on the secondary risk factors

[0055] As attached Figure 6 As shown in the risk factor image, the fluctuation of each numerical curve is manifested as follows: there is a voltage drop layer in the single cell voltage curve of a single battery cell (abnormal battery cell), and at the end of the third charging process, the voltage difference of the battery cell compared with the normal battery cell is significantly increased compared with the previous two charging processes, that is, the self-discharge abnormality shows a trend of obvious deterioration in a shorter charge and discharge cycle, and it is determined that a relatively serious self-discharge abnormality exists.

[0056] As attached Figure 7 As shown in the risk factor image, the fluctuation of each numerical curve is manifested as follows: if a single battery cell (abnormal battery cell) has an obvious high-charge and low-discharge phenomenon, it is determined that the vehicle has a connection abnormality.

[0057] Correspondingly, the fault causes the internal resistance of the battery cell to increase, resulting in a phenomenon in which the voltage during the charging process is higher than that of normal batteries. During the discharge process, as the vehicle switches back and forth between acceleration and braking, the voltage also oscillates with the discharge and feeding states, but the abnormal battery cell with increased internal resistance has a greater oscillation amplitude.

[0058] As attached Figure 8As shown in the risk factor image, the fluctuation of each numerical curve is manifested as follows: the part of the cells where the two cells (abnormal cells) are located shows obvious stratification and the signal trend is consistent. Since their cell numbers are close, it is determined that the module replacement is abnormal, specifically that the replacement of the module where the abnormal cell is located causes its single cell voltage behavior to collectively deviate from the group.

[0059] As attached Figure 9 As shown in the risk factor image, the fluctuation of each numerical curve is manifested as follows: there is an obvious stratification phenomenon in the single cell voltage numerical curve, and the voltage numerical curve corresponding to the two battery cells in the bottom layer has the largest offset, but the voltage of the stratified battery cells changes with the working conditions, and the overall voltage behavior is similar, which is judged to be consistency abnormality, that is, the consistency of each single cell in the power battery itself is poor.

[0060] As attached Figure 10 As shown in the risk factor image, the fluctuations of each value curve are as follows: the abnormal voltage evolution trend in the single cell voltage value curve of a single battery cell (abnormal battery cell) is that the voltage drops out of the group in the initial period, the value abnormality intensifies in the middle period, until the voltage drops sharply, and the voltage difference exceeds 1000mV. The abnormality is eliminated after a time interval. It is determined that the battery cell has an internal short circuit abnormality caused by self-discharge. After the problem deteriorates, the battery is repaired in time, avoiding subsequent serious accidents including thermal runaway.

[0061] This embodiment provides a power battery risk identification and traceability system based on extreme voltage, which can accurately identify and quantify power battery risks, precisely determine the power battery status, and accurately trace the source of the fault. Furthermore, this solution also features a risk traceability module that, based on the absolute risk probability determined by the risk identification module and combined with a risk factor graph, can reliably assess the data performance of power batteries that exhibit different regular changes, and correspondingly determine the risk sources corresponding to different abnormal data performances. Accurately determining the risk source helps to preemptively address risks and can provide reliable data references for improving the safety and reliability of power batteries and new energy vehicles.

[0062] The above is only an embodiment of the present invention. Common knowledge such as the specific structure and characteristics of the scheme is not described in detail here. Ordinary technicians in the relevant field are aware of all common technical knowledge in the technical field of the invention before the application date or priority date, can obtain all existing technologies in the field, and have the ability to apply conventional experimental means before that date. Ordinary technicians in the relevant field can improve and implement this scheme in combination with their own abilities under the guidance of this application. Some typical well-known structures or well-known methods should not become obstacles for ordinary technicians in the relevant field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several variations and improvements can be made, which should also be regarded as the scope of protection of the present invention. These will not affect the effect of the implementation of the present invention and the practicality of the patent.

Claims

1. A power battery risk identification and traceability system based on extreme voltage difference, characterized in that: It includes acquisition module, pre-processing module, factor extraction module, risk quantification module, risk identification module and risk tracing module; The acquisition module is used to collect historical operating data of the power battery as initial data; The preprocessing module is used to preprocess the initial data and obtain basic data; The element extraction module is used to extract risk factors, and when extracting, it calculates the extreme voltage based on the basic data and converts the extreme voltage into risk factors according to the conversion strategy. The risk factors include the first-level risk factors and secondary risk factors ;in, is the extreme difference voltage, is the speed of change of extreme voltage; is the signal amplification factor; The risk quantification module is used to quantify risk factors and form quantitative features p, the quantitative features p is the risk factor in Quantized eigenvalues within the interval; The risk identification module is used to identify the safety status based on quantitative characteristics and obtain the absolute risk probability; The risk tracing module is used to prepare a risk factor image according to the tracing strategy, and determine the risk source based on the feature changes in the security factor image.

2. A power battery risk identification and traceability system based on extreme voltage difference according to claim 1, characterized in that: In the preprocessing module, the preprocessing includes limiting the boundary value of the data signal, identifying and marking interference pulses, identifying and marking time discontinuity points, and performing mean filtering on the initial data.

3. The power battery risk identification and traceability system based on extreme voltage difference according to claim 1 is characterized in that: The traceability strategy includes: selecting the time point when the absolute risk probability exceeds a specified threshold as the risk time point; collecting the single cell voltage, current and SOC values in the local time before and after the risk time point to prepare a risk factor image; and determining the risk source based on the fluctuation of the corresponding numerical curve of each battery cell in the risk factor image.

4. The power battery risk identification and traceability system based on extreme voltage difference according to claim 3 is characterized in that: The risk sources include: sampling abnormality, self-discharge abnormality, connection abnormality, consistency abnormality, internal short circuit abnormality, module replacement abnormality and maintenance abnormality.

5. The power battery risk identification and traceability system based on extreme voltage difference according to claim 1, characterized in that: The quantitative characteristics ;in, , .

Images