Power battery offline detection method and system and portable detection equipment
By setting dynamic thresholds based on predictive models and using AI model judgment, combined with portable testing equipment, the problems of high false detection rate and high cost in off-line testing of power batteries are solved, achieving an efficient, flexible, and low-cost testing solution.
Patent Information
- Application Number
- CN202511058657.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-30
- Publication Date
- 2025-09-09
AI Technical Summary
Existing power battery offline detection technologies have high false detection rates and high costs, mainly due to their reliance on real-time analysis of local data with fixed thresholds and large fixed equipment.
A dynamic threshold setting method based on a predictive model is adopted. By calculating the variance and average value of the predicted value and the measured value of the test item, the judgment criteria are dynamically adjusted. The test items are judged in combination with the AI model, and full coverage of high and low voltage test items is achieved through portable detection equipment.
Significantly reduce the false detection rate, improve detection accuracy and flexibility, reduce equipment costs, and achieve quality management upgrade from post-judgment to pre-prevention. It is suitable for the flexible production of multiple models and batches of power batteries.
Smart Images
Figure CN120610171A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of power battery off-line detection, and in particular to a power battery off-line detection method, system and portable detection equipment. Background Art
[0002] With the rapid development of industries like new energy vehicles and energy storage systems, quality requirements for power batteries are increasing. As a core component of power batteries, the BMS (Battery Management System) is responsible for real-time monitoring of battery status and executing charging and discharging strategies. Its functional integrity is directly related to reliable operation throughout the battery's lifecycle. Therefore, battery packs typically undergo EOL (End-of-Line) testing after production and assembly and before vehicle installation to verify the safety, performance consistency, and regulatory compliance of the power batteries, ensuring that the produced battery packs meet quality standards and safety requirements.
[0003] Current detection technologies are mostly limited to real-time analysis of local data and use fixed thresholds for qualification determination, resulting in a high false detection rate. Summary of the Invention
[0004] The present application provides a power battery offline detection method, system and portable detection equipment, which can solve the technical problem of high false detection rate in current battery offline detection technology.
[0005] To achieve the above objectives, in a first aspect, the present application provides a method for detecting a power battery offline, the method comprising: Based on the prediction model, a certain detection item of multiple power batteries is predicted and the predicted values are obtained respectively.
[0006] The variance is calculated based on the error between the predicted and measured values of the test item for all power batteries. The average of the predicted values for the test item for all power batteries is calculated. Based on the variance, the average, and the determination type of the test item, a corresponding determination criterion is set, which is a preset threshold or threshold range.
[0007] According to the corresponding judgment standard, the measured value of the test item of each power battery is judged to determine whether the test item is qualified.
[0008] Furthermore, in one embodiment, the detection items include battery cell internal resistance, battery cell voltage, SOC, battery temperature, and detection equipment parameters.
[0009] Furthermore, in one embodiment, historical detection data of different detection items are used as training samples for the prediction model; for each detection item, its historical detection data includes: the measured values of other detection items related to the detection item, and the measured value of the detection item.
[0010] Furthermore, in one embodiment, the prediction of a certain detection item of multiple power batteries based on the prediction model to obtain prediction values respectively includes: The predicted value of each power battery is obtained in the same manner. For each test item, the measured values of other test items related to the test item are obtained.
[0011] The measured value is input into the prediction model, and the predicted value of the detection item is output.
[0012] Furthermore, in one embodiment, setting a corresponding determination criterion based on the variance, the average value, and the determination type of the detection item includes: When the determination type of the detection item is unidirectional determination, the determination criterion is a threshold value, and the threshold value is the sum of the average value and the variance.
[0013] When the determination type of the detection item is bidirectional determination, the determination criterion is a threshold range, the minimum value of the threshold range is the difference between the average value and the variance, and the maximum value is the sum of the average value and the variance.
[0014] Furthermore, in one embodiment, judging the measured value of the test item of each power battery according to the corresponding judgment standard to determine whether the test item is qualified includes: When the test item is the internal resistance of the battery cell, if the actual measured value is greater than its threshold, it is judged as unqualified; if it is less than or equal to its threshold, it is judged as qualified.
[0015] When the test items are cell voltage, SOC, battery temperature, or test equipment parameters, if the actual measured value is within the threshold range, it is judged as qualified; if the actual measured value is outside the threshold range, it is judged as unqualified.
[0016] Furthermore, in one embodiment, in the test data of different batches of power batteries of the same model, when the fluctuation amplitude of the actual measured value of a certain test item is greater than a preset amplitude but does not exceed its threshold or the limit of its threshold range, the item to be tested is marked as a potential fault item.
[0017] Furthermore, in one embodiment, when a certain test item fails to meet the standards multiple times in the test results of different batches of power batteries of the same model, the test priority of the item to be tested is increased.
[0018] In a second aspect, based on the above-mentioned power battery offline detection method, the present application provides a power battery offline detection system, the system comprising: The prediction module is used to predict a certain detection item of multiple power batteries based on the prediction model and obtain predicted values respectively.
[0019] The calculation module is used to calculate the variance based on the error between the predicted value and the measured value of the detection item of all power batteries, and is also used to calculate the average value of the predicted value of the detection item of all power batteries.
[0020] The threshold module is used to set a corresponding determination standard based on the variance and average value calculated by the calculation module and the determination type of the detection item. The determination standard is a preset threshold or threshold range.
[0021] The detection module is used to judge the actual measured value of the detection item of each power battery according to the corresponding judgment standard to determine whether the detection item is qualified.
[0022] In a third aspect, the present application provides a portable detection device for detecting the measured value of the above-mentioned item to be detected, the device comprising: The low voltage interface is used to receive low voltage detection item information among the power battery items to be detected.
[0023] The main control unit provides resistance to each pin of the low-voltage interface through the resistance output unit, outputs voltage to power the power battery through the voltage output unit, and controls the on and off of the high-voltage interface through the high-voltage on-off controller.
[0024] The wireless communication module is used to upload the detection data of the detection items to the cloud.
[0025] The CAN / LIN communication module is used to forward the message information of the external device to the low-voltage interface and the main control unit through the communication interface, and is also used to forward the instructions of the main control unit to the external device.
[0026] The high-voltage interface is used to receive high-voltage detection item information among the power battery items to be detected.
[0027] The beneficial effects of the technical solutions provided in the embodiments of the present application include: This application is based on a prediction model, predicting a certain test item of multiple power batteries, obtaining predicted values respectively, and then calculating the variance based on the error between the predicted value and the measured value of the test item of all power batteries. According to the variance, the average value of the predicted value of the test item of all power batteries, and the judgment type of the test item, a corresponding judgment standard is set, and the judgment standard is a threshold value or a threshold range. Finally, according to the corresponding judgment standard, the measured value of the test item of each power battery is judged to determine whether the test item is qualified. By predicting the value of the item to be detected in real time through the prediction model, and dynamically adjusting the threshold of the item to be detected according to the predicted value and the measured value of the item to be detected, it can effectively reduce the false detection rate caused by using a fixed threshold for detection, and improve the accuracy of the end-of-line detection of power batteries; secondly, the technical solution of this application can be flexibly deployed at different stations of the end-of-line detection of the power battery production line, which can reduce equipment costs compared to traditional EOL detection that relies on fixed large-scale detection equipment. BRIEF DESCRIPTION OF THE DRAWINGS
[0028] Figure 1 This is a flow chart of the power battery off-line detection method according to an embodiment of the present application.
[0029] Figure 2 This is a block diagram of a power battery offline detection system according to an embodiment of the present application.
[0030] Figure 3 This is a schematic diagram of the power battery off-line detection equipment according to an embodiment of the present application. DETAILED DESCRIPTION
[0031] In order to enable those skilled in the art to better understand the present invention, 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. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0032] First, some technical terms in this application are explained to facilitate those skilled in the art to understand this application.
[0033] (1) BMS: Battery Management System, which is responsible for monitoring and managing the working status, performance and safety of rechargeable batteries.
[0034] (2) EOL testing: End-of-Line testing refers to the final quality verification of power batteries after production and assembly and before shipment.
[0035] (3) CAN: Controller Area Network, a common automotive communication protocol used for data exchange between ECUs (Vehicle Control Units).
[0036] (4) LIN: Local Interconnect Network, a low-cost, low-speed serial communication protocol, commonly used for communication between on-board sensors and actuators.
[0037] (5) DCR: Direct Current Resistance, DC resistance test, used to measure the DC resistance inside the battery to evaluate its health status.
[0038] (6) AI: Artificial Intelligence, a discipline that studies how to make computers simulate human intelligent behavior. It covers multiple fields, including but not limited to machine learning, natural language processing, computer vision, robotics, etc. In this invention, it is mainly used for data analysis, anomaly detection, fault prediction and intelligent optimization.
[0039] (7) Random Forest Model: Random Forest is an ensemble learning model based on decision trees, which is widely used in classification and regression tasks. It improves the accuracy and stability of the model by constructing multiple decision trees and combining them.
[0040] (8) Long Short-Term Memory (LSTM): This is a special recurrent neural network (RNN) architecture designed to address the vanishing and exploding gradient problems that traditional RNNs often encounter when processing long sequences of data. LSTM effectively captures long-term dependencies in sequence data by introducing a gating mechanism to control the flow of information.
[0041] A BMS typically consists of both hardware and software, and its functions cover all aspects of battery management. Specifically, the main functions and components of a BMS include battery status monitoring, charge and discharge control, temperature control, fault diagnosis, and status estimation.
[0042] After the battery pack is assembled and before the vehicle is assembled, each power battery pack will undergo EOL testing. The entire process is as follows.
[0043] First, once the low-voltage wiring harness is connected, the BMS immediately powers on and establishes communication with the host computer. The host computer can then read back all the basic data collected by the BMS, including the programmed program version, pressure sensor values, the voltage and differential pressure of each battery cell string, and the temperature and differential pressure of each battery cell string.
[0044] After confirming the basic data is correct, the inspector triggers specific BMS functions by performing actions such as plugging, unplugging, shorting, or pressing the battery, and the host computer then reads the feedback. Specific tasks include: high-voltage interlock circuit continuity testing, CC2 fast-charge communication handshake testing, fast-charge temperature sampling path testing, wake-up pin level change detection, relay on / off function testing, and short-circuit detection logic verification, without the need for additional large-scale equipment.
[0045] Finally, with the help of external high-precision instruments such as insulation testers, voltage withstand testers, and equipotential testers, insulation withstand voltage tests, equipotential tests, voltage error tests, DCR tests, and current tests are performed.
[0046] However, current power battery end-of-line detection technologies are mostly limited to real-time analysis of local data and use fixed thresholds for qualification determination, resulting in a high false positive rate. Furthermore, most detection technologies typically use large, fixed detection equipment, which is bulky and expensive. Therefore, this application provides a power battery end-of-line detection method, system, and portable detection device that can address the high false positive rate and high cost issues currently encountered in power battery end-of-line detection technologies.
[0047] In order to make the objectives, technical solutions and advantages of this application clearer, the implementation methods of this application will be further described in detail below with reference to the accompanying drawings.
[0048] In a first aspect, an embodiment of the present application provides a method for detecting an off-line power battery.
[0049] In one embodiment, see Figure 1 As shown, the above detection method includes: S1. Based on the prediction model, predict a certain detection item of multiple power batteries and obtain predicted values respectively.
[0050] In this embodiment, multiple power batteries are of the same model.
[0051] S2. Calculate the variance based on the error between the predicted value and the measured value of the test item for all power batteries; calculate the average value of the predicted value of the test item for all power batteries; and set a corresponding determination criterion based on the variance, the average value, and the determination type of the test item, where the determination criterion is a preset threshold or threshold range.
[0052] S3. According to the corresponding judgment standard, the measured value of the test item of each power battery is judged to determine whether the test item is qualified.
[0053] In this embodiment, the predicted value of a certain detection item of multiple power batteries of the same model is obtained through the prediction model, and then the variance is calculated based on the error between the predicted value and the measured value of the detection item of all power batteries, and the average value of the predicted value of the detection item of all power batteries is calculated. The judgment standard is automatically generated according to the judgment type, including a threshold or a threshold range. The setting method of the judgment standard can adjust the threshold or the threshold range in real time according to the drift of the detection item-related parameters of the same model of power batteries in different batches and different environments. This effectively improves the accuracy of the power battery off-line detection while reducing the risk of false detection caused by the fixed threshold. It does not require the large fixed equipment required for traditional EOL detection, has strong flexible deployment capabilities, and can reduce equipment costs.
[0054] In this application, the off-line testing of power batteries includes low-voltage testing and high-voltage testing. Low-voltage testing is performed when the high-voltage system of the battery pack is not activated and only requires contact with the low-voltage terminals (such as the BMS communication interface or low-voltage sampling line). High-voltage testing involves the actual high-voltage output of the battery pack and requires closing the high-voltage relay, which poses a risk of electric shock. Therefore, the testing items for power batteries include low-voltage testing and high-voltage testing. Low-voltage testing items include cell internal resistance, cell voltage, SOC, battery temperature, and testing equipment parameters. High-voltage testing items include insulation, withstand voltage, total battery pack voltage, and total battery pack current.
[0055] Furthermore, in one embodiment, before the above step S1, the prediction model may be trained, specifically: Historical test data for different test items is collected as training samples for the prediction model, and preprocessed by outlier removal, normalization, and other operations. For each test item, its historical test data includes the measured values of other test items related to it, as well as the measured value of the test item itself.
[0056] The prediction model is trained using the historical detection data of all detection items. After training, the prediction model can predict the measured values of the corresponding detection items based on the measured values of other detection items related to each detection item.
[0057] In this embodiment, the prediction model is an AI model, which can be a random forest model or a long short-term memory network. By using the AI model to obtain the predicted value of the test item, the accuracy and robustness of the prediction can be significantly improved, thereby dynamically optimizing the judgment criteria, reducing the fixed threshold error caused by environmental or batch differences, further reducing the false detection rate, and improving the intelligence level and reliability of power battery end-of-line testing.
[0058] Furthermore, in one embodiment, in the above step S1, a certain detection item of multiple power batteries is predicted based on the prediction model to obtain prediction values respectively. The specific steps are as follows: For each detection item, the measured values of other detection items related to the detection item are obtained, the measured values are input into the prediction model, and the predicted value of the detection item is output.
[0059] In this embodiment, the predicted values of each detection item of each power battery are obtained in the same manner as above.
[0060] Furthermore, in one embodiment, in the above step S2, the method for setting the judgment criteria for each detection item is as follows: Calculate the error between the predicted value and the measured value of a certain test item of all power batteries, and calculate the variance based on the error.
[0061] Calculate the average value of the predicted values of this test item for all power batteries.
[0062] According to the above variance, average value, and the judgment type of the detection item, set the corresponding judgment criteria, specifically: When the determination type of the detection item is unidirectional determination, the determination criterion is a threshold, and the threshold of the detection item is the sum of the average value of the detection item and the variance of the detection item.
[0063] When the determination type of the detection item is bidirectional determination, the determination criterion is a threshold range, the minimum value of the threshold range of the detection item is the difference between the average value of the detection item and the variance of the detection item, and the maximum value is the sum of the average value of the detection item and the variance of the detection item.
[0064] In this embodiment, by introducing a dynamic threshold setting mechanism based on variance and predicted average, adaptive adjustment of the detection item judgment criteria is achieved, which can respond in real time to data drift under different batches, environments or working conditions, effectively avoiding misjudgment caused by fixed thresholds, thereby improving the accuracy and robustness of power battery off-line detection.
[0065] Furthermore, in one embodiment, in the above step S3, the measured value of the test item of each power battery is judged according to the corresponding judgment standard to determine whether the test item is qualified. The specific steps are: When the test item is the internal resistance of the battery cell, the judgment standard is the threshold value. If the actual measured value is greater than the threshold value, it is judged as unqualified; if it is less than or equal to the threshold value, it is judged as qualified.
[0066] When the test item is cell voltage, SOC, battery temperature, or test equipment parameters, the judgment standard is the threshold range. If the actual measured value is within the threshold range, it is judged as qualified; if the actual measured value is outside the threshold range, it is judged as unqualified.
[0067] Furthermore, in one embodiment, in the test data of different batches of power batteries of the same model, when the fluctuation amplitude of the measured value of a certain test item is greater than a preset amplitude but does not exceed its threshold or the limit of its threshold range, the item to be tested is marked as a potential fault item and the engineer is reminded to check.
[0068] The power battery offline detection method proposed in this application can use AI models to identify potential failure trends that have not yet triggered a failure judgment but have already shown abnormal data fluctuations, and provide early warning of possible quality risks, realizing the transition from "post-judgment" to "pre-prevention", thereby improving the initiative and foresight of the overall quality control of power batteries and reducing maintenance costs.
[0069] Furthermore, in one embodiment, when a certain test item fails to meet the standards multiple times in the test results of different batches of power batteries of the same model, the test priority of the item to be tested is increased.
[0070] In this embodiment, the off-line inspection results of different batches of the same model of power batteries are collected, and the inspection results of different batches are correlated and analyzed in the cloud. Based on the results of the analysis, the inspection order of the off-line inspection is dynamically adjusted. If the inspection results of a certain model of power battery are repeatedly unqualified in a certain inspection item, the priority of the inspection item is adjusted in advance to reduce repeated inspections and improve efficiency.
[0071] In the second aspect, based on the above embodiment of the power battery offline detection method, an embodiment of a power battery offline detection system is provided. Figure 2 As shown, the above system includes a prediction module, a calculation module, a threshold module, and a detection module. Specifically: The prediction module is used to predict a certain detection item of multiple power batteries based on the prediction model and obtain predicted values respectively.
[0072] The calculation module is used to calculate the variance based on the error between the predicted value and the measured value of the detection item of all power batteries, and is also used to calculate the average value of the predicted value of the detection item of all power batteries.
[0073] The threshold module is used to set a corresponding determination standard based on the variance and average value calculated by the calculation module and the determination type of the detection item. The determination standard is a preset threshold or threshold range.
[0074] The detection module is used to judge the actual measured value of the detection item of each power battery according to the corresponding judgment standard to determine whether the detection item is qualified.
[0075] In a third aspect, based on the above embodiment of the power battery offline detection method, an embodiment of a power battery offline detection device is provided. Figure 3 As shown, the above equipment includes a low-voltage interface, a main control unit, a wireless communication module, a CAN / LIN communication module, and a high-voltage interface. Specifically: The low-voltage interface is used to receive low-voltage detection item information among the items to be detected of the power battery.
[0076] The main control unit provides resistance to each pin of the low-voltage interface through the resistance output unit, outputs voltage to the power battery through the voltage output unit, and controls the on / off of the high-voltage interface through the high-voltage on / off controller. The high-voltage on / off controller can detect high-voltage interlock signals and coordinate detection with external high-voltage equipment.
[0077] The wireless communication module is used to upload the detection data of the detection items to the cloud.
[0078] The CAN / LIN communication module is used to forward the message information of the external device to the low-voltage interface and the main control unit through the communication interface, and is also used to forward the instructions of the main control unit to the external device.
[0079] The high-voltage interface is used to receive high-voltage detection item information among the power battery items to be detected.
[0080] In this embodiment, by integrating the low-voltage interface, high-voltage interface, main control unit, wireless communication module and CAN / LIN communication module into one, full coverage and integrated detection of high and low voltage detection items of the power battery are achieved, and multi-functional collaborative operations such as high-voltage interlocking, insulation withstand voltage, and battery parameter acquisition are supported; at the same time, with the help of the wireless communication module, the detection data is uploaded to the cloud in real time, and combined with the cloud AI model to perform dynamic threshold analysis and potential fault warning, which significantly improves the detection efficiency, data traceability and system intelligence level, meets the requirements of flexible production lines for miniaturization, modularization and remote control of equipment, and effectively reduces equipment investment and operation and maintenance costs.
[0081] Furthermore, in one embodiment, the low-voltage interface is also used to receive the potential information of each pin from the power battery BMS, to monitor the status of the power battery, to ensure that the battery operates within a safe and effective range, and to prevent overcharging, over-discharging, overheating, etc.; it is also used to receive CAN / LIN message information from the power battery BMS, to transmit the low-voltage detection item information of the power battery in real time.
[0082] This application realizes the full-process intelligent quality inspection of high and low voltage detection items of power batteries by constructing a power battery offline detection method, system and portable detection equipment based on AI prediction model.
[0083] Specifically, at the method level, random forest or LSTM models are used to predict key test items such as battery cell internal resistance, battery cell voltage, SOC, battery temperature, and insulation withstand voltage. The error variance and average value of the predicted value and the measured value are combined to dynamically generate a judgment threshold or threshold range to replace the traditional fixed threshold, significantly reducing the false detection rate caused by batch differences and environmental drift; at the same time, a potential fault marking mechanism is introduced to issue early warnings for test items that are not out of limit but have abnormal fluctuations, realizing the upgrade of quality management from "result judgment" to "trend prevention".
[0084] At the system level, the prediction module, calculation module, threshold module and detection module work together to support cloud data interaction and multi-batch correlation analysis. They can automatically increase the detection priority of high-frequency unqualified items, optimize the detection sequence, and improve production line efficiency.
[0085] At the equipment level, the portable detection terminal integrates low-voltage interface, high-voltage interface, main control unit, wireless communication module and CAN / LIN communication module. It supports seamless switching of high and low voltage detection items, and has functions such as high-voltage interlocking, insulation withstand voltage, electrical parameter acquisition, and BMS communication message analysis. The detection data can be uploaded to the cloud in real time to achieve remote monitoring and quality traceability.
[0086] The entire technical solution does not rely on traditional large-scale EOL testing equipment. It has the advantages of miniaturization, modularity, low cost and easy deployment. It is particularly suitable for flexible production scenarios of multiple models and multiple batches of power batteries, and significantly improves the intelligence level, detection accuracy and economic benefits of power battery end-of-line testing.
[0087] It should be noted that the serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.
[0088] The terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices. The terms "first", "second" and "third" are used to distinguish different objects, etc., and do not represent a sequence, nor do they limit the "first", "second" and "third" to different types.
[0089] In the description of the embodiments of this application, the words "exemplary," "for example," or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary," "for example," or "for example" in the embodiments of this application should not be construed as being preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary," "for example," or "for example" is intended to present the relevant concepts in a concrete manner.
[0090] In the description of the embodiments of the present application, unless otherwise specified, “ / ” means or, for example, A / B can mean A or B; “and / or” in the text is merely a description of the association relationship of associated objects, indicating that three relationships may exist, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of the present application, “multiple” refers to two or more than two.
[0091] In some processes described in the embodiments of the present application, multiple operations or steps are included that appear in a specific order. However, it should be understood that these operations or steps may not be performed in the order in which they appear in the embodiments of the present application or may be performed in parallel. The sequence numbers of the operations are only used to distinguish between different operations, and the sequence numbers themselves do not represent any order of execution. In addition, these processes may include more or fewer operations, and these operations or steps may be performed in sequence or in parallel, and these operations or steps may be combined.
[0092] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, or the part that contributes to the existing technology, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above and includes a number of instructions for enabling a terminal device to execute the methods described in each embodiment of this application.
[0093] The above are only preferred embodiments of the present application and do not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.
Claims
1. A power battery offline detection method, characterized in that: The method comprises: Based on the prediction model, a certain test item of multiple power batteries is predicted and the predicted value is obtained respectively; Calculating a variance based on the error between the predicted value and the measured value of the test item for all power batteries; calculating an average value of the predicted values of the test item for all power batteries; and setting a corresponding determination criterion based on the variance, the average value, and the determination type of the test item, where the determination criterion is a preset threshold or threshold range; According to the corresponding judgment standard, the measured value of the test item of each power battery is judged to determine whether the test item is qualified.
2. The power battery off-line detection method according to claim 1, characterized in that: The test items include battery cell internal resistance, battery cell voltage, SOC, battery temperature, and test equipment parameters.
3. The power battery off-line detection method according to claim 2, characterized in that: Use historical test data of different test items as training samples for the prediction model; For each detection item, its historical detection data includes: the measured values of other detection items related to the detection item, and the measured value of the detection item.
4. The power battery offline detection method according to claim 2, characterized in that: The method of predicting a certain detection item of multiple power batteries based on the prediction model to obtain prediction values respectively includes: The same method is used to obtain the predicted value of each power battery. For each test item, obtain the measured values of other test items related to the test item; The measured value is input into the prediction model, and the predicted value of the detection item is output.
5. The power battery off-line detection method according to claim 1, characterized in that: The setting of corresponding determination criteria according to the variance, the average value, and the determination type of the detection item includes: When the determination type of the detection item is one-way determination, the determination criterion is a threshold value, which is the sum of the average value and the variance; When the determination type of the detection item is bidirectional determination, the determination criterion is a threshold range, the minimum value of the threshold range is the difference between the average value and the variance, and the maximum value is the sum of the average value and the variance.
6. The power battery offline detection method according to claim 1, characterized in that: The step of judging the measured value of the test item of each power battery according to the corresponding judgment standard to determine whether the test item is qualified includes: When the test item is the internal resistance of the battery cell, if the actual measured value is greater than the threshold, it is judged as unqualified; if it is less than or equal to the threshold, it is judged as qualified; When the test items are cell voltage, SOC, battery temperature, or test equipment parameters, if the actual measured value is within the threshold range, it is judged as qualified; if the actual measured value is outside the threshold range, it is judged as unqualified.
7. The power battery off-line detection method according to claim 1, characterized in that: In the test data of different batches of the same model of power batteries, when the fluctuation range of the measured value of a certain test item is greater than the preset range but does not exceed its threshold or the limit of its threshold range, the item to be tested is marked as a potential fault item.
8. The power battery offline detection method according to claim 1, characterized in that: When a certain test item fails to meet the standards multiple times in the test results of different batches of the same model of power batteries, the test priority of the item to be tested is increased.
9. A power battery offline detection system based on the power battery offline detection method according to any one of claims 1 to 8, characterized in that: The system comprises: A prediction module, which is used to predict a certain detection item of multiple power batteries based on the prediction model and obtain predicted values respectively; a calculation module, configured to calculate a variance based on an error between a predicted value and a measured value of the test item for all power batteries, and further configured to calculate an average value of the predicted values of the test item for all power batteries; A threshold module, which is used to set a corresponding judgment standard based on the variance and average value calculated by the calculation module and the judgment type of the detection item, and the judgment standard is a preset threshold or threshold range; The detection module is used to judge the actual measured value of the detection item of each power battery according to the corresponding judgment standard to determine whether the detection item is qualified.
10. A portable detection device for detecting the measured value of the item to be detected according to any one of claims 1 to 8, characterized in that: The device comprises: A low-voltage interface is used to receive low-voltage detection information from power battery items to be detected; The main control unit provides resistance to each pin of the low-voltage interface through the resistance output unit, outputs voltage to the power battery through the voltage output unit, and controls the on and off of the high-voltage interface through the high-voltage on-off controller; A wireless communication module, which is used to upload the test data of the test items to the cloud; CAN / LIN communication module, which is used to forward the message information of the external device to the low-voltage interface and the main control unit through the communication interface, and is also used to forward the instructions of the main control unit to the external device; The high-voltage interface is used to receive high-voltage detection item information among the power battery items to be detected.