A method and device for cleaning automobile single battery undervoltage alarm data

Abstract

The present invention belongs to the field of automotive technology, and specifically relates to a method and device for cleaning undervoltage alarm data for automotive single-cell batteries. The present invention utilizes a single-cell undervoltage data cleaning control logic to clean alarm data uploaded by new energy vehicles in accordance with GB / T 32960.3-2016. The method selects the single-cell undervoltage alarm data and selects data related to single-cell undervoltage when thermal runaway occurs. This method is used to study changes in battery systems during thermal runaway events, improving work efficiency and data analysis accuracy.

Description

Technical Field

[0001] The present invention belongs to the field of automobile technology, and specifically provides a method and device for cleaning undervoltage alarm data of a single cell of a power battery of a new energy vehicle. Background Art

[0002] With the increasing adoption of new energy vehicles, OEMs and suppliers are faced with a significant amount of data, including vehicle alarms and battery status monitoring information. Battery undervoltage analysis is a key area of focus within the industry. As a crucial indicator for determining thermal runaway, it helps engineers accurately assess the battery's chemical state at the time of the event. However, current battery undervoltage alarms can be caused by numerous other factors, compromising the ability to determine thermal runaway conditions. Therefore, data cleansing and preprocessing can significantly improve efficiency and accuracy when analyzing data.

[0003] CN202010314133.1 relates to a battery analysis method and system based on electrochemical parameters. This method couples the collected battery macroscopic parameters with electrochemical microscopic parameters further calculated through a curve fitting algorithm, so that the characteristic parameters input into the trained machine learning model contain more effective information. This method also performs two-level data cleaning through edge computing and cloud computing, fundamentally improving the prediction accuracy of the machine learning model. Through the analysis of battery big data, fault warning and out-of-control alarms can be achieved based on outlier detection of real-time data. A more accurate prediction model can be obtained through training on massive historical data, enabling accurate estimation of the battery status.

[0004] Compared with the searched patents, the advantages of this invention are:

[0005] 1. Analyze single-cell undervoltage alarm data based on big data and clean invalid data;

[0006] 2. Invalid data is analyzed and classified.

[0007] CN202111442183.9 relates to a power battery safety risk assessment method based on real vehicle data, including the following: data preparation stage: obtaining operating data; data collation and feature extraction stage: preprocessing the operating data and extracting relevant features; algorithm model construction stage: coupling the extracted relevant feature data with the LSTM model and the equivalent circuit model through the Adaboost algorithm to construct an algorithm model; algorithm model prediction stage: training the constructed algorithm model on risk-free pure electric vehicle data, and using the trained model to predict the normal voltage of risky pure electric vehicles; and safety risk assessment stage. Utilizing big data technology, machine learning, and statistical learning algorithms, combined with an online parameter identification method of an equivalent circuit model, the method fully mines the hidden normal battery voltage information and its variation patterns from vehicle and battery data, thereby performing safety risk assessment based on the deviation between the normal voltage and the actual voltage.

[0008] Compared with the searched patents, the advantages of this invention are:

[0009] 1. Clean the undervoltage data of the monomers that have thermal runaway;

[0010] 2. Analyze and classify other non-thermal runaway undervoltage alarms

[0011] CN202010472793.2 provides an electric vehicle fire warning method and device. The method includes using vehicle data as input to various pre-trained prediction models to obtain the maximum temperature of a single cell, the minimum temperature of a single cell, and the battery pack current output by the prediction model; using preset fire determination rules, the maximum temperature of a single cell, the minimum temperature of a single cell, and the battery pack current are analyzed to determine whether the electric vehicle is about to catch fire. If so, a fire warning message is output. Because the prediction model predicts the temperature and current over a period of time when determining whether the electric vehicle is about to catch fire, the impending fire of the electric vehicle can be predicted long in advance, improving the safety of electric vehicle use.

[0012] Compared with the searched patents, the advantages of this invention are:

[0013] 1. Ability to clean initial data and narrow the scope of valid data;

[0014] 2. Analyze and classify other non-thermal runaway undervoltage alarms

[0015] In summary, the single-cell battery undervoltage warning methods disclosed in the above-mentioned patent documents do not have the characteristics of targeted single-cell battery thermal runaway data cleaning and invalid data analysis and classification, and some documents do not have big data analysis characteristics. Summary of the Invention

[0016] To solve the above problems, the present invention provides a method and device for cleaning undervoltage alarm data of a single cell power battery of a new energy vehicle. Based on the power battery upload data in GB / T 32960.3-2016, the method and device are used to clean the non-thermal runaway alarm data and false alarm data of the power battery for the battery cell undervoltage alarm through a data judgment logic and device.

[0017] A method for cleaning data of an automobile single battery undervoltage alarm is characterized in that the specific method of the data cleaning is as follows:

[0018] 1) Import single battery undervoltage alarm information;

[0019] 2) Determine whether the alarm persists: Based on the imported information, determine whether the alarm is triggered. If so, proceed to step 3); otherwise, proceed to step 13);

[0020] 3) Import alarm segment battery data;

[0021] 4) Determine whether the SOC is too low; if so, proceed to step 5); if not, proceed to step 8);

[0022] 5) Determine the consistency of the battery cells: Determine whether the voltage difference of the battery cells during the alarm period is ≥ △V, where △V is the alarm threshold for poor battery cell consistency; if so, proceed to step 6); if not, proceed to step 11);

[0023] 6) Determine whether the undervoltage and poor consistency after charging reappear. If so, proceed to step 7); if not, proceed to step 13);

[0024] 7) Record VIN code, battery balancing or poor consistency: output battery balancing performance or poor battery cell consistency, and record the VIN code of the faulty vehicle;

[0025] 8) Output the undervoltage single cell battery number;

[0026] 9) Determine the temperature of the undervoltage cell: Determine whether the temperature value measured by the temperature probe of the undervoltage cell is ≥ T, where T is the battery high temperature alarm threshold. If so, proceed to step 10); otherwise, proceed to step 11);

[0027] 10) Record battery high temperature and undervoltage abnormalities, and output the risk of thermal runaway of the battery;

[0028] 11) Determine whether the undervoltage is restored during the next charge. If so, proceed to step 13); if not, proceed to step 12);

[0029] 12) Record and output battery abnormality data;

[0030] 13) Alarm is invalid and data is flushed out: the alarm data is deemed invalid and the detected abnormal data is not counted in the data warehouse.

[0031] Furthermore, the importing of single cell undervoltage alarm information is to import the alarm information of single cell undervoltage alarm according to the alarm information of the national monitoring platform.

[0032] Furthermore, the determination of whether the alarm persists is based on the imported information to determine whether the alarm occurs 6 times or more in succession. If so, step 3) is executed; otherwise, step 13) is executed.

[0033] Furthermore, the importing of the alarm segment battery data is carried out in accordance with GB / T 32960.3-2016.

[0034] Furthermore, the determination of whether the SOC is too low is to determine whether the battery SOC is <X% during the alarm period; X% is the threshold value of battery undervoltage caused by too low SOC when consistency is considered; if so, execute step 5), if not, execute step 8).

[0035] Furthermore, the determination of whether undervoltage and poor consistency reappear after charging is to detect whether the undervoltage and poor consistency alarm reappears in the low SOC segment after the first charging to normal voltage after the alarm occurs. If so, execute step 7), if not, execute step 13).

[0036] Furthermore, the determination of whether the undervoltage is restored in the next charge is to detect whether the voltage of the undervoltage cell is restored after the first charge is completed after the alarm occurs. If so, step 13) is executed; if not, step 12) is executed.

[0037] A device for cleaning data of an automobile single battery undervoltage alarm includes the following modules:

[0038] Data acquisition module, used to collect car alarm information in the cloud;

[0039] Alarm detection module, used to detect the number of alarms;

[0040] Data import module, used to import collected data into the system;

[0041] Data analysis module, used to clean data;

[0042] Data warehouse, used to store cleaned data.

[0043] A device for cleaning undervoltage alarm data of a single automobile battery, characterized by:

[0044] 1) Collect cloud-based car alarm information through the data acquisition module;

[0045] 2) The alarm detection module detects the number of alarms;

[0046] 3) The data import module imports the collected data into the system;

[0047] 4) Clean the data in the data analysis module through a series of cleaning logic;

[0048] 5) Import the cleaned data into the corresponding data warehouse.

[0049] Furthermore, the collected data is GB / T 32960.3-2016 data.

[0050] Compared with the prior art, the present invention has the following beneficial technical effects:

[0051] The present invention can greatly improve work efficiency and increase the accuracy of data analysis through a certain degree of data cleaning preprocessing.

[0052] The present invention studies the changes in the battery system when thermal runaway occurs, thereby improving work efficiency and the accuracy of data analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0054] Figure 1 This is a logic diagram of a method for cleaning undervoltage alarm data of a single automobile battery according to the present invention;

[0055] Figure 2 The present invention is a schematic diagram of a device for cleaning undervoltage alarm data of a single automobile battery. DETAILED DESCRIPTION

[0056] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0057] It should be noted that similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are used only to distinguish the description and should not be understood as indicating or implying relative importance.

[0058] The present invention uses a single cell undervoltage data cleaning control logic to clean the alarm data uploaded by new energy vehicles in accordance with GB / T32960.3-2016, sorts the single cell undervoltage alarm data, and filters out the data related to single cell undervoltage when thermal runaway occurs. The present invention is used to study changes in the battery system when thermal runaway occurs, thereby improving work efficiency and the accuracy of data analysis.

[0059] A method for cleaning undervoltage alarm data of a single battery of a new energy vehicle comprises the following steps:

[0060] 1) Import single battery undervoltage alarm information;

[0061] 2) Determine whether the alarm persists: Based on the imported information, determine whether the alarm has occurred six times or more in a row. If so, proceed to step 3; otherwise, proceed to step 13.

[0062] 3) Import alarm segment battery data;

[0063] 4) Determine if the SOC is too low: Determine if the battery SOC is less than X% during the alarm period. X% is the threshold for battery undervoltage due to low SOC, considering consistency. If so, proceed to step 5; if not, proceed to step 8.

[0064] 5) Determine the consistency of the battery cells: Determine whether the voltage difference between the battery cells during the alarm period is ≥ △V, where △V is the alarm threshold for poor battery consistency. If so, proceed to step 6; if not, proceed to step 11.

[0065] 6) Determine whether undervoltage and poor consistency occur again after charging: After the first charge after the alarm occurs and the voltage is normal, check whether the undervoltage and poor consistency alarm occurs again in the low SOC section. If so, proceed to step 7; if not, proceed to step 13;

[0066] 7) Record VIN code, battery balancing or poor consistency: output battery balancing performance or poor battery cell consistency, record the faulty vehicle VIN code to facilitate maintenance reminder;

[0067] 8) Output undervoltage unit number;

[0068] 9) Determine the temperature of the undervoltage cell: Determine whether the temperature value measured by the temperature probe of the undervoltage cell is ≥ T, where T is the battery high temperature alarm threshold. If so, proceed to step 10; otherwise, proceed to step 11;

[0069] 10) Record battery high temperature and undervoltage abnormalities, and output thermal runaway risks;

[0070] 11) Determine whether the undervoltage is restored during the next charge: After the first charge is completed after the alarm occurs, check whether the undervoltage cell voltage is restored. If so, proceed to step 13; if not, proceed to step 12;

[0071] 12) Record and output abnormal data;

[0072] 13) Alarm is invalid and data is flushed out: the alarm data is deemed invalid and the detected abnormal data is not counted in the data warehouse.

[0073] See Figure 2 , a vehicle single battery undervoltage alarm data cleaning device, including the following modules:

[0074] Data acquisition module, used to collect car alarm information in the cloud;

[0075] Alarm detection module, used to detect the number of alarms;

[0076] Data import module, used to import collected data into the system;

[0077] Data analysis module, used to clean data;

[0078] Data warehouse, used to store cleaned data.

[0079] A device for cleaning undervoltage alarm data of a single battery of a new energy vehicle, comprising:

[0080] 1) Collect cloud-based car alarm information through the data acquisition module;

[0081] 2) The alarm detection module detects the number of alarms;

[0082] 3) The data import module imports the collected GB / T 32960.3-2016 data into the system;

[0083] 4) Clean the data in the data analysis module through a series of cleaning logic;

[0084] 5) Import the cleaned data into the corresponding data warehouse.

[0085] A specific method for cleaning single cell undervoltage alarm data is as follows:

[0086] See Figure 1 ,1) Import single cell undervoltage alarm information: According to the alarm information of the national monitoring platform, the alarm information of single cell undervoltage alarm will be imported;

[0087] 2) Determine whether the alarm persists: Based on the imported information, determine whether the alarm has occurred six times or more in a row. If so, proceed to step 3; otherwise, proceed to step 13.

[0088] 3) Import alarm segment battery data: According to GB / T 32960.3-2016, import alarm segment battery data;

[0089] 4) Determine if the SOC is too low: Determine if the battery SOC is less than X% during the alarm period. X% is the threshold for battery undervoltage due to low SOC, considering consistency. If so, proceed to step 5; if not, proceed to step 8.

[0090] 5) Determine the consistency of the battery cells: Determine whether the voltage difference between the battery cells during the alarm period is ≥ △V, where △V is the alarm threshold for poor battery consistency. If so, proceed to step 6; if not, proceed to step 11.

[0091] 6) Determine whether undervoltage and poor consistency occur again after charging: After the first charge after the alarm occurs, check whether the undervoltage and poor consistency alarm occur again in the low SOC section after the voltage returns to normal. That is, after charging until the voltage returns to normal, discharge to the low SOC section to see whether the undervoltage and poor consistency alarm occur again. If so, execute step 7; if not, execute step 13;

[0092] 7) Record VIN code, battery balancing or poor consistency: output battery balancing performance or poor battery cell consistency, record the faulty vehicle VIN code to facilitate maintenance reminder;

[0093] 8) Output undervoltage single cell position number: Output undervoltage single cell position number;

[0094] 9) Determine the temperature of the undervoltage cell: Determine whether the temperature value measured by the temperature probe of the undervoltage cell is ≥ T, where T is the battery high temperature alarm threshold. If so, proceed to step 10; otherwise, proceed to step 11;

[0095] 10) Record battery high temperature and undervoltage anomalies, and output thermal runaway risks: Record battery high temperature and undervoltage anomalies, and output battery thermal runaway risks;

[0096] 11) Determine whether the undervoltage is restored during the next charge: After the first charge is completed after the alarm occurs, check whether the undervoltage cell voltage is restored. If so, proceed to step 13; if not, proceed to step 12;

[0097] 12) Record and output abnormal data: record and output battery abnormal data;

[0098] 13) Alarm is invalid and data is flushed out: the alarm data is deemed invalid and the detected abnormal data is not counted in the data warehouse.

[0099] Although the embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and example embodiments. They can be applied to a variety of fields suitable for the present invention. Further modifications will be readily apparent to those skilled in the art. Therefore, the present invention is not limited to the specific details and illustrations shown and described herein without departing from the general concept defined by the claims and their equivalents.

[0100] The foregoing description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions, and improvements made by any person skilled in the art within the technical scope disclosed herein and within the spirit and principles of the present invention shall be covered by the scope of protection of the present invention. Furthermore, any matters not described in detail in this specification constitute prior art known to those skilled in the art.

Claims

1. A method for cleaning automobile single battery undervoltage alarm data, characterized in that: The specific method of data cleaning is as follows: 1) Import single battery undervoltage alarm information; 2) Determine whether the alarm persists: Based on the imported information, determine whether the alarm is triggered. If so, proceed to step 3); otherwise, proceed to step 13); 3) Import alarm segment battery data; 4) Determine whether the SOC is too low; if so, proceed to step 5); if not, proceed to step 8); 5) Determine the consistency of the battery cells: Determine whether the voltage difference of the battery cells during the alarm period is ≥ △V, where △V is the alarm threshold for poor battery cell consistency; if so, proceed to step 6); if not, proceed to step 11); 6) Determine whether the undervoltage and poor consistency after charging reappear. If so, proceed to step 7); if not, proceed to step 13); 7) Record VIN code, battery balancing or poor consistency: output battery balancing performance or poor battery cell consistency, and record the VIN code of the faulty vehicle; 8) Output the undervoltage single cell number; 9) Determine the temperature of the undervoltage cell: Determine whether the temperature value measured by the temperature probe of the undervoltage cell is ≥ T, where T is the battery high temperature alarm threshold. If so, proceed to step 10); otherwise, proceed to step 11); 10) Record battery high temperature and undervoltage abnormalities, and output the risk of thermal runaway of the battery; 11) Determine whether the undervoltage is restored during the next charge. If so, proceed to step 13); if not, proceed to step 12); 12) Record and output battery abnormality data; 13) Alarm invalid, data flushed: The alarm data is deemed invalid, and the detected abnormal data is not counted in the data warehouse; The determination of whether the alarm is continuous is based on the imported information to determine whether the alarm occurs 6 times or more in succession. If so, step 3) is executed; otherwise, step 13) is executed.

2. The method for cleaning automobile single battery undervoltage alarm data according to claim 1, characterized in that: The importing of single cell undervoltage alarm information is to import the alarm information of single cell undervoltage alarm according to the alarm information of the national monitoring platform.

3. The method for cleaning automobile single battery undervoltage alarm data according to claim 1, characterized in that: The importing of the alarm segment battery data is carried out according to GB / T 32960.3-2016.

4. The method for cleaning automobile single battery undervoltage alarm data according to claim 1, characterized in that: The determination of whether the SOC is too low is to determine whether the battery SOC is less than X% during the alarm period; X% is the threshold value of battery undervoltage caused by too low SOC when consistency is considered; if so, step 5) is executed, if not, step 8) is executed.

5. The method for cleaning automobile single battery undervoltage alarm data according to claim 1, characterized in that: The determination of whether undervoltage and poor consistency reappear after charging is to detect whether the undervoltage and poor consistency alarm reappears in the low SOC section after the first charging to normal voltage after the alarm occurs. If so, execute step 7), otherwise execute step 13).

6. The method for cleaning automobile single battery undervoltage alarm data according to claim 1, characterized in that: The judgment of whether the undervoltage is restored in the next charge is to detect whether the voltage of the undervoltage cell is restored after the first charge is completed after the alarm occurs. If so, step 13) is executed, and if not, step 12) is executed.

Images