Pipeline corridor control method, device and storage medium based on pipeline corridor perception network

Abstract

An embodiment of the present invention provides a tunnel control method, device, and storage medium based on a tunnel sensing network. The method includes: upon obtaining construction information, determining a first target chamber based on the construction information; controlling a fan module to perform ventilation processing on the first target chamber based on sensor information from a sensor module, so that the air quality of the first target chamber is equal to a preset quality; obtaining water level information of the tunnel chamber through the sensor module; and, when the water level information indicates that water has accumulated in a second target chamber, controlling a water pump module to pump water from the second target chamber, so that the depth of the water accumulated in the second target chamber is less than a preset depth. Ventilating the tunnel in advance based on construction information ensures smooth construction progress and high construction control efficiency. By timely obtaining water level information to control the water pump module to perform water pumping, unnecessary damage to the tunnel is avoided.

Description

Technical Field

[0001] The present invention relates to the field of, but is not limited to, information processing technology, and in particular to a tunnel control method, device, and storage medium based on a tunnel sensing network. Background Art

[0002] A utility gallery is an integrated underground urban pipeline corridor. Typically constructed as a tunnel beneath a city, it integrates various engineering pipelines, including power, communications, gas, heating, water supply, and drainage. It features specialized inspection and lifting ports, as well as monitoring systems, and is uniformly planned, designed, constructed, and managed.

[0003] At present, the management and control of the pipeline corridor is carried out by relevant personnel, which has low management efficiency. When accidents occur in the pipeline corridor, it is difficult to discover and deal with them in time, which can easily cause unnecessary losses to the pipeline corridor. Summary of the Invention

[0004] The following is a summary of the subject matter described in detail herein. This summary is not intended to limit the scope of the claims.

[0005] The main purpose of the embodiments of the present invention is to propose a tunnel control method, equipment and storage medium based on a tunnel perception network, which can improve the management and control efficiency of the tunnel, and promptly detect and handle accidents in the tunnel, thereby avoiding unnecessary losses to the tunnel.

[0006] In a first aspect, an embodiment of the present invention provides a pipe gallery control method based on a pipe gallery perception network, which is applied to a pipe gallery system, wherein the pipe gallery system includes multiple pipe gallery chambers and a pipe gallery perception network, wherein the pipe gallery chamber is provided with a sensor module, a water pump module, a fan module, and a gate control module, wherein the gate control module is provided at the inlet and outlet ends of the pipe gallery chamber, and adjacent pipe gallery chambers are connected through the gate control module, and the sensor module, the water pump module, the fan module, and the gate control module communicate based on the pipe gallery perception network. The method includes:

[0007] When construction information is obtained, a first target chamber is determined according to the construction information, where the first target chamber represents the pipe gallery chamber where construction is being carried out and where construction workers pass through. The construction information is obtained through a construction checklist.

[0008] Controlling the fan module to perform ventilation processing on the first target chamber according to the sensor information of the sensor module specifically includes: obtaining a first volume of a to-be-constructed area and a second volume of a construction-passed area in the first target chamber, and controlling the fan module of the first target chamber to perform ventilation processing based on the first volume, the sensor information, and the second volume, so that the air quality of the first target chamber is equal to a preset quality;

[0009] Acquiring water level information of the pipe gallery chamber through the sensor module;

[0010] When the water level information indicates that water accumulates in the second target chamber, the water pump module is controlled to pump water from the second target chamber so that the depth of water accumulation in the second target chamber is less than a preset depth.

[0011] In some optional embodiments, after controlling the water pump module to pump water from the second target chamber, the method further includes:

[0012] Obtaining maintenance knowledge graphs of the first target chamber and the second target chamber;

[0013] A corresponding maintenance plan is generated according to the maintenance knowledge graph and the accident information, wherein the accident information represents a liquid leakage accident and a gas leakage accident.

[0014] In some optional embodiments, before generating a corresponding maintenance plan based on the maintenance knowledge graph and accident information, the method further includes:

[0015] Calculating the safety risk coefficient of the target accident corresponding to the accident information through the consequence assessment model and risk prediction model included in the maintenance knowledge graph;

[0016] Calculating a business impact coefficient of the target accident using a business impact model included in the maintenance knowledge graph, where the business impact coefficient indicates the economic loss, social impact, and recovery time caused by the target accident;

[0017] Calculating the maintenance difficulty coefficient of the target accident using the maintenance difficulty model included in the maintenance knowledge graph;

[0018] Determining a handling priority of the target accident based on the safety risk factor, the business impact factor, and the maintenance difficulty factor;

[0019] The maintenance plans are generated in sequence according to the processing priorities.

[0020] In some optional embodiments, generating a corresponding maintenance plan based on the maintenance knowledge graph and accident information includes:

[0021] generating an accident feature vector based on the accident information and the maintenance knowledge graph, wherein the accident feature vector represents the accident type, accident severity, accident time, accident location, and environmental parameters;

[0022] Comparing the accident feature vector with multiple historical cases in the historical accident case library of the maintenance knowledge graph for similarity, and configuring the historical case with the highest similarity as a candidate case;

[0023] When the similarity between the candidate case and the accident feature vector is greater than a preset similarity threshold, obtaining a candidate solution configured for the candidate case from a maintenance solution library of the maintenance knowledge graph, and configuring the candidate solution as the maintenance solution for the accident feature vector;

[0024] When the similarity between the candidate case and the accident feature vector is less than a preset similarity threshold, obtaining a difference vector between the candidate case and the accident feature vector, where the difference vector represents a vector composed of difference data between the candidate case and the accident feature vector;

[0025] Sending the difference vector, the candidate solution, the candidate case, and the accident feature vector to an operation and maintenance terminal, so that the operation and maintenance terminal inputs a manual correction parameter;

[0026] The selected solution is corrected by the manual correction parameters to obtain a corrected solution;

[0027] The correction plan is configured as the maintenance plan of the accident feature vector.

[0028] In some optional embodiments, controlling the fan module to perform ventilation processing on the first target chamber according to the sensor information of the sensor module includes:

[0029] Determine the locations to be constructed inside the pipe gallery based on the construction list and generate a three-dimensional space model;

[0030] Determine a first volume of the area to be constructed and a second volume of the area to be constructed based on the three-dimensional spatial model and the optimal path planning model, wherein the second volume represents the volume of the pipe gallery chamber passed by the construction personnel, and the first volume represents the volume of the pipe gallery chamber to be constructed. The optimal path planning model is used to generate an optimal path for the construction personnel to reach the area to be constructed based on the three-dimensional spatial model of the area to be constructed and the pipe gallery;

[0031] acquiring the sensor information through a gas sensor provided on the optimal path in the sensor module, wherein the sensor information indicates the air quality on the optimal path;

[0032] The fan module of the first target chamber is controlled according to the first volume, the sensor information, the optimal path and the second volume so that the air quality of the area to be constructed is greater than a first threshold, the air quality of the area through which construction is conducted is greater than a second threshold, and the first threshold is greater than the second threshold.

[0033] In some optional embodiments, controlling the blower module of the first target chamber according to the first volume, the sensor information, the optimal path, and the second volume includes:

[0034] Determine a starting position and an end position according to the optimal path, wherein the end position represents the location of the construction area, and the starting position represents the location where construction personnel enter the pipe gallery;

[0035] activating the gate control module on the optimal path according to the starting position and the end position, so that the pipe gallery chambers on the optimal path are interconnected;

[0036] determining a first ventilation power of a first ventilation module and a second ventilation power of a second ventilation module according to the optimal path, the sensor information, the first volume, and the second volume, wherein the first ventilation module represents the fan module corresponding to the starting position, and the second ventilation module represents the fan module on the optimal path, and the second ventilation module does not include the first ventilation module;

[0037] Control the first ventilation module to blow air toward the starting position at the first ventilation power, and control the second ventilation module to exhaust air along the optimal path at the second ventilation power, so that the air quality of the area to be constructed is greater than the first threshold and the air quality of the area through which construction is passed is greater than the second threshold.

[0038] In some optional embodiments, the water pump module includes a first water pump unit, a second water pump unit, and a third water pump unit, the first water pump unit and the second water pump unit are arranged in the second target chamber, the third water pump unit is arranged outside the second target chamber, and the third water pump unit pumps water to the second target chamber through a vent on the second target chamber. Controlling the water pump module to pump water to the second target chamber includes:

[0039] When the water level information indicates that the water depth of the second target chamber is greater than or equal to the first depth, starting the first water pump unit to pump water from the second target chamber;

[0040] When the water level information indicates that the water depth of the second target chamber is greater than or equal to a second depth, starting the first water pump unit and the second water pump unit to pump water from the second target chamber;

[0041] When the water level information indicates that the water depth of the second target chamber is greater than or equal to a third depth, the first water pump unit, the second water pump unit, and the third water pump unit are started to pump water from the second target chamber.

[0042] In some optional embodiments, the method further includes:

[0043] When the water level information indicates that the water depth of the second target chamber is greater than or equal to the first depth, obtaining a first water depth change rate according to the water level information, determining a first pumping power of the first water pump unit according to the first water depth change rate, and controlling the first water pump unit to pump water from the second target chamber at the first pumping power;

[0044] When the water level information indicates that the water depth of the second target chamber is greater than or equal to the second depth, obtaining a second water depth change rate according to the water level information, determining a second pumping power of the first water pump unit and a third pumping power of the second water pump unit according to the second water depth change rate, and controlling the first water pump unit to pump water from the second target chamber at the second pumping power and the second water pump unit to pump water at the third pumping power;

[0045] When the water level information indicates that the water accumulation depth of the second target chamber is greater than or equal to the third depth, the third water depth change rate is obtained according to the water level information, and the fourth pumping power of the first water pump unit, the fifth pumping power of the second water pump unit, and the sixth pumping power of the third water pump unit are determined according to the third water depth change rate. The first water pump unit is controlled to pump water to the second target chamber with the fourth pumping power, the second water pump unit is controlled to pump water to the second target chamber with the fifth pumping power, and the third water pump unit is controlled to pump water to the second target chamber with the sixth pumping power.

[0046] In the second aspect, an embodiment of the present invention provides a tunnel control device based on a tunnel perception network, comprising: a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the computer program, the tunnel control method based on the tunnel perception network described in the first aspect is implemented.

[0047] In a third aspect, an embodiment of the present invention provides a computer storage medium storing computer-executable instructions, wherein the computer-executable instructions are used to execute the tunnel control method based on the tunnel perception network described in the first aspect.

[0048] The beneficial effects of the present invention include: upon obtaining construction information, determining a first target chamber based on the construction information, wherein the first target chamber represents the pipe corridor chamber where construction is being carried out and where construction workers pass through, and the construction information is obtained through a construction checklist; controlling the fan module to ventilate the first target chamber based on sensor information from the sensor module, so that the air quality of the first target chamber equals a preset quality; obtaining water level information of the pipe corridor chamber through the sensor module; and, when the water level information indicates that water has accumulated in a second target chamber, controlling the water pump module to pump water from the second target chamber so that the depth of the water accumulated in the second target chamber is less than a preset depth. Ventilating the pipe corridor in advance based on the construction information ensures smooth construction progress and high construction management efficiency. In the event of a water pipe burst, the water level information is promptly obtained to control the water pump module to pump water, thereby avoiding unnecessary damage to the pipe corridor.

[0049] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purposes and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 This is a flowchart of the steps of a pipeline corridor control method based on a pipeline corridor sensing network provided by an embodiment of the present invention;

[0051] Figure 2 This is a schematic block diagram of the structure of a pipe gallery chamber provided by an embodiment of the present invention;

[0052] Figure 3 Schematic diagram of the structure of the pipe gallery sensing network provided by an embodiment of the present invention;

[0053] Figure 4 Schematic diagram of a controller provided by one embodiment of the present invention.

[0054] Reference numerals: controller 1000 , processor 1100 , memory 1200 ;

[0055] Pipe gallery chamber 100, fan module 110, fan control unit 111, fan unit 112, fan duct 120, vent 130, environmental sensor 140, third water pump unit 150, first water pump unit 151, second water pump unit 152, water level sensor 160, door control module 170, lighting device 180. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0057] It should be noted that although the device schematics illustrate functional module divisions and the flowcharts illustrate logical sequences, in certain circumstances, the steps shown or described may be performed in a sequence that differs from the module divisions in the device or the sequence in the flowcharts. The terms "first," "second," and the like in the specification, claims, or accompanying drawings are used to distinguish similar items and are not necessarily used to describe a specific sequence or precedence.

[0058] A utility gallery is an integrated underground urban pipeline corridor. Typically constructed as a tunnel beneath a city, it integrates various engineering pipelines, including power, communications, gas, heating, water supply, and drainage. It features specialized inspection and lifting ports, as well as monitoring systems, and is uniformly planned, designed, constructed, and managed.

[0059] At present, it is difficult to detect and monitor construction activities or intrusions in the tunnel in a timely manner, which can easily cause unnecessary losses to the tunnel.

[0060] To solve the above-mentioned problems, the present application provides a tunnel control method, device and storage medium based on a tunnel sensing network.

[0061] In this application, a tunnel control method, device and storage medium based on a tunnel perception network are provided, which are described in detail one by one in the following embodiments.

[0062] like Figure 1 As shown, an embodiment of the present invention provides a tunnel control method based on a tunnel perception network, which is applied to a tunnel system. The tunnel system includes multiple tunnel chambers 100 and a tunnel perception network. The tunnel chamber 100 is provided with a sensor module, a water pump module, a fan module 110 and a door control module 170. The door control module 170 is provided at the entrance and exit ends of the tunnel chamber 100. Adjacent tunnel chambers 100 are connected through the door control module 170. The sensor module, the water pump module, the fan module 110 and the door control module 170 communicate based on the tunnel perception network. The method includes:

[0063] S100: When construction information is obtained, determine a first target chamber based on the construction information, where the first target chamber represents the pipe gallery chamber 100 where construction is being carried out and where construction workers pass through. The construction information is obtained from a construction checklist.

[0064] S200, controlling the fan module 110 to perform ventilation processing on the first target chamber according to the sensor information of the sensor module, specifically comprising: obtaining a first volume of the area to be constructed and a second volume of the area to be constructed in the first target chamber, and controlling the fan module of the first target chamber to perform ventilation processing based on the first volume, the sensor information, and the second volume, so that the air quality of the first target chamber is equal to a preset quality;

[0065] S300, obtaining water level information of the pipe gallery chamber 100 through the sensor module;

[0066] S400 : When the water level information indicates that water accumulates in the second target chamber, control the water pump module to pump water from the second target chamber so that the depth of water accumulation in the second target chamber is less than a preset depth.

[0067] Specifically, the tunnel in this application is divided into multiple tunnel chambers by several access control modules. Each tunnel chamber 100 includes a sensor module, a water pump module, a fan module, and a access control module. The sensor module includes an environmental sensor 140 and a water level sensor 160. The environmental sensor 140 includes a gas sensor (for detecting oxygen, carbon monoxide, methane, etc.), a temperature and humidity sensor, a pressure sensor, and a video surveillance camera. The gas sensor is deployed at the top (for detecting gases with a density less than air) or the bottom (for detecting gases with a density greater than air). The water level sensor 160 is installed in a low-lying area, and the camera covers the entire chamber without blind spots. The water pump module is used to remove accumulated water and supports automatic switching. A filter is installed at the water pump inlet and an electromagnetic flowmeter is installed at the outlet. The fan module 110 is installed at the top of the chamber for air replacement and ventilation. The access control module 170 includes an intelligent fire and explosion-proof door equipped with an electric door closer, an electromagnetic lock, and an emergency manual switch. The door status signal (open / closed / fault) is uploaded to the corridor sensing network in real time, and when linked with the fire alarm system, the adjacent chamber doors are automatically closed.

[0068] Reference Figure 3 The tunnel perception network architecture includes:

[0069] Backbone network: Integrates environmental monitoring, security, and communications networks into one, isolating multiple business resources; ensures bandwidth, latency, and reliability based on IPv6+; integrates wired and wireless network operations and maintenance, enables network status visualization, and facilitates fault diagnosis.

[0070] Network Access: Sensors and actuators connect to WiFi APs via WiFi IOBOXs, supporting WiFi 6 / 5G wireless backup communications. WiFi IOBOXs enable monitoring terminals within the tunnel to be wired, saving cabling and easing maintenance. Deployment of Wi-Fi IOBOXs provides terminal access (RS485, DI, DO, and AI interfaces), sensor power, and Wi-Fi wireless backhaul (supporting dual-transmit and selective receive). These devices support IP65 protection, plug-and-play, and iConnect seamless access. Compared to Wi-Fi 5, Wi-Fi 6 offers four times the speed, 10ms end-to-end latency, and four times the number of users. Wi-Fi 6 provides continuous WLAN coverage within the tunnel, eliminating blind spots and ensuring seamless roaming and handover.

[0071] Data Hub: Edge computing nodes (edge ​​computing gateway + software PLC), deployed near each chamber, perform data preprocessing (such as denoising and normalization), local rule engines (such as automatically starting a pump when the water level exceeds a threshold), and equipment status prediction (based on machine learning algorithms). Software PLCs deployed on edge computing gateways replace traditional hardware PLCs, providing autonomous control and supporting active / standby high-reliability deployment. A cloud-based management platform builds a digital twin system for the tunnel, displaying each chamber's status parameters, equipment operating curves, and 3D spatial models in real time. It supports remote control and big data analysis, as well as tunnel environmental monitoring, security management, emergency response management, and operations and maintenance management.

[0072] Through the integrated sensing gateway, environmental monitoring, equipment control terminal, security, and communication terminal set up on the edge node or terminal side: the Internet of Things gateway expands the fiber optic sensing monitoring capabilities, integrating networking, computing, and measurement; high integration, easy deployment, and cost reduction of more than 50%; real-time monitoring of pipeline corridor settlement, construction damage, and temperature anomalies, and coordinated disposal.

[0073] The tunnel chamber 100 of the present application is configured as follows: a cabinet is deployed in the equipment room of the tunnel chamber 100, where the export router (uplinked to the Mosa switch to connect to the monitoring center), the industrial switch (connected to the Wi-Fi AP), and the all-in-one IoT gateway (temperature, vibration, strain monitoring and early warning, providing edge computing operating software PLC) are centrally placed; 4 Wi-Fi 6 outdoor APs are deployed to achieve Wi-Fi coverage in the equipment room and cabin; a group of sensors, including temperature and humidity, oxygen, hydrogen sulfide, methane, and liquid level gauges, are deployed; 4 IOBOXes are deployed to achieve "braid cutting" of the monitoring terminal, including three Wi-Fi IOBOXes for lighting devices 180, water pumps, and sensors, and an Ethernet IOBOX for fans. Sensors and lighting / fan / water pump control boxes are connected to the IOBOX; the Wi-Fi IOBOX needs to be powered by the power distribution cabinet wiring in the equipment room (AC 220V), and the Wi-Fi AP is uplinked to the industrial switch in the cabinet (communication and POE power supply).

[0074] Extract key information from the construction checklist: construction time window, chamber numbers involved (such as C01-C05), construction content (such as pipe welding, cable laying), areas that need to be isolated, etc.

[0075] Combined with historical construction data, it automatically identifies high-risk operations (such as hot work) and triggers additional safety measures (such as increasing the CO detection frequency to once every five minutes).

[0076] Ventilation strategy generation: Fan modules 110 are activated for ventilation in the construction chamber and the chambers through which construction personnel pass. Pollutant concentrations in the construction chamber are continuously monitored to ensure smooth construction progress. Based on construction progress and personnel movement trajectories, the scope of the first target chamber is updated in real time to ensure ventilation and safety coverage throughout the construction area.

[0077] When air quality falls below a preset standard, the system automatically activates the fan module in a "supply + exhaust" mode, introducing fresh air through the air inlet and exhausting polluted air through the exhaust. The fan speed is adjusted based on the level of pollution (for example, increasing to high speed when dust concentration exceeds the standard) until air quality meets the standard. During ventilation, a gas sample is collected from the center of the chamber every 10 minutes to verify that the oxygen concentration is ≥19.5%, the combustible gas concentration is <10% of the LEL (Lower Explosive Limit), and the dust concentration is <8mg / m³. Specific indicators are not specified. If any indicator is not met, the ventilation time is automatically extended or the backup fan is activated until the air quality reaches the standard and stabilizes.

[0078] The sensor module collects chamber water level data in real time and combines historical data to establish a three-level threshold of "normal-warning-alarm" (such as warning value 5cm, alarm value 10cm).

[0079] When the water level reaches the warning value, the edge computing node automatically generates water level information, including the water level rise rate, water level value, predicted time to reach the alarm value, etc.

[0080] The drainage strategy is implemented in a hierarchical manner: Level 1 response (water level < 10 cm): The main water pump is started and operated alone, with a drainage flow rate of 50 m³ / h. The gate control module 170 of the chamber and the downstream chamber is simultaneously closed to prevent backflow of accumulated water. Level 2 response (water level 10 cm ≤ < 30 cm): The main and backup water pumps are operated in parallel, with a total flow rate of 80 m³ / h. Level 3 response (water level ≥ 30 cm): "Flood control emergency mode" is triggered, and all water pumps are coordinated to drain water. The digital twin system simulates the optimal drainage path (for example, draining water from the lowest chamber first).

[0081] Drainage Effectiveness Assessment: Drainage efficiency (drainage flow rate minus leakage flow rate) is calculated in real time. If the water level drops less than 2 cm / min within 30 minutes, the system automatically switches to "intermittent drainage with leak plugging priority" mode, dispatching a robot to locate the leak. After drainage is complete, a health diagnosis is performed on parameters such as the pump motor temperature and bearing vibration, automatically generating maintenance recommendations (e.g., lubricant replacement after 500 hours of cumulative operation).

[0082] If the gas sensor detects a sudden increase in CO (carbon monoxide) concentration (e.g., >500 ppm) and the temperature and humidity sensor detects a temperature rise rate >5°C / min, it automatically identifies a fire hazard and triggers the following actions: all door control modules 170 are closed to form fire zones; the fans are switched to "negative pressure smoke exhaust" mode to prioritize the exhaust of toxic fumes; and the fire sprinkler system is activated to cool down adjacent chambers. If the camera, using an AI algorithm, identifies a construction worker not wearing a helmet or entering an unauthorized area, it immediately sends an audible and visual alarm to the site and pushes a warning message to the operator's mobile app.

[0083] In some optional embodiments, after controlling the water pump module to pump water from the second target chamber, the method further includes: obtaining a maintenance knowledge graph of the first target chamber and the second target chamber; generating a corresponding maintenance plan based on the maintenance knowledge graph and accident information, wherein the accident information represents a liquid leakage accident and a gas leakage accident.

[0084] Specifically, the maintenance knowledge graph is used to generate corresponding maintenance plans, thereby guiding maintenance personnel to quickly complete the inspection of the pipeline corridor. The data architecture of the knowledge graph of this application includes: Equipment entities: including water pumps, fans, sensors, pipeline valves, etc. Each entity contains attributes such as model, installation location, operating parameters (such as water pump head, power), supplier, etc. Accident type: covers liquid leakage, gas leakage, equipment failure and other types, and records the time, location, severity (such as leakage volume, scope of impact), and initial cause (such as corrosion, external force damage) of the accident. Maintenance plan: includes maintenance personnel, tools, spare parts, plan steps, etc., and associates maintenance time, cost, and effect evaluation indicators.

[0085] The knowledge graph is configured with corresponding relationship layers: Equipment-accident relationships, such as "a certain model of water pump leaks liquid due to impeller wear." Accident-solution relationships, such as "a gas leak accident requires a three-step plan of 'valve closing, ventilation, and testing.'" Solution-resource relationships, such as "a pipeline leak plugging solution requires epoxy resin glue, patch clamps, and underwater workers."

[0086] Dynamic knowledge graph updates: Data from each sensor in the sensor module and operational data from each device (such as pump start and stop times, leak detection signals, and carbon monoxide concentrations) are collected in real time. Unstructured data such as maintenance reports, accident investigation reports, and equipment replacement records are manually entered. Natural language processing (NLP) technology is used to parse maintenance documents and extract key information (such as "In May 2023, a leak occurred at a weld in the C03 chamber pipe, which was repaired by repair welding"). Graph databases (such as Neo4j) are used to semantically correlate multi-source data and eliminate data silos.

[0087] Knowledge Verification: An expert review mechanism is introduced to manually verify automatically generated knowledge nodes (such as new maintenance solutions). The accuracy of the knowledge is verified through feedback from actual maintenance results (such as return repair rates and fault recurrence cycles), forming a closed loop of "collection-integration-verification-optimization."

[0088] Liquid leakage: The water depth change curve recorded by the water level sensor 160, and / or the pipeline flow abnormality detected by the ultrasonic flow meter (such as a sudden drop of 30%).

[0089] Gas leakage: Leakage of the gas pipeline is determined by the concentration of each gas in the pipe gallery chamber 100 detected by the gas sensor, and / or the leakage of the gas pipeline is determined by the pipeline pressure drop rate captured by the pressure sensor.

[0090] Visual data: Video footage of the leak site captured by cameras installed in the pipe gallery chamber 100 is used to calculate water flow velocity and gas diffusion range using computer vision algorithms (such as optical flow). Images of the pipeline interior walls captured by robotic inspections are used to identify features such as crack length (e.g., 5 cm) and corrosion area (e.g., 10 cm²).

[0091] Manually reported data: accident description (such as "water gushing out from the pipe interface at the top of chamber C05"), on-site photos, and preliminary handling measures reported by on-site personnel through the APP.

[0092] Feature extraction and encoding:

[0093] Numerical characteristics of liquid leakage accident characteristics: leakage duration (t), water depth (h), leakage flow rate (Q), pipeline pressure (P). Categorical characteristics of liquid leakage accident characteristics: leakage location (e.g., elbow / straight pipe section), pipeline material (e.g., ductile iron / PE), leakage form (e.g., spray / drip).

[0094] Numerical characteristics of gas leakage accident characteristics: gas concentration (C), diffusion velocity (v), ambient temperature (T), wind speed (u). Categorical characteristics of gas leakage accident characteristics: gas type (e.g., methane / carbon monoxide), leak point sealing method (e.g., flange connection / weld), explosion-proof area level (e.g., Zone 0 / Zone 1).

[0095] The numerical features of liquid leak accident characteristics and their categorical features are numerically coded accordingly to obtain the accident code corresponding to the liquid leak accident. Similarly, the numerical features of gas leak accident characteristics and their categorical features are numerically coded accordingly to obtain the accident code corresponding to the gas leak accident. This allows for rapid matching of the accident code with the maintenance knowledge graph to generate the corresponding maintenance plan.

[0096] In some optional embodiments, before generating a corresponding maintenance plan based on the maintenance knowledge graph and accident information, the method further includes: calculating the safety risk coefficient of the target accident corresponding to the accident information through the consequence assessment model and risk prediction model contained in the maintenance knowledge graph; calculating the business impact coefficient of the target accident through the business impact model contained in the maintenance knowledge graph, the business impact coefficient indicating the economic loss, social impact and recovery time caused by the target accident; calculating the maintenance difficulty coefficient of the target accident through the maintenance difficulty model contained in the maintenance knowledge graph; determining the processing priority of the target accident based on the safety risk coefficient, the business impact coefficient and the maintenance difficulty coefficient; and generating the maintenance plan in sequence according to the processing priority.

[0097] Specifically, the consequence assessment model of the present application is obtained by training multiple groups of corresponding historical data. Specifically, by comprehensively considering parameters such as accident type, degree of hazard, and scope of impact, an assessment matrix is ​​constructed through the analytic hierarchy process (AHP), which is to obtain a consequence assessment model. The consequence coefficient caused by the accident can be calculated through the consequence assessment model. The larger the consequence coefficient, the more serious the consequences of the accident. That is, the direct hazards that may be caused by the target accident are analyzed, such as the depth of accumulated water from liquid leakage, the toxicity or flammability level of gas leakage, and the frequency of personnel activities in the corridor chamber, the distribution of equipment, etc., to calculate the potential severity of the consequences of the accident on personnel safety and equipment damage. Consequence coefficient of the consequence assessment model The calculation formula includes:

[0098]

[0099] in, is the weight of the risk factor (e.g., gas toxicity weight 0.4, leakage rate weight 0.3, personnel exposure risk weight 0.3); It is the quantitative value of the risk factor (the value range is 0-10. For example, when the concentration of methane leakage reaches 30% of the LEL, the quantitative value of the risk factor of gas toxicity is 8).

[0100] The risk prediction model uses historical accident evolution data and an LSTM neural network to predict the probability of an accident worsening. This means that the risk prediction model combines real-time sensor data (such as leakage volume and diffusion speed) with the tunnel structure (such as ventilation conditions and fire protection zones) to predict the spread of an accident. For example, in the event of a gas leak, the risk model simulates the diffusion range of harmful gases in the tunnel and assesses whether an explosion or personnel poisoning risk may occur in a short period of time. The input parameters of the risk prediction model are: current leakage rate ( ), ambient temperature and humidity (temperature ,humidity )、Rescue response time( ); Output of the risk prediction model (probability of deterioration ): The probability of risk level upgrade in the next hour (for example, the probability of upgrading from "medium risk" to "high risk" is 25%).

[0101] To adapt to the input requirements of the LSTM network, the original parameters are normalized:

[0102] , , ,

[0103] in, express The normalized value of express The minimum value of express The maximum value of express The normalized value of express The minimum value of express The maximum value of express The normalized value of express The minimum value of express The maximum value of express The normalized value of express The minimum value of express The maximum value of .

[0104] The LSTM network model includes: Input layer: receiving 3D feature vectors [ , , , Hidden layer: Single or multi-layer LSTM unit, capturing the temporal dependencies in time series data (such as the trend of leakage rate over time). Fully connected layer: maps LSTM output to degradation probability. The formula for calculating degradation probability is:

[0105]

[0106] in, Represents the activation function Sigmoid or the hyperbolic tangent activation function tanh; Indicates that the input gate is The weight matrix and bias vector at the moment; Indicates that the input gate is The weight matrix and bias vector at the moment; Indicates that the LSTM hidden layer is The state of the moment.

[0107] The calculated consequence coefficient and the probability of deterioration Substitute the preset weights into the weighted calculation formula to calculate the safety risk coefficient , safety risk factor A higher value indicates a higher risk of an accident.

[0108] The business impact model calculates the business impact coefficient by comprehensively calculating economic losses, social impact and recovery time ( Business impact coefficient ( ) can be directly quantified using economic losses: Economic losses = direct losses + indirect losses. Direct losses include equipment repair costs and spare parts replacement fees. Indirect losses include production and work stoppage losses (e.g., a pipeline corridor outage causes daily losses of 1 million yuan for surrounding businesses) and user compensation costs. Substituting economic losses into the economic impact coefficient table yields the corresponding economic impact coefficient for different economic losses.

[0109] The social impact assessment adopts the fuzzy comprehensive evaluation method, and the indicators include: the number of affected residents (such as 10 points if the number is ≥1,000, 6 points if the number is 100-1,000, and 2 points if the number is less than 100); media attention (8 points if the number is reported by national media, 4 points if the number is local media, and 0 points if there is no report).

[0110] Recovery time assessment uses historical case statistics from the maintenance knowledge graph to create a "accident characteristics-recovery time" mapping table. For example, the average repair time for a DN300 pipeline rupture is 24 hours, while the average time for pressurized plugging is 4 hours.

[0111] Substitute the economic impact coefficient, social impact score, and recovery time into the business impact coefficient The business impact coefficient can be obtained by calculating the formula .

[0112] Maintenance difficulty coefficient This is the technical complexity assessment. The indicators include: maintenance process level (e.g., welding process requires a certified welder, which is scored 5 points, ordinary leak plugging, which is scored 2 points), equipment accessibility (diving operation required, which is scored 6 points, ground operation, which is scored 1 point), and the maintenance difficulty model comprehensively calculates the coefficient corresponding to the technical complexity by scoring each indicator. . Resource scarcity assessment:

[0113]

[0114] in, The current resource inventory (e.g., 5 seals of a certain model in stock); : The amount of resources required for maintenance (e.g. this model requires 3 sealing rings).

[0115] When resources are sufficient ( ≥ ), resource scarcity coefficient 1; when urgent purchase is required, 0.3-0.8 (adjusted according to the length of the procurement cycle). Comprehensive calculation and You can get the maintenance difficulty coefficient .

[0116] Processing priority determination: Security risk factor , business impact coefficient , Maintenance difficulty coefficient Normalize to the interval [0,1] through linear transformation:

[0117] , , ;

[0118] The weighted summation method is used to calculate the comprehensive priority index : .in, Represents the normalized safety risk factor The weight value of Represents the normalized business impact coefficient The weight value of Represents the normalized maintenance difficulty coefficient The specific weights are set according to the safety standards for pipe gallery operation and maintenance (for example, the safety risk factor accounts for 50%, the business impact factor accounts for 30%, and the maintenance difficulty factor accounts for 20%, with no specific restrictions).

[0119] The priority levels are shown in Table 1:

[0120] Table 1

[0121]

[0122] Special / Level 1 priority: Directly access the "High Priority Plan Library" in the knowledge graph, which includes predefined rapid response procedures (e.g., Special Response for Gas Leak: Close all adjacent valves within 10 minutes; activate a mobile exhaust truck within 30 minutes; and complete leak repair within 2 hours). This automatically triggers cross-departmental coordination mechanisms, such as notifying fire, environmental protection, and public security departments for coordinated response.

[0123] Secondary / tertiary priority: Generate a general plan by matching similar cases in the maintenance knowledge graph (as described above), and allow operation and maintenance personnel to adjust the plan details (such as replacing spare parts brands) within their authority.

[0124] Resource conflict resolution: When multiple high-priority accidents occur simultaneously, resource allocation is dynamically adjusted according to the priority index: special-level accidents will take up all available water pumps first; level-one accidents will be allocated from high to low according to the ratio of "remaining resources / required resources".

[0125] Time window management: Combined with maintenance time window constraints (such as prohibiting hot work in the pipeline corridor during urban traffic peak hours), the time nodes in the plan are automatically adjusted: the welding work was originally planned to be carried out from 9:00 to 11:00, but if there is a morning rush hour, it will be postponed to 12:00 to 14:00.

[0126] In some optional embodiments, the generating of a corresponding maintenance plan according to the maintenance knowledge graph and the accident information includes: generating an accident feature vector according to the accident information and the maintenance knowledge graph, the accident feature vector representing the accident type, accident severity, accident time, accident location and environmental parameters; comparing the accident feature vector with a plurality of historical cases in the historical accident case library of the maintenance knowledge graph for similarity, and configuring the historical case with the highest similarity as a candidate case; and obtaining the candidate case configuration from the maintenance plan library of the maintenance knowledge graph when the similarity between the candidate case and the accident feature vector is greater than a preset similarity threshold. A candidate plan is selected, and the candidate plan is configured as the maintenance plan of the accident feature vector; when the similarity between the candidate case and the accident feature vector is less than a preset similarity threshold, a difference vector between the candidate case and the accident feature vector is obtained, and the difference vector represents a vector composed of difference data between the candidate case and the accident feature vector; the difference vector, the candidate plan, the candidate case and the accident feature vector are sent to the operation and maintenance terminal, so that the operation and maintenance terminal inputs manual correction parameters; the candidate plan is corrected by the manual correction parameters to obtain a corrected plan; and the corrected plan is configured as the maintenance plan of the accident feature vector.

[0127] Specifically, by traversing the historical accident case database (consisting of several past accident records) in the maintenance knowledge graph, a feature vector of the same dimension is extracted for each historical case. This feature vector contains the accident type, severity, time of occurrence, location, and environmental parameters. From the accident information, the accident type (such as liquid leak, gas leak, or equipment failure), severity (measured by leakage volume and impact range), time of occurrence, location (specific coordinates within the tunnel chamber), and environmental parameters (temperature, humidity, wind speed, and gas concentration) are extracted. This data is structured into a format that facilitates calculation and analysis, resulting in an accident feature vector containing the accident type, severity, time of occurrence, location, and environmental parameters. The accident feature vector is constructed using either a word vector or a numeric vector. All of this data is then fused into a multidimensional vector that fully represents the accident characteristics.

[0128] The historical accident case library of the maintenance knowledge graph stores a large number of past accident cases. Each case is represented by a feature vector containing the accident type, accident severity, accident time, accident location and environmental parameters. Therefore, through simple vector comparison, the historical case with the highest similarity to the accident feature vector can be quickly matched in the historical accident case library, and the maintenance plan for the accident can be quickly generated based on the solution of the historical case, so that maintenance response can be carried out in time according to the maintenance plan after the accident to avoid causing greater losses.

[0129] In some embodiments, algorithms such as cosine similarity and Euclidean distance are used to calculate the similarity between the accident feature vector and the historical case vector. Taking cosine similarity as an example, its calculation formula is:

[0130]

[0131] in, is the accident feature vector, is the historical case vector, represents the dot product of two vectors, Represents the modulus of two vectors. The calculated similarity value The range is [-1, 1]. The closer the value is to 1, the more similar the two vectors are, that is, the more similar the accident characteristics are to the historical case. All historical cases are ranked by their similarity to the accident characteristic vectors, and the historical case with the highest similarity (i.e., the one with the highest similarity ranking) is selected as the candidate case.

[0132] When the similarity between the candidate case and the accident feature vector exceeds a preset similarity threshold (e.g., 0.8), it indicates that the current accident has a high degree of similarity with the historical case, and historical maintenance experience can be directly leveraged. Therefore, the candidate solution configured for the candidate case is retrieved from the maintenance solution library within the maintenance knowledge graph. This solution includes detailed repair steps for the historical case, required resources (personnel, tools, spare parts, etc.), safety precautions, and other information. This specific information is determined by the actual maintenance plan and is not limited here. This candidate solution is directly configured as the maintenance solution for the current accident feature vector, eliminating the need to reanalyze the accident and generate a new solution, thereby improving the efficiency of accident maintenance decision-making.

[0133] When the similarity between the candidate case and the accident feature vector falls below a preset similarity threshold, a difference vector is calculated. By comparing the two vectors element by element, the difference data along each dimension is obtained and formed into a difference vector. For example, if the leak volume in the accident feature vector is 50 cubic meters, and the leak volume in the candidate case is 30 cubic meters, the difference data along the leak volume dimension is 20 cubic meters, and so on, until a complete difference vector is formed.

[0134] The difference vector, candidate solutions, candidate cases, and accident feature vector are sent to the operation and maintenance terminal (including specific host terminals, computer terminals, and / or mobile terminals, etc., and the specific operation and maintenance terminal is not limited). The operation and maintenance personnel use their professional knowledge and experience to analyze the differences and input manual correction parameters. Correction parameters include adjusting the order of maintenance steps, adding or replacing maintenance tools, replenishing specific spare parts, etc. The specific correction method and content are not limited here. After the correction parameters are entered into the system, the system will modify the candidate solutions based on the manual correction parameters. For example, on the basis of the original solution, equipment and operating procedures for larger leaks are added to obtain a corrected solution, which is then configured as the maintenance solution for the accident feature vector.

[0135] The maintenance plan generation of this application combines intelligent matching and manual experience. It not only uses historical data to improve the efficiency of maintenance plan generation, but also ensures the applicability of the plan through manual correction, reducing the risk of unreasonable maintenance plans and even maintenance errors caused by blindly applying historical plans.

[0136] In some optional embodiments, the controlling the fan module 110 to perform ventilation processing on the first target chamber according to the sensor information of the sensor module includes: determining the position to be constructed inside the tunnel based on the construction list and generating a three-dimensional space model; determining the first volume of the area to be constructed and the second volume of the construction area according to the three-dimensional space model and the optimal path planning model, the second volume representing the volume of the tunnel chamber 100 passed by the construction personnel, and the first volume representing the volume of the tunnel chamber 100 to be constructed, and the optimal path planning model is used to generate the optimal path for the construction personnel to reach the area to be constructed according to the three-dimensional space model of the area to be constructed and the tunnel; obtaining the sensor information through the gas sensor set on the optimal path in the sensor module, and the sensor information indicates the air quality on the optimal path; controlling the fan module 110 of the first target chamber according to the first volume, the sensor information, the optimal path and the second volume, so that the air quality of the area to be constructed is greater than a first threshold, the air quality of the construction area is greater than a second threshold, and the first threshold is greater than the second threshold.

[0137] Specifically, the fan module 110 of the present application includes a fan unit 112 for extracting or blowing air and a fan control unit 111 for controlling the fan unit 112. The fan unit 112 extracts or blows air through a fan duct 120. During and before tunnel construction, the fan module 110 is controlled based on sensor information from the sensor module to ventilate the first target chamber, thereby ensuring construction safety and personnel health.

[0138] Before construction begins, a construction checklist uploaded to the system or pre-existing in the system is used to obtain detailed coordinates, shape, dimensions, and other information about the locations within the corridor where construction will begin after a preset time period. This time period can be half an hour or an hour before construction, depending on actual ventilation efficiency and construction schedules, and is not specifically limited here. Using 3D modeling software and combining it with the existing structural design drawings of the corridor, a 3D spatial model of the area to be constructed and its surroundings is constructed.

[0139] Based on the 3D spatial model, geometric algorithms are used to calculate the first volume of the area to be constructed and the second volume of the area to be constructed. For regularly shaped areas, the volume formula can be used directly; for irregular areas, methods such as gridding or numerical integration are used to approximate the volume.

[0140] At the same time, an optimal path planning model can be used to determine the optimal route for construction workers to reach the area to be constructed. Based on the three-dimensional spatial model of the utility corridor, this model comprehensively considers path length, obstacles (such as equipment and pipeline layout), and safety factors (such as ventilation conditions and the distribution of hazardous areas). Using a corresponding path planning algorithm, it generates an optimal path from the construction workers' starting location to the area to be constructed. During the path planning process, the path can be dynamically adjusted based on real-time personnel flow information and equipment status information to ensure efficient and safe passage for construction workers. It is easy to see that based on the optimal path, the first volume of the area to be constructed and the second volume of the area to be constructed can be determined.

[0141] The present application arranges multiple gas sensors on the optimal path, which are used to collect air quality data on the optimal path, including oxygen concentration, harmful gases (such as carbon monoxide CO, hydrogen sulfide The sensors transmit the collected data to the control system in real time. The control system (which obtains sensor data through the tunnel sensing network) pre-processes the data, such as filtering and denoising, and data normalization, to improve data accuracy and usability.

[0142] Sensor data is used to determine whether the current air quality along the optimal path meets construction safety standards and personnel access criteria. This assessment is made by comparing collected data such as gas and dust concentrations with preset safety thresholds to assess air quality. This can be measured through quality scores, specifically based on the degree of deviation between gas concentrations and safety thresholds; higher deviations result in lower scores. If air quality fails to meet standards, ventilation is performed in the tunnel space along the optimal path.

[0143] During ventilation, differentiated control strategies for fan module 110 are developed based on the first volume, sensor information, optimal path, and second volume. Because the construction area is the core area of ​​construction activity and has higher air quality requirements, a first threshold is set (e.g., oxygen concentration ≥ 20.5%, hazardous gas concentration ≤ 50% of the occupational exposure limit). Meanwhile, in construction areas where personnel spend relatively short periods of time, a slightly lower second threshold is set (e.g., oxygen concentration ≥ 19.8%, hazardous gas concentration ≤ 80% of the occupational exposure limit).

[0144] If the air quality in area A (this is just an example, not a specific area) on the optimal path represented by sensor data does not meet the corresponding threshold, the control system calculates the required ventilation rate based on the correlation between this area and the area to be constructed, the area through which construction will pass, and the area's volume. For example, if the air quality in the area to be constructed does not meet the standard, the control system calculates the number of fans that need to be activated (multiple or one fan in a tunnel chamber can be set, without specific restrictions), the operating time, and the speed based on the first volume and target air quality requirements, combined with the fan's ventilation capacity (such as air volume and pressure), to ensure that the air quality in the area to be constructed is greater than or equal to the first threshold within the specified time. The main and auxiliary fans are appropriately configured based on the layout of the tunnel chamber 100 and the ventilation requirements. In the area to be constructed, the main fan (i.e., the fan module of the tunnel chamber where the area to be constructed is located) is started first to quickly increase the air circulation speed; in the area where construction is passing, the auxiliary fan (i.e., the fan module of the tunnel chamber where the area to be constructed is located) is started or the fan speed is adjusted according to the actual situation. Since the gate control module on the optimal path is open and the tunnel chambers are connected, the ventilation wind blows from the area to be constructed to the area where construction is passing, and the gas is extracted by the fan in the area where construction is passing, thereby ensuring the ventilation efficiency on the optimal path; and because the area to be constructed is the air intake area, the air quality in the area to be constructed is higher than that in other areas where construction is passing, thereby ensuring that the air quality in the area where construction is passing and the area to be constructed meets the requirements at the same time.

[0145] The operating status of the fan can also be dynamically adjusted according to changes in construction progress and personnel activity range to achieve a balance between energy saving and efficient ventilation.

[0146] This method enables precise ventilation control of the tunnel's primary target chamber, effectively ensuring construction safety and personnel health. Differentiated ventilation is implemented based on the actual needs of different areas, avoiding unnecessary energy waste and reducing operating costs. It also ensures that air quality in areas awaiting construction and those undergoing construction meets safety standards, reducing safety risks for construction workers due to inhalation of harmful gases or hypoxia. The ventilation strategy can also be dynamically adjusted based on changes in the construction checklist, construction progress, and personnel movement range, enhancing the system's adaptability and intelligence.

[0147] In some optional embodiments, the controlling of the fan module 110 of the first target chamber according to the first volume, the sensor information, the optimal path and the second volume includes: determining a starting position and an end position according to the optimal path, the end position representing the position of the construction area, and the starting position representing the position where the construction personnel enter the tunnel; starting the door control module 170 on the optimal path according to the starting position and the end position, so that the tunnel chambers 100 on the optimal path are interconnected; determining a first ventilation power of a first ventilation module and a second ventilation power of a second ventilation module according to the optimal path, the sensor information, the first volume and the second volume, the first ventilation module representing the fan module 110 corresponding to the starting position, the second ventilation module representing the fan module 110 on the optimal path, and the second ventilation module not including the first ventilation module; controlling the first ventilation module to blow air toward the starting position at the first ventilation power, and controlling the second ventilation module to exhaust air on the optimal path at the second ventilation power, so that the air quality of the area to be constructed is greater than a first threshold value, and the air quality of the area through which the construction passes is greater than a second threshold value.

[0148] Specifically, the starting point of the optimal path in this application ( ) is the entrance location for construction workers, and its coordinates are ( ). The end position of the optimal path ( ) is the center point of the area to be constructed, and its coordinates are ( ), which can be obtained by directly extracting the optimal path in the three-dimensional space model.

[0149] The optimal path can be divided into Continuous line segments , each section corresponds to an independent chamber, and the coordinates of the starting point of each section are recorded , end point coordinates and length Based on the best path , activate the gates between all adjacent chambers on the path, that is, for any adjacent line segment on the path and The gates between the paths are opened. Gates on non-paths remain closed, forming a ventilation channel that overlaps with the optimal path to ensure ventilation quality. Example: If the optimal path is C01→C02→C03, the gates between C01 and C02, and C02 and C03 are opened, and the other gates are closed.

[0150] The first volume of the area to be constructed ( ) and the second volume of the construction area ( ) is calculated by the above method, where , is the volume of the i-th chamber on the path.

[0151] The first ventilation power of the first ventilation module and the second ventilation power of the second ventilation module are determined according to the optimal path, the sensor information, the first volume and the second volume, wherein the power calculation formula of the first ventilation module is: in, is the ambient air baseline concentration; is the current average concentration in the area to be constructed (obtained by weighted average of sensor data); is the safety factor; Target ventilation time; is the fan efficiency.

[0152] The fan power of the i-th chamber on the path in the second ventilation module:

[0153]

[0154] in, is the current concentration of the i-th chamber; is the path attenuation coefficient (increases with increasing distance from the starting point, e.g. =1+0.1×i); is the efficiency of the jth fan.

[0155] The first ventilation module at the endpoint is controlled to blow air at the first ventilation module power, creating a positive pressure zone and accelerating the influx of fresh air. Simultaneously, the second ventilation module along the optimal path operates at the second ventilation power (derived by the sum of the fan powers of multiple tunnel chambers), creating a negative pressure gradient and guiding airflow along the optimal path until the air quality in the construction area exceeds the first threshold and the air quality in the construction area exceeds the second threshold.

[0156] In some optional embodiments, the water pump module includes a first water pump unit 151, a second water pump unit 152 and a third water pump unit 150, the first water pump unit 151 and the second water pump unit 152 are arranged in the second target chamber, the third water pump unit 150 is arranged outside the second target chamber, and the third water pump unit 150 pumps water to the second target chamber through the vent 130 on the second target chamber; the control of the water pump module to pump water to the second target chamber includes: when the water level information indicates that the water depth of the second target chamber is greater than When the water level information indicates that the water depth in the second target chamber is greater than or equal to the first depth, the first water pump unit 151 is started to pump water to the second target chamber; when the water level information indicates that the water depth in the second target chamber is greater than or equal to the second depth, the first water pump unit 151 and the second water pump unit 152 are started to pump water to the second target chamber; when the water level information indicates that the water depth in the second target chamber is greater than or equal to the third depth, the first water pump unit 151, the second water pump unit 152 and the third water pump unit 150 are started to pump water to the second target chamber.

[0157] Specifically, the first water pump unit of this application is located in the low-lying area of ​​the second target chamber to pump out regular accumulated water. The second water pump unit is connected in parallel with the first water pump and serves as a backup pump or auxiliary pump to deal with sudden large-scale accumulated water. The third water pump unit is installed outside the chamber and is connected to a dedicated drainage pipe through a vent to pump out accumulated water in the pipe gallery chamber.

[0158] When the water level information indicates that the accumulated water in the second target chamber is between the first depth (H1) and the second depth (H2), the first water pump unit is activated to pump water from the second target chamber. When the water level information indicates that the accumulated water in the second target chamber is between the second depth (H2) and the third depth (H3), the first water pump unit and the second water pump unit are simultaneously activated to pump water from the second target chamber. When the water level information indicates that the accumulated water in the second target chamber is greater than or equal to the third depth (H3), the first, second, and third water pump units are simultaneously activated to pump water from the second target chamber. Thus, different water pumps are activated according to different accumulated water depths, ensuring that the accumulated water in the second target chamber can be pumped out in a timely manner.

[0159] In some optional embodiments, the method further includes: when the water level information indicates that the water depth of the second target chamber is greater than or equal to the first depth, obtaining a first water depth change rate according to the water level information, determining a first pumping power of the first water pump unit 151 according to the first water depth change rate, and controlling the first water pump unit 151 to pump water from the second target chamber at the first pumping power;

[0160] When the water level information indicates that the water depth of the second target chamber is greater than or equal to the second depth, obtaining a second water depth change rate according to the water level information, determining a second pumping power of the first water pump unit 151 and a third pumping power of the second water pump unit 152 according to the second water depth change rate, and controlling the first water pump unit 151 to pump water from the second target chamber at the second pumping power and the second water pump unit 152 to pump water at the third pumping power;

[0161] When the water level information indicates that the water accumulation depth of the second target chamber is greater than or equal to the third depth, the third water depth change rate is obtained according to the water level information, and the fourth pumping power of the first water pump unit 151, the fifth pumping power of the second water pump unit 152, and the sixth pumping power of the third water pump unit 150 are determined according to the third water depth change rate. The first water pump unit 151 is controlled to pump water to the second target chamber with the fourth pumping power, the second water pump unit 152 is controlled to pump water to the second target chamber with the fifth pumping power, and the third water pump unit 150 is controlled to pump water to the second target chamber with the sixth pumping power.

[0162] Specifically, the present application realizes refined control of the water pump power by real-time monitoring of the water level change rate and dynamically adjusting the water pumping power to avoid the water pump power being too small or too large.

[0163] Among them, pumping power Rate of change with water depth The relationship satisfies: .

[0164] in, is the density of water; is the acceleration due to gravity; is the pumping flow; is the current water depth; is the cross-sectional area of ​​the waterlogged area; is the pump efficiency.

[0165] And introduce the dynamic response coefficient Corrected power calculation:

[0166]

[0167] Final pumping power .

[0168] When the water depth of the second target chamber is greater than or equal to the first depth, the first water depth change rate is obtained from the water level information, and the first water depth change rate and the above-mentioned pumping power are used to calculate the water depth change rate. The calculation formula can be used to calculate the first pumping power of the first water pump unit, and the first water pump unit is controlled to pump water to the second target chamber with the first pumping power to complete the pumping. Similarly, when the water depth of the second target chamber is greater than or equal to the second depth, the second water depth change rate is obtained from the water level information, and the second water depth change rate and the above-mentioned pumping power are used to calculate the first pumping power of the first water pump unit. The calculation formula can be used to calculate the second pumping power of the first water pump unit and the third pumping power of the second water pump unit (where the first water pump unit bears the base power, and the second water pump unit supplements the excess power to meet the needs of pumping accumulated water). Pumping can be completed by controlling the first water pump unit to pump water from the second target chamber at the second pumping power and the second water pump unit to pump water from the third pumping power. Similarly, when the accumulated water depth in the second target chamber is greater than or equal to the third depth, the third water depth change rate is obtained from the water level information. Based on the third water depth change rate and the aforementioned pumping power calculation formula, the fourth pumping power of the first water pump unit, the fifth pumping power of the second water pump unit, and the sixth pumping power of the third water pump unit can be calculated (where the first water pump unit bears the base power, and the second and third water pump units supplement the excess power in sequence to meet the needs of pumping accumulated water). Pumping can be completed by controlling the first water pump unit to pump water from the second target chamber at the fourth pumping power, the second water pump unit at the fifth pumping power, and the third water pump unit at the sixth pumping power.

[0169] The beneficial effects of implementing the embodiments of the present invention include: upon obtaining construction information, determining a first target chamber based on the construction information, wherein the first target chamber represents the pipe gallery chamber 100 where construction is being carried out and where construction workers pass through, and the construction information is obtained through a construction checklist; controlling the fan module 110 to ventilate the first target chamber based on the sensor information from the sensor module, so that the air quality in the first target chamber equals a preset quality; obtaining water level information of the pipe gallery chamber 100 through the sensor module; and, if the water level information indicates that water has accumulated in a second target chamber, controlling the water pump module to pump water from the second target chamber, so that the depth of the water in the second target chamber is less than a preset depth. Ventilating the pipe gallery in advance based on the construction information ensures smooth construction progress and high construction management efficiency. In the event of a water pipe burst, the water level information is promptly obtained to control the water pump module to pump water, thereby avoiding unnecessary damage to the pipe gallery.

[0170] In addition, an embodiment of the present invention provides a tunnel control device based on a tunnel perception network, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor.

[0171] The processor and the memory may be connected via a bus or other means.

[0172] It should be noted that the computer in this embodiment may correspond to include: Figure 4 The memory and processor in the embodiment shown can constitute Figure 4 Part of the system architecture platform in the illustrated embodiment, both belong to the same inventive concept, so both have the same implementation principles and beneficial effects, and will not be described in detail here.

[0173] The non-transient software program and instructions required to implement the uplink co-channel interference elimination method of the above embodiment are stored in the memory. When executed by the processor, the corridor control method based on the corridor perception network of the above embodiment is executed, for example, the above-described Figure 1 Method steps S100 to S500 in .

[0174] In addition, an embodiment of the present invention further provides a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are used to execute the above-described device's pipe gallery control method based on the pipe gallery sensing network, for example, executing the above-described Figure 1 Method steps S100 to S500 in .

[0175] Those skilled in the art will appreciate that all or some of the steps and systems described above can be implemented as software, firmware, hardware, or any combination thereof. Some or all of the physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on computer-readable media, which may include computer storage media (or non-transitory media) and communication media (or transient media). As is well known to those skilled in the art, the term computer storage media encompasses volatile and non-volatile, removable, and non-removable media implemented in any method or technology for storing information, such as computer-readable instructions, data structures, program modules, or other data. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disks (DVDs) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. Furthermore, as is well known to those skilled in the art, communication media typically includes computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0176] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the above implementation. Those skilled in the art can also make various equivalent modifications or substitutions under the shared conditions that do not violate the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of the present invention.

Claims

1. A pipe gallery control method based on a pipe gallery sensing network is characterized in that: Applied to a pipe gallery system, the pipe gallery system includes multiple pipe gallery chambers and a pipe gallery sensing network. The pipe gallery chambers are provided with sensor modules, water pump modules, fan modules, and door control modules. The door control modules are provided at the entrance and exit ends of the pipe gallery chambers. Adjacent pipe gallery chambers are connected through the door control modules. The sensor modules, the water pump modules, the fan modules, and the door control modules communicate based on the pipe gallery sensing network. The method includes: When construction information is obtained, a first target chamber is determined according to the construction information, where the first target chamber represents the pipe gallery chamber where construction is being carried out and where construction workers pass through. The construction information is obtained through a construction checklist. Controlling the fan module to perform ventilation processing on the first target chamber according to the sensor information of the sensor module specifically includes: obtaining a first volume of a to-be-constructed area and a second volume of a construction-passed area in the first target chamber, and controlling the fan module of the first target chamber to perform ventilation processing based on the first volume, the sensor information, and the second volume, so that the air quality of the first target chamber is equal to a preset quality; Acquiring water level information of the pipe gallery chamber through the sensor module; When the water level information indicates that water has accumulated in the second target chamber, controlling the water pump module to pump water from the second target chamber so that the depth of the water accumulated in the second target chamber is less than a preset depth; Obtaining maintenance knowledge graphs of the first target chamber and the second target chamber; Calculating the safety risk coefficient of the target accident corresponding to the accident information using the consequence assessment model and the risk prediction model included in the maintenance knowledge graph, wherein the accident information represents a liquid leakage accident and a gas leakage accident; Calculating a business impact coefficient of the target accident using a business impact model included in the maintenance knowledge graph, where the business impact coefficient indicates the economic loss, social impact, and recovery time caused by the target accident; Calculating the maintenance difficulty coefficient of the target accident using the maintenance difficulty model included in the maintenance knowledge graph; Determining a handling priority of the target accident based on the safety risk factor, the business impact factor, and the maintenance difficulty factor; Generate maintenance plans in sequence according to the processing priorities: generating an accident feature vector based on the accident information and the maintenance knowledge graph, wherein the accident feature vector represents the accident type, accident severity, accident time, accident location, and environmental parameters; Comparing the accident feature vector with multiple historical cases in the historical accident case library of the maintenance knowledge graph for similarity, and configuring the historical case with the highest similarity as a candidate case; When the similarity between the candidate case and the accident feature vector is greater than a preset similarity threshold, obtaining a candidate solution configured for the candidate case from a maintenance solution library of the maintenance knowledge graph, and configuring the candidate solution as the maintenance solution for the accident feature vector; When the similarity between the candidate case and the accident feature vector is less than a preset similarity threshold, obtaining a difference vector between the candidate case and the accident feature vector, where the difference vector represents a vector composed of difference data between the candidate case and the accident feature vector; Sending the difference vector, the candidate solution, the candidate case, and the accident feature vector to an operation and maintenance terminal, so that the operation and maintenance terminal inputs a manual correction parameter; The selected solution is corrected by the manual correction parameters to obtain a corrected solution; The correction plan is configured as the maintenance plan of the accident feature vector.

2. The pipe gallery control method based on the pipe gallery sensing network according to claim 1 is characterized in that: The controlling the fan module to perform ventilation processing on the first target chamber according to the sensor information of the sensor module includes: Determine the locations to be constructed inside the pipe gallery based on the construction list and generate a three-dimensional space model; Determine a first volume of the area to be constructed and a second volume of the area to be constructed based on the three-dimensional spatial model and the optimal path planning model, wherein the second volume represents the volume of the pipe gallery chamber passed by the construction personnel, and the first volume represents the volume of the pipe gallery chamber to be constructed. The optimal path planning model is used to generate an optimal path for the construction personnel to reach the area to be constructed based on the three-dimensional spatial model of the area to be constructed and the pipe gallery; acquiring the sensor information through a gas sensor provided on the optimal path in the sensor module, wherein the sensor information indicates the air quality on the optimal path; The fan module of the first target chamber is controlled according to the first volume, the sensor information, the optimal path and the second volume so that the air quality of the area to be constructed is greater than a first threshold, the air quality of the area through which construction is conducted is greater than a second threshold, and the first threshold is greater than the second threshold.

3. The pipe gallery control method based on the pipe gallery sensing network according to claim 2 is characterized in that: The method of controlling the blower module of the first target chamber according to the first volume, the sensor information, the optimal path, and the second volume includes: Determine a starting position and an end position according to the optimal path, wherein the end position represents the location of the construction area, and the starting position represents the location where construction personnel enter the pipe gallery; activating the gate control module on the optimal path according to the starting position and the end position, so that the pipe gallery chambers on the optimal path are interconnected; determining a first ventilation power of a first ventilation module and a second ventilation power of a second ventilation module according to the optimal path, the sensor information, the first volume, and the second volume, wherein the first ventilation module represents the fan module corresponding to the starting position, and the second ventilation module represents the fan module on the optimal path, and the second ventilation module does not include the first ventilation module; Control the first ventilation module to blow air toward the starting position at the first ventilation power, and control the second ventilation module to exhaust air along the optimal path at the second ventilation power, so that the air quality of the area to be constructed is greater than the first threshold and the air quality of the area through which construction is passed is greater than the second threshold.

4. The pipe gallery control method based on the pipe gallery sensing network according to claim 1 is characterized in that: The water pump module includes a first water pump unit, a second water pump unit, and a third water pump unit, wherein the first water pump unit and the second water pump unit are arranged in the second target chamber, and the third water pump unit is arranged outside the second target chamber, and the third water pump unit pumps water to the second target chamber through a vent on the second target chamber; and controlling the water pump module to pump water to the second target chamber includes: When the water level information indicates that the water depth of the second target chamber is greater than or equal to the first depth, starting the first water pump unit to pump water from the second target chamber; When the water level information indicates that the water depth of the second target chamber is greater than or equal to a second depth, starting the first water pump unit and the second water pump unit to pump water from the second target chamber; When the water level information indicates that the water depth of the second target chamber is greater than or equal to a third depth, the first water pump unit, the second water pump unit, and the third water pump unit are started to pump water from the second target chamber.

5. The pipe gallery control method based on the pipe gallery sensing network according to claim 4 is characterized in that: The method further comprises: When the water level information indicates that the water depth of the second target chamber is greater than or equal to the first depth, obtaining a first water depth change rate according to the water level information, determining a first pumping power of the first water pump unit according to the first water depth change rate, and controlling the first water pump unit to pump water from the second target chamber at the first pumping power; When the water level information indicates that the water depth of the second target chamber is greater than or equal to the second depth, obtaining a second water depth change rate according to the water level information, determining a second pumping power of the first water pump unit and a third pumping power of the second water pump unit according to the second water depth change rate, and controlling the first water pump unit to pump water from the second target chamber at the second pumping power and the second water pump unit to pump water at the third pumping power; When the water level information indicates that the water accumulation depth of the second target chamber is greater than or equal to the third depth, the third water depth change rate is obtained according to the water level information, and the fourth pumping power of the first water pump unit, the fifth pumping power of the second water pump unit, and the sixth pumping power of the third water pump unit are determined according to the third water depth change rate. The first water pump unit is controlled to pump water to the second target chamber with the fourth pumping power, the second water pump unit is controlled to pump water to the second target chamber with the fifth pumping power, and the third water pump unit is controlled to pump water to the second target chamber with the sixth pumping power.

6. A pipe gallery control device based on a pipe gallery sensing network, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for controlling a pipe corridor based on a pipe corridor sensing network according to any one of claims 1 to 5 is implemented.

7. A computer storage medium, characterized in that The computer storage medium stores computer-executable instructions, and the computer-executable instructions are used to execute the pipeline corridor control method based on the pipeline corridor perception network described in any one of claims 1 to 5.

Images