An automated FOTA testing method and system for a truck lithium battery system

Abstract

The present invention discloses an automated FOTA testing method and system for a truck lithium battery system, which relates to the technical field of battery management. The method includes collecting BMS data of the to-be-tested truck lithium battery system to a test host through a CAN bus; establishing a test sample library based on the BMS data, and segmenting the test sample library according to the battery working state; extracting features from each data segment in the test sample library to obtain a feature vector group; inputting the feature vector group into a preset FOTA upgrade evaluation model to output a FOTA upgrade recommended time window; performing firmware upgrade of the battery management system within the FOTA upgrade recommended time window, and simultaneously recording the BMS data during the upgrade process as new test samples. By collecting, segmenting, and extracting features from the BMS data of the truck lithium battery system in real time, and combining with the FOTA upgrade evaluation model trained with historical upgrade data, the present invention can accurately predict a suitable upgrade time window.

Description

Technical Field

[0001] The present invention relates to the technical field of battery management, and particularly to an automated FOTA testing method and system for a lithium battery system of a truck. Background Art

[0002] With the popularization of electric vehicles, especially heavy vehicles, the application of lithium battery systems in trucks has gradually become the mainstream. To ensure the reliability and performance of the battery system, the battery management system (BMS), as the core unit for controlling and monitoring the state of the battery pack, has been widely studied and applied. The BMS system can monitor parameters such as the voltage, temperature, and SOC (State of Charge) of the battery to ensure that the battery operates within an appropriate temperature and voltage range. However, with the long-term use of the vehicle, the battery performance gradually decays, and a non-linear cumulative effect of battery parameter changes may occur under different working conditions, thereby affecting the working efficiency and safety of the system. For this reason, Firmware Over-The-Air (FOTA) has gradually been applied to the BMS system to remotely and real-time update the firmware, improve the adaptability of the battery system in different scenarios, and optimize its performance. However, traditional FOTA methods have problems such as inappropriate upgrade time and scenario selection, often unable to adapt to the complex truck usage environment, and prone to upgrade failures or battery anomalies.

[0003] Existing automated FOTA testing methods for lithium battery systems have some technical bottlenecks in practical applications. First, traditional methods rely on a preset static test process, lacking automated segmented processing and feature extraction of dynamic battery data under different working conditions, and it is difficult to efficiently and accurately capture key state information of the BMS. Second, the existing system's selection of the upgrade time window relies on empirical judgment or simple algorithms, failing to effectively combine key parameters such as the voltage fluctuation of the battery pack, temperature uniformity, and SOC decay rate, resulting in difficulties in ensuring the FOTA upgrade efficiency and reliability. In addition, the evaluation models of some systems fail to fully utilize real-time BMS data to construct a sample library, lacking sufficient data support, making the accuracy and adaptability of the upgrade evaluation model limited. Therefore, how to use real-time battery data to construct a sample library, segmentally extract features, and achieve intelligent time window selection has become an urgent problem in this field. Summary of the Invention

[0004] In view of the problems existing in the above-mentioned automated FOTA testing method for lithium battery systems, the present invention is proposed.

[0005] Therefore, the present invention provides an automated FOTA testing method for a lithium battery system of a truck, which can solve the problems mentioned in the background art.

[0006] To solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides an automated FOTA testing method for a truck lithium battery system, which includes:

[0008] Collect the BMS data of the truck lithium battery system to be tested to the test host through the CAN bus;

[0009] Based on the BMS data, establish a test sample library, and segment the test sample library according to the battery working state;

[0010] Extract features from each data segment in the test sample library to obtain a feature vector group;

[0011] Input the feature vector group into a preset FOTA upgrade evaluation model to output a FOTA upgrade recommended time window;

[0012] Execute the firmware upgrade of the battery management system within the FOTA upgrade recommended time window, and record the BMS data during the upgrade as a new test sample.

[0013] As a preferred solution of the automated FOTA testing method for the truck lithium battery system of the present invention, wherein: the segmenting the test sample library according to the battery working state includes:

[0014] Segment the BMS data according to the charge and discharge state. When the charge and discharge current switches from zero to a positive value, it is marked as the starting point of the charging segment. When the charge and discharge current changes from a positive value to zero, it is marked as the ending point of the charging segment. When the charge and discharge current switches from zero to a negative value, it is marked as the starting point of the discharging segment. When the charge and discharge current changes from a negative value to zero, it is marked as the ending point of the discharging segment. The interval where the current remains zero and the duration exceeds a preset value is marked as the static segment;

[0015] In the charging process data segment, record the charging rate, single charging duration, charging cut-off voltage, and ambient temperature parameters, and establish sub-data segments for fast charging mode and slow charging mode respectively;

[0016] In the discharging process data segment, record the discharging rate, single discharging duration, discharging cut-off voltage, and power demand parameters, and divide the data into high-power segments, medium-power segments, and low-power segments according to the discharging power;

[0017] In the static process data segment, record the static duration, leakage current value, and temperature change rate parameters, and divide it into short-term static segments and long-term static segments according to the static duration.

[0018] As a preferred solution of the automatic FOTA test method for the truck lithium battery system described in the present invention, wherein: the feature extraction includes voltage volatility, temperature uniformity, SOC decay rate, and comprehensive evaluation;

[0019] Construct the feature vector group according to the obtained four eigenvalue; the feature vector group includes the voltage volatility as the first-dimensional feature of the feature vector group, the temperature uniformity as the second-dimensional feature of the feature vector group, the SOC decay rate as the third-dimensional feature of the feature vector group, and the comprehensive evaluation as the fourth-dimensional feature of the feature vector group.

[0020] As a preferred solution of the automatic FOTA test method for the truck lithium battery system described in the present invention, wherein: the calculation of the voltage volatility is shown in the following formula:

[0021]

[0022] wherein, R v (t) is the voltage volatility after temperature compensation, α is the temperature sensitivity coefficient, T(t) is the current temperature, T ref is the reference temperature, ΔV(t) is the voltage change at time t, and Δt is the sampling time interval;

[0023] The calculation of the temperature uniformity is shown in the following formula:

[0024]

[0025] wherein, U(T) is the temperature uniformity index considering voltage change, T i is the temperature of the i-th region, T avg is the average temperature, β is the voltage influence factor, is the voltage change rate, n is the total number of temperature detection points, T max is the allowable maximum operating temperature;

[0026] The calculation of the SOC decay rate is shown in the following formula:

[0027]

[0028] wherein, R SOC (t) is the SOC decay rate, is the original value of the SOC change rate, and f(T, V) is the temperature-voltage coupling influence function, as shown in the following formula:

[0029]

[0030] wherein, k1 and k2 are coupling coefficients, and T0 and V0 are standard operating points.

[0031] As a preferred solution of the automated FOTA testing method for the truck lithium battery system according to the present invention, wherein: the comprehensive evaluation is obtained by weighting the voltage volatility, the temperature uniformity, and the SOC decay rate;

[0032] The comprehensive evaluation can be expressed by the following formula:

[0033] EI = w1×R v (t) + w2×U(T) + w3×R SOC (t)

[0034] Wherein, EI is the comprehensive evaluation index, and w1, w2, and w3 are dynamic weight coefficients, which are updated by an adaptive algorithm:

[0035]

[0036] Wherein, γ is the learning rate, EI real is the actual observed value, EI predict is the predicted value, w i (t + 1) is the weight value at the next moment, and w i (t) is the weight value at the current moment.

[0037] As a preferred solution of the automated FOTA testing method for the truck lithium battery system according to the present invention, wherein: the output FOTA upgrade recommendation time window includes:

[0038] Processing the feature vector group;

[0039] Inputting the processed feature vector group into the FOTA upgrade evaluation model to obtain a preliminary time window;

[0040] Screening the preliminary time window to obtain a recommended time window.

[0041] As a preferred solution of the automated FOTA testing method for the truck lithium battery system according to the present invention, wherein: processing the feature vector group is shown by the following formula:

[0042]

[0043] Wherein, A(t) is the attention weight matrix, Q(t), V(t) are the query, key, and value matrices, H(t - 1) is the historical state information, and γ1 is the historical information weight coefficient.

[0044] Second, in order to further solve the safety problems existing in the automated FOTA testing method for the truck lithium battery system, the embodiment of the present invention provides an artificial intelligence-based vehicle fault diagnosis and maintenance evaluation system, which includes:

[0045] A data acquisition module, configured to collect BMS data of a lithium battery system of a truck to be tested to a test host through a CAN bus;

[0046] A segmentation processing module, configured to establish a test sample library based on the BMS data, and segment the test sample library according to the battery operating state;

[0047] A feature extraction module, configured to extract features from each data segment in the test sample library to obtain a feature vector group;

[0048] A time window determination module, configured to input the feature vector group into a preset FOTA upgrade evaluation model, and output a FOTA upgrade recommended time window;

[0049] An upgrade module, configured to execute firmware upgrade of the battery management system within the FOTA upgrade recommended time window, and record the BMS data during the upgrade process as a new test sample.

[0050] In a third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program is executed by the processor, any step of the automated FOTA test method for a truck lithium battery system as described in the first aspect of the present invention is implemented.

[0051] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program is executed by the processor, any step of the automated FOTA test method for a truck lithium battery system as described in the first aspect of the present invention is implemented.

[0052] The beneficial effects of the present invention are as follows: by collecting, segmenting and extracting features from the BMS data of the truck lithium battery system in real time, a feature vector group covering battery voltage fluctuation, temperature uniformity and SOC decay rate is established, and combined with the FOTA upgrade evaluation model trained with historical upgrade data, it can accurately predict the appropriate upgrade time window. This design has multiple beneficial effects: First, based on multi-dimensional feature extraction and temporal attention mechanism, it can dynamically identify the battery state change and accurately match the best upgrade timing, avoiding upgrade failure caused by unstable battery operating state; Second, the automated data verification and dynamic threshold adjustment mechanism improve the adaptability of the evaluation model and ensure the high success rate and safety of FOTA upgrade; Finally, this solution synchronously records the newly generated BMS data during the upgrade, providing reliable data support for subsequent tests, and realizing the automation and intelligence of the test process. Description of the Drawings

[0053] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings. Among them:

[0054] Figure 1 It is a flowchart of an automated FOTA test method for a truck lithium battery system.

[0055] Figure 2 It is a partial screenshot of the test report for the automated FOTA test method of the truck lithium battery system. Detailed implementation manners

[0056] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will describe the detailed implementation manners of the present invention in conjunction with the drawings of the specification.

[0057] In the following description, many specific details are set forth to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0058] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that can be included in at least one implementation manner of the present invention. The "in one embodiment" that appears in different places in this specification does not all refer to the same embodiment, nor is it a separate or alternative embodiment that excludes other embodiments.

[0059] Embodiment 1

[0060] Refer to Figure 1 , which is the first embodiment of the present invention. This embodiment provides an automated FOTA test method for a truck lithium battery system, including the following steps:

[0061] S1: Collect the BMS data of the to-be-tested truck lithium battery system to the test host through the CAN bus. The BMS data includes battery pack voltage, battery pack temperature, SOC value, and charge and discharge status;

[0062] Collect the BMS data of the to-be-tested truck lithium battery system to the test host through the CAN bus. The BMS data includes battery pack voltage, battery pack temperature, SOC value, and charge and discharge status, which specifically includes the following content:

[0063] During the CAN bus acquisition process, the CAN communication baud rate is set to 500 kbps, standard frame format is used for data transmission, and the CAN message identifier uses the extended frame ID. In the BMS data, the battery pack voltage acquisition frequency is 100 ms / time, the acquisition accuracy is 0.1 V, and the acquisition range is 0 - 850 V; the battery pack temperature acquisition frequency is 500 ms / time, the acquisition accuracy is 0.1 °C, and the acquisition range is -40 °C to 85 °C; the SOC value is obtained by the coulomb counting method, the calculation frequency is 1 s / time, the value range is 0 - 100%, and the accuracy is 0.1%; the charge and discharge status includes three status identifiers: charging, discharging, and standing still; the test host is connected to the CAN bus through a USB-CAN converter. The transmit buffer of the converter is configured with 64 groups of messages, the receive buffer is configured with 128 groups of messages, and it has an automatic retransmission function;

[0064] During the data acquisition process, the test host performs CRC verification on the received CAN messages. When the verification fails, it automatically requests retransmission. If the verification fails continuously for 3 times, an error log is recorded;

[0065] S2: Based on the BMS data, establish a test sample library, and divide the test sample library into a charging process data segment, a discharging process data segment, and a standing still process data segment according to the battery working status;

[0066] The test sample library is stored using a distributed database structure. The main data table records the timestamp, data segment type identifier, and associated ID, and the sub-data tables store the specific parameter values of the BMS data respectively;

[0067] The BMS data is segmented according to the charge and discharge status. When the charge and discharge current switches from zero to a positive value, it is marked as the starting point of the charging segment. When the charge and discharge current changes from a positive value to zero, it is marked as the end point of the charging segment; when the charge and discharge current switches from zero to a negative value, it is marked as the starting point of the discharging segment. When the charge and discharge current changes from a negative value to zero, it is marked as the end point of the discharging segment; the interval where the current remains zero and the duration exceeds 300 s is marked as the standing still segment;

[0068] In the charging process data segment, parameters such as the charging rate, single charging duration, charging cut-off voltage, and ambient temperature are recorded, and sub-data segments are established for fast charging mode (rate ≥ 1C) and slow charging mode (rate < 1C) respectively;

[0069] In the discharging process data segment, parameters such as the discharging rate, single discharging duration, discharging cut-off voltage, and power demand are recorded, and the data is divided into a high-power segment (≥ 80% rated power), a medium-power segment (30% - 80% rated power), and a low-power segment (< 30% rated power) according to the discharging power;

[0070] In the static process data segment, parameters such as static time, leakage current value, temperature change rate, etc. are recorded, and the static process data are divided into short-term static segment (<24h) and long-term static segment (≥24h) according to the static time.

[0071] During the data segment division process, a 50ms state switching judgment delay is set to avoid misjudgment caused by transient fluctuations. At the same time, additional transition interval data of 10s before and after the start and end of each data segment is recorded;

[0072] Perform data integrity check on each divided data segment, including data continuity, rationality of parameter value range and logical correctness of state switching. Unqualified data segments are marked as invalid and the reasons are recorded.

[0073] S3: extracting features from each data segment in the test sample library to obtain a feature vector group including voltage fluctuation rate, temperature uniformity and SOC decay rate;

[0074] Feature extraction is performed on each data segment in the test sample library to obtain a feature vector group including voltage fluctuation rate, temperature uniformity and SOC decay rate, which specifically includes the following contents:

[0075] First, in the process of extracting the voltage fluctuation rate feature, based on the accumulation of the absolute value of the voltage change in the standard time window, the product of the temperature difference and the temperature sensitivity coefficient is used as a correction term to establish a voltage fluctuation rate index after temperature compensation. This index increases the fluctuation rate value when the actual battery temperature is higher than the reference temperature, and decreases it when it is lower, thereby more accurately reflecting the impact of temperature on voltage fluctuation. It can be expressed as follows:

[0076]

[0077] Among them, R v (t) is the voltage fluctuation rate after temperature compensation, α is the temperature sensitivity coefficient, T(t) is the current temperature, T ref is the reference temperature, ΔV(t) is the voltage change at time t, and Δt is the sampling time interval.

[0078] Secondly, for the temperature uniformity characteristics, an improved uniformity calculation method is used. The sum of the squares of the deviations between each temperature detection point and the average temperature is used as the basis. The value is divided by the product of the number of detection points and the square of the highest temperature, and then corrected by the exponential decay term of the voltage change rate. Finally, the uniformity index is obtained by subtracting the corrected result from 1. In this way, the impact of voltage change on temperature distribution is quantified through the exponential decay term. When the voltage change rate is large, the temperature uniformity evaluation value decreases accordingly. The obtained temperature characteristics can be expressed as follows:

[0079]

[0080] Among them, U(T) is the temperature uniformity index considering voltage variation, T i is the temperature of the i-th region, T avg is the average temperature, β is the voltage influence factor, is the voltage change rate, n is the total number of temperature detection points, T max is the maximum allowable operating temperature.

[0081] Extract the SOC decay rate characteristics and establish a temperature-voltage two-factor influence model. Based on the change rate of SOC over time, it is corrected through the temperature-voltage coupling influence function. This coupling function consists of the square term of temperature deviation and the proportional term of voltage deviation, and is adjusted by the coupling coefficient respectively to realize the joint influence evaluation of temperature and voltage on the SOC decay rate. It can be expressed by the following formula:

[0082]

[0083] Among them, R SOC (t) is the SOC decay rate, is the original value of the SOC change rate, f(T, V) is the temperature-voltage coupling influence function, as shown in the following formula:

[0084]

[0085] Among them, k1 and k2 are the coupling coefficients, and T0 and V0 are the standard operating points.

[0086] Finally, the above three characteristics are combined in a weighted manner to form a comprehensive evaluation index. The weight coefficients adopt an adaptive update mechanism. According to the deviation between the predicted value and the actual observed value, the weight size is adjusted through the learning rate. When the predicted value is lower than the actual value, the weight of the corresponding characteristic is increased, and vice versa, so that the evaluation system can be adaptively optimized. The comprehensive evaluation index can be expressed by the following formula:

[0087] EI = w1 × R v (t) + w2 × U(T) + w3 × R SOC (t)

[0088] Among them, EI is the comprehensive evaluation index, w1, w2, and w3 are the dynamic weight coefficients, which are updated through an adaptive algorithm:

[0089]

[0090] Among them, γ is the learning rate, EI real is the actual observed value, EI predict is the predicted value.

[0091] Construct an n-dimensional basic feature vector F based on the obtained four eigenvalues, including the voltage volatility R v (t) after temperature compensation as the first-dimensional eigenvalue, the improved temperature uniformity index U(T) as the second-dimensional eigenvalue, and the SOC decay rate R SOC (t) of temperature-voltage coupling as the third-dimensional eigenvalue, and the comprehensive evaluation index EI as the fourth-dimensional eigenvalue.

[0092] S4: Input the feature vector group into a preset FOTA upgrade evaluation model, which is trained based on historical upgrade data, and output the FOTA upgrade recommended time window;

[0093] In the FOTA upgrade evaluation model, first use an improved temporal attention mechanism to process the feature vector group. This mechanism generates an attention weight matrix by calculating the mutual relationship between the query matrix, key matrix, and value matrix and combining historical state information. Specifically, the product of the query matrix and the key matrix is subjected to scale normalization processing, and then multiplied by the value matrix weighted and corrected by historical information to capture the time-dependent relationship of the feature sequence. This mechanism can effectively identify key temporal patterns in historical data and improve the model's perception ability of time correlation. The temporal attention mechanism is shown in the following formula:

[0094]

[0095] where, A(t) is the attention weight matrix, Q(t), V(t) are the query, key, and value matrices, H(t - 1) is the historical state information, and γ1 is the historical information weight coefficient.

[0096] It should be noted that the core structure of the evaluation model includes an input layer that receives the feature vector group, performs data standardization and temporal alignment, uses a multi-layer perceptron to extract feature combination relationships, an implicit layer uses the ELU activation function, and an output layer gives the time window prediction value and confidence score, and a residual connection structure is introduced to alleviate the problem of difficult training of deep neural networks.

[0097] Furthermore, further screen the time window output by the evaluation model:

[0098] During the screening process of the FOTA upgrade time window, the limiting requirements of multiple key indicators need to be met simultaneously:

[0099] The time period with the comprehensive evaluation index EI higher than the preset benchmark threshold will be selected. Meanwhile, within this time period, it is required that the fluctuation range of the battery voltage remains in a relatively stable state, that is, the fluctuation does not exceed 0.5 volts per minute. The temperature distribution uniformity index of the battery pack needs to be maintained above 0.85 to ensure a stable temperature field distribution. The state of charge SOC of the battery needs to be maintained within a moderate range of 30% to 80% to avoid safety risks caused by overcharging and over-discharging. In addition, the model trained based on historical data also needs to predict that the upgrade success probability within this time window is not less than 95%. Only the time periods that meet all these conditions will be determined by the system as suitable upgrade time windows.

[0100] Furthermore, by comparing the difference between the actual upgrade success rate and the target success rate, the scoring threshold is dynamically adjusted. When the actual success rate is lower than the target value, the threshold standard is increased through the adjustment coefficient; otherwise, the threshold requirement is appropriately reduced. This mechanism enables the evaluation criteria to be automatically optimized according to the actual operation effect, improving the adaptability of the system. The specific adjustment is shown in the following formula:

[0101] S th (t + 1) = S th (t) × [1 + η(P success -P target )]

[0102] where η is the adjustment coefficient, P success is the actual success rate, and P target is the target success rate.

[0103] Through the above evaluation model and time window selection mechanism, the optimal FOTA upgrade timing can be determined intelligently.

[0104] S5: Execute the firmware upgrade of the battery management system within the FOTA upgrade recommended time window, and record the BMS data during the upgrade process as new test samples.

[0105] This embodiment also provides an automated FOTA test system for a truck lithium battery system, including:

[0106] A data acquisition module for collecting the BMS data of the truck lithium battery system to be tested to the test host through the CAN bus;

[0107] A segmented processing module for establishing a test sample library based on the BMS data and segmenting the test sample library according to the battery working state;

[0108] A feature extraction module for extracting features from each data segment in the test sample library to obtain a feature vector group;

[0109] A time window determination module, configured to input the feature vector group into a preset FOTA upgrade evaluation model and output a FOTA upgrade recommended time window;

[0110] An upgrade module, configured to perform firmware upgrade of the battery management system within the FOTA upgrade recommended time window, and record the BMS data during the upgrade process as new test samples.

[0111] This embodiment also provides a computer device, applicable to the case of the automated FOTA test method for a truck lithium battery system, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the automated FOTA test method for a truck lithium battery system as proposed in the above embodiment.

[0112] This computer device may be a terminal. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner. The wireless manner can be implemented through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device may be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0113] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the method for realizing the automated FOTA test for the truck lithium battery system as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, abbreviated as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, abbreviated as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, abbreviated as EPROM), programmable read-only memory (Programmable Red-Only Memory, abbreviated as PROM), read-only memory (Read-Only Memory, abbreviated as ROM), magnetic memory, flash memory, magnetic disk or optical disc.

[0114] Embodiment 2

[0115] This is the second embodiment of the present invention. To further verify the advancement of the present invention, experimental simulations / contrast data with the prior art for the method of realizing the automated FOTA test for the truck lithium battery system are given.

[0116] In this test, 10 heavy truck lithium battery systems operating under different working conditions are selected as the test objects. The rated capacity of the battery system is 280 kWh, and the rated voltage is 750 V. Before the test, the test equipment is calibrated, including the HIOKI 3390 power analyzer (accuracy class 0.1), the Fluke Ti480 PRO infrared thermal imager (temperature resolution 0.05 °C), and the NI USB-8473s high-performance USB-CAN interface (supporting a maximum communication rate of 1 Mbps). The test environment temperature is controlled at 25 ± 2 °C, and the relative humidity is 45 ± 5%.

[0117] In the data acquisition stage, the BMS data is collected in real time through the configured CAN bus network (baud rate 500 kbps). To ensure the data quality, a dual-channel redundant acquisition system is deployed on each test vehicle. The main channel is responsible for regular data acquisition, and the backup channel is used for data backup and cross-verification. Through the independently developed data acquisition software, high-precision acquisition of key parameters such as the voltage, temperature, and SOC of the battery pack is realized. Among them, the voltage data uses a sampling period of 100 ms, the temperature data uses a sampling period of 500 ms, and the SOC calculation frequency is 1 s / time.

[0118] During the testing process, each vehicle underwent a complete charge-discharge cycle and a static process. Through an intelligent scheduling algorithm, the test vehicles were operated under different working conditions, including fast charging mode (2C charging rate), standard charging mode (0.5C charging rate), high-power discharge condition (peak power 220kW), medium-power cruising condition (80kW continuous output), and low-power idle condition (<20kW). At the same time, static tests with different durations were set up, including four static conditions of 4h, 12h, 24h, and 48h.

[0119] To avoid interference and anomalies during the data acquisition process, a multiple data verification mechanism was implemented: First, CRC verification at the hardware level to ensure the integrity of CAN communication data; second, data validity verification at the software level, including numerical range checks, rate-of-change limitations, and logical relationship verification; finally, manual inspections to regularly check and maintain the operating status of the acquisition system.

[0120] In the feature extraction section, an improved sliding time window algorithm was used to process the original data. The window length was dynamically adjusted according to different parameter characteristics: a 1-minute window for voltage volatility, a 5-minute window for temperature uniformity, and a 15-minute window for SOC decay rate. Adaptive filtering algorithms were used to eliminate data noise and improve the accuracy of feature extraction. To improve the calculation efficiency, a parallel computing architecture was implemented, and the feature extraction tasks were assigned to multiple computing nodes for simultaneous processing.

[0121] The following are the relevant experimental data obtained, as shown in Table 1 below and Figure 2 shown:

[0122] Table 1 Experimental Data Table

[0123]

[0124] Through the statistical analysis of the test data, it can be concluded that:

[0125] The data shows that the test vehicles (TV-001, 002, 004, 005, 007, 009, 010) with a voltage volatility lower than 0.4V / min all achieved successful upgrades, and the upgrade success rate reached 100%. In contrast, among the test vehicles (TV-003, 006, 008) with a voltage volatility exceeding 0.4V / min, 2 had upgrade failures. This indicates that the setting of the voltage volatility threshold proposed in the present invention has significant predictive value.

[0126] The temperature uniformity index of the successfully upgraded vehicles is generally higher than 0.88, while that of the failure cases is lower than 0.85. In particular, the temperature uniformity of TV-009 reached 0.93, and its upgrade process was the smoothest, taking only 17.6 minutes. This verifies the good practicality of the setting of the temperature uniformity threshold of 0.85.

[0127] Data shows that there is an obvious positive correlation between the SOC decay rate and the upgrade time consumption. For the test vehicles (TV-001, 004, 005, 007, 009, 010) with a decay rate lower than 0.5% / h, the average upgrade time consumption is 18.5 minutes, while for the vehicles with a decay rate exceeding 0.5% / h, the average time consumption increases to 24.5 minutes. This finding provides an important basis for optimizing the selection of the upgrade time window.

[0128] Through analysis, it can be found that there is a significant linear correlation between the comprehensive evaluation index (EI) and the upgrade success probability (correlation coefficient R 2 = 0.94). When EI ≥ 0.85, the actual upgrade success rate reaches 100%; while in the cases where EI < 0.85, the upgrade failure rate increases significantly. This verifies that the comprehensive evaluation model proposed by the present invention has accurate prediction ability.

[0129] Under the condition of meeting all threshold requirements, the average time consumption of successful upgrade cases is 19.0 minutes, and the standard deviation is only 0.6 minutes, showing good time consistency. Compared with the traditional method which generally requires 30 - 40 minutes for upgrade time, the present invention has achieved a significant efficiency improvement.

[0130] From the above analysis, it can be seen that the FOTA upgrade evaluation system proposed by the present invention can accurately predict the upgrade success probability and effectively improve the upgrade efficiency. Especially under the synergistic effect of the three key features of voltage volatility, temperature uniformity, and SOC decay rate, the system demonstrates excellent prediction accuracy and practical value. The experimental data fully verifies the innovation and practical application effect of the present invention.

[0131] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. An automated FOTA test method for a truck lithium battery system, characterized by: include: Collect the BMS data of the lithium battery system of the truck to be tested to the test host through the CAN bus; Establish a test sample library based on the BMS data, and process the test sample library in sections according to the battery working status; Extracting features from each data segment in the test sample library to obtain a feature vector group; Inputting the feature vector group into a preset FOTA upgrade evaluation model, and outputting a FOTA upgrade recommended time window; Performing a battery management system firmware upgrade within the FOTA upgrade recommended time window, and recording the BMS data during the upgrade process as a new test sample; The feature extraction includes voltage fluctuation rate, temperature uniformity, SOC decay rate and comprehensive evaluation; Constructing the eigenvector group according to the obtained four eigenvalues; The feature vector group includes the voltage fluctuation rate as a first dimension feature of the feature vector group, the temperature uniformity as a second dimension feature of the feature vector group, the SOC decay rate as a third dimension feature of the feature vector group, and the comprehensive evaluation as a fourth dimension feature of the feature vector group.

2. The automated FOTA test method for a truck lithium battery system according to claim 1, characterized in that: The segmenting of the test sample library according to the battery working status comprises: The BMS data is processed in sections according to the charge and discharge status. When the charge and discharge current switches from zero to a positive value, it is marked as the starting point of the charge section. When the charge and discharge current changes from a positive value to zero, it is marked as the end point of the charge section. When the charge and discharge current switches from zero to a negative value, it is marked as the starting point of the discharge section. When the charge and discharge current changes from a negative value to zero, it is marked as the end point of the discharge section. The interval in which the current is continuously zero and the duration exceeds a preset value is marked as a static section. In the charging process data segment, the charging rate, single charging time, charging cut-off voltage and ambient temperature parameters are recorded, and sub-data segments are established for the fast charging mode and the slow charging mode respectively; In the discharge process data segment, the discharge rate, single discharge duration, discharge cut-off voltage and power requirement parameters are recorded, and the data is divided into a high power segment, a medium power segment (and a low power segment) according to the discharge power; In the static process data segment, the static time, leakage current value and temperature change rate parameters are recorded, and the static process data is divided into a short-term static segment and a long-term static segment according to the static time.

3. The automated FOTA test method for a truck lithium battery system according to claim 2, characterized in that: The voltage fluctuation rate is calculated as follows: Among them, R v (t) is the voltage fluctuation rate after temperature compensation, α is the temperature sensitivity coefficient, T(t) is the current temperature, T ref is the reference temperature, ΔV(t) is the voltage change at time t, and Δt is the sampling time interval; The temperature uniformity is calculated as follows: Among them, U(T) is the temperature uniformity index considering voltage changes, T i is the temperature of the ith region, T avg is the average temperature, β is the voltage influence factor, is the voltage change rate, n is the total number of temperature detection points, T max is the maximum allowable operating temperature; The calculation of the SOC decay rate is shown in the following formula: Among them, R SOC (t) is the SOC decay rate, is the original value of the SOC change rate, and f(T,V) is the temperature-voltage coupling influence function, as shown in the following formula: Among them, k1 and k2 are coupling coefficients, and T0 and V0 are standard working points.

4. The automated FOTA test method for a truck lithium battery system according to claim 3, characterized in that: The comprehensive evaluation is obtained by weighting the voltage fluctuation rate, the temperature uniformity and the SOC decay rate; The comprehensive evaluation can be expressed as follows: EI=w1×R v (t)+w2×U(T)+w3×R SOC (t) Among them, EI is a comprehensive evaluation index, w1, w2, and w3 are dynamic weight coefficients, which are updated through an adaptive algorithm: Among them, γ is the learning rate, EI real is the actual observed value, EI predict is the predicted value, w i (t+1) is the weight value at the next moment, w i (t) is the weight value at the current moment.

5. The automated FOTA test method for a truck lithium battery system according to claim 4, characterized in that: The output FOTA upgrade recommended time window includes: Processing the feature vector group; Inputting the processed feature vector group into the FOTA upgrade assessment model to obtain a preliminary time window; The preliminary time window is screened to obtain a recommended time window.

6. The automated FOTA test method for a truck lithium battery system according to claim 5, characterized in that: The feature vector group is processed as shown below: Among them, A(t) is the attention weight matrix, Q(t), V(t) is the query, key, and value matrix, H(t-1) is the historical state information, and γ1 is the weight coefficient of the historical information.

7. An automated FOTA test system for a truck lithium battery system, based on the automated FOTA test method for a truck lithium battery system according to any one of claims 1 to 6, characterized in that: include: Data acquisition module, used to collect BMS data of the lithium battery system of the truck to be tested to the test host through the CAN bus; A segment processing module, used to establish a test sample library based on the BMS data, and segment the test sample library according to the battery working status; A feature extraction module, used to extract features from each data segment in the test sample library to obtain a feature vector group; A time window determination module, used for inputting the feature vector group into a preset FOTA upgrade evaluation model, and outputting a FOTA upgrade recommended time window; The upgrade module is used to perform a battery management system firmware upgrade within the FOTA upgrade recommended time window, and record the BMS data during the upgrade process as a new test sample.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the automated FOTA testing method for a truck lithium battery system according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the automated FOTA testing method for a truck lithium battery system according to any one of claims 1 to 6 are implemented.

Images