CN-122018486-A - Multi-path load intelligent diagnosis and protection method for vehicle body controller

CN122018486ACN 122018486 ACN122018486 ACN 122018486ACN-122018486-A

Abstract

The invention discloses a multipath load intelligent diagnosis and protection method of a vehicle body controller, and relates to the technical field of vehicle controllers; the method comprises the steps of synchronously obtaining a current and voltage signal of a target load and a bus voltage value of a vehicle-mounted power supply, carrying out decoupling analysis on the current and voltage sequence under a load cooperative working condition to obtain a load characteristic vector, comparing the load characteristic vector with a preset normal characteristic threshold value, judging whether the target load has abnormal change, analyzing the bus voltage sequence to obtain a power supply characteristic vector, inputting the two types of characteristic vectors into a load power supply coupling noise identification model to obtain a correlation judgment result, and executing protection action of the target load according to the correlation judgment result. According to the method, through the association analysis of the load and the power supply characteristics, the power supply coupling noise and the independent fault of the load can be accurately distinguished, the error protection caused by power supply fluctuation is avoided, the corresponding protection action can be matched aiming at the real fault, and the accuracy of vehicle-mounted load diagnosis is improved.

Inventors

ZHOU XIANGDONG
HE LIHUA
ZHOU CHEN

Assignees

江苏罗思韦尔电气有限公司

Dates

Publication Date: 20260512
Application Date: 20260202

Claims (7)

1. The intelligent diagnosis and protection method for the multipath load of the vehicle body controller is characterized by comprising the following steps of: synchronously acquiring a current signal and a voltage signal of the target load, and a bus voltage sequence of a vehicle-mounted power supply bus voltage value, wherein the target load is any one load of all loads; Carrying out load collaborative condition decoupling analysis on the current and voltage sequence to obtain a load characteristic vector; Comparing the target load characteristic vector with a preset normal characteristic threshold value, and judging whether the target load has abnormal change or not; if abnormal changes exist, carrying out multidimensional feature analysis on the bus voltage sequence to obtain a power supply feature vector; Inputting the load characteristic vector and the power supply characteristic vector into a preset load power supply coupling noise identification model to obtain a correlation judgment result; and performing protection action on the target load according to the association judgment result.
2. The method for intelligently diagnosing and protecting the multiple loads of the vehicle body controller according to claim 1, wherein the step of performing load collaborative condition decoupling analysis on the current-voltage sequence to obtain the load characteristic vector comprises the following steps: Carrying out load operation mode clustering on the current and voltage sequences to obtain a sub-current and voltage sequence set formed by sub-current and voltage sequences respectively corresponding to each load exclusive operation mode; constructing a modal dynamic constraint function for a target sub-current voltage sequence, wherein the target sub-current voltage sequence is any sub-current voltage sequence in the sub-current voltage sequence set; performing time sequence feature analysis and parameter extraction on the target sub-current voltage sequence according to the modal dynamic constraint function to obtain a modal constraint feature sequence; Calculating the trend matching degree of the mode constraint feature sequences corresponding to different load exclusive operation modes to obtain cross-mode association features; And carrying out weighted fusion on the modal constraint feature sequence and the cross-modal association feature according to the duty ratio of the sub-current voltage sequence corresponding to each load exclusive operation mode in the current voltage sequence to obtain a load feature vector.
3. The method for intelligently diagnosing and protecting the multiple loads of the vehicle body controller according to claim 2, wherein constructing the modal dynamic constraint function on the target sub-current voltage sequence comprises the following steps: The modal dynamic constraint function is Wherein, the method comprises the steps of, Let t be the instantaneous rate of change of current versus voltage, Is the voltage value at the time t, Is a reference voltage in the exclusive operation mode of the load.
4. The method for intelligently diagnosing and protecting the multiple loads of the vehicle body controller according to claim 2, wherein the step of performing time sequence feature analysis and parameter extraction on the target sub-current voltage sequence according to the modal dynamic constraint function to obtain a modal constraint feature sequence comprises the following steps: Substituting the time sequence parameters of the target sub-current voltage sequence into a modal dynamic constraint function to obtain a constraint function value sequence corresponding to the target sub-current voltage sequence; Performing time sequence segmentation on the constraint function value sequence to obtain a constraint function segmentation sequence set; Traversing the maximum value and the minimum value of the target constraint function segmentation sequence to obtain the fluctuation extremum characteristic, wherein the target constraint function segmentation sequence is any constraint function segmentation sequence in the constraint function segmentation sequence set; calculating a difference sequence of adjacent function values in the objective constraint function segmentation sequence, and counting the fluctuation interval duty ratio of the difference sequence to obtain a time sequence stability characteristic; performing linear fitting on the target constraint function piecewise sequence to obtain trend change characteristics based on fitting slope change; Integrating the fluctuation extremum feature, the time sequence stability feature and the trend change feature to obtain feature parameters corresponding to the target constraint function segmentation sequence; and arranging the characteristic parameters of all constraint function segment sequences according to a time sequence order to obtain a modal constraint characteristic sequence.
5. The method for intelligently diagnosing and protecting the multiple loads of the vehicle body controller according to claim 2, wherein the step of calculating the trend matching degree of the modal constraint feature sequences corresponding to different load-specific operation modes to obtain the cross-modal correlation features comprises the following steps: Performing trend fitting on the modal constraint characteristic sequences corresponding to the exclusive operation modes of each load to obtain a trend fitting curve set formed by trend fitting curves corresponding to the modal constraint characteristic sequences; calculating the trend similarity between any two trend fitting curves in the trend fitting curve set to obtain a similarity coefficient set; Constructing a similarity matrix according to the similarity coefficient set, wherein matrix elements of the similarity matrix are similarity coefficients corresponding to two load exclusive operation modes; and performing feature dimension reduction processing on the similarity matrix to obtain cross-modal associated features.
6. The method for intelligently diagnosing and protecting multiple loads of a vehicle body controller according to claim 1, wherein the step of performing multidimensional feature analysis on the bus voltage sequence to obtain a power supply feature vector comprises the steps of: performing transient and steady state time sequence decoupling on the bus voltage sequence to obtain a transient voltage subsequence and a steady state voltage subsequence; Extracting the change rate characteristic and the convergence trend characteristic of the transient voltage subsequence to obtain a transient voltage dynamic characteristic; traversing the steady-state voltage subsequences through preset sliding windows, calculating a voltage average value in each window and taking the voltage average value as a voltage baseline of the window to obtain a window baseline sequence; calculating the difference value of adjacent voltage baselines in the window baseline sequence to obtain a baseline deviation sequence; Extracting fluctuation amplitude features and integral offset trend features of the baseline deviation sequence to obtain steady-state voltage dynamic features; calculating the association degree of the transient voltage subsequence and the steady-state voltage subsequence in the time sequence dimension to obtain a time domain coupling characteristic; And fusing the transient voltage dynamic characteristic, the steady voltage dynamic characteristic and the time domain coupling characteristic to obtain a power supply characteristic vector.
7. The method for intelligently diagnosing and protecting multiple loads of a vehicle body controller according to claim 1, wherein the protecting the target load according to the association determination result comprises: If the association judgment result is association, judging the abnormal change as coupling noise caused by power supply fluctuation, generating a noise suppression instruction, and terminating fault judgment of the target load; And if the association judgment result is non-association, judging the abnormal change as an independent fault of the target load or a power supply loop thereof, and performing fault processing and load protection actions according to the fault type matched with the load characteristic vector.

Description

Multi-path load intelligent diagnosis and protection method for vehicle body controller Technical Field The invention belongs to the technical field of whole vehicle controllers, and particularly relates to an intelligent diagnosis and protection method for multiple loads of a vehicle body controller. Background Along with the improvement of the complexity of the vehicle electrical system, the multi-path load diagnosis and protection technology of the vehicle body controller gradually evolves from single fault detection to multi-load collaborative monitoring, the early stage takes overcurrent and overvoltage protection of a hardware circuit as a core, software judgment logic based on voltage and current thresholds is subsequently integrated, the identification of typical faults such as open load and short circuit is realized, and in recent years, the method further combines with simple time sequence feature analysis to try to adapt to dynamic scenes such as vehicle start-stop and load switching, and the whole is developed towards the directions of multi-parameter fusion and automatic protection, so that the method gradually becomes one of basic technologies for guaranteeing the stable operation of the vehicle electrical system. However, in a complex and dynamic real vehicle electrical environment, system transient noise and real load faults cannot be effectively distinguished, so that problems of high false alarm rate and low diagnosis accuracy are caused, when a vehicle is started, an air conditioner or power steering and other large loads are started and stopped, severe fluctuation of a power supply network is easily misjudged as a load open circuit or short circuit fault by a diagnosis system, false fault frequent and invalid maintenance are caused, and gray faults with insignificant current changes such as increased contact resistance and slow degradation of load performance are caused, so that the accuracy of vehicle-mounted load diagnosis is low. Disclosure of Invention The invention aims to solve the problem that the diagnosis accuracy of a vehicle-mounted load is low because the transient noise and the real load fault of a system cannot be effectively distinguished in a complex dynamic real vehicle electrical environment, and provides a multi-path load intelligent diagnosis and protection method of a vehicle body controller. The invention provides a multipath load intelligent diagnosis and protection method of a vehicle body controller, which comprises the following steps: synchronously acquiring a current signal and a voltage signal of the target load, and a bus voltage sequence of a vehicle-mounted power supply bus voltage value, wherein the target load is any one load of all loads; Carrying out load collaborative condition decoupling analysis on the current and voltage sequence to obtain a load characteristic vector; Comparing the target load characteristic vector with a preset normal characteristic threshold value, and judging whether the target load has abnormal change or not; if abnormal changes exist, carrying out multidimensional feature analysis on the bus voltage sequence to obtain a power supply feature vector; Inputting the load characteristic vector and the power supply characteristic vector into a preset load power supply coupling noise identification model to obtain a correlation judgment result; and performing protection action on the target load according to the association judgment result. Optionally, performing load collaborative condition decoupling analysis on the current-voltage sequence to obtain a load feature vector includes: Carrying out load operation mode clustering on the current and voltage sequences to obtain a sub-current and voltage sequence set formed by sub-current and voltage sequences respectively corresponding to each load exclusive operation mode; constructing a modal dynamic constraint function for a target sub-current voltage sequence, wherein the target sub-current voltage sequence is any sub-current voltage sequence in the sub-current voltage sequence set; performing time sequence feature analysis and parameter extraction on the target sub-current voltage sequence according to the modal dynamic constraint function to obtain a modal constraint feature sequence; Calculating the trend matching degree of the mode constraint feature sequences corresponding to different load exclusive operation modes to obtain cross-mode association features; And carrying out weighted fusion on the modal constraint feature sequence and the cross-modal association feature according to the duty ratio of the sub-current voltage sequence corresponding to each load exclusive operation mode in the current voltage sequence to obtain a load feature vector. Optionally, constructing the modal dynamic constraint function on the target sub-current voltage sequence includes: The modal dynamic constraint function is Wherein, the method comprises the steps of,Let t be the instantaneous rate of