Portable new energy automobile battery thermal runaway early warning detection device and method

Abstract

The invention relates to the technical field of new energy automobile battery thermal runaway early warning detection, in particular to a portable new energy automobile battery thermal runaway early warning detection device and method, and the device comprises a simulation charging pile low-voltage auxiliary power supply and a wake-up vehicle BMS; receiving and binding related data; bMS battery data is collected, data preprocessing and feature extraction are carried out, and a plurality of numerical features are obtained; carrying out parallel calculation on the plurality of numerical features to obtain multi-dimensional independent abnormal scores, and fusing the independent dimensional abnormal scores into a thermal runaway early warning confidence value; according to the thermal runaway early warning method, a multi-dimensional abnormal scoring system is adopted, comprehensive evaluation is carried out in combination with three kinds of characteristics of voltage, temperature and faults, the accuracy and reliability of thermal runaway early warning are improved, and the reliability of thermal runaway early warning is improved through a trend persistence strengthening factor and collaborative coefficient mechanism. Early warning can be given out at the early stage of risk accumulation, early intervention is achieved, and accident expansion is avoided.

Description

Technical Field

[0001] This invention relates to a portable early warning and detection device and method for thermal runaway of new energy vehicle batteries, belonging to the field of early warning and detection of thermal runaway of new energy vehicle batteries. Background Technology

[0002] With the rapid development of the new energy vehicle industry, the number of pure electric vehicles continues to rise. Power batteries are the core power components of new energy vehicles. Thermal runaway of power batteries is one of the most serious safety hazards of new energy vehicles. Its essence is that a violent thermal reaction occurs inside the battery, which triggers a chain reaction of rapid temperature rise, smoke, fire and even explosion. Moreover, this process is characterized by suddenness, rapid spread and high degree of harm.

[0003] Of particular concern is the significantly increased probability and difficulty of preventing battery thermal runaway in new energy vehicles during transportation scenarios (including but not limited to passenger roll-on / roll-off ships, ocean-going roll-on / roll-off ships, and land transportation) and parking scenarios (including but not limited to roadsides, roadside parking lots, centralized parking lots in buildings, and private car parking spaces). During transportation, new energy vehicles are densely packed, ventilation is limited, and environmental temperature fluctuates greatly (such as high temperature and humidity in shipping and exposure to direct sunlight or low temperature in land transportation). Furthermore, there is a lack of real-time and effective monitoring methods throughout the entire process. Once a battery in one vehicle experiences thermal runaway, it can easily spread rapidly to surrounding vehicles, causing large-scale safety accidents. In parking scenarios, vehicles are stationary, and in some cases, there is a lack of dedicated personnel (such as roadside parking spaces and private car parking spaces). Existing monitoring systems often rely on onboard equipment, which is prone to issues such as monitoring dormancy and failure to transmit early warning signals in a timely manner when stationary. This can lead to the neglect of early signs of thermal runaway, ultimately resulting in serious consequences such as fires.

[0004] Currently, various technical solutions have been developed for early warning and detection of thermal runaway in new energy vehicle batteries. Among them, the existing implementations most similar to this invention are mainly divided into two categories, which are described in detail below: Option 1: Built-in early warning scheme in the vehicle's battery management system (BMS). This scheme is the basic early warning method currently standard in new energy vehicles. Its core principle is to use temperature sensors, pressure sensors, and gas sensors installed inside the power battery pack to collect parameters such as voltage, internal resistance, ambient temperature, and gas levels of individual battery cells and the battery pack in real time. The collected parameters are compared with preset safety thresholds. When a parameter exceeds the threshold range, it is determined that the battery is abnormal. The abnormal signal sends a warning to the driver through the vehicle's instrument panel and on-board terminal (such as audible and visual alarms, text reminders). On the other hand, the collected battery abnormality data and warning information can be simultaneously transmitted to the vehicle manufacturer's management platform through the on-board remote information processing terminal (T-BOX), enabling the vehicle manufacturer to remotely monitor the vehicle's battery status. This scheme does not require any additional external equipment and relies solely on the vehicle's own hardware and software logic.

[0005] Option 2: Intelligent Diagnostic Instrument for Power Batteries. This option is currently the most widely used for diagnosing the lifespan of power batteries in new energy vehicles. Its core principle is to connect one end of the diagnostic instrument to the DC charging interface of the new energy vehicle and the other end to the charging gun of the DC charging station. During charging, it acquires parameters such as voltage, current, and temperature of the power battery according to the GB / T 27930 standard. After charging is complete, the data is reported to the backend server, and the health status of the new energy vehicle's power battery is assessed and diagnosed based on these parameters.

[0006] While the two similar existing solutions described above can achieve early warning and detection of battery thermal runaway to a certain extent, they still have significant shortcomings and deficiencies in terms of adapting to all scenarios of new energy vehicle transportation and parking, and improving the accuracy and timeliness of early warnings, as detailed below: Limitations of Solution 1: This solution can only monitor when the vehicle's power supply is normal and the onboard system is operational. When the vehicle is stationary for extended periods (e.g., a private car parked for a long time) or the onboard system is dormant during transportation (e.g., the vehicle is powered off for energy saving), the detectors inside the battery pack collect data intermittently, and the onboard T-BOX cannot transmit data and warning signals in real time, resulting in monitoring blind spots in parking and transportation scenarios. Furthermore, this solution can only monitor internal battery pack parameters and cannot perceive the impact of the external environment (e.g., temperature inside the transport compartment, smoke diffusion in the parking lot) on battery thermal runaway, easily leading to delayed warnings or false alarms. In addition, even if the warning information is transmitted to the vehicle manufacturer's management platform via the T-BOX, this platform primarily serves after-sales service and vehicle maintenance management, and cannot synchronize warning signals to transportation management personnel and parking lot staff. This creates a barrier to warning information transmission, hindering rapid emergency response in various scenarios. Moreover, the vehicle manufacturer's management platform can only receive data and lacks emergency linkage mechanisms related to transportation and parking scenarios, making it impossible to trigger on-site response measures in a timely manner.

[0007] Limitations of Option 2: This diagnostic tool requires simultaneous connection to both the vehicle's DC charging port and the DC charging station's charging gun, and can only be used during vehicle charging. It cannot follow the vehicle's movement to achieve full-process monitoring, and therefore cannot cover mobile transportation scenarios such as passenger ro-ro ships, ocean ro-ro ships, and land transportation. In addition, the diagnostic tool is designed for single-charge trigger data acquisition, without periodic wake-up or continuous monitoring, and cannot continuously and fully track the battery status. Furthermore, the risk of battery thermal runaway in transportation and parking scenarios is random, requiring long-term, continuous status monitoring of the battery, which this solution cannot meet.

[0008] The two existing solutions are independent of each other and cannot achieve seamless monitoring between transportation and parking scenarios. That is, when a vehicle switches from transportation to parking, different monitoring systems need to be switched, which is cumbersome and prone to monitoring interruption. At the same time, the existing solutions cannot be adapted to all transportation and parking sub-scenarios (such as roadside parking spaces, ocean roll-on / roll-off ships, etc.), and the monitoring range is limited, making it difficult to meet the safety monitoring needs of new energy vehicles in multiple scenarios and throughout the entire process. Summary of the Invention

[0009] To address the problems in the prior art, this invention provides a portable early warning and detection device and method for thermal runaway of new energy vehicle batteries.

[0010] The technical solution adopted by this invention to solve its technical problem is: a portable method for early warning and detection of thermal runaway in new energy vehicle batteries, applied to a battery thermal runaway early warning and detection equipment terminal, comprising: S1. Connect to the DC charging interface of the new energy vehicle to simulate the low-voltage auxiliary power supply of the charging pile and wake up the vehicle's BMS. S2. Receive and bind vehicle license plate, portable battery thermal runaway early warning detection device MAC address, parking space area number and parking space number data; S3. Collect BMS battery data; S4. Based on the collected BMS battery data, perform data preprocessing and feature extraction to obtain multiple numerical features; S5. Perform parallel computation on multiple numerical features to obtain multi-dimensional independent anomaly scores; S6. Merge the independent dimensional anomaly scores into a single thermal runaway early warning confidence value; S7. Based on the confidence value of the thermal runaway warning, execute the warning output and report the warning data to the server.

[0011] Furthermore, the BMS battery data collected in S3 includes abnormal fault codes, maximum cell voltage, maximum cell temperature, total voltage, SOC, and insulation resistance.

[0012] Furthermore, the numerical characteristics obtained in S4 include voltage characteristics, temperature characteristics, and fault characteristics; The voltage characteristics include the highest voltage value, lowest voltage value, average value, maximum voltage difference, and standard deviation of the individual cell; The temperature characteristics include the highest temperature value of the individual unit, the lowest high temperature value, the device interface temperature value, and the maximum temperature rise rate. The fault characteristics include the current list of valid fault codes and the historical persistence of fault codes.

[0013] Furthermore, during parallel computation in S5, the score for each dimension is calculated and synthesized using different formulas; The formula for calculating the voltage imbalance score in the voltage characteristic scoring rules is as follows: V=max(V_cell)-min(V_cell); V_avg = mean(V_cell); Imbalance = V / V_avg*100%; The formula for calculating the undervoltage score in the voltage characteristic scoring rules is as follows: V_min = min(V_cell); Undervoltage fraction = (2.8 - V_min) * 80; The calculation formula for overvoltage score in the voltage characteristic scoring rules is as follows: V_max = max(V_cell); If V_max is greater than 4.25V, the overvoltage fraction = (V_max - 4.25) * 100; If V_max is less than or equal to 4.25V and greater than 4.15V, the overvoltage fraction = (V_max - 4.15) * 50; The formula for calculating the temperature rise rate score in the temperature characteristic scoring rules is as follows: ΔT / Δt=(T_current-T_previous) / Δt_minutes; In the fault feature scoring rules, fault features are scored and confirmed by fault codes fed back by the BMS. The identified faults include voltage faults, temperature faults, and insulation faults. The specific scoring principles include one fault being considered mild, two faults being considered severe, and three faults being considered critical. The scores are added together by comparison.

[0014] Further, S6 includes: Assign a base weight to the activated dimension, where voltage is 35%, temperature is 40%, and fault is 25%. Calculate the weighted sum of the scores for all activated dimensions; Select the corresponding synergy coefficient; Calculate the confidence score; The formula for calculating the confidence score includes: Confidence score = reinforcement factor coefficient x synergy coefficient x (V_score x 0.35 + T_score x 0.4 + F_score x 0.25).

[0015] Furthermore, the early warning data in S7 includes the early warning level and corresponding handling suggestions for the early warning level; The warning levels include normal, attention, low risk, medium risk, high risk, and emergency.

[0016] Furthermore, the battery thermal runaway early warning detection device starts timing after being removed from the vehicle's DC charging port. If it is not reconnected to the vehicle's DC charging port within 3 minutes, the device will automatically deregister and unbind the vehicle's information.

[0017] Furthermore, after the battery thermal runaway early warning detection device terminal is removed from the vehicle's DC charging port, it sends a brief low-voltage wake-up signal to the BMS at preset time intervals to wake up the BMS and quickly collect data.

[0018] A portable early warning and detection device for thermal runaway of new energy vehicle batteries is applied to a terminal of a battery thermal runaway early warning and detection equipment to implement the aforementioned portable early warning and detection method for thermal runaway of new energy vehicle batteries. The device includes: The device's physical interface is used to connect to the vehicle's charging port; The BMS communication module is used to wake up the vehicle's BMS and receive bound data. The processor module is used to perform data preprocessing and feature extraction, multi-dimensional independent anomaly scoring, and multi-dimensional comprehensive score and confidence calculation on the received data. The audible and visual alarm module is used to issue audible and visual alarms. Wireless communication module, used for data transmission with the server.

[0019] Furthermore, the BMS communication module includes a simulated wake-up module, a data receiving and binding module, and a data acquisition module; The simulated wake-up module is used to simulate the low-voltage auxiliary power supply of the charging pile and wake up the vehicle's BMS. The data receiving and binding module is used to receive and bind vehicle license plate, portable battery thermal runaway early warning detection device MAC address, parking space area number and parking space number data; The data acquisition module is used to collect BMS battery data; The processor module includes a preprocessing and feature extraction module, a multidimensional independent anomaly score calculation module, and an anomaly score fusion processing module. The preprocessing and feature extraction module is used for data preprocessing and feature extraction to obtain multiple numerical features; The multidimensional independent anomaly scoring module is used to perform parallel calculations on multiple numerical features to obtain multidimensional independent anomaly scores. The anomaly score fusion processing module is used to fuse independent dimensional anomaly scores into a thermal runaway early warning confidence value.

[0020] The beneficial effects of this invention are: 1. Supports centralized and independent parking scenarios and is applicable to various transportation scenarios, such as shipping and land transportation. It has flexible deployment capabilities and does not require modification of existing parking facilities.

[0021] 2. Supports multiple domestic and international charging and communication standards, such as GB / T 27930, ISO 15118, CHAdeMO, etc., applicable to domestic and export vehicles, improving the equipment's versatility.

[0022] 3. Adopt a multi-dimensional anomaly scoring system, combining voltage, temperature, and fault characteristics for comprehensive evaluation, to improve the accuracy and reliability of thermal runaway early warning.

[0023] 4. By leveraging the trend persistence reinforcement factor and synergy coefficient mechanism, early warnings can be issued in the early stages of risk accumulation, enabling early intervention and preventing the escalation of accidents.

[0024] 5. The testing equipment starts up immediately upon connection and supports Bluetooth binding and automatic deregistration mechanisms, reducing the complexity of manual operation and improving efficiency.

[0025] 6. The device has multiple built-in communication module interfaces, supports flexible networking methods, adapts to different network environments, and ensures reliable data reporting. Attached Figure Description

[0026] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a scenario example diagram illustrating the portable early warning and detection method for thermal runaway of new energy vehicle batteries according to the present invention. Figure 2 This is a schematic diagram of the structure of the portable battery thermal runaway early warning detection device in the portable new energy vehicle battery thermal runaway early warning device of the present invention; Figure 3 This is a flowchart of a portable early warning and detection method for thermal runaway of new energy vehicle batteries according to the present invention. Figure 4 for Figure 3 Flowchart of the method for extracting medium-temperature features; Figure 5 for Figure 3 Flowchart of the method for extracting medium voltage features; Figure 6 for Figure 3 Flowchart of the method for extracting fault features in China; Figure 7 for Figure 3 Flowchart of the method for multi-dimensional comprehensive score analysis, confidence calculation, and confidence-early warning mapping; Figure 8 for Figure 7Enlarged view of part A in the middle; Figure 9 for Figure 7 A magnified view of part B in the middle. Detailed Implementation

[0027] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0028] like Figure 1 As shown, the embodiments of this application can be applied to multiple portable battery thermal runaway early warning and detection devices that are connected to a server through a communication gateway. Specifically, they can be applied to scenarios involving new energy vehicles on land, in shipping, and in various parking lots. The server can also be connected to a monitoring center's large screen.

[0029] This application provides a portable method for early warning and detection of thermal runaway in new energy vehicle batteries, applied to a battery thermal runaway early warning and detection equipment terminal, including the following steps: Step S1: Connect the DC charging interface of the new energy vehicle to simulate the low-voltage auxiliary power supply of the charging pile and wake up the vehicle BMS; Step S2: Receive and bind vehicle license plate, portable battery thermal runaway early warning detection device MAC address, parking space area number, and parking space number data; Step S3: Collect BMS battery data; The collected BMS battery data includes abnormal fault codes, maximum cell voltage, maximum cell temperature, total voltage, SOC, and insulation resistance.

[0030] Based on multi-source data such as battery voltage, temperature, and BMS fault messages collected by portable battery thermal runaway early warning detection equipment, a layered fusion calculation logic is used to output a thermal runaway early warning confidence value in the range of 0-100. The higher the value, the greater the risk of thermal runaway in the current battery state, providing tiered decision support for management personnel. The early warning confidence algorithm completes a full assessment at a fixed interval, such as 1 minute, and is divided into four stages, specifically steps S4-S7 as follows: Step S4: Based on the collected BMS battery data, perform data preprocessing and feature extraction to obtain multiple numerical features; The numerical characteristics include voltage characteristics, temperature characteristics, and fault characteristics; Voltage characteristics include, but are not limited to, the highest voltage value, the lowest voltage value, the average value, the maximum voltage difference, and the standard deviation (unevenness) of a single cell. Temperature characteristics include, but are not limited to, the highest temperature value of a single unit, the lowest high temperature value, the temperature value of the device interface, and the maximum temperature rise rate (ΔT / Δt); Fault characteristics include the current list of valid fault codes and the persistence of fault code history.

[0031] Step S5: Perform parallel calculations on multiple numerical features to obtain multi-dimensional independent anomaly scores; Multiple numerical features in the first stage are computed in parallel, each outputting an anomaly score of 0-100. The score for each dimension is calculated and combined using different formulas.

[0032] The voltage characteristic scoring rule (V_score) includes voltage imbalance score, voltage undervoltage score, voltage overvoltage score, and voltage comprehensive dimension score; The formula for calculating the voltage imbalance score is as follows: V = max(V_cell) - min(V_cell), and the maximum voltage difference of a single cell is obtained from this formula; V_avg=mean(V_cell), and the average value of the single cell voltage is determined by this formula; Imbalance = V / V_avg*100%, expressed as a percentage; It is important to note that: V_cell is an array or list containing the collected voltage values ​​of individual cells in the battery pack; mean() represents the function to calculate the arithmetic mean, max() represents the function to calculate the maximum value; and min() represents the function to calculate the minimum value.

[0033]

[0034] The formula for calculating the undervoltage score is as follows: The undervoltage threshold is set at 2.8V. For every 0.1V drop, 8 points are added, with a maximum score of 100 points. The calculation formula is as follows: V_min = min(V_cell); Undervoltage fraction = (2.8 - V_min) * 80.

[0035] The formula for calculating the voltage overvoltage score is: Overvoltage thresholds are set at 4.15V and 4.25V. For every 0.1V increase between 4.15V and 4.25V, 5 points are added, up to a maximum of 50 points. For voltages above 4.25V, 10 points are added for every 0.1V increase. The calculation formula is as follows: V_max = max(V_cell); If V_max is greater than 4.25V, the overvoltage fraction = (V_max - 4.25) * 100; If V_max is less than or equal to 4.25V and greater than 4.15V, the overvoltage fraction = (V_max - 4.15) * 50; The formula for calculating the comprehensive voltage dimension score is as follows: The max() function calculates the maximum values ​​of voltage imbalance score, undervoltage score, and overvoltage score. If both undervoltage and overvoltage exist, V_score is directly assigned a value of 100.

[0036] The temperature characteristic scoring rule (T_score) includes a score for the rate of temperature rise; Temperature characteristics are scored by combining the highest temperature value and the rate of temperature rise. The formula for calculating the rate of temperature rise is as follows: ΔT / Δt=(T_current-T_previous) / Δt_minutes Notes: T_current is the temperature value measured at the current time; T_previous is the temperature value measured at the previous time; Δt_minutes is the time interval between the two measurements; ΔT / Δt is the rate of temperature change (temperature rise rate).

[0037] The calculation method for the fault characteristic scoring rule (F_score): Fault characteristics are mainly confirmed by scoring based on fault codes fed back by the BMS. The main faults identified are: voltage fault (40 points), temperature fault (40 points), and insulation fault (20 points). The specific scoring principle is: one fault is mild, two are severe, and three are critical. The scores are added together. For example, if there are both voltage and temperature faults, the score is 80 points, which is a severe fault.

[0038] Step S6: Merge the independent dimensional anomaly scores into a single thermal runaway early warning confidence value; This step merges the three independent anomaly scores—voltage V_score, temperature T_score, and fault F_score—into a single, comprehensive thermal runaway early warning confidence value (ranging from 0 to 100). The specific judgment method is as follows: Step S61. Detect whether it is necessary to enter the "multi-dimensional comprehensive score analysis" algorithm. The judgment conditions are: voltage > 20 points, temperature > 25 points, fault > 15 points; Step S62. Assign a basic weight to the activated dimension, where voltage is 35%, temperature is 40%, and fault is 25%; Step S63. Calculate the weighted sum of the scores for all activated dimensions; Step S64. Select the synergy coefficient according to the table below, where multiple dimensions are activated, increasing the risk;

[0039] Regarding the trend persistence reinforcement factor coefficient, the system detects and analyzes whether the current comprehensive risk value has shown a continuous upward trend over the past few periods. If so, it is considered that the risk is accumulating and worsening, and an additional reinforcement factor coefficient of 1.3 will be applied to the current result to provide an early warning.

[0040] Step S65. Calculate the confidence score; Confidence score = reinforcement factor coefficient x synergy coefficient x (V_score x 0.35 + T_score x 0.4 + F_score x 0.25).

[0041] For example: if V_score is 70 and T_score is 50, then Confidence score = 1.3 x (1.3 x (70 x 0.35 + 50 x 0.4)) = 75 Step S7: Based on the confidence value of the thermal runaway warning, execute the warning output and report the warning data to the server.

[0042] The early warning data includes the warning level and corresponding handling recommendations. The warning levels include Normal, Attention, Low Risk, Medium Risk, High Risk, and Emergency. The handling recommendations for each warning level are shown in the table below:

[0043] Step S8: After the portable battery thermal runaway early warning detection device is removed from the vehicle's DC charging port, a timer begins. If the device is not reconnected to the vehicle's DC charging port within 3 minutes, the device will automatically deregister and unbind the vehicle's information.

[0044] Preferably, after the portable battery thermal runaway early warning and detection device is removed from the vehicle's DC charging port, another operating method can be continued, specifically: S81. After the device is removed from the DC charging port, it does not immediately unbind. Instead, the device enters a low-power sleep mode and establishes intermittent wake-up communication with the vehicle BMS. For example, the wake-up interval can be set: wake up once every hour for the first 24 hours of inactivity, and wake up once every 3 hours after 24 hours. Users can adjust this manually.

[0045] S82. During intermittent wake-up, the device briefly simulates low-voltage power supply to wake up the BMS through the residual contact of the DC charging port or the added wireless wake-up module, and quickly collects core parameters such as individual unit voltage and temperature. After the collection is completed, the wake-up power is immediately cut off, the device returns to hibernation, and the data is synchronously uploaded to the server. If abnormal data is collected, an early warning is immediately triggered. The early warning methods include local audible and visual alarms and server push notifications, and the wake-up interval is shortened, such as once every 10 minutes, to continuously track abnormalities.

[0046] S83. If the user decides that they no longer want to monitor the vehicle, they can manually cancel the binding relationship with the vehicle and clear the relevant information through the device button or mobile APP to avoid binding too many vehicles to the device and affecting its use.

[0047] This design balances portability and continuous monitoring needs, avoiding the problem of no monitoring after the device is removed.

[0048] This invention proposes a multi-dimensional anomaly scoring algorithm for thermal runaway early warning confidence. It independently scores parameters such as voltage, temperature, and fault codes, and combines weight allocation, synergy coefficient, and trend persistence enhancement factor to obtain a comprehensive confidence value of 0-100, realizing the fusion analysis of multi-dimensional data. It adopts a confidence level early warning mapping mechanism to divide the 0-100 confidence value into 6 early warning levels and corresponding clear handling suggestions to achieve graded response and decision support.

[0049] This invention relates to a portable thermal runaway early warning and detection device adapted to the DC charging interface of new energy vehicles. It can simulate the low-voltage auxiliary power supply of the charging pile to wake up the vehicle's BMS and collect battery data at configurable intervals, thereby realizing continuous battery monitoring when the vehicle is stationary or powered off.

[0050] This invention designs a unique binding and precise positioning mechanism for a combination of detection equipment, vehicles, and parking spaces / charging piles. It associates the device's MAC address with the vehicle's VIN / license plate, parking space / charging pile number, and physical location, and combines GPS / BeiDou / electronic maps to achieve precise vehicle positioning down to the floor / area / parking space.

[0051] This application provides a portable method for early warning and detection of thermal runaway in new energy vehicle batteries, including the following steps: Step S11: After the new energy vehicle is stably parked in the parking space, the on-site staff or the vehicle owner will connect the portable battery thermal runaway early warning detection device to the vehicle's DC charging interface. Step S12: On-site staff or vehicle owners use handheld PADs or mobile phones to take photos and scan codes to bind the vehicle license plate, the MAC address of the portable battery thermal runaway early warning detection device, the parking space area number, and the parking space number, and then enter the information into the portable battery thermal runaway early warning detection device for local storage and upload to the server. Step S13: After successful information binding, the portable battery thermal runaway early warning and detection device enters the data acquisition and monitoring state; Step S11: After the portable battery thermal runaway early warning detection device is removed from the vehicle's DC charging port, a timer is started. If the device is not reconnected to the vehicle's DC charging port within 3 minutes, the device will automatically deregister and unbind the vehicle's information.

[0052] It should be noted that: for centralized parking, the parking lot map in the monitoring center is displayed visually; for independent parking, the equipment and vehicles are located using GPS and Beidou coordinate information, and the location is displayed visually on the back-end management system.

[0053] This application provides a portable early warning and detection device for thermal runaway of new energy vehicle batteries, applied to a battery thermal runaway early warning and detection equipment terminal, used to implement the aforementioned portable early warning and detection method for thermal runaway of new energy vehicle batteries, such as... Figure 2 As shown, the device includes: The device's physical interface is used to connect to the vehicle's charging port; The physical interface of the equipment and the interface for interfacing with new energy vehicles comply with the GB / T20234.3 standard. For exported vehicles, it meets international standards such as European standard (CCS2), North American standard (CCS1), and Japanese standard (CHAdeMO), and its protection level is ≥IP54 (dustproof and waterproof).

[0054] The BMS communication module is used to wake up the vehicle's BMS and receive bound data. The battery thermal runaway early warning detection equipment is standardized and securely attached to the DC charging interface of the new energy vehicle. The reliability of the connection between the equipment and the vehicle's DC charging interface is confirmed by the voltage signals corresponding to CC1 and CC2.

[0055] The testing equipment simulates the low-voltage auxiliary power supply (12V / 24V) of the charging pile, wakes up the vehicle BMS, and establishes a secure handshake communication process with the vehicle BMS.

[0056] Regarding communication protocols, the communication protocols between equipment and vehicle BMS in new energy vehicles in China comply with the relevant provisions of the national standard GB / T27930. New energy vehicles exported from China must have BMS communication protocols between their equipment and vehicles that comply with relevant international standards such as ISO 15118 / DIN SPEC 70121 and CHAdeMO 1.2 / 2.0.

[0057] The BMS communication module includes a simulated wake-up module, a data receiving and binding module, and a data acquisition module. The simulated wake-up module is used to simulate the low-voltage auxiliary power supply of the charging pile and wake up the vehicle's BMS; Preferably, the detection device can also have a built-in adaptive voltage adjustment module and an integrated multi-protocol compatibility module. The adaptive voltage adjustment module automatically identifies the BMS wake-up voltage threshold of a vehicle model by detecting the BMS feedback signal through a pre-stored mainstream vehicle model wake-up voltage database or through initial communication probing. The automatic adjustment output voltage is adjustable from 12V to 24V in 0.5V steps to ensure that different brands and vehicle models can successfully wake up the BMS and solve the voltage incompatibility problem.

[0058] The testing equipment incorporates mainstream charging pile and BMS communication protocols such as GB / T27930 and ISO15118. It can automatically identify the communication protocols supported by the vehicle's BMS, switch to the corresponding communication mode, and establish a secure handshake. It also adds a handshake failure retry mechanism with 3 retries and 10-second intervals between each retry. If the handshake fails, it automatically records the vehicle model information and sends it back to the server, while switching to the backup communication method to ensure the success rate of data acquisition.

[0059] By adopting a combination of adaptive voltage and automatic switching of multiple protocols, the compatibility problem of multiple vehicle models is solved, and the versatility of the equipment is improved.

[0060] The data receiving and binding module is used to receive and bind vehicle license plate, portable battery thermal runaway early warning detection device MAC address, parking space area number, and parking space number data. The data acquisition module is used to collect BMS battery data; The device wakes up the BMS and collects battery data from the BMS. The data includes, but is not limited to, abnormal fault codes, maximum cell voltage, maximum cell temperature, total voltage, SOC, insulation resistance, etc. The data is collected in accordance with the format specified in the above communication protocol.

[0061] The processor module is used to perform data preprocessing and feature extraction, multi-dimensional independent anomaly scoring, and multi-dimensional comprehensive score and confidence calculation on the received data.

[0062] The processor module includes a preprocessing and feature extraction module, a multidimensional independent anomaly score calculation module, and an anomaly score fusion processing module. The preprocessing and feature extraction module is used for data preprocessing and feature extraction to obtain multiple numerical features; The multidimensional independent anomaly score calculation module is used to perform parallel calculations on multiple numerical features to obtain multidimensional independent anomaly scores; The anomaly score fusion processing module is used to fuse independent dimensional anomaly scores into a single thermal runaway early warning confidence value.

[0063] The audible and visual alarm module is used to issue audible and visual alarms. The battery thermal runaway early warning and detection equipment features local audible and visual alarms, and also has reserved interfaces for integration with parking area broadcast systems and existing fire alarm systems. Wireless communication module, used for data transmission with the server.

[0064] The battery thermal runaway early warning and detection device has multiple built-in wireless communication module interfaces, including LoRa, 4G / 5G, WiFi, and Bluetooth; it can set a unique device ID via Bluetooth and bind the device ID with vehicle information such as parking space number, license plate, and vehicle VIN code; For centralized parking, the detection equipment can communicate with a local area network gateway to form a network and report data to the monitoring center or cloud server through the gateway; the local area network gateway includes, but is not limited to, Bluetooth gateway, LoRa gateway, and WiFi gateway.

[0065] For stand-alone parking spaces, the detection equipment reports data to the cloud server via 4G / 5G communication.

[0066] In addition, the battery thermal runaway early warning and detection equipment is also powered by a built-in battery, which can support continuous monitoring for ≥30 days on a single charge.

[0067] LED three-color indicator: green (normal), yellow (fault), red (alarm).

[0068] The detection method of this invention can be applied to various parking and transportation scenarios for new energy vehicles, including centralized parking scenarios such as parking lots inside buildings, transportation scenarios such as passenger roll-on / roll-off ships, ocean-going roll-on / roll-off ships, and land transportation, as well as independent parking scenarios such as street parking and private car parking spaces. After a new energy vehicle enters a designated parking space, staff (in centralized parking and transportation scenarios) or vehicle owners (independent scenarios such as private cars) connect the portable battery thermal runaway early warning detection device to the vehicle's DC charging interface, and bind the device with the parking space number and vehicle information according to the actual scenario. After the device completes a secure handshake and data reading with the vehicle's battery management system (BMS), it performs local analysis and calculation on the collected battery data. At the same time, it supports uploading data to the backend server via 4G / 5G, local area network, and other network methods. Relevant monitoring information can be viewed on the monitoring management platform or mobile devices, thereby realizing early warning and periodic assessment of the risk of battery thermal runaway in new energy vehicles in all scenarios.

[0069] Although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A portable method for early warning and detection of thermal runaway in new energy vehicle batteries, characterized in that: Applications include: Terminals of battery thermal runaway early warning and detection equipment, including: S1. Connect to the DC charging interface of the new energy vehicle to simulate the low-voltage auxiliary power supply of the charging pile and wake up the vehicle's BMS. S2. Receive and bind vehicle license plate, portable battery thermal runaway early warning detection device MAC address, parking space area number and parking space number data; S3. Collect BMS battery data; S4. Based on the collected BMS battery data, perform data preprocessing and feature extraction to obtain multiple numerical features; S5. Perform parallel computation on multiple numerical features to obtain multi-dimensional independent anomaly scores; S6. Merge the independent dimensional anomaly scores into a single thermal runaway early warning confidence value; S7. Based on the confidence value of the thermal runaway warning, execute the warning output and report the warning data to the server.

2. The portable early warning and detection method for thermal runaway of new energy vehicle batteries according to claim 1, characterized in that: The BMS battery data collected in S3 includes abnormal fault codes, maximum cell voltage, maximum cell temperature, total voltage, SOC, and insulation resistance.

3. The portable early warning and detection method for thermal runaway of new energy vehicle batteries according to claim 1, characterized in that: The numerical characteristics obtained in S4 include voltage characteristics, temperature characteristics, and fault characteristics; The voltage characteristics include the highest voltage value, lowest voltage value, average value, maximum voltage difference, and standard deviation of the individual cell; The temperature characteristics include the highest temperature value of the individual unit, the lowest high temperature value, the device interface temperature value, and the maximum temperature rise rate. The fault characteristics include the current list of valid fault codes and the historical persistence of fault codes.

4. The portable early warning and detection method for thermal runaway of new energy vehicle batteries according to claim 1, characterized in that: During parallel computation in S5, the score for each dimension is calculated and combined using different formulas. The formula for calculating the voltage imbalance score in the voltage characteristic scoring rules is as follows: V=max(V_cell)-min(V_cell); V_avg = mean(V_cell); Imbalance = V / V_avg*100%; The formula for calculating the undervoltage score in the voltage characteristic scoring rules is as follows: V_min = min(V_cell); Undervoltage fraction = (2.8 - V_min) * 80; The calculation formula for overvoltage score in the voltage characteristic scoring rules is as follows: V_max = max(V_cell); If V_max is greater than 4.25V, the overvoltage fraction = (V_max - 4.25) * 100; If V_max is less than or equal to 4.25V and greater than 4.15V, the overvoltage fraction = (V_max - 4.15) * 50; The formula for calculating the temperature rise rate score in the temperature characteristic scoring rules is as follows: ΔT / Δt=(T_current-T_previous) / Δt_minutes; In the fault feature scoring rules, fault features are scored and confirmed by fault codes fed back by the BMS. The identified faults include voltage faults, temperature faults, and insulation faults. The specific scoring principles include one fault being considered mild, two faults being considered severe, and three faults being considered critical. The scores are added together by comparison.

5. The portable early warning and detection method for thermal runaway of new energy vehicle batteries according to claim 1, characterized in that: S6 includes: Assign a base weight to the activated dimension, where voltage is 35%, temperature is 40%, and fault is 25%. Calculate the weighted sum of the scores for all activated dimensions; Select the corresponding synergy coefficient; Calculate the confidence score; The formula for calculating the confidence score includes: Confidence score = reinforcement factor coefficient x synergy coefficient x (V_score x 0.35 + T_score x 0.4 + F_score x 0.25).

6. The portable early warning and detection method for thermal runaway of new energy vehicle batteries according to claim 1, characterized in that: The warning data in S7 includes the warning level and corresponding handling suggestions for the warning level; The warning levels include normal, attention, low risk, medium risk, high risk, and emergency.

7. The portable early warning and detection method for thermal runaway of new energy vehicle batteries according to claim 1, characterized in that: The battery thermal runaway early warning detection device starts timing after being removed from the vehicle's DC charging port. If it is not reconnected to the vehicle's DC charging port within 3 minutes, the device will automatically deregister and unbind the vehicle's information.

8. The portable early warning and detection method for thermal runaway of new energy vehicle batteries according to claim 1, characterized in that: After the battery thermal runaway early warning detection device terminal is removed from the vehicle's DC charging port, it sends a brief low-voltage wake-up signal to the BMS at preset time intervals to wake up the BMS and quickly collect data.

9. The portable new energy vehicle battery thermal runaway early warning and detection device according to claim 1, characterized in that: An apparatus for use in a battery thermal runaway early warning and detection device terminal, for implementing the portable new energy vehicle battery thermal runaway early warning and detection method according to any one of claims 1-8, the apparatus comprising: The device's physical interface is used to connect to the vehicle's charging port; The BMS communication module is used to wake up the vehicle's BMS and receive bound data. The processor module is used to perform data preprocessing and feature extraction, multi-dimensional independent anomaly scoring, and multi-dimensional comprehensive score and confidence calculation on the received data. The audible and visual alarm module is used to issue audible and visual alarms. Wireless communication module, used for data transmission with the server.

10. The portable new energy vehicle battery thermal runaway early warning and detection device according to claim 9, characterized in that: The BMS communication module includes a simulated wake-up module, a data receiving and binding module, and a data acquisition module. The simulated wake-up module is used to simulate the low-voltage auxiliary power supply of the charging pile and wake up the vehicle's BMS. The data receiving and binding module is used to receive and bind vehicle license plate, portable battery thermal runaway early warning detection device MAC address, parking space area number and parking space number data; The data acquisition module is used to collect BMS battery data; The processor module includes a preprocessing and feature extraction module, a multidimensional independent anomaly score calculation module, and an anomaly score fusion processing module. The preprocessing and feature extraction module is used for data preprocessing and feature extraction to obtain multiple numerical features; The multidimensional independent anomaly scoring module is used to perform parallel calculations on multiple numerical features to obtain multidimensional independent anomaly scores. The anomaly score fusion processing module is used to fuse independent dimensional anomaly scores into a thermal runaway early warning confidence value.

Images