Comprehensive pipe gallery monitoring and early warning analysis method based on multimode sensing fusion reasoning

Abstract

The invention discloses a comprehensive pipe gallery monitoring and early warning analysis method based on multimode sensing fusion reasoning, relates to the technical field of comprehensive pipe gallery monitoring and early warning, and solves the technical problem that in the prior art, only the risk at the current moment is processed, the potential chain reaction of the risk is ignored, and the risk is spread. According to the method, risk event evolution chains are constructed through development nodes of risk events, and a comprehensive pipe gallery risk evolution network is constructed according to overlapped nodes and index description sets of the risk event evolution chains; the early warning state score of the comprehensive pipe gallery is analyzed; analyzing an early warning consequence score of the comprehensive pipe gallery according to the consequence score of the risk event corresponding to the alarm node; based on the early warning state score and the early warning consequence score, the early warning grade of the comprehensive pipe gallery is analyzed; according to the invention, through the comprehensive pipe gallery risk evolution network, potential risk chain reactions are predicted in advance, and early warning is carried out in time.

Description

Technical Field

[0001] The present invention belongs to the field of integrated pipe gallery monitoring and early warning, and specifically is an integrated pipe gallery monitoring and early warning analysis method based on multi-mode perception fusion reasoning. Background Art

[0002] The urban underground comprehensive pipe corridor contains various municipal pipelines, which is an important carrier for providing a safe operating environment for urban energy transportation and an important infrastructure for ensuring the safe operation of the city. As the comprehensive pipe corridor gradually becomes the supporting infrastructure for urban municipal construction, the scale of the pipe corridor, the types and number of pipelines contained in the corridor are also increasing, and the safety and stability of the internal space of the comprehensive pipe corridor have become more important.

[0003] The invention patent application with publication number CN110397080A discloses a monitoring and early warning system for an integrated pipe corridor. Intelligent sensing components and monitoring equipment are arranged in the integrated pipe corridor. The monitoring and early warning system includes: a data and information module, a numerical calculation module, an artificial intelligence analysis module and a monitoring and early warning module. The data and information module is used to collect and organize the data collected by the intelligent sensing components and the monitoring equipment; the numerical calculation module is used to perform real-time inversion analysis on the data; the artificial intelligence analysis module is used to perform intelligent analysis on the characteristics and laws of the data collected by the intelligent sensing components and the monitoring equipment; the monitoring and early warning module is used to perform safety analysis on the integrated pipe corridor, establish a pipe corridor safety evaluation model, and provide a safety evaluation and health diagnosis report.

[0004] The above-mentioned prior art uses an artificial intelligence analysis module for intelligent analysis, and combines a numerical calculation module to perform real-time inversion analysis on data, builds a monitoring and early warning system for a comprehensive pipe gallery, and realizes real-time diagnosis and early warning of the health status of the comprehensive pipe gallery. However, when an abnormality occurs in the comprehensive pipe gallery, it often causes a series of other safety hazards, and the prior art only processes the abnormality at the current moment, while ignoring the potential chain reaction caused by the abnormality, leading to the problem of risk spread. Therefore, the present invention provides a monitoring and early warning analysis method for a comprehensive pipe gallery based on multi-mode perception fusion reasoning. Summary of the invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a comprehensive pipeline corridor monitoring, early warning and analysis method based on multi-mode perception fusion reasoning, which is used to solve the technical problem that the prior art only deals with the risks at the current moment, but ignores the potential chain reactions of the risks, leading to the spread of risks.

[0006] To achieve the above object, the first aspect of the present invention provides a comprehensive pipe gallery monitoring and early warning analysis method based on multi-mode perception fusion reasoning, comprising the following steps:

[0007] Step 1: Extract risk events of the integrated pipeline corridor from the historical risk data of the integrated pipeline corridor and construct a risk event set of the integrated pipeline corridor; construct the risk event evolution chain according to the development nodes of the risk events; the historical risk data of the integrated pipeline corridor includes accident cases, standards, literature and disaster mechanism models of the integrated pipeline corridor; the chain structure includes the first node, evolution node and tail node;

[0008] Step 2: Extract the multi-source heterogeneous monitoring data of the integrated pipeline corridor, perform feature extraction on the multi-source heterogeneous monitoring data, and construct an indicator description set that characterizes the status of the chain nodes; the multi-source heterogeneous monitoring data includes equipment monitoring data, equipment control signals, video image data, and personnel inspection record data;

[0009] Step 3: Based on the overlapping nodes and indicator description sets of each risk event evolution chain, the indicator status changes of the same node are associated with different risk events to form a comprehensive pipeline corridor risk evolution network;

[0010] Step 4: Analyze the early warning status score of the integrated pipeline corridor; the early warning status score is determined based on the alarm status score and the risk event development trend prediction correction score; the risk event development trend prediction correction score is determined based on the change trend of the monitoring indicator data, the control system status, and the status of related nodes in the risk evolution network;

[0011] Step 5: Analyze the warning consequence score of the integrated pipeline corridor according to the consequence score of the risk event corresponding to the alarm node; analyze the warning level of the integrated pipeline corridor based on the warning status score and the warning consequence score.

[0012] Preferably, the indicator description set includes original data and feature data processed based on the original data, forming a multi-source multi-modal data set to describe the state of the chain nodes.

[0013] Preferably, the analysis of the warning status score of the integrated pipe gallery includes:

[0014] Extract the alarm status score corresponding to the current monitoring indicator data from the abnormal status score table of the corresponding monitoring indicator;

[0015] Analyze the changing trends of monitoring indicator data through historical data and current data of monitoring indicators, and assign scores to different changing trends;

[0016] and obtaining the corresponding control system status and the status of related nodes in the risk evolution network, and assigning scores to different statuses;

[0017] Based on the score of the change trend and the score of the status, the risk event development trend prediction correction score is obtained;

[0018] The alarm status score is corrected by correcting the risk event development trend score to obtain the early warning status score of the integrated utility corridor.

[0019] Preferably, the abnormal condition score table of the monitoring indicators includes:

[0020] A number of monitoring indicator data and the corresponding abnormal conditions are extracted from the historical data, and the corresponding alarm status scores are assigned according to the abnormal degree of the abnormal condition. The range of the monitoring indicator data corresponding to each abnormal condition is summarized to obtain the abnormal condition score table of each monitoring indicator.

[0021] Preferably, analyzing the warning consequence score of the integrated pipe gallery according to the consequence score of the risk event corresponding to the alarm node includes:

[0022] Extract the consequences of the risk events of the integrated pipeline corridor from the historical data, sort out and summarize the consequences of the risk events, obtain a set of consequence indicators, and assign scores to the consequence indicators respectively;

[0023] Based on several consequence indicator sets of several risk events pointed to by the alarm node, the largest consequence score in the several consequence indicator sets is the consequence score of the alarm node; among which, the alarm node is the first node whose monitoring indicator data is not within the preset threshold range.

[0024] Preferably, the analysis of the warning level of the integrated pipe gallery based on the warning status score and the warning consequence score includes:

[0025] Based on the warning status score and warning consequence score of the integrated pipeline corridor, the warning level of the integrated pipeline corridor is obtained through hierarchical matrix analysis; among them, the warning level is iterated in real time according to multi-source dynamic monitoring data and spatiotemporal big data.

[0026] Compared with the prior art, the present invention has the following beneficial effects:

[0027] 1. The present invention dynamically monitors and classifies the integrated pipe corridor based on multi-source heterogeneous dynamic data obtained from multiple channels, which helps to reduce the risk of accidents in the integrated pipe corridor, reduce casualties and economic losses; dynamically deduce the potential consequences of comprehensive events based on spatiotemporal big data, and reflect them in the warning level, so as to realize the intelligent prediction of the risk status of the integrated pipe corridor; and by deducing the evolution chain of comprehensive risk events and forming risk nodes, the multi-source heterogeneous data of the integrated pipe corridor are sorted and summarized to construct an indicator set, and the integrated pipe corridor risk evolution network is formed based on the fusion of risk nodes and indicator sets, so as to realize the mapping of indicators and risk nodes, and to warn potential risks in advance through the evolution of risk events;

[0028] 2. The present invention constructs a risk event evolution chain through the development nodes of risk events, and constructs an indicator description set that characterizes the status of the chain nodes; according to the overlapping nodes and indicator description sets of each risk event evolution chain, the indicator status changes of the same node are associated and mapped with different risk events to form a comprehensive pipeline corridor risk evolution network; the early warning status score of the comprehensive pipeline corridor is analyzed; the early warning consequence score of the comprehensive pipeline corridor is analyzed according to the consequence score of the risk event corresponding to the alarm node; based on the early warning status score and the early warning consequence score, the early warning level of the comprehensive pipeline corridor is analyzed; by associating and mapping the node indicator status changes of different risk events, the present invention can predict potential risk chain reactions in advance, timely warn and take measures to prevent further expansion of risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0030] Figure 1 It is a schematic diagram of the process of the present invention;

[0031] Figure 2 It is a schematic diagram of the combustion and explosion evolution chain of the gas pipeline leakage of the present invention;

[0032] Figure 3 This is a schematic diagram of the evolution network of flooding risk of the integrated pipeline corridor of the present invention. DETAILED DESCRIPTION

[0033] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] See also Figure 1 The first aspect of the present invention provides a method for monitoring and early warning analysis of a comprehensive pipe gallery based on multi-mode perception fusion reasoning, comprising the following steps:

[0035] Step 1: Extract the risk events of the integrated pipeline corridor from the historical risk data of the integrated pipeline corridor and construct a set of risk events of the integrated pipeline corridor; construct the risk event evolution chain according to the development nodes of the risk events; wherein the chain structure includes the first node, the evolution node and the tail node;

[0036] It should be noted that the historical risk data of the integrated pipe corridor includes accident cases, standards, literature and disaster mechanism models of the integrated pipe corridor. Risk events are extracted from these historical data to determine the risk event set of the integrated pipe corridor; such as {gas pipeline leakage and explosion, pipe cabin flooding, pipe corridor collapse, casualties, ...};

[0037] See also Figure 2 , construct the risk event evolution chain according to the development and evolution of risk events; such as the occurrence process of gas pipeline leakage and explosion: the first node (the fuse that causes gas pipeline leakage): such as misoperation / human damage / pipeline corrosion / ..., evolution node (the evolution process after the gas leakage): methane diffusion->reaching the explosion limit->spark->gas pipeline leakage and explosion, the tail node (a series of consequences caused by gas pipeline leakage and explosion): such as casualties, pipeline corridor collapse, economic losses, supply interruption, social impact,...

[0038] Step 2: Extract the multi-source heterogeneous monitoring data of the integrated pipeline corridor, extract features from the multi-source heterogeneous monitoring data, and construct an indicator description set that characterizes the status of the chain nodes;

[0039] Among them, multi-source heterogeneous monitoring data includes equipment monitoring data, equipment control signals, video image data, and personnel inspection record data;

[0040] The indicator description set includes the original data and the feature data processed based on the original data, forming a multi-source and multi-modal data set to describe the status of the chain nodes. For example, the indicator description set describing the waterlogging event in the pipe cabin includes the liquid level meter monitoring data, humidity data, video recognition of the water level in the pipe cabin based on the monitoring video, and information reported by personnel inspections.

[0041] Step 3: Based on the overlapping nodes and indicator description sets of each risk event evolution chain, the indicator status changes of the same node are associated with different risk events to form a comprehensive pipeline corridor risk evolution network;

[0042] For example: Figure 3 The figure shows the risk evolution network of waterlogging in the integrated pipe gallery. One part of the risk evolution network is that when the temperature and humidity alarm / hydrogen sulfide alarm / liquid level alarm is triggered, pipeline corrosion will occur, which will cause drainage pipeline leakage / water supply pipeline leakage; drainage pipeline leakage / water supply pipeline leakage will lead to drainage failure / flooding.

[0043] Step 4: Analyze the early warning status score of the integrated pipeline corridor; the early warning status score is determined based on the alarm status score and the risk event development trend prediction correction score; the risk event development trend prediction correction score is determined based on the change trend of the monitoring indicator data, the control system status, and the status of related nodes in the risk evolution network;

[0044] Specifically, a number of monitoring indicator data and corresponding abnormal conditions are extracted from historical data, corresponding alarm status scores are assigned according to the abnormal degree of the abnormal condition, and the range of monitoring indicator data corresponding to each abnormal condition is summarized to obtain an abnormal condition score table for each monitoring indicator;

[0045] Extract the alarm status score corresponding to the current monitoring indicator data from the abnormal status score table of the corresponding monitoring indicator; for example, the methane alarm level can be defined according to the alarm concentration, with a methane concentration of 1% VOL, a first-level alarm; a methane concentration of 2% VOL, a second-level alarm; and a methane concentration of 5% VOL, a third-level alarm. Table 1 below is a score table for gas pipeline leakage and explosion risk events:

[0046] Table 1 Gas pipeline leakage and explosion risk event score table

[0047]

[0048]

[0049] The larger the score, the higher the risk level. Table 2 below is an explanation table of the alarm status score:

[0050] Table 2 Alarm status score description table

[0051]

[0052] Extract the alarm status score corresponding to the current monitoring indicator data from the abnormal status score table of the corresponding monitoring indicator;

[0053] Analyze the changing trends of monitoring indicator data through historical data and current data of monitoring indicators, and assign scores to different changing trends;

[0054] and obtaining the corresponding control system status and the status of related nodes in the risk evolution network, and assigning scores to different statuses;

[0055] Among them, the control system status is obtained through equipment signals or manual inspection reports;

[0056] Based on the score of the change trend and the score of the status, the risk event development trend prediction correction score is obtained; as shown in Table 3, the event development trend prediction correction score table:

[0057]

[0058] The alarm status score is corrected by correcting the risk event development trend score to obtain the early warning status score of the integrated utility corridor.

[0059] For example: assuming that the alarm level corresponding to the methane alarm node is level three, the corresponding alarm score is 5. If the future change trend of the methane concentration is a downward trend, and the control system status and the status of the relevant nodes in the risk evolution network are normal, the overall status is normal. It can be queried from Table 3 that the event development trend prediction correction score is -1, and the alarm status score is 4 after the correction; if the control system status or the status of the relevant nodes in the risk evolution network is faulty, the overall status is faulty. It can be queried from Table 3 that the event development trend prediction correction score is 0, and the alarm status score is 5 after the correction.

[0060] Step 5: Analyze the warning consequence score of the integrated pipeline corridor according to the consequence score of the risk event corresponding to the alarm node; analyze the warning level of the integrated pipeline corridor based on the warning status score and the warning consequence score.

[0061] Specifically, the consequences corresponding to the risk events of the integrated pipeline corridor are extracted from the historical data, the consequences of the risk events are sorted out and summarized, and a set of consequence indicators is obtained. Scores are assigned to the consequence indicators respectively; the assigned scores range from 1 to 5; the larger the score, the more serious the consequence;

[0062] Based on the several consequence indicator sets of several risk events pointed to by the alarm node, the largest consequence score in the several consequence indicator sets is the consequence score of the alarm node; where the alarm node is the first node whose monitoring indicator data is not within the preset threshold range. The following Table 4 is a score table for the risk event consequence indicator set;

[0063] Table 4 Scoring table of risk event consequence indicator set

[0064]

[0065]

[0066]

[0067] For example: the alarm node {liquid level alarm, humidity alarm, personnel inspection, ...} points to the first node {water supply pipeline leakage, drainage pipeline leakage, ...} points to the risk event {pipe cabin waterlogging}, and points to the direct consequences {personnel casualties, economic losses, supply interruption, social impact, traffic impact, ...}; at the same time, {pipe cabin waterlogging} causes pipeline corrosion, pipeline corridor corrosion, and ancillary facilities corrosion, resulting in the node {water supply pipeline leakage, drainage pipeline leakage, thermal pipeline leakage, gas pipeline leakage, control system failure, ...} and the risk events it points to {pipe cabin waterlogging, fire, explosion, collapse, ...} The risk increases, pointing to the indirect consequences {personnel casualties, economic losses, supply interruption, social impact, traffic impact, ...};

[0068] Assuming that the consequence score indicator set of the pipe cabin flooding is {casualties: population density spatiotemporal big data ρ<0.005; supply interruption (interruptible users; number of households associated with the natural gas pipeline<10,000); traffic impact: (secondary trunk road; moderate congestion)}, the consequence score of the pipe cabin flooding risk event is 4.

[0069] Based on the warning status score and warning consequence score of the integrated pipeline corridor, the warning level of the integrated pipeline corridor is obtained through hierarchical matrix analysis, and corresponding measures are taken according to the warning level; among them, the warning level is iterated in real time according to multi-source dynamic monitoring data and spatiotemporal big data.

[0070] The details are shown in Table 5 below:

[0071] Table 5 Warning level table of integrated pipeline corridor

[0072]

[0073] Among them, the severity of the warning level is level 1>level 2>level 3>level 4; based on the warning status and potential consequences, risk events, event nodes, warning levels and corresponding control recommendations are output.

[0074] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0075] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A comprehensive pipe gallery monitoring and early warning analysis method based on multi-mode perception fusion reasoning, characterized in that: The following steps are involved: Step 1: Extract risk events of the integrated pipeline corridor from the historical risk data of the integrated pipeline corridor and construct a risk event set of the integrated pipeline corridor; construct the risk event evolution chain according to the development nodes of the risk events; the historical risk data of the integrated pipeline corridor includes accident cases, standards, literature and disaster mechanism models of the integrated pipeline corridor; the chain structure includes the first node, evolution node and tail node; Step 2: Extract the multi-source heterogeneous monitoring data of the integrated pipeline corridor, perform feature extraction on the multi-source heterogeneous monitoring data, and construct an indicator description set that characterizes the status of the chain nodes; the multi-source heterogeneous monitoring data includes equipment monitoring data, equipment control signals, video image data, and personnel inspection record data; Step 3: Based on the overlapping nodes and indicator description sets of each risk event evolution chain, the indicator status changes of the same node are associated with different risk events to form a comprehensive pipeline corridor risk evolution network; Step 4: Analyze the early warning status score of the integrated pipeline corridor; the early warning status score is determined based on the alarm status score and the risk event development trend prediction correction score; the risk event development trend prediction correction score is determined based on the change trend of the monitoring indicator data, the control system status, and the status of related nodes in the risk evolution network; Step 5: Analyze the warning consequence score of the integrated pipeline corridor according to the consequence score of the risk event corresponding to the alarm node; analyze the warning level of the integrated pipeline corridor based on the warning status score and the warning consequence score.

2. According to claim 1, a comprehensive pipe gallery monitoring and early warning analysis method based on multi-mode perception fusion reasoning is characterized in that: The indicator description set includes original data and feature data processed based on the original data, forming a multi-source multi-modal data set that describes the status of chain nodes.

3. According to claim 1, a comprehensive pipe gallery monitoring and early warning analysis method based on multi-mode perception fusion reasoning is characterized in that: The analysis of the early warning status score of the comprehensive pipe gallery includes: Extract the alarm status score corresponding to the current monitoring indicator data from the abnormal status score table of the corresponding monitoring indicator; Analyze the changing trends of monitoring indicator data through historical data and current data of monitoring indicators, and assign scores to different changing trends; and obtaining the corresponding control system status and the status of related nodes in the risk evolution network, and assigning scores to different statuses; Based on the score of the change trend and the score of the status, the risk event development trend prediction correction score is obtained; The alarm status score is corrected by correcting the risk event development trend score to obtain the early warning status score of the integrated utility corridor.

4. According to claim 3, a comprehensive pipe gallery monitoring and early warning analysis method based on multi-mode perception fusion reasoning is characterized in that: The abnormal condition score table of the monitoring indicators includes: A number of monitoring indicator data and the corresponding abnormal conditions are extracted from the historical data, and the corresponding alarm status scores are assigned according to the abnormal degree of the abnormal condition. The range of the monitoring indicator data corresponding to each abnormal condition is summarized to obtain the abnormal condition score table of each monitoring indicator.

5. According to claim 1, a comprehensive pipe gallery monitoring and early warning analysis method based on multi-mode perception fusion reasoning is characterized in that: The analysis of the early warning consequence score of the integrated pipe gallery according to the consequence score of the risk event corresponding to the alarm node includes: Extract the consequences of the risk events of the integrated pipeline corridor from the historical data, sort out and summarize the consequences of the risk events, obtain a set of consequence indicators, and assign scores to the consequence indicators respectively; Based on several consequence indicator sets of several risk events pointed to by the alarm node, the maximum consequence score in the several consequence indicator sets is the consequence score of the alarm node.

6. According to claim 1, a comprehensive pipe gallery monitoring and early warning analysis method based on multi-mode perception fusion reasoning is characterized in that: The warning level of the integrated pipe gallery is analyzed based on the warning status score and the warning consequence score, including: Based on the warning status score and warning consequence score of the integrated pipeline corridor, the warning level of the integrated pipeline corridor is obtained through hierarchical matrix analysis.

Images