A lithium ion battery active safety early warning method based on end-cloud cooperation

Abstract

The application discloses a kind of lithium ion battery active safety early warning methods based on end cloud cooperation, it is related to the technical field of vehicle power battery pack, solve the technical problem of low accuracy and identification efficiency in prior art fault identification;Including: obtaining the real-time operation data of lithium ion battery pack from cloud;Battery voltage distance abnormality evaluation;Battery voltage entropy abnormality evaluation;Battery voltage difference abnormality evaluation;Battery internal resistance estimation;According to voltage distance anomaly, voltage difference fluctuation anomaly, voltage entropy anomaly, comprehensive judgment is made to make active safety early warning, while early warning result will also be compared with battery internal resistance estimation value and verified, improve the accuracy of active safety early warning;The application can quickly identify a variety of abnormal risks, improve the operation efficiency and safety reliability of battery management system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle power battery packs, in particular to a lithium ion battery active safety early warning method based on end-cloud cooperation. BACKGROUND

[0002] With the development of electric vehicles, electronic products, energy storage systems and other technologies, lithium ion batteries have become widely used battery technologies. However, lithium batteries also have certain safety hazards, such as overcharge, overdischarge, short circuit and other abnormal conditions that may cause the battery to heat runaway, and even cause fire and explosion accidents.

[0003] Currently, the main solutions to lithium battery safety hazards include: (1) passive safety protection: through battery management system (BMS) to monitor the voltage, current, temperature and other parameters of the battery, and trigger overcharge, overdischarge, overcurrent and other protection measures once abnormal conditions are detected, but this method can only respond to abnormalities and cannot provide early warning; (2) chemical / material improvement: by optimizing the positive and negative electrode materials, separators, electrolytes and other chemical components to improve the intrinsic safety of the battery, but due to factors such as cost and energy density, it is difficult to fully replace existing battery technology; (3) macro design improvement: through modularization, series-parallel connection and other battery system design optimization to improve overall safety, but it cannot solve the safety hazards of the battery itself; based on the above shortcomings, the present application proposes a lithium ion battery active safety early warning method based on end-cloud cooperation. SUMMARY

[0004] The present application aims to at least solve one of the technical problems existing in the prior art; for this purpose, the present application proposes a lithium ion battery active safety early warning method based on end-cloud cooperation, which can realize fine monitoring and early warning of the battery pack, take timely measures before abnormal conditions occur, thereby improving the safety of lithium batteries in use, and has important application value.

[0005] To achieve the above purpose, the first aspect of the present application provides a lithium ion battery active safety early warning method based on end-cloud cooperation, comprising the following steps:

[0006] Step one: obtaining real-time running data of lithium ion battery pack from the cloud;

[0007] Step two: battery monomer voltage distance abnormality evaluation, specifically including:

[0008] According to the data screening and calculation rules, the z-score value of each battery cell is calculated;

[0009] When the z-score distance value of the battery cell is observed for the first time > 3, it is determined that the battery cell is an abnormal battery cell, and the current time is recorded as the starting time of the abnormal battery cell;

[0010] From the starting time, judge whether the subsequent maximum distance battery is in the recorded abnormal battery set. If the battery is in the recorded set and its distance value is greater than 3, the abnormal frequency of the battery is increased by 1.

[0011] Select the battery with abnormal frequency ≥100 times as an abnormal battery.

[0012] Step three: battery voltage entropy abnormal evaluation, specifically including:

[0013] According to the data screening and calculation rules, the entropy value of the minimum voltage battery in each time window is calculated.

[0014] If the voltage entropy value of the battery in the 3-day time window is 0, and the frequency of the battery as the minimum voltage battery is ≥100 times, the battery is determined to be abnormal.

[0015] Step four: battery voltage difference abnormal evaluation, specifically including:

[0016] According to the data screening and calculation rules, the voltage difference between the maximum voltage and the minimum voltage of the battery pack is calculated.

[0017] Abnormal judgment is made on the voltage difference.

[0018] Step five: battery internal resistance estimation.

[0019] Step six: active safety warning: according to the voltage distance abnormality, voltage difference fluctuation abnormality, and voltage entropy abnormality, comprehensive judgment is made to make active safety warning; wherein the warning result is compared and verified with the battery internal resistance estimation value.

[0020] Further, the data screening and calculation rules in step two are as follows:

[0021] From the charging state, select the charging data that meets the conditions, wherein the selection conditions include: 1. Current greater than 0; 2. Maximum voltage greater than or equal to 3.78V;

[0022] For each charging data, calculate the average value of the voltage of all batteries And the voltage standard deviation s; then calculate the z-score value of each battery, the calculation formula is as follows:

[0023] Where x i represents the voltage of the i-th battery, z i represents the z-score value of the i-th battery; z-score reflects the deviation of the voltage of each battery from the average voltage, which is used to judge the abnormality of the battery.

[0024] Further, the data screening and calculation rules in step three are as follows:

[0025] The charging data with a current greater than or equal to 3A is screened from the charging state to ensure that the battery pack has entered a stable charging working condition; a sliding window is used to set a time window of 3 days, and the entropy value of the minimum voltage cell in each time window is calculated, and the calculation formula is as follows:

[0026]

[0027] where n represents a total of n cells, x i represents the voltage of the i th cell, p(x i ) represents the data probability density of the voltage of the i th cell, and H(x) represents the entropy value of the i th cell.

[0028] Further, the data screening and calculation rules in step four are as follows:

[0029] The charging data meeting the conditions is screened from the charging state, and the screening condition is that the maximum voltage is between 3.78V and 3.82V; for each charging data meeting the condition, the voltage difference between the maximum voltage and the minimum voltage of the battery pack is calculated.

[0030] Further, the voltage difference is judged for abnormality, which specifically includes:

[0031] When the voltage difference is greater than or equal to 20mV is observed for the first time, the observation time is recorded as the starting abnormal time; from the starting abnormal time, if the voltage difference at the subsequent time is still greater than or equal to 20mV, the abnormal frequency is added by 1;

[0032] When the abnormal frequency reaches 100 times or more, and at least once in the time window, the voltage difference is greater than or equal to 60mV, it is determined that the voltage difference fluctuation is abnormal.

[0033] Further, the battery monomer internal resistance estimation specifically includes:

[0034] Selecting the adjacent two data whose running state changes from the static state to the charging state, and the time difference between the two data is less than or equal to 3 minutes; screening the charging data in the charging state, and the screening condition is that the current is greater than or equal to 5A; the SOC is greater than or equal to 20%; and the running state of the next data is the charging state;

[0035] The internal resistance of the battery monomer is calculated for the adjacent two data meeting the condition, and the calculation formula is as follows:

[0036]

[0037] where R i represents the estimated internal resistance of the battery monomer i, represents the voltage of the battery monomer i in the charging state, V represents the voltage of the battery monomer i in the static state, and I represents the current in the charging state.

[0038] Further, the early warning result in step six specifically comprises:

[0039] In the battery monomer voltage distance abnormality evaluation calculation, if the battery cell is the battery cell with the largest voltage distance from the beginning to the end, and the average voltage distance of the battery cell is greater than or equal to 3, it is determined that the device is a particularly serious device candidate;

[0040] If the device has abnormal fluctuations in the battery monomer voltage difference abnormality evaluation, it is directly determined that the device is a particularly serious device candidate;

[0041] If the device has abnormalities in both the battery monomer voltage distance abnormality evaluation and the battery monomer voltage entropy abnormality evaluation, and the abnormal battery cells are the same, it is determined that the device is a particularly serious device candidate;

[0042] Output the early warning result: perform intersection analysis on all particularly serious device candidates to obtain the final particularly serious device, and output the particularly serious device; output the remaining devices in the abnormal device as general serious devices.

[0043] Further, the real-time running data includes running time, SOC, SOH, running state, total voltage of the battery pack, current, and cell voltage; wherein the running state includes charging state, discharging state and static state.

[0044] Compared with the prior art, the beneficial effects of the present application are:

[0045] The present application obtains real-time running data of the lithium ion battery pack from the cloud, performs battery monomer voltage distance abnormality evaluation, battery monomer voltage entropy abnormality evaluation, battery monomer voltage difference abnormality evaluation and battery monomer internal resistance estimation; according to the voltage distance abnormality, voltage difference fluctuation abnormality and voltage entropy abnormality, the active safety early warning is comprehensively judged; at the same time, the early warning result is also compared and verified with the battery monomer internal resistance estimation value, so as to improve the accuracy of the active safety early warning; the present application can quickly identify various abnormal risks, and improve the operation efficiency and safety reliability of the battery management system. BRIEF DESCRIPTION OF DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0047] Figure 1A flowchart of a lithium ion battery active safety early warning method based on end-cloud cooperation. DETAILED DESCRIPTION

[0048] The technical solutions of the present application will be described below in conjunction with embodiments. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0049] Please refer to Figure 1 The first aspect of the present application provides a lithium ion battery active safety early warning method based on end-cloud cooperation, comprising the following steps:

[0050] Step one: obtain real-time running data of lithium ion battery pack from the cloud; the real-time running data includes running time, SOC, SOH, running state, total voltage of the battery pack, current, and cell voltage; wherein the running state includes charging state, discharging state, and standing state;

[0051] Step two: battery monomer voltage distance abnormality evaluation, specifically including:

[0052] S21: data screening and calculation rules: screen the charging data that meets the conditions from the charging state, wherein the screening conditions include: 1. current greater than 0; 2. maximum voltage greater than or equal to 3.78V;

[0053] For each piece of charging data, calculate the average value of the voltage of all cells And the voltage standard deviation s; then calculate the z-score value of each cell, and the calculation formula is as follows:

[0054] Where x i represents the voltage of the i-th cell, and z i The z-score value of the i-th cell; z-score reflects the deviation of the voltage of each cell from the average voltage, which can be used to judge the abnormality of the cell; a too large distance value means that the voltage of the cell is much higher or much lower than that of other cells, which may have hidden dangers;

[0055] S22: Abnormality judgment and statistics: when the z-score distance value of the cell is observed for the first time > 3, it is determined that the cell is an abnormal cell, and the current time is recorded as the starting time of the abnormal cell;

[0056] From the starting time, judge whether the subsequent maximum distance cell is in the recorded abnormal cell set, if the cell is in the recorded set and its distance value > 3, then the abnormal frequency of the cell + 1;

[0057] Select the abnormal frequency of the battery cell as the abnormal battery cell. If multiple battery cells have an abnormal frequency of ≥100 times, then select the single cell with the most warning times as the abnormal battery cell according to the battery cell warning times;

[0058] Step three: battery cell voltage entropy abnormality evaluation, specifically including:

[0059] S31: Data screening and calculation rules: select charging data with a current of ≥3A from the charging state to ensure that the battery pack has entered a stable charging working condition;

[0060] Using a sliding window, set the time window to 3 days, calculate the entropy value of the minimum voltage cell in each time window, and the calculation formula is as follows:

[0061]

[0062] Where n represents a total of n battery cells, x i represents the voltage of the i-th battery cell, p(x i ) represents the data probability density of the i-th battery cell voltage, and H(x) represents the entropy value of the i-th battery cell. Voltage entropy reflects the uniformity of the voltage distribution of the battery cell, and can find the uneven voltage distribution inside the battery pack, providing a basis for proactive safety warning;

[0063] S32: Abnormality judgment: if the voltage entropy value of the battery cell in the 3-day time window is 0, it means that the voltage of the battery cell is completely fixed and has no fluctuation;

[0064] At the same time, if the frequency of the battery cell as the minimum voltage cell is ≥100 times, it means that this voltage fixed condition has lasted for a long time; according to the above two conditions, the battery cell can be determined as an abnormal battery cell.

[0065] Step four: battery cell voltage difference abnormality evaluation, specifically including:

[0066] S41: Data screening and calculation rules: select charging data that meet the conditions from the charging state, where the selection conditions are: the maximum voltage is between 3.78V and 3.82V; for each charging data that meets the conditions, calculate the voltage difference between the maximum voltage and the minimum voltage of the battery pack;

[0067] S42: Abnormality judgment: when the voltage difference is first observed to be greater than or equal to 20mV, the observation time is recorded as the initial abnormal time; from the initial abnormal time, if the voltage difference at the subsequent time is still greater than or equal to 20mV, the abnormal frequency is added by 1;

[0068] When the abnormal frequency reaches 100 times and above, and there is at least one voltage difference greater than or equal to 60mV in the time window, it is determined that there is a voltage difference fluctuation anomaly; such a voltage difference fluctuation anomaly reflects an imbalance in the internal voltage distribution of the battery pack, which may cause local overcharging or overdischarging, and requires close attention;

[0069] Step five: battery monomer internal resistance estimation, specifically including:

[0070] S51: data screening and calculation rules: select the adjacent two data whose running state changes from static state to charging state, and the time difference between the two data is less than or equal to 3 minutes;

[0071] Screen the charging data in this charging state, and the screening condition is: current ≥ 5A; SOC ≥ 20%; the running state of the next data is charging state;

[0072] Calculate the internal resistance of the battery monomer for the adjacent two data meeting the condition, and the calculation formula is as follows:

[0073]

[0074] Wherein, R i represents the estimated internal resistance of the battery monomer i, represents the voltage of the battery monomer i in the charging state, represents the voltage of the battery monomer i in the static state, and I represents the current in the charging state;

[0075] Step six: active safety warning, specifically including:

[0076] S61: according to the voltage distance anomaly, voltage difference fluctuation anomaly and voltage entropy anomaly, comprehensive judgment is made to make active safety warning, and the warning result is divided into two kinds: very serious and general serious; The warning result will also be compared and verified with the battery monomer internal resistance estimation value to improve the accuracy of the active safety warning;

[0077] S62: warning result, specifically including:

[0078] In the battery monomer voltage distance anomaly evaluation and calculation, if the cell is the cell with the largest voltage distance from the beginning to the end, and the average voltage distance of the cell is greater than or equal to 3, then the device is determined as a very serious device candidate;

[0079] If the device has abnormal fluctuation in the battery monomer voltage difference anomaly evaluation, it is directly determined that the device is a very serious device candidate;

[0080] If the device has abnormality in both the battery monomer voltage distance anomaly evaluation and the battery monomer voltage entropy anomaly evaluation, and the abnormal cells are the same, then the device is determined as a very serious device candidate;

[0081] Outputting the warning result: performing intersection analysis on all the particularly serious equipment candidates to obtain the final particularly serious equipment, and outputting the particularly serious equipment; outputting the remaining equipment in the abnormal equipment as the generally serious equipment.

[0082] The application can quickly identify various abnormal risks, improve the operation efficiency and safety and reliability of the battery management system, and solve the technical problems of low fault identification accuracy and identification efficiency in the prior art.

[0083] Part of the data in the above formula is calculated by removing the dimension, the formula is obtained by software simulation of a large amount of collected data closest to the real situation; the preset parameters and the preset threshold in the formula are set by the person skilled in the art according to the actual situation or obtained by a large amount of data simulation.

[0084] In the description of the present specification, the description referring to the terms "one embodiment", "example", "specific example" and the like means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are contained in at least one embodiment or example of the present application. In the present specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0085] The preferred embodiments of the application disclosed above are only used to help explain the application. The preferred embodiments do not describe all the details and limit the application to the specific embodiments. Obviously, many modifications and variations can be made according to the content of the present specification. The present specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the application, so that those skilled in the art can well understand and utilize the application. The application is limited only by the claims and their full scope and equivalents.

Claims

1. A lithium ion battery active safety early warning method based on end-cloud cooperation, characterized in that, Comprising the following steps: Step one: obtain real-time running data of lithium ion battery pack from the cloud; Step two: battery cell voltage distance abnormality evaluation, specifically including: According to the data screening and calculation rules, the z-score value of each cell is calculated; When the z-score distance value of the cell is observed for the first time > 3, it is determined that the cell is an abnormal cell, and the current time is recorded as the starting time of the abnormal cell; From the starting time, judge whether the subsequent maximum distance cell is in the recorded abnormal cell set, if the cell is in the recorded set and its distance value > 3, then the abnormal frequency of the cell + 1; Select the cell with abnormal frequency ≥100 times as the abnormal cell; Step three: battery cell voltage entropy abnormality evaluation, specifically including: According to the data screening and calculation rules, the entropy value of the minimum voltage cell in each time window is calculated; If the voltage entropy value of the cell in the 3-day time window is 0, and the frequency of the cell as the minimum voltage cell is ≥100 times, then the cell is determined to be abnormal; Step four: battery cell voltage difference abnormality evaluation, specifically including: According to the data screening and calculation rules, the voltage difference between the maximum voltage and the minimum voltage of the battery pack is calculated; Abnormal judgment is made on the voltage difference; Step five: battery cell internal resistance estimation; Step six: proactive safety warning: according to the voltage distance abnormality, voltage difference fluctuation abnormality, voltage entropy abnormality, comprehensive judgment is made to make proactive safety warning; The warning result is compared and verified with the battery cell internal resistance estimation value. 2.The lithium ion battery active safety early warning method based on end-cloud cooperation according to claim 1, characterized in that, The data screening and calculation rules in step two are as follows: From the charging state, select the charging data that meets the conditions, where the selection conditions include:

1. Current greater than 0; 2. Maximum voltage greater than or equal to 3.78V; For each charge data, calculate the average voltage of all cells and the voltage standard deviation s; then for each cell, calculate its z-score value, which is calculated as follows: where x i represents the voltage of the i-th battery cell, z i represents the z-score value of the i-th battery cell; the z-score reflects the deviation of each battery cell voltage from the average voltage, which is used to determine battery cell abnormalities. 3.The lithium ion battery active safety early warning method based on end-cloud collaboration of claim 1, wherein, The data screening and calculation rules in step three are as follows: From the charging state, select the charging data with current ≥3A to ensure that the battery pack has entered a stable charging condition; Use sliding window method, set time window to 3 days, calculate the entropy value of the minimum voltage cell in each time window, the calculation formula is as follows: where n represents a total of n cells, x i represents the voltage of the i-th cell, p(x i ) represents the data probability density of the i-th cell voltage, and H(x) represents the entropy value of the i-th cell. 4.The lithium ion battery active safety early warning method based on end-cloud cooperation of claim 1, wherein, The data screening and calculation rules in step four are as follows: From the charging state, select the charging data that meets the conditions, where the selection conditions are: the maximum voltage is between 3.78V and 3.82V; For each charging data that meets the conditions, calculate the voltage difference between the maximum voltage and the minimum voltage of the battery pack.

5. The active safety early warning method for lithium ion batteries based on end-cloud collaboration according to claim 4, characterized in that, Abnormal judgment is made on the voltage difference, specifically including: When the voltage difference is observed for the first time > 20mV, the observation time is recorded as the starting abnormal time; From the starting abnormal time, if the voltage difference at the subsequent time is still > 20mV, the abnormal frequency is added by 1; When the abnormal frequency reaches 100 times or more, and at least once in the time window, the voltage difference is greater than or equal to 60mV, it is determined to be voltage difference fluctuation abnormality. 6.The active safety early warning method for lithium ion battery based on end-cloud collaboration according to claim 1, characterized in that, Battery cell internal resistance estimation, specifically including: Select the adjacent two data of running state from static state to charging state, and the time difference of the two data is less than or equal to 3 minutes; Select the charging data in the charging state, the selection condition is: current ≥5A; SOC ≥20%; The running state of the next data is charging state; The internal resistance of the battery monomer is calculated for the two adjacent data meeting the condition, and the calculation formula is as follows: wherein R i represents an estimated internal resistance of the battery cell i, represents a voltage of the battery cell i in a charged state, represents a voltage of the battery cell i in a resting state, and I represents a current in a charged state. 7.The active safety early warning method for lithium ion battery based on end-cloud collaboration according to claim 1, characterized in that, The early warning result in step six specifically includes: In the battery monomer voltage distance abnormality evaluation calculation, if the battery cell is the battery cell with the largest voltage distance from the beginning to the end, and the average voltage distance of the battery cell is greater than or equal to 3, the device is determined as a particularly serious device candidate; If the device has abnormal fluctuations in the battery monomer voltage difference abnormality evaluation, the device is directly determined as a particularly serious device candidate; If the device has abnormalities in both the battery monomer voltage distance abnormality evaluation and the battery monomer voltage entropy abnormality evaluation, and the abnormal battery cells are the same, the device is determined as a particularly serious device candidate; Output the early warning result: obtain the final particularly serious device by performing intersection analysis on all particularly serious device candidates, and output the particularly serious device; output the remaining devices in the abnormal device as general serious devices. 8.The method of claim 1, wherein, The real-time running data includes running time, SOC, SOH, running state, total voltage of the battery pack, current, and cell voltage. The running state includes charging state, discharging state, and standing state.

Images