Intelligent cabin power supply fault detection system and method for realizing functional safety
The AI fault prediction module monitors the power status in real time and performs multi-level fault response, solving the system paralysis problem caused by single-point failures in traditional smart cockpit power systems, achieving power supply stability and fault tolerance, and meeting the ISO 26262 standard.
Patent Information
- Application Number
- CN202511009677.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-22
- Publication Date
- 2025-09-26
AI Technical Summary
Traditional smart cockpit power systems lack real-time protection and are prone to system paralysis due to single-point failures. They cannot meet the requirements of the ISO 26262 functional safety standard and find it difficult to take into account the differentiated power requirements of multiple modules.
A system including an AI fault prediction module, a safety monitoring MCU module, a SOC module, and a fan module is used. The voltage, current, and temperature are monitored in real time through the AD acquisition module. Fault prediction is performed using software-preset thresholds, and control instructions are issued according to the fault level to implement a multi-level fault response mechanism.
Ensure the power supply stability and redundancy of the intelligent cockpit system, meet the ISO 26262 functional safety standard, and achieve fault self-diagnosis and rapid recovery.
Smart Images
Figure CN120703629A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of automotive electronic intelligent cockpit power supplies, and in particular to an intelligent cockpit power supply fault detection system and method for achieving functional safety. Background Art
[0002] Traditional smart cockpit power systems lack real-time protection for advanced smart cockpit functions (such as multi-screen linkage, voice interaction, and sensor power supply), and are prone to system paralysis due to single-point failures (such as chip overheating and line short circuit).
[0003] Existing power management solutions do not fully meet the automotive safety integrity level (ASIL) requirements of the ISO 26262 functional safety standard, and are unable to achieve self-diagnosis and rapid recovery from faults. The power requirements of multiple modules in the smart cockpit (such as the MCU, display, and sensors) vary widely, making it difficult for traditional linear voltage regulation or simple DC-DC conversion to achieve a balance between efficiency and dynamic response. Summary of the Invention
[0004] The purpose of the present invention is to overcome the shortcomings of the existing technology and provide a smart cockpit power supply fault detection system and method that achieves functional safety, ensuring the power supply stability, redundancy and fault tolerance of the smart cockpit system to meet the requirements of the automotive safety integrity level in the ISO26262 functional safety standard.
[0005] One aspect of the present invention is achieved as follows: an intelligent cockpit power failure detection system that achieves functional safety includes three independent power supply modules, an AI fault prediction module, a SOC module, a safety monitoring MCU module, and a fan module;
[0006] The AI fault prediction module has a built-in AD acquisition module, which is used to sample the voltage and current of each power supply and the temperature of each power supply module and high-power device in real time, and transmit the collected voltage, current and temperature signals to the AI fault prediction module;
[0007] The AI fault prediction module is electrically connected to the three independent power supply modules and is used to monitor the power supply status of the entire system and make predictions based on the monitoring data. Specifically, the AI fault prediction chip determines the power supply fault prediction information based on the output of each power supply module and the requirements of the subsequent functional modules through software-preset voltage and current upper and lower thresholds and maximum temperature, and compares them with the sampled data of the AD acquisition module, and transmits the fault prediction results to the safety monitoring MCU module.
[0008] The safety monitoring MCU module is used to receive fault prediction information from the AI fault prediction module, determine the fault safety level, and issue corresponding control instructions to control the working status of the power supply and functional modules according to the safety level;
[0009] The SOC module is electrically connected to the safety monitoring MCU module, and is used to receive control instructions from the safety monitoring MCU module and adjust the operating frequency according to the control instructions;
[0010] The fan module is electrically connected to the safety monitoring MCU module and is used to cool the system according to the control instructions issued by the safety monitoring MCU module.
[0011] Furthermore, the three independent power supply modules include a multi-phase BUCK converter and a multi-channel LDO unit, an LDO unit and a dynamic load matrix unit; the multi-phase BUCK converter and the multi-channel LDO unit are used to provide voltage and current requirements for the subsequent functional modules; the LDO unit is used to power the security monitoring MCU module and the AI fault prediction module; the dynamic matrix unit is used to dynamically distribute and control power to the peripheral modules.
[0012] Furthermore, the AI fault prediction module is electrically connected to the safety monitoring MUC module through an IIC interface.
[0013] Furthermore, the SOC module is electrically connected to the security monitoring MCU module through the SPI interface and the GPIO interface.
[0014] Another aspect of the present invention is achieved as follows: a method for detecting power failure in an intelligent cockpit to achieve functional safety, comprising the following steps:
[0015] 1) The AI fault prediction module uses a built-in high-precision AD acquisition module to sample the voltage and current of each power supply, as well as the temperature of each power supply module and high-power device in real time, and feeds the collected voltage, current, and temperature signals to the AI fault prediction module. The AI fault prediction module performs fault prediction based on the collected voltage, current, and temperature data.
[0016] 2) The AI fault prediction module determines power supply overvoltage, undervoltage, overcurrent, and short circuit faults based on the output of each power supply and the requirements of the subsequent functional modules, using the software-preset voltage upper and lower thresholds and current upper thresholds, and comparing them with the voltage and current data collected by the AD acquisition module;
[0017] 3) The AI fault prediction module presets a maximum temperature value through software and compares it with the temperature data collected by the AD acquisition module. When the temperature reaches a certain range, it triggers system frequency reduction or shutdown protection;
[0018] 4) The AI fault prediction module transmits the fault prediction results to the safety monitoring MCU module; the safety monitoring MCU module is used to receive the fault information from the AI fault prediction module, determine the fault safety level, and issue corresponding control instructions based on the safety level to control the working status of the power supply and functional modules.
[0019] Furthermore, the real-time sampling of the voltage of each power supply in step 1) specifically includes: output voltage response quantization, voltage drop depth ΔV dip =V dip / V normal ×100%, which quantifies the depth of the dynamic voltage deviation from the rated value by calculating the voltage drop depth as a percentage of the rated voltage. When the voltage drop depth is less than 5%, no action is required. When the voltage drop depth is between 5% and 10%, a log is recorded and an alert is reported to the safety monitoring MCU module, requiring the load current to be reduced. When the voltage drop depth is greater than 10%, the power supply output has failed, and a serious fault is reported to the safety monitoring MCU module, requiring the power supply to be forcibly shut down.
[0020] Furthermore, the real-time sampling of the current of each power supply in step 1) specifically includes: load current step detection, using the forward differential calculation method, di / dt = {I(t+Δt)-I(t)} / Δt, calculating the current change rate, and judging the stability of the load operation; detection time Δt = 1us, load current is 1A, when the current change rate is less than 1×10 5 A / s, no action is required; when the current change rate is greater than 1×10 5 A / s, and less than 2×10 5 A / s, record the log and report to the safety monitoring MCU module for early warning; when the current change rate is greater than 2×10 5 A / s, the load is working abnormally, reporting a serious fault to the safety monitoring MCU module, and forcibly shutting down the load device corresponding to this power supply.
[0021] Furthermore, the real-time sampling of the temperature of each power supply module and high-power device in step 1) specifically includes: when the temperature continues to rise and approaches the upper limit of the preset temperature, a warning signal is sent to the safety monitoring MCU module, and the MCU controls the fan speed to cool the system; when the temperature is greater than the upper limit of the preset temperature, it is determined that the power device is abnormal, and a serious fault is reported to the safety monitoring MCU module, and the power output or the corresponding load is turned off.
[0022] Furthermore, the step 4) specifically includes: the security monitoring MCU module determines the severity of the fault based on the fault information provided by the AI fault prediction module:
[0023] When the AI fault prediction module reports multiple severe faults, indicating multiple power supply or load short circuits or failure of multiple functional modules, the safety monitoring MCU module determines the fault level as Level 3, a catastrophic fault. It shuts down the faulty circuit, leaving only the safety monitoring MCU module active to maintain eCall power supply. A hard reset of the system restores the system to normal operation.
[0024] When the AI fault prediction module reports a single severe fault or multiple warning fault signals, the safety monitoring MCU module determines the fault level as a Level 2 severe fault. At this time, the peripheral power supply is turned off, the SOC module maintains basic power supply, and reports the fault code.
[0025] When the AI fault prediction module reports a single early warning fault signal, the safety monitoring MCU determines the fault level as Level 1, which is a recoverable fault; adjusts the operating frequency of the SOC module, dynamically adjusts the voltage, shuts down non-critical loads, and reduces system power.
[0026] Furthermore, in the fault level Level 2 and fault level Level 1, the safety monitoring MCU module indirectly monitors whether the fault is recovered. If the fault persists, it enters the safe state and the system shuts down; if the fault is recovered, it enters the continuous fault detection state and restores the system to normal working mode.
[0027] The present invention adopts the above technical solution, and compared with the existing technology, the beneficial effect is as follows: the present invention uses the AI fault prediction module to collect voltage, current and temperature data of three independent power supply modules through the built-in AD acquisition module, and then the AI fault prediction module compares with the preset voltage threshold, current threshold and temperature maximum value to predict the fault, and transmits the predicted information to the safety monitoring MCU module. The MCU module determines the fault safety level based on the fault information of the AI fault prediction module, and issues corresponding control instructions to control the working status of the power supply and functional modules according to the safety level. The three independent power supplies ensure that single point failures do not affect the operation of core functions. The multi-level fault response mechanism of the present invention ensures the power supply stability, redundancy and fault tolerance of the intelligent cockpit system to meet the requirements of the ISO 26262 functional safety standard for ASIL (Automotive Safety Integrity Level). BRIEF DESCRIPTION OF THE DRAWINGS
[0028] Figure 1 System block diagram of the present invention.
[0029] Figure 2 Schematic diagram of the paths of three independent power supply modules of the present invention.
[0030] Figure 3 Flowchart of the multi-stage fault response mechanism of the present invention. DETAILED DESCRIPTION
[0031] like Figure 1 The figure shows an intelligent cockpit power fault detection system that implements functional safety, including three independent power supply modules, an AI fault prediction module, a SOC module, a safety monitoring MCU module, and a fan module.
[0032] The AI fault prediction module has a built-in AD acquisition module, which is used to sample the voltage and current of each power supply and the temperature of each power supply module and high-power device in real time, and transmit the collected voltage, current and temperature signals to the AI fault prediction module;
[0033] The AI fault prediction module is electrically connected to the three independent power supply modules and is used to monitor the power status of the entire system and make predictions based on the monitoring data. Specifically, the AI fault prediction chip determines the power supply fault prediction information based on the output of each power supply module and the requirements of the subsequent functional modules through software-preset voltage and current upper and lower thresholds, as well as the maximum temperature. By comparing the data with the sampling data of the AD acquisition module, the AI fault prediction chip determines the power supply fault prediction information and transmits the fault prediction results to the safety monitoring MCU module.
[0034] The safety monitoring MCU module is used to receive fault prediction information from the AI fault prediction module, determine the fault safety level, and issue corresponding control instructions based on the safety level to control the working status of the power supply and functional modules;
[0035] The SOC module is electrically connected to the safety monitoring MCU module, and is used to receive control instructions from the safety monitoring MCU module and adjust the operating frequency according to the control instructions;
[0036] The fan module is electrically connected to the safety monitoring MCU module and is used to cool the system according to the control instructions issued by the safety monitoring MCU module.
[0037] The three independent power supply modules include a multi-phase BUCK converter and a multi-channel LDO unit, an LDO unit and a dynamic load matrix unit; the multi-phase BUCK converter and the multi-channel LDO unit are used to provide voltage and current requirements for the subsequent functional modules; the LDO unit is used to power the security monitoring MCU module and the AI fault prediction module; the dynamic matrix unit is used to dynamically distribute and control power to the peripheral modules.
[0038] The AI fault prediction module is electrically connected to the safety monitoring MUC module through the IIC interface; the SOC module is electrically connected to the safety monitoring MCU module through the SPI interface and GPIO interface.
[0039] A method for detecting power failure in a smart cockpit to achieve functional safety includes the following steps:
[0040] 1) The AI fault prediction module uses a built-in high-precision AD acquisition module to sample the voltage and current of each power supply, as well as the temperature of each power supply module and high-power device in real time, and feeds the collected voltage, current, and temperature signals to the AI fault prediction module. The AI fault prediction module performs fault prediction based on the collected voltage, current, and temperature data.
[0041] Real-time sampling of the voltage of each power supply includes: output voltage response quantization, voltage drop depth ΔV dip =V dip / V normal ×100%, which quantifies the depth of the dynamic voltage deviation from the rated value by calculating the voltage drop depth as a percentage of the rated voltage. When the voltage drop depth is less than 5%, no action is required. When the voltage drop depth is between 5% and 10%, a log is recorded and an alert is sent to the safety monitoring MCU module to reduce the load current. When the voltage drop depth is greater than 10%, the power supply output is faulty, and a serious fault is reported to the safety monitoring MCU module, requiring the power supply to be forcibly shut down.
[0042] The real-time sampling of the current of each power supply specifically includes: load current step detection, using the forward differential calculation method, di / dt = {I(t+Δt)-I(t)} / Δt, to calculate the current change rate and judge the stability of the load operation; the detection time Δt = 1us, the load current is 1A, when the current change rate is less than 1×10 5 A / s, no action is required; when the current change rate is greater than 1×10 5 A / s, and less than 2×10 5 A / s, record the log and report to the safety monitoring MCU module for early warning; when the current change rate is greater than 2×10 5 A / s, the load is working abnormally, reporting a serious fault to the safety monitoring MCU module, and forcibly shutting down the load device corresponding to this power supply.
[0043] Real-time sampling of the temperature of each power supply module and high-power device specifically includes: when the temperature continues to rise and approaches the preset temperature upper limit, a warning signal is sent to the safety monitoring MCU module, and the MCU controls the fan speed to cool the system; when the temperature is greater than the preset temperature upper limit, it is determined that the power device is abnormal, and a serious fault is reported to the safety monitoring MCU module, and the power output or the corresponding load is turned off.
[0044] 2) The AI fault prediction module determines power supply overvoltage, undervoltage, overcurrent, and short circuit faults based on the output of each power supply and the requirements of the subsequent functional modules, using the software-preset voltage upper and lower thresholds and current upper thresholds, and comparing them with the voltage and current data collected by the AD acquisition module;
[0045] 3) The AI fault prediction module presets a maximum temperature value through software and compares it with the temperature data collected by the AD acquisition module. When the temperature reaches a certain range, it triggers system frequency reduction or shutdown protection;
[0046] 4) The AI fault prediction module transmits the fault prediction results to the safety monitoring MCU module; the safety monitoring MCU module is used to receive the fault information from the AI fault prediction module, determine the fault safety level, and issue corresponding control instructions based on the safety level to control the working status of the power supply and functional modules.
[0047] When the AI fault prediction module reports multiple severe faults, indicating multiple power supply or load short circuits or failure of multiple functional modules, the safety monitoring MCU module determines the fault level as Level 3, a catastrophic fault. It shuts down the faulty circuit, leaving only the safety monitoring MCU module active to maintain eCall power supply. A hard reset of the system restores the system to normal operation.
[0048] When the AI fault prediction module reports a single severe fault or multiple warning fault signals, the safety monitoring MCU module determines the fault level as a Level 2 severe fault. At this time, the peripheral power supply is turned off, the SOC module maintains basic power supply, and reports the fault code.
[0049] When the AI fault prediction module reports a single early warning fault signal, the safety monitoring MCU determines the fault level as Level 1, which is a recoverable fault; adjusts the operating frequency of the SOC module, dynamically adjusts the voltage, shuts down non-critical loads, and reduces system power.
[0050] In fault levels Level 2 and Level 1, the safety monitoring MCU module indirectly monitors whether the fault is recovered. If the fault persists, it enters the safe state and the system shuts down; if the fault is recovered, it enters the continuous fault detection state and restores the system to normal working mode.
[0051] When the present invention works, Figure 2 As shown, the power input is divided into three independent paths: a main power supply path 100 , a safety island path 200 , and a peripheral path 300 .
[0052] The main power supply path 100 includes a multi-phase BUCK converter and multiple LDOs 101 for providing voltage and current requirements for subsequent functional modules;
[0053] The latter module includes the SOC module 102, which is used to realize the operation of the entire system function;
[0054] DDR memory chip 103, used for system data storage;
[0055] Other functional modules 104 are modules for implementing other functions (such as navigation, BT / WIFI, Ethernet, etc.).
[0056] The safety island path 200 includes: an independent low-dropout linear regulator LDO 201, which is used to power the safety monitoring MCU module and the AI fault prediction module;
[0057] Safety monitoring MCU module 202, used to receive fault information, determine the fault safety level, and issue corresponding control instructions to control the working status of the power supply and functional modules according to the safety level;
[0058] The AI fault prediction module 203 is used to monitor the power status of the entire system, analyze and predict the faults based on the monitoring data, and provide the fault prediction results to the safety monitoring MCU module 202;
[0059] eCall 204 is used to implement the emergency call function.
[0060] The peripheral path 300 includes: a dynamic load matrix power supply 301 for dynamically distributing and controlling power to the peripheral modules;
[0061] The peripheral modules include: a display screen 302 (main screen, instrument, sub-screen, etc.) for display and human-computer interaction; a HUD 303 for head-up display of instrument information; and a camera 304 for monitoring the surrounding environment of the vehicle.
[0062] like Figure 3 As shown, the safety MCU detects faults in the three independent power supply paths; when no fault is detected, the system operates normally;
[0063] When a fault is detected, take appropriate countermeasures based on the fault level:
[0064] When the safety monitoring MCU module determines that the fault level is Level 3 (for example, a short circuit occurs, a multi-level function fails, etc.), it shuts down the faulty circuit, leaving only the safety monitoring MCU module, maintaining the eCall power supply, and returning to normal through a system hard reset.
[0065] When the safety monitoring MCU module determines that the fault mode is Level 2 (such as continuous overvoltage, overcurrent, etc.), it turns off the power supply of the peripherals, the SOC maintains the basic power supply, and reports the fault code;
[0066] When the safety monitoring MCU module determines that the fault mode is Level 1 (such as instantaneous under-voltage, slight overtemperature, etc.), it adjusts the SOC operating frequency, dynamically adjusts the voltage, turns off non-critical loads, and reduces system power. For Level 2 and Level 1 fault modes, the safety monitoring MCU module indirectly monitors whether the fault has been recovered. If the fault persists, it enters a safe state and shuts down the system. If the fault is recovered, it enters a continuous fault detection state and restores the system to normal operating mode.
[0067] The present invention is not limited to the above-mentioned embodiments. On the basis of the technical solutions disclosed in the present invention, those skilled in the art can make some substitutions and modifications to some of the technical features therein according to the disclosed technical content without creative labor, and these substitutions and modifications are all within the protection scope of the present invention.
Claims
1. A smart cockpit power failure detection system that achieves functional safety, characterized by: It includes three independent power supply modules, AI fault prediction module, SOC module, safety monitoring MCU module and fan module; The AI fault prediction module has a built-in AD acquisition module, which is used to sample the voltage and current of each power supply and the temperature of each power supply module and high-power device in real time, and transmit the collected voltage, current and temperature signals to the AI fault prediction module; The AI fault prediction module is electrically connected to the three independent power supply modules and is used to monitor the power supply status of the entire system and make predictions based on the monitoring data. Specifically, the AI fault prediction chip determines the power supply fault prediction information based on the output of each power supply module and the requirements of the subsequent functional modules through software-preset voltage and current upper and lower thresholds and maximum temperature, and compares them with the sampled data of the AD acquisition module, and transmits the fault prediction results to the safety monitoring MCU module. The safety monitoring MCU module is used to receive fault prediction information from the AI fault prediction module, determine the fault safety level, and issue corresponding control instructions to control the working status of the power supply and functional modules according to the safety level; The SOC module is electrically connected to the safety monitoring MCU module, and is used to receive control instructions from the safety monitoring MCU module and adjust the operating frequency according to the control instructions; The fan module is electrically connected to the safety monitoring MCU module and is used to cool the system according to the control instructions issued by the safety monitoring MCU module.
2. The intelligent cockpit power processing system for achieving functional safety according to claim 1, characterized in that: The three independent power supply modules include a multi-phase BUCK converter and a multi-channel LDO unit, an LDO unit and a dynamic load matrix unit; the multi-phase BUCK converter and the multi-channel LDO unit are used to provide voltage and current requirements for the subsequent functional modules; the LDO unit is used to power the security monitoring MCU module and the AI fault prediction module; the dynamic matrix unit is used to dynamically distribute and control power to the peripheral modules.
3. The intelligent cockpit power processing system for achieving functional safety according to claim 1, characterized in that: The AI fault prediction module is electrically connected to the safety monitoring MUC module through the IIC interface.
4. The intelligent cockpit power processing system for achieving functional safety according to claim 1, characterized in that: The SOC module is electrically connected to the safety monitoring MCU module through the SPI interface and the GPIO interface.
5. A method for detecting power failure in an intelligent cockpit to achieve functional safety, characterized in that: The following steps are involved: 1) The AI fault prediction module uses a built-in high-precision AD acquisition module to sample the voltage and current of each power supply, as well as the temperature of each power supply module and high-power device in real time, and feeds the collected voltage, current, and temperature signals to the AI fault prediction module. The AI fault prediction module performs fault prediction based on the collected voltage, current, and temperature data. 2) The AI fault prediction module determines power supply overvoltage, undervoltage, overcurrent, and short circuit faults based on the output of each power supply and the requirements of the subsequent functional modules, using the software-preset voltage upper and lower thresholds and current upper thresholds, and comparing them with the voltage and current data collected by the AD acquisition module; 3) The AI fault prediction module presets a maximum temperature value through software and compares it with the temperature data collected by the AD acquisition module. When the temperature reaches a certain range, it triggers system frequency reduction or shutdown protection; 4) The AI fault prediction module transmits the fault prediction results to the safety monitoring MCU module; the safety monitoring MCU module is used to receive the fault information from the AI fault prediction module, determine the fault safety level, and issue corresponding control instructions based on the safety level to control the working status of the power supply and functional modules.
6. The method for detecting power failure in an intelligent cockpit to achieve functional safety according to claim 5, characterized in that: The real-time sampling of the voltage of each power supply in step 1) specifically includes: output voltage response quantization, voltage drop depth ΔV dip =V dip / V normal ×100%, which quantifies the depth of the dynamic voltage deviation from the rated value by calculating the voltage drop depth as a percentage of the rated voltage. When the voltage drop depth is less than 5%, no action is required. When the voltage drop depth is between 5% and 10%, a log is recorded and an alert is reported to the safety monitoring MCU module, requiring the load current to be reduced. When the voltage drop depth is greater than 10%, the power supply output has failed, and a serious fault is reported to the safety monitoring MCU module, requiring the power supply to be forcibly shut down.
7. The method for detecting power failure in a smart cockpit to achieve functional safety according to claim 5, characterized in that: The real-time sampling of the current of each power supply in step 1) specifically includes: load current step detection, using the forward differential calculation method, di / dt = {I(t+Δt)-I(t)} / Δt, calculating the current change rate, and judging the stability of the load operation; detection time Δt = 1us, load current is 1A, when the current change rate is less than 1×10 5 A / s, no action is required; when the current change rate is greater than 1×10 5 A / s, and less than 2×10 5 A / s, record the log and report to the safety monitoring MCU module for early warning; when the current change rate is greater than 2×10 5 A / s, the load is working abnormally, reporting a serious fault to the safety monitoring MCU module, and forcibly shutting down the load device corresponding to this power supply.
8. The method for detecting power failure in a smart cockpit to achieve functional safety according to claim 5, characterized in that: The real-time sampling of the temperature of each power supply module and high-power device in the step 1) specifically includes: when the temperature continues to rise and approaches the upper limit of the preset temperature, a warning signal is sent to the safety monitoring MCU module, and the MCU controls the fan speed to cool the system; when the temperature is greater than the upper limit of the preset temperature, it is determined that the power device is abnormal, and a serious fault is reported to the safety monitoring MCU module, and the power output or the corresponding load is turned off.
9. The method for detecting power failure in an intelligent cockpit to achieve functional safety according to claim 5, characterized in that: The step 4) specifically includes: the security monitoring MCU module determines the severity of the fault based on the fault information provided by the AI fault prediction module: When the AI fault prediction module reports multiple severe faults, indicating multiple power supply or load short circuits or failure of multiple functional modules, the safety monitoring MCU module determines the fault level as Level 3, a catastrophic fault. It shuts down the faulty circuit, leaving only the safety monitoring MCU module active to maintain eCall power supply. A hard reset of the system restores the system to normal operation. When the AI fault prediction module reports a single severe fault or multiple warning fault signals, the safety monitoring MCU module determines the fault level as a Level 2 severe fault. At this time, the peripheral power supply is turned off, the SOC module maintains basic power supply, and reports the fault code. When the AI fault prediction module reports a single early warning fault signal, the safety monitoring MCU determines the fault level as Level 1, which is a recoverable fault; adjusts the operating frequency of the SOC module, dynamically adjusts the voltage, shuts down non-critical loads, and reduces system power.
10. The method for detecting power failure in an intelligent cockpit to achieve functional safety according to claim 9, characterized in that: In the fault level 2 and fault level 1, the safety monitoring MCU module indirectly monitors whether the fault is recovered. If the fault persists, the system enters a safe state and shuts down. If the fault is recovered, the system enters a continuous fault detection state and returns to normal operation mode.