CN-121980441-A - Intelligent inspection abnormality identification method, system, equipment and medium for cable tunnel

CN121980441ACN 121980441 ACN121980441 ACN 121980441ACN-121980441-A

Abstract

The invention discloses a cable tunnel intelligent inspection anomaly identification method, a system, equipment and a medium, which belong to the technical field of data processing and comprise the steps of synchronously acquiring operation data of a cable tunnel through heterogeneous sensors to obtain multi-mode operation data comprising space structure information, gas component information, sound signals, electromagnetic signals and environment state information, respectively carrying out feature extraction and anomaly modeling to generate binary anomaly marking data, carrying out feature level fusion through a two-level algorithm, judging whether equipment is abnormal or not based on the fused features to finish equipment anomaly detection, carrying out feature extraction and independent anomaly modeling on each mode data through a first-level algorithm to generate a preliminary binary anomaly mark, solving the problem of high single sensor index misjudgment rate, generating uniform multi-mode fusion feature parameters, improving the accuracy and reliability of equipment anomaly identification under a complex environment, and overcoming the limitations of isolated analysis and lack of intelligent fusion of a traditional monitoring method.

Inventors

YAN QING
QI WANBI
LIU XING
CHEN KEYU
WANG WEIJUN
HU KAIQIANG
LI CHAOJU
WANG QIUFENG
SUN HE
LIU PENG
DONG WEI
Sheng xinglong
XIE YANGHUA
DU XUE
LI SHISHUANG
ZENG JUNXIONG
LI CHEN
WANG DESHENG
LIU YANG
DU HAO
HU RONGJUN
WU SHOUCHANG
ZHANG XIAO
HUANG HUI
GUO JUFU
CHEN LE

Assignees

贵州电网有限责任公司

Dates

Publication Date: 20260505
Application Date: 20251218

Claims (10)

1. The intelligent inspection abnormality identification method for the cable tunnel is characterized by comprising the following steps of, Synchronously acquiring operation data of the cable tunnel through heterogeneous sensors to obtain multi-mode operation data comprising space structure information, gas component information, sound signals, electromagnetic signals and environmental state information; Based on the collected multi-mode operation data, respectively carrying out feature extraction and anomaly modeling to generate binary anomaly flag data; and carrying out feature level fusion on the binary abnormal mark data through a secondary algorithm, judging whether the equipment is abnormal or not based on the fused features, and finishing equipment abnormality detection.
2. The method for identifying abnormal intelligent patrol of a cable tunnel according to claim 1, wherein the step of synchronously collecting operation data of the cable tunnel by heterogeneous sensors to obtain multi-mode operation data including spatial structure information, gas composition information, sound signals, electromagnetic signals and environmental state information comprises, Collecting space structure information through a laser radar, collecting gas state information through a gas sensor, and collecting sound information generated by operation through an acoustic sensor; the electromagnetic radiation sensor is used for collecting the generated electromagnetic radiation information, and the temperature and humidity sensor is used for collecting the environmental state information; Multimodal operation data is generated based on the collected spatial structure information, gas composition information, acoustic signals, electromagnetic signals, and environmental state information.
3. The intelligent inspection anomaly identification method for the cable tunnel according to claim 2, wherein the feature extraction and anomaly modeling are respectively performed based on the acquired multi-mode operation data, and the generation of binary anomaly flag data comprises, Based on the acquired multi-mode operation data, calculating an electric cloud registration characteristic parameter through a first-level algorithm, extracting gas characteristics and calculating a gas concentration characteristic parameter; extracting acoustic features, electromagnetic radiation features and environmental state features, and calculating state environmental feature parameters; And performing preliminary anomaly detection based on the electric cloud registration characteristic parameters, the gas concentration characteristic parameters, the acoustic characteristic, the electromagnetic radiation characteristic, the environment state characteristic and the state environment characteristic parameters to generate binary anomaly flag data.
4. The intelligent inspection anomaly identification method for cable tunnels of claim 3, wherein the feature level fusion of binary anomaly flag data is performed by a two-level algorithm, and whether the anomaly exists in the equipment is judged based on the fused features, and the completion of equipment anomaly detection comprises, Based on binary abnormal sign data, carrying out feature level fusion through a secondary algorithm, and outputting multi-mode data fusion feature parameters; Judging the equipment state based on the multi-mode data fusion characteristic parameters, and outputting abnormal alarm information according to the equipment state; And outputting suggested measures according to the abnormality alarming information, and carrying out abnormality investigation according to the suggested measures.
5. The intelligent inspection anomaly identification method for cable tunnels of claim 4, wherein the computing the electrical cloud registration feature parameters, extracting gas features, and computing gas concentration feature parameters based on the acquired multi-modal operation data comprises, Three-dimensional point cloud data are acquired through a laser radar, and the expression is as follows: Wherein, the For the vector data registration error, For the coordinate vector of the ith three-dimensional point acquired by the laser radar in the current scanning period, N is the nth data in the three-dimensional point cloud data, , Registering a characteristic parameter anomaly threshold value for the electric cloud; The gas concentration detection vector expression is: Wherein, the As a vector of the concentration of the gas, As a measure of the concentration of sulfur dioxide gas, As a measure of the concentration of hydrogen sulfide gas, As a measure of the concentration of carbon monoxide gas, Is a concentration measurement value of sulfur hexafluoride gas, Is a transposition; Extracting the characteristic of the change rate of the gas concentration, wherein the expression is as follows: Wherein, the In order to be a rate of change of the concentration of the gas, , For the concentration value of the kth gas measured at time t, To be at the time of The measured concentration value of the kth gas, In order to provide for the time interval of time, , And judging a threshold value for the characteristic parameter of the gas concentration.
6. The method for identifying the intelligent inspection anomalies of the cable tunnel according to claim 5, wherein the steps of extracting acoustic features, electromagnetic radiation features and environmental state features, calculating state environmental feature parameters include, The abnormal sound feature expression is: Wherein, the For sound pressure level characteristic data, D for analysis frame duration, As a function of the characteristics of the sound, , Judging a threshold value for the acoustic feature; The partial discharge detection is acquired through an electromagnetic radiation sensor, and the expression is as follows: Wherein, the In order to detect the data of the partial discharge, In order to indicate the function, For the original signal amplitude acquired by the electromagnetic radiation sensor at time t, In order to detect the threshold value of the discharge, , Judging a threshold value for electromagnetic radiation characteristics; the temperature and humidity sensors are collected through the temperature and humidity sensors to perform feature extraction, and the expression is as follows: Wherein, the 、 Respectively a first weight coefficient and a second weight coefficient, Is the comprehensive index of the temperature and the humidity, In order to be a value of the temperature, As the value of the humidity to be measured, , A threshold is determined for the environmental state.
7. The intelligent inspection anomaly identification method for cable tunnels according to claim 6, wherein the feature level fusion by a two-level algorithm based on binary anomaly flag data comprises, Carrying out multi-mode data fusion on the abnormal data characteristics, wherein the expression is as follows: Wherein, the Is the characteristic data of the multi-mode data fusion, , , Respectively a first weight, a second weight and a third weight, In order to indicate the function, For the vector data registration error, , For the electrical cloud registration feature parameter anomaly threshold, In order to be a rate of change of the concentration of the gas, , A threshold value is judged for the characteristic parameter of the gas concentration, , A threshold value is determined for the environmental state, , And judging a threshold value for the characteristic data of the multi-mode data fusion.
8. A system for identifying cable tunnel intelligent inspection anomalies, which is characterized by comprising a data acquisition module, a feature extraction module and a data fusion and decision module, The data acquisition module synchronously acquires operation data of the cable tunnel through heterogeneous sensors to obtain multi-mode operation data comprising space structure information, gas component information, sound signals, electromagnetic signals and environmental state information; the feature extraction module is used for respectively carrying out feature extraction and anomaly modeling based on the acquired multi-mode operation data to generate binary anomaly flag data; The data fusion and decision module is used for carrying out feature level fusion on binary abnormal mark data through a secondary algorithm, judging whether the equipment has abnormality or not based on the fused features, and finishing equipment abnormality detection.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of a method for identifying an intelligent patrol anomaly of a cable tunnel according to any one of claims 1 to 7.
10. A computer readable storage medium having stored thereon a computer program, characterized in that the computer program when executed by a processor implements the steps of a cable tunnel intelligent patrol anomaly identification method according to any one of claims 1 to 7.

Description

Intelligent inspection abnormality identification method, system, equipment and medium for cable tunnel Technical Field The invention relates to the technical field of data processing, in particular to a method, a system, equipment and a medium for identifying intelligent inspection anomalies of a cable tunnel. Background Cable tunnels are often located underground, and environments are complex and variable, including confined spaces, limited lighting, high temperature and humidity, and the like. Harmful gases such as methane, hydrogen sulfide and the like can exist in the tunnel, and serious safety threat is formed to patrol personnel. The facilities are dense, and the facilities in the cable tunnel are numerous, including cables, brackets, illumination, ventilation and the like, so that the traditional manual inspection is difficult to cover the whole surface. Defects of cables and tunnel facilities are often strong in concealment, such as equipment internal cracks, insulation layer falling and the like, and are difficult to find by a traditional inspection method. Conventional monitoring techniques typically detect only a single type of anomaly and have difficulty in fully reflecting the status of the device. The traditional monitoring technology lacks intelligent analysis capability and is difficult to fuse complex data, so that whether one device is abnormal or not needs a plurality of data to comprehensively judge, and the traditional monitoring technology is single data index judgment, so that the problems of high misjudgment rate and inaccurate judgment often occur. Disclosure of Invention The present invention has been made in view of the above-described problems. Therefore, the invention solves the technical problem that the traditional monitoring technology can only detect single type of abnormality and is difficult to comprehensively reflect the equipment state. The traditional monitoring technology lacks intelligent analysis capability, is difficult to fuse complex data, is single data index judgment, and often has the problems of high misjudgment rate and inaccurate judgment. In order to solve the technical problems, the invention provides a method for identifying the abnormal condition of intelligent inspection of a cable tunnel, which comprises the following steps, Synchronously acquiring operation data of the cable tunnel through heterogeneous sensors to obtain multi-mode operation data comprising space structure information, gas component information, sound signals, electromagnetic signals and environmental state information; Based on the collected multi-mode operation data, respectively carrying out feature extraction and anomaly modeling to generate binary anomaly flag data; and carrying out feature level fusion on the binary abnormal mark data through a secondary algorithm, judging whether the equipment is abnormal or not based on the fused features, and finishing equipment abnormality detection. The invention relates to a cable tunnel intelligent inspection abnormality identification method, which comprises the steps of synchronously collecting operation data of a cable tunnel through heterogeneous sensors, obtaining multi-mode operation data comprising space structure information, gas composition information, sound signals, electromagnetic signals and environmental state information, Collecting space structure information through a laser radar, collecting gas state information through a gas sensor, and collecting sound information generated by operation through an acoustic sensor; the electromagnetic radiation sensor is used for collecting the generated electromagnetic radiation information, and the temperature and humidity sensor is used for collecting the environmental state information; Multimodal operation data is generated based on the collected spatial structure information, gas composition information, acoustic signals, electromagnetic signals, and environmental state information. The invention relates to a cable tunnel intelligent inspection abnormality identification method, which comprises the steps of respectively carrying out feature extraction and abnormality modeling based on acquired multi-mode operation data, generating binary abnormality mark data, Based on the acquired multi-mode operation data, calculating an electric cloud registration characteristic parameter through a first-level algorithm, extracting gas characteristics and calculating a gas concentration characteristic parameter; extracting acoustic features, electromagnetic radiation features and environmental state features, and calculating state environmental feature parameters; And performing preliminary anomaly detection based on the electric cloud registration characteristic parameters, the gas concentration characteristic parameters, the acoustic characteristic, the electromagnetic radiation characteristic, the environment state characteristic and the state environment characteristic parameters to genera