CN-121993743-A - Gas pipe network hidden danger identification method and system based on unmanned detection vehicle

CN121993743ACN 121993743 ACN121993743 ACN 121993743ACN-121993743-A

Abstract

The invention relates to the technical field of gas pipe network detection and discloses a gas pipe network hidden danger identification method and system based on an unmanned detection vehicle, wherein the method comprises the steps of generating a patrol path and a detection mode of the unmanned detection vehicle based on GIS coordinates and historical hidden danger data of the gas pipe network; the method comprises the steps of acquiring multi-source detection data through a multi-sensor system configured on an unmanned detection vehicle during inspection, extracting abnormal characteristics of the multi-source detection data, primarily identifying hidden danger based on the abnormal characteristics to determine primary hidden danger points and corresponding classification, carrying out surrounding detection and multi-azimuth measurement on the primary hidden danger points to determine whether the primary hidden danger points are leakage sources or not, evaluating surrounding environment risk values, starting emergency response based on the surrounding environment risk values, and updating an inspection path. The invention improves the utilization efficiency of the inspection resource and the hidden danger discovery rate.

Inventors

LIU XINHONG
SU SAILONG
CUI WUWEI
ZHANG CHUNHUI

Assignees

河北泽宏科技股份有限公司

Dates

Publication Date: 20260508
Application Date: 20251231

Claims (10)

1. The gas pipe network hidden danger identification method based on the unmanned detection vehicle is characterized by comprising the following steps of: receiving GIS coordinates and historical hidden danger data of a gas pipe network, generating a patrol path and a detection mode of the unmanned detection vehicle based on the GIS coordinates and the historical hidden danger data, and controlling the unmanned detection vehicle to patrol according to the patrol path and the detection mode; acquiring multi-source detection data through a multi-sensor system configured on an unmanned detection vehicle during inspection, wherein the multi-source detection data comprises geospatial data, visible light and infrared image data, gas concentration gradient data and acoustic characteristic data; performing space-time alignment and fusion processing on the multi-source detection data, extracting abnormal characteristics of the multi-source detection data, and performing primary identification of hidden danger based on the abnormal characteristics to determine initial hidden danger points and corresponding classification; performing surrounding detection and multi-azimuth measurement on the initial hidden danger point, determining whether the initial hidden danger point is a leakage source, and if so, evaluating a surrounding environment risk value; And starting an emergency response based on the surrounding environment risk value, and updating a patrol path based on the leakage source.
2. The gas pipe network hidden danger identification method based on the unmanned detection vehicle according to claim 1, wherein generating the inspection path of the unmanned detection vehicle based on the GIS coordinates and the historical hidden danger data comprises: the history hidden danger data comprises hidden danger high-incidence area coordinates; Dividing a gas pipe network into a plurality of areas based on the GIS coordinates, wherein each area only comprises a hidden danger high-emission area coordinate and each area is not overlapped; Acquiring the starting coordinates of the unmanned detection vehicle, and generating the shortest routing inspection path traversing each area according to the starting coordinates; Generating an in-area inspection path of a corresponding area according to the hidden danger high-emission area coordinates in each area, wherein the initial coordinates of the in-area inspection path are hidden danger high-emission area coordinates; And fusing the shortest inspection path and the inspection path in the area to obtain the inspection path of the unmanned inspection vehicle.
3. The gas pipe network hidden danger identification method based on the unmanned detection vehicle according to claim 2, wherein the generating the detection mode of the unmanned detection vehicle based on the GIS coordinates and the historical hidden danger data comprises the following steps: The detection mode comprises a patrol frequency and a patrol mode; If the GIS coordinates are hidden danger high-frequency area coordinates, setting the inspection frequency as a first inspection frequency, otherwise, setting the inspection frequency as a second inspection frequency, wherein the first inspection frequency is higher than the second inspection frequency; And acquiring meteorological environment data of a gas pipe network, and determining a patrol mode according to the meteorological environment data, wherein the patrol mode comprises a daytime patrol mode and a night patrol mode.
4. The method for identifying hidden dangers of a gas pipe network based on an unmanned detection vehicle according to claim 1, wherein the multi-sensor system comprises a laser radar, a multi-spectrum camera, an inertial navigation system, a gas concentration sensor array and an acoustic sensor.
5. The method for identifying hidden danger of a gas pipe network based on an unmanned detection vehicle according to claim 1, wherein the method for extracting abnormal characteristics of the multi-source detection data comprises the following steps: Extracting image features of the geospatial data, visible light and infrared image data to obtain earth surface subsidence features and third party construction features; Extracting gas characteristics from the gas concentration gradient data to obtain gas concentration mutation characteristics and gas continuous leakage trend characteristics; And extracting acoustic features from the acoustic feature data to obtain gas leakage spectrum features and abnormal sound source features.
6. The method for identifying hidden danger of a gas pipe network based on an unmanned detection vehicle according to claim 5, wherein the step of initially identifying hidden danger based on the abnormal characteristics to determine initial hidden danger points and corresponding classification comprises the steps of: Constructing a mapping relation library of abnormal characteristics and hidden danger types, wherein the hidden danger types comprise pipeline leakage, third party construction interference, earth surface subsidence threat and abnormal sound source interference; Inputting the extracted earth surface subsidence characteristics, third party construction characteristics, gas concentration mutation characteristics, gas continuous leakage trend characteristics, gas leakage spectrum characteristics and abnormal sound source characteristics into a mapping relation library, and matching to obtain corresponding primary hidden danger types; And constructing a grading index system based on the confidence coefficient value and the characteristic intensity of each abnormal characteristic, calculating the comprehensive risk score of the primary hidden danger point according to the grading index system, dividing the primary hidden danger point into three risk grades of high, medium and low according to the comprehensive risk score, and associating corresponding hidden danger classification labels.
7. The method for identifying hidden danger of a gas pipe network based on an unmanned detection vehicle according to claim 1, wherein the steps of performing surrounding detection and multi-azimuth measurement on the initial hidden danger point to determine whether the initial hidden danger point is a leakage source comprise: Automatically executing annular multidirectional movement detection with the initial hidden danger point as a center and the radius of 2-5 meters on the initial hidden danger point, and repeatedly collecting gas concentration and acoustic data at least 8 equally dividing square points of an annular path; Based on multi-azimuth measurement data, calculating three-dimensional coordinates of initial hidden danger points by adopting a triangular positioning algorithm, constructing a leakage source probability model by combining a gas concentration gradient change curve and an acoustic signal attenuation law, judging the initial hidden danger points as leakage sources when the output value of the leakage source probability model is larger than a preset probability threshold value, recording accurate three-dimensional coordinates, a gas concentration peak value and a diffusion direction of the leakage sources, and determining the initial hidden danger points as non-leakage sources if the output value is smaller than or equal to the preset probability threshold value.
8. The method for identifying hidden danger of a gas pipe network based on an unmanned detection vehicle according to claim 1, wherein when the initial hidden danger point is determined to be a leakage source, the method for identifying the hidden danger of the gas pipe network based on the unmanned detection vehicle is characterized by evaluating a surrounding environment risk value, and comprises the steps of analyzing population density, building distribution and traffic flow in a preset range around the leakage source, evaluating leakage diffusion trend based on real-time wind speed and wind direction, and determining the surrounding environment risk value.
9. The unmanned inspection vehicle-based gas pipe network hidden danger identification method according to claim 1, wherein the emergency response is started based on the surrounding environment risk value, and the inspection path is updated based on the leakage source, comprising: When the surrounding environment risk value is higher than a first emergency threshold value, automatically triggering a first-stage emergency response, including uploading the leakage source coordinate and a risk assessment report to a monitoring center in real time, synchronously starting an audible and visual alarm device, and pushing evacuation early warning information to intelligent terminals within a surrounding 500-meter range through a vehicle-mounted communication module; when the surrounding environment risk value is between the second emergency threshold value and the first emergency threshold value, a second-level emergency response is started, the leakage source is continuously monitored, the transmission state data are uploaded, and the nearest rush-repair team is scheduled to go to the site; when the surrounding environment risk value is lower than a second emergency threshold value, starting three-level emergency response, generating a hidden part management work order and incorporating a conventional maintenance plan; when the existence of the leakage source is determined, marking the leakage source as a key rechecking node, preferentially inserting the current inspection sequence, adjusting the inspection sequence of the subsequent path, and increasing the inspection frequency of the leakage source.
10. A gas pipe network hidden trouble recognition system based on an unmanned detection vehicle, for applying the gas pipe network hidden trouble recognition method based on an unmanned detection vehicle according to any one of claims 1 to 9, the system comprising: The inspection control module is configured to receive GIS coordinates and history hidden danger data of the gas pipe network, generate an inspection path and a detection mode of the unmanned detection vehicle based on the GIS coordinates and the history hidden danger data, and control the unmanned detection vehicle to carry out inspection according to the inspection path and the detection mode; The system comprises a data acquisition module, a detection module and a detection module, wherein the data acquisition module is configured to acquire multi-source detection data through a multi-sensor system configured on an unmanned detection vehicle when the inspection is performed, and the multi-source detection data comprises geospatial data, visible light and infrared image data, gas concentration gradient data and acoustic characteristic data; the hidden danger identification module is configured to perform space-time alignment and fusion processing on the multi-source detection data, extract abnormal characteristics of the multi-source detection data, and perform preliminary identification of hidden danger based on the abnormal characteristics to determine initial hidden danger points and corresponding classification thereof; The hidden danger approval response module is configured to perform surrounding detection and multi-azimuth measurement on the initial hidden danger point, determine whether the initial hidden danger point is a leakage source, evaluate a surrounding environment risk value if the initial hidden danger point is the leakage source, start emergency response based on the surrounding environment risk value, and update a patrol path based on the leakage source.

Description

Gas pipe network hidden danger identification method and system based on unmanned detection vehicle Technical Field The invention relates to the technical field of gas pipe network detection, in particular to a gas pipe network hidden danger identification method and system based on an unmanned detection vehicle. Background With the continuous expansion of the scale of urban gas pipe networks and the increase of service life, the hidden dangers of gas leakage, third party construction damage, surface subsidence and the like are increased increasingly, and serious threat is formed to public safety. The traditional manual inspection mode has the problems of low efficiency, incomplete coverage, delayed response, large interference by environment and human factors and the like, and is difficult to meet the real-time and accurate requirements of safe operation of a gas pipe network in a modern city under high-density and complex environment. In recent years, an unmanned detection vehicle is used as an intelligent and automatic inspection platform, and becomes an important technical means for identifying hidden danger of a gas pipe network by virtue of flexible deployment, all-weather operation and multi-sensor fusion capability. However, the existing unmanned vehicle-based gas pipe network detection method has the defects that firstly, the routing inspection path planning lacks depth fusion of historical hidden danger data and a geographic information system, so that resource allocation is unreasonable, key areas are not covered, secondly, multisource heterogeneous sensor data (such as gas concentration, images, acoustics, space positions and the like) lacks an effective space-time alignment and fusion mechanism, high-precision abnormal feature extraction is difficult to realize, thirdly, the hidden danger identification process relies on single sensor judgment, primary judgment-verification-classification-response closed loop logic is lacking, real leakage sources and interference signals cannot be effectively distinguished, an emergency response strategy and risk assessment are disjointed, and dynamic scheduling and intelligent decision are difficult to support. Disclosure of Invention The invention aims to provide a gas pipe network hidden danger identification method and system based on an unmanned detection vehicle, aiming at solving one or more of the problems. The invention provides a gas pipe network hidden danger identification method based on an unmanned detection vehicle, which comprises the following steps: receiving GIS coordinates and historical hidden danger data of a gas pipe network, generating a patrol path and a detection mode of the unmanned detection vehicle based on the GIS coordinates and the historical hidden danger data, and controlling the unmanned detection vehicle to patrol according to the patrol path and the detection mode; acquiring multi-source detection data through a multi-sensor system configured on an unmanned detection vehicle during inspection, wherein the multi-source detection data comprises geospatial data, visible light and infrared image data, gas concentration gradient data and acoustic characteristic data; performing space-time alignment and fusion processing on the multi-source detection data, extracting abnormal characteristics of the multi-source detection data, and performing primary identification of hidden danger based on the abnormal characteristics to determine initial hidden danger points and corresponding classification; performing surrounding detection and multi-azimuth measurement on the initial hidden danger point, determining whether the initial hidden danger point is a leakage source, and if so, evaluating a surrounding environment risk value; And starting an emergency response based on the surrounding environment risk value, and updating a patrol path based on the leakage source. Preferably, generating a routing inspection path of the unmanned inspection vehicle based on the GIS coordinates and the historical hidden trouble data includes: the history hidden danger data comprises hidden danger high-incidence area coordinates; Dividing a gas pipe network into a plurality of areas based on the GIS coordinates, wherein each area only comprises a hidden danger high-emission area coordinate and each area is not overlapped; Acquiring the starting coordinates of the unmanned detection vehicle, and generating the shortest routing inspection path traversing each area according to the starting coordinates; Generating an in-area inspection path of a corresponding area according to the hidden danger high-emission area coordinates in each area, wherein the initial coordinates of the in-area inspection path are hidden danger high-emission area coordinates; And fusing the shortest inspection path and the inspection path in the area to obtain the inspection path of the unmanned inspection vehicle. Preferably, the generating a detection mode of the unmanned detection ve