Vehicle stopping electrical safety monitoring system

Abstract

The invention discloses a vehicle stopping electrical safety monitoring system, and relates to the technical field of new energy vehicle control, the vehicle stopping electrical safety monitoring system comprises an intelligent sensing layer, a digital twin decision-making layer, a hierarchical execution layer and a safety escape layer, and the intelligent sensing layer comprises a multi-mode sensor array and is used for collecting physical parameters, chemical parameters and environmental parameters of a vehicle electrical system in real time. According to the method, a high-precision three-dimensional model, an embedded electrolyte leakage model and a wire harness aging model are constructed, the high-precision three-dimensional model comprises a battery compartment geometric structure and a wire harness layout, the former simulates a diffusion path and predicts a short-circuit risk based on fluid dynamics, and the latter realizes fault evolution prediction based on a physical mechanism by combining vibration fatigue and insulation material aging analysis. The hysteresis of traditional threshold value alarm is broken through, a dynamic threshold value table is generated in combination with environment temperature, the charge state and historical fault data, the problem that a traditional fixed threshold value cannot adapt to complex working conditions is solved, and the false alarm rate and the missing alarm rate are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy vehicle control, and particularly to a vehicle stop electrical safety monitoring system. Background Art

[0002] With the acceleration of the global energy transformation, especially the significant increase in the penetration rate of new energy electric vehicles such as pure electric and hybrid vehicles, their core power systems rely on high-energy-density batteries, high-voltage electrical architectures, and complex wiring harness networks. However, during the rapid development of new energy electric vehicles, electrical system safety issues have become increasingly prominent. Problems such as battery thermal runaway, electrolyte leakage, and aging of high-voltage wiring harnesses have become technical bottlenecks restricting the development of the industry. According to statistics, among the safety accidents of new energy electric vehicles globally, approximately 60% are directly related to battery system failures, and 30% involve abnormalities in high-voltage electrical circuits, revealing serious deficiencies in traditional safety monitoring technologies.

[0003] Published patent: A vehicle high-voltage electrical safety intelligent monitoring system, monitoring method, and vehicle (Publication No.: CN116176277A), including a microprocessor controller and a fault detection device; the fault detection device is used to detect fault information of the high-voltage electrical system, and the microprocessor controller obtains the fault information according to the detection frequency set by the priority of fault detection, performs fusion processing on the fault information according to the divided fault levels, and determines a power-off instruction, which can realize the safety monitoring of the working states of vehicle high-voltage fuses, contactors, and each connection node, and can also realize the real-time monitoring function of the overall insulation performance of the high-voltage electrical system, improving the comprehensiveness of the monitoring of the vehicle high-voltage system and further enhancing the safety of vehicle operation.

[0004] The existing technology only relies on a single parameter for monitoring, unable to effectively identify compound faults, resulting in limited detection ability for complex faults. The fixed threshold strategy leads to a high false alarm rate in specific environments. After an emergency stop, there is a lack of real-time perception ability of the occupant state and the environment, and no dynamic escape assistance mechanism is integrated, resulting in incomplete safety measures. There is a lack of effective means for monitoring key indicators such as gas evolution and deformation inside the battery, making it difficult to timely warn of potential risks. Summary of the Invention

[0005] The purpose of the present invention is to propose a vehicle stop electrical safety monitoring system to solve the deficiencies existing in the prior art.

[0006] To achieve the above object, the present invention adopts the following technical solutions: a vehicle stop electrical safety monitoring system, including an intelligent perception layer, a digital twin decision layer, a hierarchical execution layer, and a safety escape layer. The intelligent perception layer includes a multi-modal sensor array for real-time collection of physical parameters, chemical parameters, and environmental parameters of the vehicle electrical system. The digital twin decision layer is based on a three-dimensional physical model of the vehicle electrical system, synchronizes the data of the intelligent perception layer in real time, and dynamically adjusts the fault threshold. The three-dimensional physical model is embedded with an electrolyte leakage model and a wire harness aging model. The hierarchical execution layer triggers a three-level response mechanism according to the diagnosis results of the digital twin decision layer, including local current limiting, high-voltage circuit isolation, and emergency braking lock-up. The safety escape layer integrates UWB radar, cameras, and emergency batteries for occupant status perception, environmental analysis, and hierarchical escape response.

[0007] As a further description of the above technical solution:

[0008] The multi-modal sensor array is composed of fiber Bragg grating sensors, ultrasonic sensors, self-powered composite sensors, voltage / current sensors, and temperature sensors. The fiber Bragg grating sensors are embedded between battery modules to monitor the micro-deformation caused by thermal expansion and generate a deformation-time curve. The ultrasonic sensors detect the gas evolution inside the battery and generate a gas evolution spectrogram. The self-powered composite sensors collect energy through the piezoelectric-thermoelectric effect and monitor the vibration amplitude of the wire harness and the temperature difference at key nodes.

[0009] As a further description of the above technical solution:

[0010] The electrolyte leakage model is based on fluid dynamics to simulate the diffusion path of the electrolyte and predict the short-circuit risk. The wire harness aging model analyzes the vibration spectrum based on the rainflow counting method and calculates the aging rate of the insulating material according to the Arrhenius equation to predict the remaining life of the wire harness.

[0011] As a further description of the above technical solution:

[0012] The three-level response mechanism of the hierarchical execution layer includes a first-level response, a second-level response, and a third-level response. When a single sensor detects an abnormality, the first-level response is triggered to reduce the current in the fault area and start the liquid cooling system. When multiple sensors jointly alarm, the second-level response is triggered to isolate the high-voltage circuit through a solid-state relay and retain the power supply of the low-voltage system. When the electrolyte leakage model predicts the short-circuit risk, the third-level response is triggered to cut off the high-voltage power supply of the whole vehicle, make the vehicle perform emergency braking, and force the vehicle to stop.

[0013] As a further description of the above technical solution:

[0014] The UWB radar detects the breathing frequency and heartbeat signal of the occupant through the Doppler effect. The camera identifies the occupant's posture and environmental obstacles, including the status of seat belt jamming and the degree of door deformation. The emergency battery powers the dual CAN buses, electromagnetic pulse window-breaking device, micro-explosion device, UWB radar, camera, audio, door electronic lock, and in-vehicle emergency lighting to ensure continuous operation after power failure.

[0015] As a further description of the above technical solution:

[0016] The electromagnetic pulse window-breaking device emits high-energy electromagnetic pulses to break the window closest to the occupant. The micro-explosion device serves as a backup window-breaking solution and is activated when the electromagnetic pulse window-breaking device fails. It is equipped with dual CAN buses, and the main control system transmits instructions through two independent buses. When one fails, the other can still transmit data.

[0017] As a further description of the above technical solution:

[0018] The hierarchical escape mechanism of the safety escape layer includes a first-level escape response and a second-level escape response. When the occupant is awake and the door / window is not severely deformed, the first-level escape response is triggered, and the escape path is prompted by voice and the door is automatically unlocked. When it is detected that the occupant is unconscious or the door fails, the second-level escape response is activated, triggering electromagnetic pulse window-breaking and emergency lighting.

[0019] As a further description of the above technical solution:

[0020] The system operation process is as follows:

[0021] S1. Data acquisition

[0022] The fiber Bragg grating sensor monitors the thermal expansion micro-deformation of the battery module, generates a deformation-time curve, and converts the deformation amount through the wavelength shift amount;

[0023] The ultrasonic sensor scans the internal gas state of the battery every 5 seconds, emits ultrasonic waves, generates a gas evolution spectrogram through the reflected wave signal, and calculates the gas evolution amount;

[0024] The self-powered composite sensor synchronously collects the vibration acceleration of the high-voltage wire harness and the temperature difference at key nodes, and generates a vibration spectrogram using the piezoelectric effect and the Seebeck effect;

[0025] The voltage / current sensor real-time collects the dynamic data of voltage and current during the charging and discharging process of the battery;

[0026] The temperature sensor records the temperature data of the battery cells and modules, compensating for the influence of temperature on the internal resistance of the battery;

[0027] S2. Data preprocessing

[0028] Eliminate the noise interference of sensor data through Kalman filtering, extract the effective signal, combine the deformation amount, gas evolution rate, vibration spectrum and temperature difference data, and construct the battery health index to provide input for subsequent modeling;

[0029] S3. Modeling

[0030] The digital twin decision-making layer receives the deformation amount, gas evolution spectrogram, gas type, gas evolution rate, wire harness vibration acceleration and temperature difference, voltage and current, and temperature data of battery cells and modules. Based on the sensor data and battery physical characteristics, a high-precision three-dimensional model is constructed, and the electrolyte leakage model and wire harness aging model are embedded in the high-precision three-dimensional model;

[0031] The electrolyte leakage model simulates the electrolyte diffusion path through fluid dynamics, and combines thermodynamics and electrochemical models to predict the short-circuit risk;

[0032] The wire harness aging model analyzes the metal fatigue life caused by vibration based on the rain flow counting method, calculates the thermal aging rate of insulating materials using the Arrhenius equation, and generates the wire harness aging coefficient by weighting;

[0033] S4. Data Analysis

[0034] Electrolyte leakage risk diagnosis: fuse the deformation data and gas evolution spectrogram to judge the thermal runaway risk; predict the short-circuit probability after 3 seconds through the electrochemical model;

[0035] Wire harness aging diagnosis: combine the vibration spectrum, temperature difference data and aging coefficient to evaluate the contact failure risk and predict the remaining life;

[0036] Dynamic threshold adjustment: generate a dynamic threshold table according to the ambient temperature, state of charge and historical fault records to improve the diagnosis accuracy;

[0037] S5. Fault Diagnosis

[0038] Thermal runaway risk: the triggering condition is that the deformation amount exceeds the limit and the gas evolution rate is abnormal;

[0039] Wire harness abnormality: the triggering condition is that the vibration amplitude > 2g or the temperature difference > 10°C;

[0040] Electrolyte leakage: the triggering condition is that the digital twin model predicts the short-circuit risk;

[0041] S6. Fault Classification Response Execution

[0042] Dynamically control the safety protection operation through a three-level response mechanism:

[0043] First-level response: potential risk. When a single sensor is abnormal, reduce the current in the fault area and direct the enhanced coolant flow;

[0044] Secondary response: Moderate failure. When multiple sensors jointly give an alarm, cut off the faulty high-voltage circuit, maintain low-voltage power supply, and enter the limp-home mode;

[0045] Tertiary response: Emergency failure. When it is predicted that the electrolyte will leak, immediately cut off the high-voltage bus power supply, trigger the vehicle-wide power-off and emergency braking;

[0046] S7. Escape response

[0047] Occupant status perception:

[0048] UWB radar: Detect the occupant's breathing frequency, heartbeat signal, and centimeter-level positioning;

[0049] Camera: Analyze the occupant's posture, seat belt status, and the deformation degree of the doors / windows;

[0050] Trigger hierarchical escape:

[0051] Primary response: Active escape. When the occupant is conscious and the doors are available, display the escape route and unlock the electronic locks;

[0052] Secondary response: Mechanical-assisted escape. When the occupant is unconscious or the doors are deformed, trigger the electromagnetic pulse window-breaking or micro-explosion device, turn on the LED lights, and provide continuous voice guidance.

[0053] The present invention has the following beneficial effects:

[0054] 1. In the present invention, a three-dimensional architecture of "intelligent perception layer - digital twin decision layer - hierarchical execution layer - safety escape layer" is constructed, breaking through the limitations of traditional single monitoring or execution systems. Each layer is linked through a data closed-loop to achieve full-process coverage from risk monitoring to safety disposal, forming a complete safety closed-loop of "monitoring - diagnosis - execution - escape". Through the innovative combination of "layered architecture + multi-modal perception + digital twin + hierarchical response + active escape", it breaks through the limitations of single-point monitoring, static threshold alarm, and passive safety disposal of traditional vehicle electrical safety systems, realizing full-cycle safety protection from "post-event response" to "pre-event prediction - in-event control - post-event guarantee", significantly improving the reliability of the new energy vehicle electrical system and the survival ability of occupants, and having the effects of high technical integration, strong scenario adaptability, and high safety redundancy.

[0055] 2. In the present invention, the safety evacuation floor is equipped with an emergency battery and a dual CAN bus to ensure that the key evacuation functions can still operate continuously after the main system power failure, solving the safety blind spot that the traditional system fails immediately after power failure and enhancing the survival ability in extreme scenarios. Multi-modal sensors such as fiber Bragg grating, ultrasonic, and self-powered composite sensors are used to monitor parameters such as battery micro-deformation, internal gas evolution, wire harness vibration, and temperature difference respectively, breaking through the limitation of single-parameter monitoring. Through data fusion analysis, a battery health index is constructed to improve the accuracy of early fault identification. The UWB radar is integrated with the camera to accurately judge the occupant's wakefulness and the availability of the evacuation path, solving the problem that the traditional evacuation system relies on manual operation or misjudgment of a single sensor. According to the occupant's status, a hierarchical evacuation response is triggered. When the occupant is awake, a first-level response is triggered, and the voice / screen guides the evacuation path and automatically unlocks the door, retaining the ability of independent evacuation. When the occupant is unconscious or the door fails, a second-level response is triggered, with dual redundancy of electromagnetic pulse window breaking and micro-explosion, combined with emergency lighting and voice commands, realizing the upgrade from "passive monitoring" to "active intervention" and breaking through the randomness and insufficient environmental adaptability of the traditional window-breaking device.

[0056] 3. In the present invention, a high-precision three-dimensional model including the geometric structure of the battery compartment and the wire harness layout is constructed, with an embedded electrolyte leakage model and a wire harness aging model. The former simulates the diffusion path based on fluid dynamics and predicts the short-circuit risk, and the latter combines vibration fatigue and insulation material aging analysis to achieve the prediction of fault evolution based on physical mechanisms, breaking through the lag of traditional threshold alarms. A dynamic threshold table is generated by combining environmental temperature, state of charge, and historical fault data to solve the problem that traditional fixed thresholds cannot adapt to complex working conditions, reducing the false alarm rate and missed alarm rate. A three-level response mechanism is set up, which dynamically adjusts the intervention intensity according to the severity of the fault, breaking through the rigid response of the traditional "either on or off", taking into account both safety and functional availability. Cross-domain technologies such as fiber optic sensing, digital twin, energy harvesting, and biometric detection are integrated to construct a technical closed-loop of "physical perception - digital mapping - intelligent decision - mechanical execution", breaking the single technical dependence of the traditional automotive safety system. The data before and after the fault is recorded for model iteration, the model parameters are corrected through historical cases, and the algorithm is dynamically updated in combination with the aging historical data to achieve the self-learning ability of the system, solving the problem of accuracy decay after long-term operation of the traditional fixed algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 is the system architecture diagram of the present invention;

[0058] Figure 2 is the system flow chart of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0059] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0060] Referring to Figure 1-2 , an embodiment provided by the present invention: a vehicle stop electrical safety monitoring system, including an intelligent perception layer, a digital twin decision-making layer, a hierarchical execution layer, and a safety escape layer. The intelligent perception layer includes a multi-modal sensor array for real-time collection of physical parameters, chemical parameters, and environmental parameters of the vehicle electrical system. The digital twin decision-making layer is based on a three-dimensional physical model of the vehicle electrical system, synchronizes the data of the intelligent perception layer in real time, and dynamically adjusts the fault threshold. The three-dimensional physical model is embedded with an electrolyte leakage model and a wire harness aging model. The hierarchical execution layer triggers a three-level response mechanism according to the diagnosis results of the digital twin decision-making layer, including local current limiting, high-voltage circuit isolation, and emergency braking lock-up. The safety escape layer integrates UWB radar, cameras, and emergency batteries for occupant status perception, environmental analysis, and hierarchical escape response.

[0061] The multi-modal sensor array is composed of fiber Bragg grating sensors, ultrasonic sensors, self-powered composite sensors, voltage / current sensors, and temperature sensors. The fiber Bragg grating sensors are embedded between battery modules to monitor the micro-deformation caused by thermal expansion and generate a deformation-time curve. The ultrasonic sensors detect the gas evolution inside the battery and generate a gas evolution spectrogram. The self-powered composite sensors collect energy through the piezoelectric-thermoelectric effect and monitor the vibration amplitude of the wire harness and the temperature difference at key nodes. The electrolyte leakage model is based on fluid dynamics to simulate the diffusion path of the electrolyte and predict the short-circuit risk. The wire harness aging model analyzes the vibration spectrum based on the rain flow counting method and calculates the aging rate of the insulating material according to the Arrhenius equation to predict the remaining life of the wire harness. The three-level response mechanism of the hierarchical execution layer includes a first-level response, a second-level response, and a third-level response. When a single sensor detects an abnormality, the first-level response is triggered to reduce the current in the fault area and start the liquid cooling system. When multiple sensors jointly alarm, the second-level response is triggered to isolate the high-voltage circuit through a solid-state relay and retain the power supply of the low-voltage system. When the electrolyte leakage model predicts a short-circuit risk, the third-level response is triggered to cut off the high-voltage power supply of the whole vehicle, cause the vehicle to perform emergency braking, and force the vehicle to stop.

[0062] The UWB radar detects the breathing frequency and heartbeat signal of the occupant through the Doppler effect. The camera identifies the occupant's posture and environmental obstacles, including the status of seat belt jamming and the degree of door deformation. The emergency battery powers the dual CAN buses, electromagnetic pulse window-breaking device, micro-explosion device, UWB radar, camera, audio, door electronic lock, and in-vehicle emergency lighting to ensure continuous operation after power failure. The electromagnetic pulse window-breaking device emits high-energy electromagnetic pulses to break the window closest to the occupant. The micro-explosion device serves as a backup window-breaking solution and is activated when the electromagnetic pulse window-breaking device fails. There is a dual CAN bus. The main control system transmits instructions through two independent buses. When one fails, the other can still transmit data. The hierarchical escape mechanism of the safety escape layer includes a primary escape response and a secondary escape response. When the occupant is awake and the door / window is not severely deformed, the primary escape response is triggered, and the escape route is prompted by voice and the door is automatically unlocked. When it is detected that the occupant is unconscious or the door fails, the secondary escape response is activated, triggering the electromagnetic pulse window-breaking and emergency lighting.

[0063] The working process of the intelligent perception layer includes three stages: data acquisition, signal processing, and data transmission. A multimodal sensor array composed of fiber Bragg grating sensors, ultrasonic sensors, self-powered composite sensors, voltage / current sensors, and temperature sensors is used to collect the physical and chemical parameters of the vehicle's electrical system in real time through the multimodal sensor array. High-precision monitoring is achieved by combining advanced sensing technologies, providing a data basis for subsequent fault diagnosis and decision-making. The fiber Bragg grating sensor monitors the thermal expansion micro-deformation and stress distribution between battery modules in real time, generating a deformation-time curve. A grating structure is formed through periodic refractive index changes. When the battery deforms due to temperature changes or internal pressure, the period of the grating changes accordingly. The wavelength shift is detected by a demodulator and converted into a deformation amount. The formula is ΔL = k·Δλ, where ΔL is the thermal expansion micro-deformation amount of the battery module, Δλ is the wavelength shift of the fiber Bragg grating, and k is the deformation-wavelength conversion coefficient, which is determined by the grating material characteristics and demodulator calibration. When the expansion deformation of the battery module caused by temperature rise exceeds the set value, a warning signal is triggered. The ultrasonic sensor scans the internal gas state of the battery every 5 seconds, monitors the gas evolution inside the battery, generates a gas evolution spectrogram, and determines the gas type through a pattern matching algorithm. Ultrasonic waves with a transmission frequency range of 20 - 200 kHz are emitted, and when they penetrate the battery shell and encounter a gas / liquid interface, they are reflected. The reflected wave signal is received, and the gas evolution amount is calculated by the time difference method. When the gas evolution rate is greater than the set value, it is determined that there is a risk of electrolyte leakage. The self-powered composite sensor monitors the vibration amplitude of the high-voltage wire harness and the temperature difference at key nodes, synchronously collects the vibration acceleration and temperature difference of the wire harness, and generates a vibration spectrogram. When the wire harness vibrates, piezoelectric materials generate charges through the piezoelectric effect and are converted into voltage signals. Using the Seebeck effect, electricity is generated by the temperature difference at both ends of the wire harness to charge the built-in supercapacitor. The comprehensive formula of the Seebeck effect is V 输出 = β·ΔT, V输出 V is the voltage output by the sensor, β is the Seebeck coefficient determined by the temperature difference characteristics of the materials at both ends of the wire harness, ΔT is the temperature difference at the key nodes of the wire harness, and it supports continuous operation for a period of time after power-off. When the vibration amplitude > 2g or the temperature difference > 10 °C, it is marked as an abnormal node. The voltage / current sensor collects the voltage and current data during the battery charging and discharging process in real time to monitor the dynamic changes. The temperature sensor is embedded in the battery cell and the wire harness node to synchronously record the temperature data of the battery cell and the module, which is used to compensate for the influence of temperature on the battery internal resistance. The data collected by the multi-modal sensor array is processed, and the noise interference is eliminated through Kalman filtering. The formula is is the state estimate value after denoising at the k-th moment, is the predicted state value at the k-th moment, K k is the Kalman gain, which controls the weight of the sensor data and the predicted value, z k is the original sensor observation value at the k-th moment, H is the observation matrix that maps the state to the sensor measurement space. Combining the deformation, gas, and vibration data, the abnormality is judged by the joint judgment of multiple sensor data. When deformation and gas abnormalities occur, it is judged as a thermal runaway risk. When vibration and temperature difference abnormalities occur, it is judged as wire harness aging or poor contact. The data is uploaded to the digital twin decision layer through the CAN bus, which supports two modes: real-time synchronization and offline caching. The self-powered sensor maintains data transmission through a supercapacitor after power-off.

[0064] The digital twin decision layer is the key link in the intelligent management of the entire vehicle electrical system. It is mainly responsible for receiving the data from the intelligent perception layer and using a variety of technologies and models for analysis, diagnosis, and prediction to provide decision-making basis for the hierarchical execution layer. It receives the deformation amount, gas evolution spectrogram, gas type, gas evolution rate, wire harness vibration acceleration and temperature difference, voltage and current, and the temperature data of the battery cell and the module collected and preprocessed by the intelligent perception layer. At the same time, combining the state of charge and health state data of the battery, it prepares for subsequent analysis. Combining the physical characteristics of the vehicle electrical system, including the battery anode and cathode materials, electrolyte composition, battery module spacing, pressure relief valve position, mechanical fatigue life of copper conductors, and deformation characteristics of insulating materials under temperature difference, a high-precision three-dimensional model is constructed using simulation tools. This three-dimensional model is embedded with an electrolyte leakage model and a wire harness aging model. Inputting the wire harness material properties and wire harness vibration amplitude parameters, the vibration spectrum is analyzed based on the rain flow counting method to calculate the fatigue life of the metal conductor. The formula is Δσ is the stress amplitude caused by vibration, extracted from the vibration spectrum analysis, C, m are material constants determined by the fatigue characteristics of the wire harness metal. Capturing the mechanical fatigue signal and evaluating the thermal stress, using the Arrhenius equation to evaluate the aging rate of the insulating material on the wire harness surface, and then a weighted comprehensive operation is performed by the metal fatigue life and the insulation aging rate to obtain the wire harness aging coefficient. The formula is C老化 = w1·L 金属疲劳 + w2·α 绝缘老化 , C 老化 is the wiring harness aging coefficient, used to quantify the overall aging degree of the wiring harness, L 金属疲劳 is the fatigue life of the metal conductor, calculated by the rainflow counting method, α 绝缘老化 is the aging rate of the insulating material, calculated by the Arrhenius equation, w1, w2 are weight coefficients, optimized and determined from historical wiring harness aging data, satisfying w1 + w2 = 1. Input the wiring harness material properties and the wiring harness vibration amplitude parameters, analyze the metal fatigue caused by vibration based on the rainflow counting method, calculate the thermal aging rate of the wiring harness insulating material through temperature difference data, use the historical wiring harness aging data to dynamically update the aging threshold, thereby constructing a wiring harness aging model for predicting the aging rate and contact failure risk of the wiring harness caused by vibration and temperature difference, obtaining the prediction of the remaining life of the wiring harness, and analyzing the vibration stress concentration area according to the wiring harness layout in the 3D model. Input the battery pack structure data and the physical properties parameters of the electrolyte, simulate the flow path of the electrolyte in the battery compartment through computational fluid dynamics, predict the impact of local temperature rise on adjacent modules after leakage through thermodynamic analysis, evaluate the short-circuit current characteristics after the electrolyte contacts the electrode through an electrochemical model, fuse the gas evolution data monitored by an ultrasonic sensor and the deformation data of a fiber Bragg grating sensor, and correct the model parameters through historical fault cases, thereby constructing an electrolyte leakage model for predicting the diffusion path, short-circuit risk and spatio-temporal evolution law after electrolyte leakage, predicting the short-circuit risk after 3 seconds. The electrolyte leakage model is based on the geometric structure of the battery compartment in the 3D model to determine the leakage diffusion boundary conditions. The 3D model updates the input parameters of the electrolyte leakage model and the wiring harness aging model according to the data collected by the real-time access intelligent perception layer. The change data of the battery internal resistance will trigger the electrochemical short-circuit prediction of the electrolyte leakage model. When the current undergoes a step change, calculate the battery internal resistance through the voltage response. The calculation formula is internal resistance = voltage change value / current change value. Calculate the voltage change value and the current change value through the voltage and current data of the battery. Since the battery internal resistance decreases with the increase of temperature, the temperature threshold will be increased in a low-temperature environment. The 3D model accesses the sensor data in real time, analyzes the sensor data, compares the sensor data with the 3D model to judge whether the data is abnormal, and continuously updates the input parameters of the electrolyte leakage and wiring harness aging models. Among them, the change data of the battery internal resistance in the 3D model will trigger the electrochemical short-circuit prediction of the electrolyte leakage model. Mark the electrolyte diffusion path and the high-risk area of wiring harness aging in the 3D model to assist engineers in intuitive diagnosis, and record the data 10 minutes before and after the fault for accident analysis and model optimization. The change data of the battery internal resistance will trigger the electrochemical short-circuit prediction of the electrolyte leakage model. The formula is ρ is the electrolyte density, defined by the physical property parameters of the electrolyte. u is the electrolyte flow velocity vector, solved through the boundary conditions of the battery compartment geometry. t is the time. The fault threshold is adjusted according to the real-time environmental parameters of temperature and state of charge. Based on the environmental temperature, state of charge, and historical fault records, a dynamic threshold table is generated to provide a more accurate judgment basis for fault diagnosis and achieve dynamic fault diagnosis.

[0065] The hierarchical execution layer, based on the data input from the intelligent perception layer and the digital twin decision layer, dynamically controls the safety protection operations of the vehicle's electrical system through a three-level response mechanism, achieving a step-by-step response from early warning to emergency stop. According to the type, quantity of sensor anomalies, and the digital twin prediction results, the response level is matched, and the response intensity is gradually increased from local current limiting to full vehicle power off, balancing safety and functional availability. The potential risk is the first-level response, and the trigger condition is the abnormal data collected by a single sensor. The current in the fault area is reduced through power electronic devices to suppress the risk of thermal runaway, and the coolant flow rate of the abnormal battery cell is directionally enhanced. The moderate fault is the second-level response, and the trigger condition is the combined alarm of multiple sensors. The high-voltage circuit isolation is executed, and the fault high-voltage circuit is cut off through a solid-state relay to maintain the low-voltage power supply functions such as window unlocking and lighting, and support the vehicle to enter the limp mode. The emergency fault is the third-level response, indicating that the trigger condition is the prediction of electrolyte leakage by the digital twin model. The high-voltage bus power supply is immediately cut off, and the whole vehicle is powered off at high voltage, causing the vehicle to perform emergency braking and forcing the vehicle to stop. At the same time, the door electronic lock is opened. The first-level response retains the vehicle's power function, the second-level response limits the power output, and the third-level response completely cuts off the power. The response level automatically upgrades with the severity of the fault. In the third-level response, the high-voltage power off takes precedence over mechanical braking to avoid secondary accidents caused by power-off delay.

[0066] The safety escape layer, through real-time status perception and hierarchical response mechanisms, can quickly identify the occupant status and environmental obstacles when a vehicle collision or electrical failure occurs, dynamically trigger escape operations, and ensure the safe evacuation of occupants. Combining the penetrative positioning of UWB radar and camera vision analysis, it accurately judges the vital signs of occupants and environmental obstacles, dynamically selects an escape route based on the occupant's consciousness state and the availability of doors / windows. Independent power supply and dual CAN bus communication ensure the continuous operation of the escape function after power failure. The UWB radar monitoring detects the breathing frequency and heartbeat signal of occupants through the Doppler effect, scans for minute movements of occupants, and judges coma or loss of mobility. With centimeter-level positioning accuracy, it can track the real-time position of occupants in the vehicle. The camera analyzes the occupant's posture, detects the status of seat belt jams and head injuries, identifies whether the window is blocked by obstacles and the degree of door deformation. It is equipped with a hierarchical response escape mechanism. The first-level response is active escape, and the trigger condition is that the occupant is conscious, has a normal breathing frequency, passes the motion response detection, and the camera determines that the door / window is not severely deformed. The escape route is displayed through the in-vehicle audio or screen, and the electronic lock is automatically released to support the occupant to manually open the door and evacuate. The second-level response is mechanical-assisted escape, and the trigger condition is that the occupant is detected to be in a coma, with a breathing frequency < 8 times / minute and no motion response, and the camera determines that the door is severely deformed or the electronic lock fails. The electromagnetic pulse window-breaking device is triggered to impact and break the window closest to the occupant. If the electromagnetic pulse window-breaking device fails, the window is broken through a micro-explosion device with controllable blasting force and a fragment splash range < 30 cm. The in-vehicle LED lights are automatically turned on to assist in escaping in a dark environment, and instructions such as "Please escape through the left window" are continuously broadcast. The emergency battery powers the UWB radar, camera, window-breaking device, emergency lighting, audio, display screen, electromagnetic pulse window-breaking device, micro-explosion device, and dual CAN bus to ensure continuous operation for ≥ 5 minutes after power failure, with priority given to ensuring the power of the window-breaking device and CAN bus. The escape system and the main control system transmit instructions through two independent buses. If one bus fails, the other bus can still transmit key instructions such as window-breaking and unlocking.

[0067] The system operation process is as follows:

[0068] S1. Data acquisition

[0069] The fiber Bragg grating sensor monitors the thermal expansion micro-deformation of the battery module, generates a deformation-time curve, and converts the wavelength shift into a deformation amount;

[0070] The ultrasonic sensor scans the internal gas state of the battery every 5 seconds, emits ultrasonic waves, generates a gas evolution spectrogram through the reflected wave signal, and calculates the gas evolution amount;

[0071] The self-powered composite sensor synchronously collects the vibration acceleration of the high-voltage harness and the temperature difference at key nodes, and generates a vibration spectrogram using the piezoelectric effect and the Seebeck effect;

[0072] The voltage / current sensor collects the dynamic data of voltage and current in real time during the charging and discharging process of the battery;

[0073] The temperature sensor records the temperature data of the battery cells and modules, compensating for the influence of temperature on the internal resistance of the battery;

[0074] S2. Data preprocessing

[0075] The noise interference of the sensor data is eliminated through Kalman filtering, and the effective signal is extracted. Combining the deformation amount, gas evolution rate, vibration spectrum and temperature difference data, a battery health index is constructed to provide input for subsequent modeling;

[0076] S3. Modeling

[0077] The digital twin decision-making layer receives the deformation amount, gas evolution spectrogram, gas type, gas evolution rate, wire harness vibration acceleration and temperature difference, voltage and current, temperature data of battery cells and modules. Based on the sensor data and battery physical characteristics, a high-precision three-dimensional model is constructed, and an electrolyte leakage model and a wire harness aging model are embedded in the high-precision three-dimensional model;

[0078] The electrolyte leakage model simulates the electrolyte diffusion path through fluid dynamics, and combines thermodynamics and electrochemical models to predict the short-circuit risk;

[0079] The wire harness aging model analyzes the metal fatigue life caused by vibration based on the rain flow counting method, calculates the thermal aging rate of insulating materials using the Arrhenius equation, and generates a wire harness aging coefficient by weighting;

[0080] S4. Data analysis

[0081] Electrolyte leakage risk diagnosis: Integrate deformation data and gas evolution spectrogram to judge the thermal runaway risk; Predict the short-circuit probability after 3 seconds through the electrochemical model;

[0082] Wire harness aging diagnosis: Combine vibration spectrum, temperature difference data and aging coefficient to evaluate the contact failure risk and predict the remaining life;

[0083] Dynamic threshold adjustment: Generate a dynamic threshold table according to the ambient temperature, state of charge and historical fault records to improve the diagnosis accuracy;

[0084] S5. Fault diagnosis

[0085] Thermal runaway risk: The triggering conditions are that the deformation amount exceeds the limit and the gas evolution rate is abnormal;

[0086] Wire harness abnormality: The triggering conditions are that the vibration amplitude > 2g or the temperature difference > 10°C;

[0087] Electrolyte leakage: The triggering condition is that the digital twin model predicts the short-circuit risk;

[0088] S6, Fault Classification Response Execution

[0089] Dynamically control safety protection operations through a three - level response mechanism:

[0090] Level 1 response: Potential risk. When a single sensor is abnormal, reduce the current in the fault area and directionally increase the coolant flow;

[0091] Level 2 response: Moderate fault. When multiple sensors jointly alarm, cut off the fault high - voltage circuit, maintain low - voltage power supply, and enter the limp - home mode;

[0092] Level 3 response: Emergency fault. When electrolyte leakage is predicted, immediately cut off the high - voltage bus power supply, trigger a full - vehicle power - off and emergency braking;

[0093] S7, Escape Response

[0094] Occupant status perception:

[0095] UWB radar: Detect the occupant's breathing frequency, heartbeat signal and centimeter - level positioning;

[0096] Camera: Analyze the occupant's posture, seat - belt status and the deformation degree of the doors / windows;

[0097] Trigger hierarchical escape:

[0098] Level 1 response: Active escape. When the occupant is conscious and the door is available, display the escape route and unlock the electronic lock;

[0099] Level 2 response: Mechanical - assisted escape. When the occupant is unconscious or the door is deformed, trigger the electromagnetic pulse to break the window or the micro - blasting device, and the LED lights are lit and continuous voice guidance is provided.

[0100] Finally, it should be noted that the above - mentioned are only the preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, for those skilled in the art, they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. Vehicle stop electrical safety monitoring system, characterized in that: It includes an intelligent perception layer, a digital twin decision layer, a hierarchical execution layer, and a safety escape layer. The intelligent perception layer includes a multi-modal sensor array for real-time collection of physical parameters, chemical parameters, and environmental parameters of the vehicle electrical system. The digital twin decision layer is based on a three-dimensional physical model of the vehicle electrical system, synchronizes the data of the intelligent perception layer in real time, and dynamically adjusts the fault threshold. The three-dimensional physical model is embedded with an electrolyte leakage model and a wire harness aging model. The hierarchical execution layer triggers a three-level response mechanism according to the diagnosis results of the digital twin decision layer, including local current limiting, high-voltage circuit isolation, and emergency braking lock-up. The safety escape layer integrates UWB radar, cameras, and emergency batteries for occupant status perception, environmental analysis, and hierarchical escape response.

2. The vehicle stop electrical safety monitoring system according to claim 1, wherein: The multi-modal sensor array is composed of fiber Bragg grating sensors, ultrasonic sensors, self-powered composite sensors, voltage / current sensors, and temperature sensors. The fiber Bragg grating sensors are embedded between battery modules to monitor the micro-deformation caused by thermal expansion and generate a deformation-time curve. The ultrasonic sensors detect the gas evolution inside the battery and generate a gas evolution spectrogram. The self-powered composite sensors collect energy through the piezoelectric-thermoelectric effect and monitor the vibration amplitude of the wire harness and the temperature difference at key nodes.

3. The vehicle stop electrical safety monitoring system according to claim 1, characterized in that: The electrolyte leakage model is based on fluid dynamics to simulate the diffusion path of the electrolyte and predict the short-circuit risk. The wire harness aging model analyzes the vibration spectrum based on the rainflow counting method and calculates the aging rate of the insulating material using the Arrhenius equation to predict the remaining life of the wire harness.

4. The vehicle stop electrical safety monitoring system according to claim 1, characterized in that: The three-level response mechanism of the hierarchical execution layer includes a first-level response, a second-level response, and a third-level response. When a single sensor detects an abnormality, the first-level response is triggered to reduce the current in the fault area and start the liquid cooling system. When multiple sensors jointly alarm, the second-level response is triggered to isolate the high-voltage circuit through a solid-state relay and retain the power supply of the low-voltage system. When the electrolyte leakage model predicts a short-circuit risk, the third-level response is triggered to cut off the high-voltage power supply of the whole vehicle, cause the vehicle to perform emergency braking, and force the vehicle to stop.

5. The vehicle stop electrical safety monitoring system according to claim 1, characterized in that: The UWB radar detects the breathing frequency and heartbeat signal of the occupant through the Doppler effect. The camera identifies the occupant's posture and environmental obstacles, including the status of seat belt jamming and the degree of door deformation. The emergency battery powers the dual CAN bus, electromagnetic pulse window-breaking device, micro-explosion device, UWB radar, camera, audio, door electronic lock, and in-vehicle emergency lighting to ensure continuous operation after power failure.

6. The vehicle stop electrical safety monitoring system according to claim 5, wherein: The electromagnetic pulse window-breaking device emits high-energy electromagnetic pulses to break the window closest to the occupant. The micro-explosion device is used as a backup window-breaking solution and is activated when the electromagnetic pulse window-breaking device fails. It is equipped with a dual CAN bus, and the main control system transmits instructions through two independent buses. When one fails, the other can still transmit data.

7. The vehicle stop electrical safety monitoring system according to claim 1, wherein: The hierarchical escape mechanism of the safety escape layer includes a first-level escape response and a second-level escape response. When the occupant is awake and the door / window is not severely deformed, the first-level escape response is triggered, and the escape path is prompted by voice and the door is automatically unlocked. When it is detected that the occupant is unconscious or the door fails, the second-level escape response is activated, triggering the electromagnetic pulse window-breaking and emergency lighting.

8. The vehicle stop electrical safety monitoring system according to claim 1, characterized in that: The system operation process is as follows: S1. Data collection The fiber Bragg grating sensor monitors the thermal expansion micro-deformation of the battery module, generates a deformation-time curve, and converts the wavelength shift into a deformation amount; The ultrasonic sensor scans the internal gas state of the battery every 5 seconds, emits ultrasonic waves, generates a gas evolution spectrogram through the reflected wave signal, and calculates the gas evolution amount; The self-powered composite sensor synchronously collects the vibration acceleration of the high-voltage harness and the temperature difference at the key nodes, and generates a vibration spectrogram using the piezoelectric effect and the Seebeck effect; The voltage / current sensor real-time collects the voltage and current dynamic data during the charging and discharging process of the battery; The temperature sensor records the temperature data of the battery cells and modules, and compensates for the influence of temperature on the internal resistance of the battery; S2. Data preprocessing The noise interference of the sensor data is eliminated through Kalman filtering, the effective signals are extracted, and a battery health index is constructed by combining the deformation amount, gas evolution rate, vibration spectrum, and temperature difference data, providing input for subsequent modeling; S3. Modeling The digital twin decision layer receives the deformation amount, gas evolution spectrogram, gas type, gas evolution rate, vibration acceleration and temperature difference of the harness, voltage and current, temperature data of the battery cells and modules, and constructs a high-precision three-dimensional model based on the sensor data and the physical characteristics of the battery. The high-precision three-dimensional model embeds an electrolyte leakage model and a harness aging model; The electrolyte leakage model simulates the diffusion path of the electrolyte through fluid dynamics, and predicts the short-circuit risk by combining thermodynamics and electrochemistry models; The harness aging model analyzes the metal fatigue life caused by vibration based on the rain flow counting method, calculates the thermal aging rate of the insulating material using the Arrhenius equation, and generates a harness aging coefficient by weighting; S4. Data analysis Electrolyte leakage risk diagnosis: Integrate the deformation data and the gas evolution spectrogram to judge the thermal runaway risk; Predict the short-circuit probability after 3 seconds through the electrochemistry model; Harness aging diagnosis: Combine the vibration spectrum, temperature difference data and aging coefficient to evaluate the contact failure risk and predict the remaining life; Dynamic threshold adjustment: Generate a dynamic threshold table according to the ambient temperature, state of charge and historical fault records to improve the diagnosis accuracy; S5. Fault diagnosis Thermal runaway risk: The trigger condition is that the deformation amount exceeds the limit and the gas evolution rate is abnormal; Harness abnormality: The trigger condition is that the vibration amplitude > 2g or the temperature difference > 10°C; Electrolyte leakage: The trigger condition is that the digital twin model predicts the short-circuit risk; S6. Fault classification response execution Dynamically control the safety protection operation through a three-level response mechanism: First-level response: Potential risk. When a single sensor is abnormal, reduce the current in the fault area and direct the enhanced coolant flow; Second-level response: Moderate fault. When multiple sensors jointly alarm, cut off the fault high-voltage circuit, maintain the low-voltage power supply, and enter the limp mode; Third-level response: Emergency fault. When predicting electrolyte leakage, immediately cut off the high-voltage bus power supply, trigger a full vehicle power-off and emergency braking; S7. Escape response Occupant status perception: UWB radar: Detect the breathing frequency, heartbeat signal and centimeter-level positioning of the occupants; Camera: Analyze the occupant posture, seat belt status and the deformation degree of the doors / windows; Trigger hierarchical escape: First-level response: Active escape. When the occupants are conscious and the doors are available, display the escape route and unlock the electronic locks; Secondary response: In case of mechanical assisted escape, when the occupant is unconscious or the car door is deformed, an electromagnetic pulse window breaker or a micro-explosion device is triggered, and the LED light is lit and continuous voice guidance is provided.

Images