CN-121977175-A - Vehicle-mounted hydrogen system leakage identification method, system, equipment, medium and product

CN121977175ACN 121977175 ACN121977175 ACN 121977175ACN-121977175-A

Abstract

The application discloses a vehicle-mounted hydrogen system leakage identification method, a system, equipment, a medium and a product, and relates to the field of hydrogen fuel cell automobile safety monitoring; and if the frequency domain characteristic data is positioned in a preset leakage energy duty ratio interval and the time domain characteristic data meets preset pulse judging leakage conditions, judging that the vehicle-mounted hydrogen system leaks and giving an alarm. The application monitors based on the pipeline vibration and deformation signals, realizes the rapid, accurate and non-invasive leakage identification of the vehicle-mounted hydrogen system through the dual-feature fusion judgment of the frequency domain and the time domain, effectively solves the problems of slow response, easy interference, difficult positioning, risk brought by invasive measurement and the like, and improves the real-time performance and the reliability of monitoring.

Inventors

LI JIANWEI
LIU JIE
LIU LINA
GAO FENG
ZHAO DONGDONG
WANG RUIHAI
GAO LEI
HAO DONG

Assignees

北京理工大学

Dates

Publication Date: 20260505
Application Date: 20260403

Claims (10)

1. The vehicle-mounted hydrogen system leakage identification method is characterized by comprising the following steps of: Acquiring a leakage response signal of a vehicle-mounted hydrogen system pipeline, wherein the leakage response signal comprises a vibration signal and a deformation strain signal; Preprocessing the leakage response signal to obtain a preprocessed leakage response signal; Frequency domain feature extraction and time domain feature extraction are respectively carried out on the preprocessing leakage response signals to obtain frequency domain feature data and time domain feature data, wherein the frequency domain feature data comprises energy duty ratio of the preprocessing leakage response signals in a preset frequency band; If the frequency domain characteristic data are located in a preset leakage energy duty ratio interval and the time domain characteristic data meet preset pulse judging leakage conditions, judging that the vehicle-mounted hydrogen system leaks and alarming, wherein the preset pulse judging leakage conditions comprise that the statistical characteristic of a difference value sequence of adjacent pulse peak amplitudes is larger than or equal to a preset difference value threshold, and the statistical characteristic of a pulse period sequence is smaller than a preset period threshold, and the statistical characteristic comprises a mean value, a median value or a set percentile value.
2. The method for identifying a leak in a vehicle-mounted hydrogen system according to claim 1, wherein the preprocessing leak response signal is subjected to a preprocessing operation to obtain a preprocessing leak response signal, and specifically comprises: based on the vibration signal and the deformation strain signal in the leakage response signal, extracting dominant frequencies of the vibration signal and the deformation strain signal through real-time frequency spectrum analysis respectively; The dominant frequency of the vibration signal and the dominant frequency of the deformation strain signal are subjected to dynamic filtering treatment through an adaptive Kalman filtering method respectively to obtain a preprocessing leakage response signal, wherein the preprocessing leakage response signal comprises a vibration signal after filtering and a deformation strain signal after filtering, and the adaptive Kalman filtering method comprises the step of increasing a process noise covariance matrix of the Kalman filtering method through the following formula if the dominant frequency is greater than or equal to a first frequency threshold and less than or equal to a second frequency threshold: ; Wherein, the Representing a process noise covariance matrix; Representing a basic process noise covariance matrix; Representing an indication function; Representing the adjustment coefficient; Representing the dominant frequency; representing a first frequency threshold; representing a second frequency threshold; And if the dominant frequency is greater than the second frequency threshold and less than or equal to the third frequency threshold, adjusting the process noise covariance matrix to be a basic process noise covariance matrix, and adjusting the measurement noise covariance matrix to be less than or equal to the basic measurement noise covariance matrix.
3. The method for identifying leakage of a vehicle-mounted hydrogen system according to claim 2, wherein the step of extracting frequency domain features from the preprocessed leakage response signal to obtain frequency domain feature data comprises: And applying a hanning window to the vibration signal after filtering and the deformation strain signal after filtering respectively through the following formulas to obtain corresponding windowed signals: ; ; Wherein, the , And the time-representing the post-filter vibration signal, The time represents the deformation strain signal after filtering; The nth sample point A windowed signal of the class signal; The nth sample point A class signal; N represents the total number of sampling points; Based on the windowed signals corresponding to the vibration signals after filtering and the deformation strain signals after filtering, respectively carrying out frequency domain transformation by a fast Fourier transformation method to obtain corresponding frequency domain complex sequences; Based on the frequency domain complex number sequence, frequency domain characteristic data is obtained through calculation according to the following formula: ; ; Wherein, the Represent the first The energy duty ratio of the class signal in a preset frequency band; representing the kth frequency component after fourier transform; representing an actual frequency value corresponding to the kth frequency component; representing the sampling frequency; representing a lower limit frequency of a preset frequency band; Representing the upper limit frequency of a preset frequency band; represent the first A frequency domain complex sequence of the class signal; represent the first Power spectral density of a frequency domain complex sequence of a signal-like.
4. The method for identifying leakage of a vehicle-mounted hydrogen system according to claim 2, wherein the extracting of the time domain features from the preprocessed leakage response signal to obtain time domain feature data specifically comprises: Detecting wave peaks through a sliding window maximum method based on the filtered vibration signal and the filtered deformation strain signal in the preprocessed leakage response signal, so as to obtain a corresponding wave peak amplitude sequence; based on the peak amplitude sequence, a difference value sequence of adjacent pulse peak amplitudes is calculated by the following formula: ; Wherein, the Represent the first The difference value of the peak amplitude of the ith adjacent pulse of the class signal; represent the first The amplitude of the i+1th pulse peak of the class signal; represent the first The amplitude of the ith pulse peak of the class signal; i= {1,2,.., -1}; Represent the first The total number of pulse peaks detected by the class signal; the pulse cycle sequence is calculated by the following formula: ; Wherein, the Represent the first The ith pulse period of the class signal; represent the first The time of the (i+1) th pulse peak of the class signal; represent the first The instant of the i-th pulse peak of the class signal.
5. The in-vehicle hydrogen system leakage identification method according to claim 1, characterized by further comprising, after determining that the in-vehicle hydrogen system leakage has occurred: Acquiring actual arrival time of each acceleration sensor in a sensor array axially arranged along a vehicle-mounted hydrogen system pipeline for detecting a vibration signal in a leakage response signal; calculating the arrival time difference of the vibration signals between any two acceleration sensors based on the actual arrival time, the length of a pipeline between the acceleration sensors and the vibration wave speed of the material; judging the azimuth of the leakage point relative to each acceleration sensor based on the arrival time difference of the vibration signals between any two acceleration sensors and the inherent propagation time difference of the pipeline length between the corresponding acceleration sensors, and calculating the preliminary position coordinates of the leakage point along the axial direction of the pipeline to obtain the preliminary positioning positions of a plurality of groups of acceleration sensor pairs; and determining the final positioning position of the leakage point through a weighted fusion algorithm based on the preliminary positioning positions of the plurality of groups of acceleration sensor pairs.
6. The method for identifying leakage of a vehicle-mounted hydrogen system according to claim 5, wherein determining a final leakage point location by a weighted fusion algorithm based on preliminary location locations of a plurality of sets of acceleration sensor pairs, specifically comprises: Based on the preliminary positioning positions of a plurality of groups of acceleration sensor pairs, taking the preliminary positioning positions meeting preset validity conditions as effective preliminary positioning positions to obtain an effective preliminary positioning position set, wherein the preset validity conditions comprise that the absolute value of the difference value between the arrival time difference of the vibration signals of the acceleration sensor pairs and the inherent propagation time difference of the pipeline length between the corresponding acceleration sensors is smaller than or equal to a preset time difference threshold value, and the signal to noise ratio of the vibration signals of the acceleration sensor pairs is larger than or equal to a preset signal to noise ratio threshold value; The weight of the acceleration sensor pair corresponding to each effective preliminary positioning position in the effective preliminary positioning position set is determined by the following formula: ; Wherein, the Representing the weight of the q-th acceleration sensor pair in the effective preliminary positioning position set; representing the signal-to-noise ratio of the vibration signal of the q-th acceleration sensor pair in the set of effective preliminary positioning locations, ; Representing the first of a set of valid preliminary positioning locations Signal to noise ratio of vibration signals of the individual acceleration sensor pairs, P represents the total number of acceleration sensor pairs in the effective preliminary positioning position set; The final leak location is determined by the following formula: ; Wherein, the Representing the final location of the leak point; and representing the effective preliminary positioning position of the q-th acceleration sensor pair in the effective preliminary positioning position set.
7. An in-vehicle hydrogen system leakage identification system, characterized in that the in-vehicle hydrogen system leakage identification system applies the in-vehicle hydrogen system leakage identification method according to any one of claims 1 to 6, the in-vehicle hydrogen system leakage identification system comprising: The system comprises a signal acquisition module, a control module and a control module, wherein the signal acquisition module is used for acquiring a leakage response signal of a vehicle-mounted hydrogen system pipeline, wherein the leakage response signal comprises a vibration signal and a deformation strain signal; the signal preprocessing module is used for preprocessing the leakage response signal to obtain a preprocessed leakage response signal; The device comprises a preprocessing leakage response signal, a feature extraction module, a frequency domain feature extraction module and a time domain feature extraction module, wherein the preprocessing leakage response signal is subjected to frequency domain feature extraction and time domain feature extraction respectively to obtain frequency domain feature data and time domain feature data; The leakage identification module is used for judging that the vehicle-mounted hydrogen system leaks and alarming if the frequency domain characteristic data is located in a preset leakage energy duty ratio interval and the time domain characteristic data meets a preset pulse judgment leakage condition, wherein the preset pulse judgment leakage condition comprises that the statistical characteristic of a difference value sequence of adjacent pulse peak amplitudes is greater than or equal to a preset difference value threshold, the statistical characteristic of a pulse period sequence is smaller than a preset period threshold, and the statistical characteristic comprises a mean value, a median value or a set percentile value.
8. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor executes the computer program to implement the in-vehicle hydrogen system leak identification method of any one of claims 1-6.
9. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when executed by a processor, implements the in-vehicle hydrogen system leak identification method according to any one of claims 1 to 6.
10. A computer program product comprising a computer program, characterized in that the computer program, when executed by a processor, implements the in-vehicle hydrogen system leak identification method as claimed in any one of claims 1-6.

Description

Vehicle-mounted hydrogen system leakage identification method, system, equipment, medium and product Technical Field The application relates to the technical field of hydrogen fuel cell automobile safety monitoring, in particular to a vehicle-mounted hydrogen system leakage identification method, a system, equipment, a medium and a product. Background In the context of low carbon conversion of global energy structures, hydrogen energy has become one of the most potential clean energy sources by virtue of its zero carbon emissions, high energy density and multi-scenario applicability. Hydrogen fuel cell automobiles are an important direction for hydrogen energy applications, which are being accelerated for large-scale applications. However, the unique physicochemical properties of hydrogen also present serious safety challenges, its molecular weight is extremely small, diffusion rate in air after leakage is extremely fast, and at the same time, the hydrogen flammability range is extremely wide, the ignition energy is low, and weak electrostatic sparks can ignite. Thus, the large-scale popularization of hydrogen fuel cell automobiles is always faced with potential leakage-explosion risks. The method for monitoring the hydrogen leakage of the vehicle-mounted hydrogen system rapidly and accurately is an important means for guaranteeing the safety application of the fuel cell automobile. The monitoring scheme in the related technology mainly comprises two aspects, namely, monitoring the atmospheric hydrogen concentration above a vehicle-mounted hydrogen system, if a plurality of hydrogen concentration sensors are adopted for joint monitoring, and comprehensively evaluating by combining with a temperature sensor, the leakage condition of a target space is determined, but the method can realize monitoring only by diffusing hydrogen to the sensor position, in the running process, the interference of environmental wind is large, the response is slow, and the leakage positioning is difficult to realize, and secondly, monitoring the physical parameters of the hydrogen in a pipeline to judge whether the hydrogen leaks, if the pressure, the flow and the temperature of the hydrogen are abnormal, a neural network model is adopted to determine whether the vehicle-mounted hydrogen system leaks, but the method is greatly influenced by the fluctuation of the gas, has lower precision, and needs to arrange a large number of temperature, pressure and flow sensors for invasive measurement, thereby increasing the risk of gas blockage or leakage. Therefore, it is needed to propose a vehicle-mounted hydrogen system leakage identification method to solve the problems of slow monitoring response, easy interference, difficult positioning, additional risks caused by invasive measurement and the like, so as to realize rapid and accurate leakage monitoring in the driving process. Disclosure of Invention The application aims to provide a vehicle-mounted hydrogen system leakage identification method, a system, equipment, a medium and a product, which can realize rapid, anti-interference and non-invasive leakage identification and positioning of a vehicle-mounted hydrogen system by utilizing pipeline vibration and deformation physical signals caused by leakage through dual-feature fusion judgment, thereby solving the problems of slow monitoring response, easiness in environmental interference, difficulty in positioning and additional risks brought by invasive measurement in the prior art. In order to achieve the above object, the present application provides the following solutions: In a first aspect, the present application provides a method for identifying leakage of a vehicle-mounted hydrogen system, comprising: and acquiring a leakage response signal of the vehicle-mounted hydrogen system pipeline, wherein the leakage response signal comprises a vibration signal and a deformation strain signal. And preprocessing the leakage response signal to obtain a preprocessed leakage response signal. And respectively carrying out frequency domain feature extraction and time domain feature extraction on the preprocessing leakage response signals to obtain frequency domain feature data and time domain feature data, wherein the frequency domain feature data comprises the energy duty ratio of the preprocessing leakage response signals in a preset frequency band, and the time domain feature data comprises a difference value sequence and a pulse period sequence of adjacent pulse crest amplitudes extracted based on the preprocessing leakage response signals. If the frequency domain characteristic data are located in a preset leakage energy duty ratio interval and the time domain characteristic data meet preset pulse judging leakage conditions, judging that the vehicle-mounted hydrogen system leaks and alarming, wherein the preset pulse judging leakage conditions comprise that the statistical characteristic of a difference value sequence of adjacent pulse p