Pipe gallery control method and device based on pipe gallery sensing network and storage medium

Abstract

The embodiment of the invention provides a pipe gallery control method and device based on a pipe gallery sensing network and a storage medium, and the method comprises the steps: determining a first target cavity according to construction information under the condition of obtaining the construction information; according to the sensor information of the sensor module, the fan module is controlled to conduct ventilation treatment on the first target cavity, so that the air quality of the first target cavity is equal to the preset quality; the water level information of the pipe gallery chamber is obtained through the sensor module; and under the condition that the water level information represents that the water is accumulated in the second target chamber, the water pump module is controlled to perform water pumping treatment on the second target chamber, so that the water accumulation depth of the second target chamber is smaller than the preset depth. Ventilation is conducted on the pipe gallery in advance through the construction information, it is guaranteed that construction is conducted smoothly, the construction management and control efficiency is high, the water pump module is controlled to conduct water pumping treatment by obtaining the water level information in time, and therefore unnecessary losses caused to the pipe gallery are avoided.

Description

Technical Field

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

[0002] An underground utility tunnel is an integrated underground corridor for urban pipelines. An underground utility tunnel is usually a tunnel space built underground in a city, integrating various engineering pipelines such as electricity, communication, gas, heating, water supply and drainage, etc., with dedicated inspection openings, hoisting openings, and monitoring systems, and implementing unified planning, unified design, unified construction, and unified management.

[0003] Currently, the control of underground utility tunnels is carried out by corresponding personnel, with low control efficiency, and it is difficult to detect and handle accidents in underground utility tunnels in a timely manner, thus easily causing unnecessary losses to underground utility tunnels. Summary of the Invention

[0004] The following is an overview of the subject matter described in detail in this document. This overview is not intended to limit the scope of protection of the claims.

[0005] The main objective of the embodiments of the present invention is to provide a control method, device, and storage medium for an underground utility tunnel based on an underground utility tunnel sensing network, which can improve the control efficiency of underground utility tunnels, and can detect and handle accidents in underground utility tunnels in a timely manner, avoiding unnecessary losses to underground utility tunnels.

[0006] In a first aspect, an embodiment of the present invention provides a control method for an underground utility tunnel based on an underground utility tunnel sensing network, which is applied to an underground utility tunnel system. The underground utility tunnel system includes multiple underground utility tunnel chambers and an underground utility tunnel sensing network. The underground utility tunnel chambers are provided with a sensor module, a water pump module, a fan module, and a door control module. The door control module is arranged at the inlet and outlet ends of the underground utility tunnel chambers, and adjacent underground utility tunnel chambers are connected through the door control module. The sensor module, the water pump module, the fan module, and the door control module communicate based on the underground utility tunnel sensing network. The method includes: When construction information is obtained, a first target chamber is determined according to the construction information. The first target chamber represents the underground utility tunnel chamber where construction is carried out and through which construction personnel pass. The construction information is obtained through a construction list; Controlling the fan module to perform ventilation treatment on the first target chamber according to the sensor information of the sensor module, specifically including: obtaining a first volume of the area to be constructed and a second volume of the area passed through during construction in the first target chamber, and controlling the fan module of the first target chamber to perform ventilation treatment based on the first volume, the sensor information, and the second volume, so that the air quality in the first target chamber is equal to a preset quality; Obtaining the water level information of the underground utility tunnel chamber through the sensor module; When the water level information indicates that there is accumulated water in the second target chamber, control the water pump module to pump water from the second target chamber so that the depth of the accumulated water in the second target chamber is less than a preset depth.

[0007] In some alternative embodiments, after controlling the water pump module to pump water from the second target chamber, the method further includes: Obtain the maintenance knowledge graph of the first target chamber and the second target chamber; Generate a corresponding maintenance plan according to the maintenance knowledge graph and the accident information, where the accident information indicates a liquid leakage accident and a gas leakage accident.

[0008] In some alternative embodiments, before generating a corresponding maintenance plan according to the maintenance knowledge graph and the accident information, the method further includes: Calculate the safety risk coefficient of the target accident corresponding to the accident information through the consequence assessment model and the risk prediction model included in the maintenance knowledge graph; Calculate the business impact coefficient of the target accident through the 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; Calculate the maintenance difficulty coefficient of the target accident through the maintenance difficulty model included in the maintenance knowledge graph; Determine the processing priority of the target accident according to the safety risk coefficient, the business impact coefficient, and the maintenance difficulty coefficient; Generate the maintenance plan in sequence according to the processing priority.

[0009] In some alternative embodiments, generating a corresponding maintenance plan according to the maintenance knowledge graph and the accident information includes: Generate an accident feature vector according to the accident information and the maintenance knowledge graph, where the accident feature vector characterizes the accident type, accident severity, accident occurrence time, accident occurrence location, and environmental parameters; Compare the similarity of the accident feature vector with multiple historical cases in the historical accident case library of the maintenance knowledge graph, and configure the historical case with the highest similarity as the candidate case; When the similarity between the candidate case and the accident feature vector is greater than a preset similarity threshold, obtain the candidate plan configured for the candidate case from the maintenance plan library of the maintenance knowledge graph, and configure the candidate plan as the maintenance plan of the accident feature vector; When the similarity between the candidate case and the accident feature vector is less than the preset similarity threshold, obtain the difference vector between the candidate case and the accident feature vector, where the difference vector represents the vector composed of the difference data between the candidate case and the accident feature vector; Send the difference vector, the candidate solution, the candidate case, and the accident feature vector to the operation and maintenance terminal, so that the operation and maintenance terminal inputs manual correction parameters; Obtain the corrected solution after correcting the candidate solution with the manual correction parameters; Configure the corrected solution as the maintenance solution for the accident feature vector.

[0010] In some alternative embodiments, the controlling the air exchange process of the first target chamber by the fan module according to the sensor information of the sensor module includes: Determine the positions to be constructed inside the pipe gallery based on the construction list and generate a three-dimensional space model; Determine the first volume of the area to be constructed and the second volume of the area passed through during construction according to the three-dimensional space model and the optimal path planning model. The second volume represents the volume of the pipe gallery chamber passed through 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 the optimal path for the construction personnel to reach the area to be constructed according to the area to be constructed and the three-dimensional space model of the pipe gallery; Obtain the sensor information through the gas sensors arranged on the optimal path in the sensor module, where the sensor information indicates the air quality on the optimal path; Control the fan module 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 in the area to be constructed is greater than the first threshold, and the air quality in the area passed through during construction is greater than the second threshold, where the first threshold is greater than the second threshold.

[0011] In some alternative embodiments, the controlling the fan module of the first target chamber according to the first volume, the sensor information, the optimal path, and the second volume includes: Determine the starting position and the ending position according to the optimal path. The ending position represents the position of the construction area, and the starting position represents the position where the construction personnel enter the pipe gallery; Start the door control module on the optimal path according to the starting position and the ending position, so that the pipe gallery chambers on the optimal path are interconnected; Determine the first ventilation power of the first ventilation module and the second ventilation power of the second ventilation module according to the optimal path, the sensor information, the first volume, and the second volume. The first ventilation module represents the fan module corresponding to the starting position, and the second ventilation module represents the fan modules on the optimal path. The second ventilation module does not include the first ventilation module. Control the first ventilation module to blow air towards the starting position at the first ventilation power, and control the second ventilation module to extract air on the optimal path at the second ventilation power, so that the air quality in the area to be constructed is greater than the first threshold and the air quality in the area passed through during construction is greater than the second threshold.

[0012] In some alternative 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, and the third water pump unit is arranged outside the second target chamber. The third water pump unit performs water extraction treatment on the second target chamber through a ventilation opening on the second target chamber. Controlling the water pump module to perform water extraction treatment on the second target chamber includes: When the water level information indicates that the accumulated water depth in the second target chamber is greater than or equal to the first depth, start the first water pump unit to perform water extraction treatment on the second target chamber; When the water level information indicates that the accumulated water depth in the second target chamber is greater than or equal to the second depth, start the first water pump unit and the second water pump unit to perform water extraction treatment on the second target chamber; When the water level information indicates that the accumulated water depth in the second target chamber is greater than or equal to the third depth, start the first water pump unit, the second water pump unit, and the third water pump unit to perform water extraction treatment on the second target chamber.

[0013] In some alternative embodiments, the method further includes: When the water level information indicates that the accumulated water depth in the second target chamber is greater than or equal to the first depth, obtain a first water depth change rate according to the water level information, determine a first water extraction power of the first water pump unit according to the first water depth change rate, and control the first water pump unit to perform water extraction treatment on the second target chamber at the first water extraction power. When the water level information indicates that the accumulated water depth in the second target chamber is greater than or equal to the second depth, obtain the second water depth change rate according to the water level information, determine the second pumping power of the first water pump unit and the third pumping power of the second water pump unit according to the second water depth change rate, and control the first water pump unit to perform pumping treatment on the second target chamber at the second pumping power and the second water pump unit to perform pumping treatment on the second target chamber at the third pumping power; When the water level information indicates that the accumulated water depth in the second target chamber is greater than or equal to the third depth, obtain the third water depth change rate according to the water level information, determine 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 according to the third water depth change rate, and control the first water pump unit to perform pumping treatment on the second target chamber at the fourth pumping power, the second water pump unit to perform pumping treatment on the second target chamber at the fifth pumping power, and the third water pump unit to perform pumping treatment on the second target chamber at the sixth pumping power.

[0014] In a second aspect, an embodiment of the present invention provides a corridor control device based on a corridor perception network, including: a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the corridor control method based on the corridor perception network described in the first aspect is implemented.

[0015] In a third aspect, an embodiment of the present invention provides a computer storage medium storing computer-executable instructions for executing the corridor control method based on the corridor perception network described in the first aspect.

[0016] The beneficial effects of the present invention include: when construction information is obtained, determine a first target chamber according to the construction information, where the first target chamber represents the corridor chamber where construction is carried out and construction personnel pass through, and the construction information is obtained through a construction list; control the fan module to perform ventilation treatment on the first target chamber according to the sensor information of the sensor module so that the air quality in the first target chamber is equal to a preset quality; obtain the water level information of the corridor chamber through the sensor module; when the water level information indicates that accumulated water is generated in the second target chamber, control the water pump module to perform pumping treatment on the second target chamber so that the accumulated water depth in the second target chamber is less than the preset depth. By ventilating the corridor in advance through construction information, the smooth progress of construction is ensured, the construction management and control efficiency is high, and when a water pipe burst accident occurs, the water pump module is controlled to perform pumping treatment by timely obtaining the water level information, thereby avoiding unnecessary losses to the corridor.

[0017] Other features and advantages of the present invention will be described in the following specification, and in part will be obvious from the specification, or will be understood by implementing the present invention. The objectives and other advantages of the present invention can be achieved and obtained by the structures specifically pointed out in the specification, claims and drawings. Description of the Drawings

[0018] Figure 1 is a step - flow block diagram of a utility tunnel control method based on a utility tunnel perception network provided by an embodiment of the present invention; Figure 2 is a structural schematic block diagram of a utility tunnel chamber provided by an embodiment of the present invention; Figure 3 is a structural schematic diagram of a utility tunnel perception network provided by an embodiment of the present invention; Figure 4 is a schematic diagram of a controller provided by an embodiment of the present invention.

[0019] Reference numerals: Controller 1000, Processor 1100, Memory 1200; Utility tunnel chamber 100, Fan module 110, Fan control unit 111, Fan unit 112, Fan duct 120, Vent 130, Environment 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 Embodiment

[0020] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0021] It should be noted that although the functional modules are divided in the device schematic diagram and the logical sequence is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order from the module division in the device or the sequence in the flowchart. Terms such as "first", "second", etc. in the specification, claims or the above - mentioned drawings are used to distinguish similar objects and do not necessarily need to describe a specific order or sequence.

[0022] A utility tunnel is an underground urban pipeline comprehensive corridor. Usually, a tunnel space is built underground in a city, integrating various engineering pipelines such as electricity, communication, gas, heating, water supply and drainage, etc. It is equipped with special inspection openings, hoisting openings and monitoring systems, and implements unified planning, unified design, unified construction and management.

[0023] At present, it is difficult to detect and monitor in time when construction and other activities are carried out inside the utility tunnel, or when an intrusion occurs, which is likely to cause unnecessary losses to the utility tunnel.

[0024] To solve the above problems, the present application provides a utility tunnel control method, device and storage medium based on a utility tunnel perception network.

[0025] In the present application, a utility tunnel control method, device and storage medium based on a utility tunnel perception network are provided, and will be described in detail one by one in the following embodiments.

[0026] As Figure 1 shown, an embodiment of the present invention provides a utility tunnel control method based on a utility tunnel perception network, which is applied to a utility tunnel system. The utility tunnel system includes a plurality of utility tunnel chambers 100 and a utility tunnel perception network. The utility 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 arranged at the inlet and outlet ends of the utility tunnel chamber 100. Adjacent utility 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 utility tunnel perception network. The method includes: S100. When construction information is obtained, determine a first target chamber according to the construction information. The first target chamber represents the utility tunnel chamber 100 where construction is carried out and construction personnel pass through. The construction information is obtained through a construction list; S200. Control the fan module 110 to perform ventilation treatment on the first target chamber according to the sensor information of the sensor module. Specifically, it includes: obtaining a first volume of the area to be constructed and a second volume of the area passed by construction in the first target chamber, and controlling the fan module of the first target chamber to perform ventilation treatment 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; S300. Obtain the water level information of the utility tunnel chamber 100 through the sensor module; S400. When the water level information indicates that water accumulates in a second target chamber, control the water pump module to pump water from the second target chamber, so that the depth of the accumulated water in the second target chamber is less than a preset depth.

[0027] Specifically, the pipe gallery of the present application is divided into multiple pipe gallery chambers by several gating modules. Each pipe gallery chamber 100 includes a sensor module, a water pump module, a fan module, and a gating module. Among them, the sensor module includes an environmental sensor 140 and a water level sensor 160. The environmental sensor 140 includes gas sensors (detecting oxygen, carbon monoxide, methane, etc.), temperature and humidity sensors, pressure sensors, and video surveillance cameras. The gas sensors are deployed at the top of the chamber (for detecting gases with a density less than air) or at 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 dead angles. The water pump module is used to drain accumulated water and supports automatic switching; a filter screen is provided at the water pump inlet, and an electromagnetic flowmeter is installed at the outlet. The fan module 110 is arranged at the top of the chamber for air replacement and ventilation. The gating 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 / close / fault) is uploaded to the pipe gallery perception network in real time. When linked with the fire alarm system, the adjacent chamber doors are automatically closed.

[0028] Referring to Figure 3 , the pipe gallery perception network architecture includes: Backbone network: Integration of multiple networks such as environmental monitoring, security, and communication, isolation of multi-service resources; Bandwidth, delay, and reliability guarantee based on IPv6+; Integrated operation and maintenance of wired and wireless networks, network status visible, fault diagnosis.

[0029] Access network: Sensors and actuators are connected to the WiFi AP through the WiFi IOBOX, supporting WiFi6 / 5G wireless backup communication. Through the WiFi IOBOX, the monitoring terminals in the pipe gallery are "unbraided", saving wiring and being easy to maintain; Deploying the Wi-Fi IOBOX provides terminal access (RS485, DI, DO, AI interfaces), sensor power supply, and Wi-Fi wireless backhaul (supporting dual-transmission and selected-reception), supporting IP65 protection, plug-and-play, and iConnect seamless access. Wi-Fi 6 provides 4 times the network speed, 10ms end-to-end delay, and 4 times the number of users compared to Wi-Fi 5. The WLAN in the pipe gallery has continuous coverage, no dead angles, and seamless roaming handover.

[0030] Data center: Edge computing nodes (edge computing gateway + software PLC), deployed near each chamber, realizing data preprocessing (such as denoising, normalization), local rule engine (such as automatically starting the water pump when the water level exceeds the threshold), and equipment status prediction (based on machine learning algorithms); The edge computing gateway deploys software PLC, replacing the traditional hardware PLC, being autonomous and controllable, and supporting primary and backup high-reliability deployment. Cloud management platform: Build a digital twin system for the pipe gallery, real-time display of the status parameters of each chamber, equipment operation curves, and three-dimensional space models, supporting remote control and big data analysis; Pipe gallery environmental monitoring, security management, emergency management, operation and maintenance management.

[0031] By setting up a communication and sensing integrated gateway, environmental monitoring device, equipment control terminal, security device, and communication terminal on the edge node or terminal side: The Internet of Things gateway expands the fiber optic sensing monitoring capability, integrating networking, computing, and measurement; with high integration, easy deployment, and a cost reduction of more than 50%; real-time monitoring of utility tunnel settlement, construction damage, and abnormal temperature, and linkage disposal.

[0032] In the utility tunnel chamber 100 of this application, the following are set: A cabinet is deployed in the equipment room of the utility tunnel chamber 100, which centrally houses an outlet router (the upstream access to the Moxa switch is connected to the monitoring center), an industrial switch (connected to the Wi-Fi AP), and a communication and sensing integrated Internet of Things gateway (monitoring and warning of temperature, vibration, and strain, and providing edge computing operation software PLC); 4 Wi-Fi 6 outdoor APs are deployed to achieve Wi-Fi coverage in the equipment room and compartments; 1 set of sensors is deployed, including temperature and humidity sensors, oxygen sensors, hydrogen sulfide sensors, methane sensors, and level gauges; 4 IOBOXes are deployed to realize the "unbraiding" of monitoring terminals, including three Wi-Fi IOBOXes for lighting devices 180, water pumps, and sensors, and an Ethernet IOBOX for the fan, and the sensors and lighting / fan / water pump control boxes are connected to the IOBOX; The Wi-Fi IOBOX needs to be wired from the power distribution cabinet in the equipment room to provide power supply (AC 220V), and the Wi-Fi AP is connected upstream to the industrial switch in the cabinet (for communication and POE power supply).

[0033] Extract key information from the construction list: construction time window, involved chamber numbers (such as C01 - C05), construction content (such as pipeline welding, cable laying), areas to be isolated, etc.

[0034] Combined with historical construction data, automatically identify high-risk operation types (such as hot work operations), and trigger additional safety measures (such as increasing the CO detection frequency to 1 time per 5 minutes).

[0035] Generate a ventilation strategy: Start the fan module 110 for ventilation in the construction chamber and the chambers passed by the construction personnel, and continuously monitor the pollutant concentration in the construction chamber to ensure the smooth progress of the construction. Combine the construction progress and the personnel movement trajectory to update the first target chamber range in real time to ensure that ventilation and safety protection cover the entire construction area.

[0036] When the air quality is lower than the preset standard, the system automatically starts the fan module to adopt the "air supply + exhaust" linkage mode, introduces fresh air through the air inlet, and exhausts polluted air through the exhaust port; adjusts the fan speed according to the degree of pollution (such as increasing to high speed when the dust concentration exceeds the standard) until the air quality meets the standard; during the ventilation process, collect gas samples in the middle of the chamber every 10 minutes to verify the oxygen concentration ≥ 19.5%, the combustible gas concentration < 10% of the LEL (lower explosion limit), the dust concentration < 8mg / m³ and other indicators, and the specific indicators are not limited. If an indicator does not meet the standard, the ventilation time will be automatically extended or the standby fan will be started until the air quality meets the standard and is stable.

[0037] The sensor module collects the 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).

[0038] 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.

[0039] Drainage strategies are implemented in different levels: Level 1 response (water level <10cm): Start the main water pump to run alone, with a drainage flow of 50m³ / h, and at the same time close the gate control module 170 of the chamber and the downstream chamber to prevent backflow of accumulated water. Level 2 response (10cm≤water level <30cm): The main pump and the backup water pump run in parallel, with a total flow of 80m³ / h. Level 3 response (water level ≥30cm): Trigger the "flood emergency mode", all water pumps are linked to drain water, and the optimal drainage path is simulated through the digital twin system (such as giving priority to draining the water in the lowest chamber).

[0040] Drainage effect evaluation: Calculate drainage efficiency (drainage flow - leakage flow) in real time. If the water level drops less than 2 cm / min within 30 minutes, it will automatically switch to "intermittent drainage + plugging priority" mode and dispatch robots to find the leak. After drainage is completed, perform health diagnosis on parameters such as pump motor temperature and bearing vibration, and automatically generate maintenance recommendations (such as changing lubricating oil after 500 hours of cumulative operation).

[0041] When the gas sensor detects a sudden rise in CO (carbon monoxide) concentration (such as > 500ppm) and the temperature and humidity sensor detects a temperature rise rate > 5℃ / min, it is automatically judged as a fire hazard and the following operations are triggered: close all door control modules 170 to form a fire partition; switch the fan to the "negative pressure smoke exhaust" mode to give priority to exhausting toxic smoke; link the fire sprinkler system to cool down the chambers adjacent to the fire source. When the camera recognizes through the AI algorithm that the construction worker is not wearing a helmet or enters an unauthorized area, it immediately sends an audible and visual alarm to the site and pushes the warning information to the mobile phone APP of the operation and maintenance personnel.

[0042] In some alternative embodiments, after controlling the water pump module to perform water pumping treatment on 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 according to the maintenance knowledge graph and accident information, where the accident information characterizes liquid leakage accidents and gas leakage accidents.

[0043] Specifically, the maintenance knowledge graph is used to generate corresponding maintenance plans to guide maintenance personnel to quickly complete the inspection and repair of the pipe gallery. The data architecture of the knowledge graph in 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), and supplier. Accident types: covering types such as liquid leakage, gas leakage, and equipment failures, recording the accident occurrence time, location, severity (such as leakage volume, influence range), and initial cause (such as corrosion, external force damage). Maintenance plans: including maintenance personnel, tools, spare parts, plan steps, etc., associated with maintenance time consumption, cost, and effect evaluation indicators.

[0044] The knowledge graph is provided with corresponding relationship layers: Equipment-accident relationship: such as "a certain model of water pump causes liquid leakage due to impeller wear". Accident-plan relationship: such as "a gas leakage accident requires implementing a three-step plan of 'closing the valve - ventilating - detecting'". Plan-resource relationship: such as "a pipeline leak plugging plan requires using epoxy resin glue, patch clamps, and underwater operators".

[0045] Dynamic update of the knowledge graph: Real-time collect data of each sensor in the sensor module and operation data of each device (such as water pump start-stop times, leakage detection signals, carbon monoxide concentration). Manually input unstructured data such as maintenance reports, accident investigation reports, and equipment replacement records. Parse maintenance documents through natural language processing (NLP) technology to extract key information (such as "a leak occurred at the welded joint of the pipeline in Chamber C03 in May 2023 and was repaired by welding"). Use a graph database (such as Neo4j) to realize semantic association of multi-source data and eliminate data islands.

[0046] Knowledge verification: Introduce an expert review mechanism to manually verify automatically generated knowledge nodes (such as new maintenance plans). Verify the accuracy of the knowledge through actual maintenance effect feedback (such as repair rate, fault recurrence period) to form a closed loop of "collection - fusion - verification - optimization".

[0047] Liquid leakage: Through the accumulated water depth change curve recorded by the water level sensor 160, and / or abnormal pipeline flow detected by an ultrasonic flowmeter (such as a sudden drop of 30%).

[0048] Gas leakage: Determine the leakage of the gas pipeline based on the concentration of each gas in the pipe gallery chamber 100 detected by the gas sensor, and / or determine the leakage of the gas pipeline based on the pipeline pressure drop rate captured by the pressure sensor.

[0049] Visual data: The video of the leakage site taken by the camera set in the pipe gallery chamber 100. Calculate the water flow velocity and the gas diffusion range through computer vision algorithms (such as the optical flow method). The inner wall image of the pipeline obtained through robot inspection, and thus identify features such as the length of the pipeline crack (such as 5 cm) and the corrosion area (such as 10 cm²) based on the inner wall image.

[0050] Manually reported data: The accident description reported by on-site personnel through the APP (such as "Water is spraying out from the pipeline interface at the top of Chamber C05"), on-site photos, and preliminary disposal measures.

[0051] Feature extraction and encoding: Numerical features of liquid leakage accident characteristics: Leakage duration (t), accumulated water depth (h), leakage flow rate (Q), pipeline pressure (P). Categorical features of liquid leakage accident characteristics: Leakage location (such as elbow / straight pipe section), pipeline material (such as ductile iron / PE), leakage form (such as jet / drip).

[0052] Numerical features of gas leakage accident characteristics: Gas concentration (C), diffusion velocity (v), ambient temperature (T), wind speed (u). Categorical features of gas leakage accident characteristics: Gas type (such as methane / carbon monoxide), leakage point sealing form (such as flange connection / welding), explosion-proof area level (such as Zone 0 / Zone 1).

[0053] Perform corresponding numerical encoding on the numerical features and categorical features of liquid leakage accident characteristics to obtain the accident code corresponding to the liquid leakage accident. Similarly, perform corresponding numerical encoding on the numerical features and categorical features of gas leakage accident characteristics to obtain the accident code corresponding to the gas leakage accident. Thus, the accident code can be quickly matched with the maintenance knowledge graph to generate the corresponding maintenance plan.

[0054] In some alternative embodiments, before generating a corresponding maintenance plan according to the maintenance knowledge graph and accident information, the method further includes: calculating a safety risk coefficient of a target accident corresponding to the accident information through a consequence assessment model and a risk prediction model included in the maintenance knowledge graph; calculating a business impact coefficient of the target accident through 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 a maintenance difficulty coefficient of the target accident through a maintenance difficulty model included in the maintenance knowledge graph; determining a processing priority of the target accident according to 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.

[0055] Specifically, the consequence assessment model of the present application is trained through multiple groups of corresponding historical data. Specifically, by comprehensively considering parameters such as accident type, harm degree, and influence range, an evaluation matrix is constructed through the Analytic Hierarchy Process (AHP), that is, a consequence assessment model is obtained. Through the consequence assessment model, a consequence coefficient caused by the accident can be calculated. The larger the consequence coefficient, the more serious the consequences caused by the accident, that is, analyzing the possible direct hazards caused by the target accident, such as the accumulated water depth of liquid leakage, the toxicity or flammability level of gas leakage, and combining the personnel activity frequency and equipment distribution in the pipe gallery chamber, calculating the potential consequence severity of the accident on personnel safety and equipment damage. The consequence coefficient of the consequence assessment model The calculation formula includes:

[0056] Wherein, is the risk factor weight (such as the gas toxicity weight is 0.4, the leakage rate weight is 0.3, and the personnel exposure risk weight is 0.3); is the risk factor quantization value (the value range is 0-10, for example, when the methane leakage concentration reaches 30% of the LEL, the risk factor quantization value of gas toxicity is 8).

[0057] The risk prediction model is to predict the accident deterioration probability through historical accident evolution data using the LSTM neural network. That is, through the risk prediction model, combined with real-time sensor data (such as leakage volume, diffusion speed) and pipe gallery structure (such as ventilation conditions, fire prevention zones), the accident spread trend is predicted. For example, when gas leaks, the risk model simulates the diffusion range of harmful gases in the pipe gallery to evaluate whether there may be risks of explosion or personnel poisoning in a short time. The input parameters of the risk prediction model are: the current leakage rate ( ), the environmental temperature and humidity (temperature , humidity ), the rescue response time ( ); The output result of the risk prediction model (deterioration probability ): The probability of risk level upgrade within the next 1 hour (e.g., the probability of upgrading from "medium risk" to "high risk" is 25%).

[0058] To adapt to the input requirements of the LSTM network, the original parameters are normalized: , , ,

[0059] Among them, represents 's normalized value, represents 's minimum value, represents 's maximum value; represents 's normalized value, represents 's minimum value, represents 's maximum value; represents 's normalized value, represents 's minimum value, represents 's maximum value; represents 's normalized value, represents 's minimum value, represents 's maximum value.

[0060] The LSTM network model includes: Input layer: Receives a 3D feature vector , , , . Hidden layer: One or more layers of LSTM cells that capture the temporal dependencies in the time series data (such as the trend of leakage rate over time). Fully connected layer: Maps the LSTM output to the deterioration probability. The deterioration probability calculation formula is:

[0061] Among them, represents the activation function Sigmoid or the hyperbolic tangent activation function tanh; represents the weight matrix and bias vector of the input gate at moment; represents the weight matrix and bias vector of the input gate at moment; Indicates the state of the LSTM hidden layer at the moment.

[0062] Substitute the calculated consequence coefficient and the deterioration probability into the weighted calculation formula of the preset weight, so as to calculate the safety risk coefficient . The higher the value of the safety risk coefficient , the higher the accident risk.

[0063] The business impact model calculates the business impact coefficient ( ) through the comprehensive calculation of economic loss, social impact and recovery time. The business impact coefficient ( ) can be directly quantitatively evaluated through economic loss: Economic loss = direct loss + indirect loss. Direct loss: Equipment damage repair cost, spare part replacement cost. Indirect loss: Production suspension loss (such as the daily loss of 1 million yuan for surrounding enterprises due to the shutdown of the pipe gallery) and user compensation cost. Substitute the economic loss into the economic impact coefficient table to obtain the economic impact coefficient corresponding to different economic losses.

[0064] The social impact assessment adopts the fuzzy comprehensive evaluation method, and the indicators include: the number of affected residents (such as 10 points for the number of people ≥ 1000, 6 points for 100 - 1000 people, and 2 points for < 100 people); media attention (8 points for national media reports, 4 points for local media, and 0 points for no reports).

[0065] The recovery time assessment is based on the statistics of historical cases in the maintenance knowledge graph to establish a mapping table of "accident characteristics - recovery time". Example: The average repair time for a DN300 pipeline fracture is 24 hours, and the average time for pressure - holding plugging is 4 hours.

[0066] Substitute the economic impact coefficient, social impact score and recovery time into the business impact coefficient calculation formula to obtain the business impact coefficient .

[0067] The maintenance difficulty coefficient is the technical complexity assessment. The indicators include: maintenance process level (such as 5 points for welding process requiring a certified welder, 2 points for ordinary plugging), equipment accessibility (6 points for requiring diving operation, 1 point for ground operation). The maintenance difficulty model calculates the coefficient corresponding to the technical complexity by comprehensively scoring each indicator. Resource scarcity assessment:

[0068] Among them, is the current resource inventory (such as 5 inventory of a certain type of sealing ring); :The amount of resources required for maintenance (e.g., 3 seals of this model are needed).

[0069] When resources are sufficient ( ≥ ), the resource scarcity coefficient is 1; when emergency procurement is required, it is 0.3 - 0.8 (adjusted according to the length of the procurement cycle). By comprehensively calculating and the maintenance difficulty coefficient can be obtained.

[0070] Determination of processing priority: Normalize the safety risk coefficient , the business impact coefficient , and the maintenance difficulty coefficient to the interval [0, 1] through linear transformation: , , ; Use the weighted summation method to calculate the comprehensive priority index : . Among them, represents the weight value of the normalized safety risk coefficient ; represents the weight value of the normalized business impact coefficient ; represents the weight value of the normalized maintenance difficulty coefficient . The specific weight values are set according to the safety standards for the operation and maintenance of utility tunnels (for example, the safety risk coefficient accounts for 50%, the business impact coefficient accounts for 30%, and the maintenance difficulty coefficient accounts for 20%, which are not specifically limited).

[0071] The priority levels are divided as shown in Table 1: Table 1

[0072] Special / First-level priority: Directly call the "High-priority Pre-plan Library" in the knowledge graph, which includes predefined rapid disposal processes (such as the special response to gas leakage: close all adjacent valves within 10 minutes; start the mobile exhaust fan within 30 minutes; complete leak stoppage within 2 hours). Automatically trigger the cross-departmental collaboration mechanism, such as notifying the fire department, environmental protection department, public security department, etc. for joint disposal.

[0073] Second / Third-level priority: Generate a conventional plan through similarity case matching in the maintenance knowledge graph (as described above), and allow the operation and maintenance personnel to adjust the plan details within their authority (such as changing the spare part brand).

[0074] Resource conflict resolution: When multiple high-priority accidents occur simultaneously, the resource allocation is dynamically adjusted according to the priority index: special-level accidents have priority to occupy all available water pumps; first-level accidents are allocated from high to low according to the ratio of "remaining resource volume / required resource volume".

[0075] Time window management: Combining the maintenance time window constraint (such as prohibiting pipe gallery hot work during the peak urban traffic period), automatically adjust the time nodes in the plan: the original plan was to carry out welding operations from 9:00 to 11:00, if it encounters the morning peak, it will be postponed to 12:00 - 14:00.

[0076] In some alternative embodiments, generating the corresponding maintenance plan according to the maintenance knowledge graph and accident information includes: generating an accident feature vector according to the accident information and the maintenance knowledge graph, the accident feature vector characterizing the accident type, accident severity, accident occurrence time, accident occurrence location and environmental parameters; comparing the similarity between the accident feature vector and multiple historical cases in the historical accident case library of the maintenance knowledge graph, and configuring the historical case with the highest similarity as the candidate case; in the case where the similarity between the candidate case and the accident feature vector is greater than a preset similarity threshold, obtaining the candidate plan configured by the candidate case from the maintenance plan library of the maintenance knowledge graph, and configuring the candidate plan as the maintenance plan of the accident feature vector; in the case where the similarity between the candidate case and the accident feature vector is less than the preset similarity threshold, obtaining the difference vector between the candidate case and the accident feature vector, the difference vector characterizing the vector composed of the difference data between the candidate case and the accident feature vector; sending the difference vector, the candidate plan, the candidate case and the accident feature vector to the operation and maintenance terminal so that the operation and maintenance terminal inputs manual correction parameters; obtaining the corrected plan after correcting the candidate plan through the manual correction parameters; configuring the corrected plan as the maintenance plan of the accident feature vector.

[0077] Specifically, by traversing the historical accident case library in the maintenance knowledge graph (which consists of several past accident records), feature vectors of the same dimension are extracted for each historical case, that is, feature vectors containing accident type, accident severity, accident occurrence time, accident occurrence location, and environmental parameters are obtained. The accident type (such as liquid leakage, gas leakage, equipment failure), accident severity (which can be measured by leakage volume, influence range, etc.), accident occurrence time, accident occurrence location (specific coordinates in the pipe gallery chamber), and environmental parameters (temperature, humidity, wind speed, gas concentration, etc.) are extracted from the accident information. These data are structured to form a format convenient for calculation and analysis, so as to obtain accident feature vectors containing accident type, accident severity, accident occurrence time, accident occurrence location, and environmental parameters. Among them, the accident feature vectors are constructed in the form of word vectors or numerical vectors. All the data are fused into a multi-dimensional vector, and the accident features are completely represented by this multi-dimensional vector.

[0078] 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 accident type, accident severity, accident occurrence time, accident occurrence location, and environmental parameters. Thus, by simply comparing vectors, the historical case with the highest similarity to the accident feature vector can be quickly matched in the historical accident case library, and a maintenance plan for the accident can be quickly generated based on the solution of the historical case, so as to carry out maintenance response in a timely manner according to the maintenance plan after the accident occurs and avoid causing greater losses.

[0079] 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:

[0080] Where, is the accident feature vector, is the historical case vector, represents the dot product of the two vectors, respectively represent the norms of the two vectors. The calculated similarity value ranges between [-1, 1]. The closer the value is to 1, the more similar the two vectors are, that is, the more similar the accident features are to the historical case. The similarities between all historical cases and the accident feature vector are sorted, and the historical case with the highest similarity (that is, the historical case with the most forward similarity ranking) is selected as the candidate case.

[0081] When the similarity between the candidate case and the accident feature vector is greater than the preset similarity threshold (such as 0.8), it means that the current accident has a high similarity with the historical case, and the historical maintenance experience can be directly borrowed. Therefore, the candidate solution configured for the candidate case is obtained from the maintenance solution library of the maintenance knowledge graph. The candidate solution contains detailed maintenance steps for historical cases, required resources (personnel, tools, spare parts, etc.), safety precautions and other information. The specific information is determined according to 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, thereby eliminating the steps of re-analyzing the accident and generating a new solution, thereby improving the maintenance decision efficiency of the accident.

[0082] When the similarity between the candidate case and the accident feature vector is less than the preset similarity threshold, the difference vector between the two is calculated. By comparing the two vectors element by element, the difference data in each dimension is obtained, and these difference data form a difference vector. For example, if the leakage volume in the accident feature vector is 50 cubic meters, and the leakage volume in the candidate case is 30 cubic meters, then the difference data in the leakage volume dimension is 20 cubic meters, and so on, until a complete difference vector is formed.

[0083] 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 terminals are not limited). The operation and maintenance personnel use their professional knowledge and experience to analyze the differences and enter 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 corrects the candidate solutions according to the manual correction parameters, such as adding equipment and operating steps to deal with larger leakage on the basis of the original solution, obtaining a corrected solution, and configuring it as the maintenance solution for the accident feature vector.

[0084] The maintenance plan generation of this application combines intelligent matching with manual experience. It not only utilizes historical data to improve the efficiency of maintenance plan generation, but also ensures the applicability of the plan through manual correction, thereby reducing the risk of unreasonable maintenance plans and even maintenance errors caused by blindly applying historical plans.

[0085] In some alternative embodiments, controlling the air exchange process of the first target chamber by the fan module 110 according to the sensor information of the sensor module includes: determining the positions to be constructed inside the pipe gallery 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 area passed through during construction according to the three-dimensional space model and the optimal path planning model, where the second volume represents the volume of the pipe gallery chamber 100 passed through by the construction personnel, the first volume represents the volume of the pipe gallery 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 area to be constructed and the three-dimensional space model of the pipe gallery; obtaining the sensor information through the gas sensors arranged on the optimal path in the sensor module, where 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 in the area to be constructed is greater than a first threshold, and the air quality in the area passed through during construction is greater than a second threshold, and the first threshold is greater than the second threshold.

[0086] Specifically, the fan module 110 of the present application includes a fan unit 112 for exhausting or blowing air and a fan control unit 111 for controlling the fan unit 112. The fan unit 112 exhausts or blows air through a fan duct 120. During and before the pipe gallery construction, the fan module 110 is controlled according to the sensor information of the sensor module to perform air exchange processing on the first target chamber, thereby ensuring construction safety and personnel health.

[0087] Before construction, detailed coordinates, shapes, dimensions, etc. of the positions to be constructed inside the pipe gallery after a preset time period are obtained by uploading to the system or pre-storing in the system the construction list; the specific preset time period can be half an hour or one hour before construction, and the specific time is determined according to the actual air exchange efficiency and construction arrangement, and no specific limitation is made here. Through three-dimensional modeling software and combined with the pre-stored structural design drawings of the pipe gallery in the system, a three-dimensional space model including the area to be constructed and the surrounding environment is constructed.

[0088] According to the three-dimensional space model, geometric algorithms are used to calculate the first volume of the area to be constructed and the second volume of the area passed through during construction. For areas with regular shapes, the volume formula can be directly used for calculation; for irregular areas, methods such as grid segmentation or numerical integration are used to approximately calculate the volume.

[0089] Meanwhile, the optimal path planning model can be used to determine the optimal path for construction workers to reach the area to be constructed. Based on the three-dimensional space model of the pipe gallery, this model comprehensively considers factors such as path length, passage obstacles (such as equipment and pipeline layout), and safety factors (such as ventilation conditions and distribution of hazardous areas). Through corresponding path planning algorithms, it generates the optimal path from the starting position of the construction workers to the area to be constructed. During the path planning process, real-time personnel flow information, equipment status information, etc. can also be combined to dynamically adjust the path to ensure the efficient and safe passage of construction workers. It is easy to know that based on the optimal path, the first volume of the area to be constructed and the second volume of the area passed through during construction can be determined.

[0090] In this application, a plurality of gas sensors are arranged on the optimal path. The gas sensors are used to collect air quality data on the optimal path, including oxygen concentration, harmful gas (such as carbon monoxide CO, hydrogen sulfide etc., without specific limitation) concentration, dust particle concentration and other information. The sensors transmit the collected data to the control system in real time. The control system (wherein the control system obtains sensor data through the pipe gallery sensing network) preprocesses the data, such as filtering and denoising, data normalization, etc., to improve the accuracy and availability of the data.

[0091] Based on the obtained sensor data, it is judged whether the current air quality on the optimal path meets the construction safety standards and personnel passage standards. Specifically, the judgment is made by comparing the collected data of various gas concentrations, dust concentrations, etc. with the preset safety thresholds, so as to evaluate the air quality status (which can be measured by quality scoring, etc. Specifically, the score can be determined by the deviation degree of the gas concentration from the safety threshold. The higher the deviation degree, the lower the score). When the air quality does not meet the standard, the pipe gallery space on the optimal path is ventilated.

[0092] When ventilation is carried out, a differential control strategy for the fan module 110 is formulated according to the first volume, sensor information, optimal path and second volume. Since the area to be constructed is the core area of construction activities and has higher requirements for air quality, a first threshold is set (such as oxygen concentration ≥ 20.5%, harmful gas concentration ≤ 50% of the occupational exposure limit); while the personnel stay time in the area passed through during construction is relatively short, a slightly lower second threshold is set (such as oxygen concentration ≥ 19.8%, harmful gas concentration ≤ 80% of the occupational exposure limit).

[0093] When the air quality in area A (only an example, without specific area limitation) on the optimal path of sensor data characterization fails to reach the corresponding threshold, the control system calculates the required ventilation volume according to the degree of association between this area and the area to be constructed and the areas through which the construction passes, as well as the volume of the area. For example, if the air quality in the area to be constructed does not meet the standard, according to the first volume and the target air quality requirements, combined with the ventilation capacity of the fan (such as parameters like air volume and air pressure), the number of fans to be turned on (multiple or one fan can be set in a pipe gallery chamber, without specific limitation), the running time, and the rotational speed are calculated 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. According to the layout and ventilation requirements of the pipe gallery chamber 100, the main fan and the auxiliary fan are reasonably configured. In the area to be constructed, the main fan (i.e., the fan module in the pipe gallery chamber where the area to be constructed is located) is preferentially started to quickly increase the air circulation speed; in the areas through which the construction passes, the auxiliary fan (i.e., the fan module in the pipe gallery chamber where the areas through which the construction passes are located) is started according to the actual situation or the rotational speed of the fan is adjusted. Since the door control modules on the optimal path are opened and the pipe gallery chambers are interconnected, the ventilation air blows from the area to be constructed to the areas through which the construction passes, and the gas is extracted by the fans in the areas through which the construction passes, thus ensuring the ventilation efficiency on the optimal path; also, because the area to be constructed is the air inlet area, the air quality in the area to be constructed is higher than that in other areas through which the construction passes, so as to ensure that the air quality in both the areas through which the construction passes and the area to be constructed meets the requirements.

[0094] The operating state of the fan can also be dynamically adjusted according to the changes in the construction progress and the range of personnel activities to achieve the balance between energy conservation and efficient ventilation.

[0095] Through the above method, precise ventilation control of the first target chamber of the pipe gallery can be achieved, effectively ensuring construction safety and personnel health. And differential ventilation is carried out according to the actual requirements of different areas to avoid unnecessary energy waste and reduce the operating cost. At the same time, it ensures that the air quality in the area to be constructed and the areas through which the construction passes meets the safety standards, reducing the safety risks caused by construction personnel inhaling harmful gases or suffering from oxygen deficiency. It can also dynamically adjust the ventilation strategy according to the construction list, construction progress, and changes in the range of personnel activities, improving the adaptability and intelligent level of the system.

[0096] In some alternative embodiments, the fan module 110 of the first target chamber controlled according to the first volume, the sensor information, the optimal path, and the second volume includes: determining a starting position and an ending position according to the optimal path, where the ending position represents the position of the construction area, and the starting position represents the position where the construction personnel enter the pipe gallery; starting the door control module 170 on the optimal path according to the starting position and the ending position, so that the pipe gallery 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, where the first ventilation module represents the fan module 110 corresponding to the starting position, the second ventilation module represents the fan module 110 on the optimal path, and the second ventilation module does not include the first ventilation module; controlling the first ventilation module to blow air towards the starting position at the first ventilation power, and controlling the second ventilation module to extract air on the optimal path at the second ventilation power, so that the air quality in the area to be constructed is greater than a first threshold, and the air quality in the area passed through during construction is greater than a second threshold.

[0097] Specifically, in this application, the starting position of the optimal path ( ) is the position of the construction personnel entrance, and the coordinates are ( ). The ending position of the optimal path ( ) is the center point of the area to be constructed, and the coordinates are ( ), which can be directly obtained by extracting the optimal path in the three-dimensional space model.

[0098] The optimal path can be divided into continuous line segments according to the boundaries of the pipe gallery chambers 100 (that is, according to the starting positions of each pipe gallery chamber) , each segment corresponding to an independent chamber, and recording the starting coordinates , ending coordinates and lengths of each segment. Based on the optimal path , activate the door control between all adjacent chambers on the path, that is, open the door control between any two adjacent line segments and on the path. The door controls not on the path remain closed, forming a ventilation channel that coincides with the optimal path to ensure ventilation quality. Example: If the optimal path is C01 → C02 → C03, then open the door controls between C01 and C02, and between C02 and C03, and close other door controls.

[0099] The first volume ( ) of the area to be constructed and the second volume ( ) Calculated by the above method, where , is the volume of the i-th chamber on the path.

[0100] Determine the first ventilation power of the first ventilation module and the second ventilation power of the second ventilation module according to the optimal path, sensor information, the first volume and the second volume. Among them, the calculation formula for the power of the first ventilation module is: Where is the reference concentration of ambient air; is the current average concentration of the area to be constructed (obtained by weighted average of sensor data); is the safety factor; is the target air change time; is the fan efficiency.

[0101] The fan power of the i-th chamber on the path in the second ventilation module:

[0102] Where is the current concentration of the i-th chamber; is the path attenuation coefficient (increasing with the increase of the distance from the starting point, such as = 1 + 0.1×i); is the fan efficiency of the j-th.

[0103] Control the first ventilation module at the end position to blow air with the power of the first ventilation module to form a positive pressure area and accelerate the injection of fresh air. At the same time, the second ventilation module on the optimal path exhausts air with the second ventilation power (obtained by superimposing the fan powers of multiple pipe gallery chambers) to form a negative pressure gradient, so as to guide the air flow to flow along the optimal path until the air quality of the area to be constructed is greater than the first threshold and the air quality of the area passed by the construction is greater than the second threshold.

[0104] In some alternative 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 disposed in the second target chamber, and the third water pump unit 150 is disposed outside the second target chamber. The third water pump unit 150 performs water pumping treatment on the second target chamber through a ventilation opening 130 on the second target chamber. Controlling the water pump module to perform water pumping treatment on the second target chamber includes: when the water level information indicates that the accumulated water depth in the second target chamber is greater than or equal to a first depth, starting the first water pump unit 151 to perform water pumping treatment on the second target chamber; when the water level information indicates that the accumulated water depth in the second target chamber is greater than or equal to a second depth, starting the first water pump unit 151 and the second water pump unit 152 to perform water pumping treatment on the second target chamber; when the water level information indicates that the accumulated water depth in the second target chamber is greater than or equal to a third depth, starting the first water pump unit 151, the second water pump unit 152, and the third water pump unit 150 to perform water pumping treatment on the second target chamber.

[0105] Specifically, the first water pump unit of the present application is located at a low-lying area of the second target chamber and is used to pump conventional accumulated water; the second water pump unit is connected in parallel with the first water pump and serves as a standby pump or an auxiliary pump to cope with sudden large amounts of accumulated water. The third water pump unit is installed outside the chamber and is connected to a dedicated drainage pipe through a ventilation opening to pump the accumulated water in the pipe gallery chamber.

[0106] 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 started to perform water pumping treatment on 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 started to perform water pumping treatment on 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 water pump unit, the second water pump unit, and the third water pump unit are simultaneously started to perform water pumping treatment on the second target chamber. Thus, different water pumps are started according to different accumulated water depths to ensure that the accumulated water in the second target chamber can be pumped out in time.

[0107] In some alternative embodiments, the method further includes: when the water level information indicates that the accumulated water depth in 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 perform water pumping treatment on the second target chamber at the first pumping power; When the water level information indicates that the accumulated water depth in the second target chamber is greater than or equal to the second depth, obtain the second water depth change rate according to the water level information, determine the second pumping power of the first water pump unit 151 and the third pumping power of the second water pump unit 152 according to the second water depth change rate, and control the first water pump unit 151 to perform water pumping treatment on the second target chamber at the second pumping power and the second water pump unit 152 to perform water pumping treatment on the second target chamber at the third pumping power; When the water level information indicates that the accumulated water depth in the second target chamber is greater than or equal to the third depth, obtain the third water depth change rate according to the water level information, determine 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 according to the third water depth change rate, and control the first water pump unit 151 to perform water pumping treatment on the second target chamber at the fourth pumping power, the second water pump unit 152 to perform water pumping treatment on the second target chamber at the fifth pumping power, and the third water pump unit 150 to perform water pumping treatment on the second target chamber at the sixth pumping power.

[0108] Specifically, the present application realizes fine control of the water pump power by monitoring the water level change rate in real time and dynamically adjusting the water pump pumping power, avoiding too small or too large water pump power.

[0109] Among them, the pumping power and the water depth change rate satisfy the relationship: .

[0110] Among them, is the density of water; is the acceleration of gravity; is the pumping flow rate; is the current water depth; is the cross-sectional area of the water accumulation area; is the water pump efficiency.

[0111] And introduce a dynamic response coefficient to correct the power calculation:

[0112] The final pumping power .

[0113] When the accumulated water depth in the second target chamber is greater than or equal to the first depth, obtain the first water depth change rate from the water level information, and according to the first water depth change rate and the above-mentioned pumping power The first pumping power of the first water pump unit can be calculated using the calculation formula, and the pumping treatment of the second target chamber can be completed by controlling the first water pump unit to pump water at the first pumping power. Similarly, when the water accumulation depth in 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 based on the second water depth change rate and the above-mentioned pumping power 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 (wherein, the first water pump unit undertakes the basic power, and the second water pump unit supplements the remaining power to meet the extraction of the accumulated water). The pumping treatment of the second target chamber can be completed by controlling the first water pump unit to pump water at the second pumping power and the second water pump unit to pump water at the third pumping power. Similarly, it can be known that when the water accumulation 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, and based on the third water depth change rate and the above-mentioned 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 (wherein, the first water pump unit undertakes the basic power, and the second water pump unit and the third water pump unit supplement the remaining power in sequence to meet the extraction of the accumulated water). The pumping treatment of the second target chamber can be completed by controlling the first water pump unit to pump water at the fourth pumping power, the second water pump unit to pump water at the fifth pumping power, and the third water pump unit to pump water at the sixth pumping power.

[0114] The beneficial effects of implementing the embodiments of the present invention include: in the case of obtaining construction information, determining a first target chamber according to the construction information, where the first target chamber represents the corridor chamber 100 where construction is carried out and construction personnel pass through, and the construction information is obtained through a construction list; controlling the fan module 110 to perform ventilation treatment on the first target chamber according to the sensor information of the sensor module, so that the air quality in the first target chamber is equal to the preset quality; obtaining the water level information of the corridor chamber 100 through the sensor module; in the case where the water level information indicates that water accumulates in the second target chamber, controlling the water pump module to perform pumping treatment on the second target chamber, so that the water accumulation depth in the second target chamber is less than the preset depth. By ventilating the corridor in advance through construction information, the smooth progress of construction is ensured, the construction management and control efficiency is high, and in the event of a water pipe burst accident, the water pump module is controlled to perform pumping treatment by timely obtaining the water level information, thereby avoiding unnecessary losses to the corridor.

[0115] In addition, an embodiment of the present invention provides a corridor control device based on a corridor perception network, and the device includes: a memory, a processor, and a computer program stored on the memory and executable on the processor.

[0116] The processor and the memory can be connected through a bus or other means.

[0117] It should be noted that the computer in this embodiment can correspond to the memory and processor in the embodiment shown as Figure 4 and can form part of the system architecture platform in the embodiment shown as Figure 4 Since the two belong to the same inventive concept, they have the same implementation principle and beneficial effects, which will not be elaborated here.

[0118] The non-transitory software programs and instructions required to implement the uplink co-frequency interference cancellation method of the above embodiment are stored in the memory. When executed by the processor, they execute the pipe gallery control method based on the pipe gallery sensing network of the above embodiment. For example, they execute the method steps S100 to S500 described above in Figure 1 .

[0119] In addition, an embodiment of the present invention also provides a computer-readable storage medium. The computer-readable storage medium stores computer-executable instructions. When the computer-executable instructions are used to execute the pipe gallery control method based on the pipe gallery sensing network of the above device, for example, they execute the method steps S100 to S500 described above in Figure 1 .

[0120] Those of ordinary skill in the art can understand that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and their appropriate combinations. Some physical components or all physical components can be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or can be implemented as hardware, or can be implemented as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include a computer storage medium (or non-transitory medium) and a communication medium (or transitory medium). As is well known to those of ordinary skill in the art, the term computer storage medium includes 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 disc (DVD), or other optical disc storage, magnetic cassette, 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. In addition, as is well known to those of ordinary skill in the art, a communication medium typically includes computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and can include any information delivery medium.

[0121] The above is a specific description of the preferred embodiment of the present invention. However, the present invention is not limited to the above-mentioned embodiments. Those skilled in the art can also make various equivalent deformations or substitutions without departing from the spirit of the present invention, and these equivalent deformations or substitutions are all included within the scope defined by the claims of the present invention.

Claims

1. A method for controlling an utility tunnel based on an utility tunnel perception network, 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 a sensor module, a water pump module, a fan module, and a door control module. The door control module is arranged at the inlet and outlet ends of the pipe gallery chambers, and adjacent pipe gallery chambers are connected through the door control module. The sensor module, the water pump module, the fan module, and the door control module communicate based on the pipe gallery sensing network. The method includes: When construction information is obtained, determine a first target chamber according to the construction information. The first target chamber represents the pipe gallery chamber where construction is carried out and through which construction personnel pass. The construction information is obtained through a construction list; Control the fan module to perform ventilation treatment on the first target chamber according to the sensor information of the sensor module. Specifically, it includes: obtaining a first volume of the area to be constructed and a second volume of the area passed by construction in the first target chamber, and controlling the fan module of the first target chamber to perform ventilation treatment based on the first volume, the sensor information, and the second volume, so that the air quality in the first target chamber is equal to a preset quality; Obtain the water level information of the pipe gallery chamber through the sensor module; When the water level information indicates that water accumulates in a second target chamber, control the water pump module to perform water pumping treatment on the second target chamber, so that the water accumulation depth in the second target chamber is less than a preset depth.

2. The corridor control method based on the corridor perception network according to claim 1, wherein After controlling the water pump module to perform water pumping treatment on the second target chamber, the method further includes: Obtain the maintenance knowledge graph of the first target chamber and the second target chamber; Generate a corresponding maintenance plan according to the maintenance knowledge graph and accident information. The accident information represents liquid leakage accidents and gas leakage accidents.

3. The corridor control method based on the corridor perception network according to claim 2, characterized in that, Before generating a corresponding maintenance plan according to the maintenance knowledge graph and accident information, the method further includes: Calculate 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; Calculate the business impact coefficient of the target accident through the business impact model included in the maintenance knowledge graph. The business impact coefficient indicates the economic loss, social impact, and recovery time caused by the target accident; Calculate the maintenance difficulty coefficient of the target accident through the maintenance difficulty model included in the maintenance knowledge graph; Determine the processing priority of the target accident according to the safety risk coefficient, the business impact coefficient, and the maintenance difficulty coefficient; Generate the maintenance plan in sequence according to the processing priority.

4. The corridor control method based on the corridor perception network according to claim 2, characterized in that, Generating a corresponding maintenance plan according to the maintenance knowledge graph and accident information includes: Generate an accident feature vector according to the accident information and the maintenance knowledge graph. The accident feature vector represents the accident type, accident severity, accident occurrence time, accident occurrence location, and environmental parameters; Compare the similarity between the accident feature vector and multiple historical cases in the historical accident case library of the maintenance knowledge graph, and configure the historical case with the highest similarity as the candidate case; When the similarity between the candidate case and the accident feature vector is greater than the preset similarity threshold, obtain the candidate solution configured for the candidate case from the repair solution library of the repair knowledge graph, and configure the candidate solution as the repair solution of the accident feature vector; When the similarity between the candidate case and the accident feature vector is less than the preset similarity threshold, obtain the difference vector between the candidate case and the accident feature vector, where the difference vector represents the vector composed of the difference data between the candidate case and the accident feature vector; Send the difference vector, the candidate solution, the candidate case, and the accident feature vector to the operation and maintenance terminal so that the operation and maintenance terminal inputs manual correction parameters; Obtain the corrected solution after correcting the candidate solution through the manual correction parameters; Configure the corrected solution as the repair solution of the accident feature vector.

5. The corridor control method based on a corridor perception network according to claim 1, characterized in that The controlling the air exchange process of the first target chamber by the fan module according to the sensor information of the sensor module includes: Determine the to-be-constructed position inside the pipe gallery based on the construction list and generate a three-dimensional space model; Determine the first volume of the to-be-constructed area and the second volume of the construction passing area according to the three-dimensional space model and the optimal path planning model, where the second volume represents the volume of the pipe gallery chamber passed by the construction personnel, the first volume represents the volume of the pipe gallery chamber to be constructed, and the optimal path planning model is used to generate the optimal path for the construction personnel to reach the to-be-constructed area according to the to-be-constructed area and the three-dimensional space model of the pipe gallery; Obtain the sensor information through the gas sensor arranged on the optimal path in the sensor module, where the sensor information indicates the air quality on the optimal path; Control the fan module 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 to-be-constructed area is greater than the first threshold and the air quality of the construction passing area is greater than the second threshold, and the first threshold is greater than the second threshold.

6. The method for controlling an utility tunnel based on an utility tunnel perception network according to claim 5, wherein The controlling the fan module of the first target chamber according to the first volume, the sensor information, the optimal path, and the second volume includes: Determine the starting position and the ending position according to the optimal path, where the ending position represents the position of the construction area and the starting position represents the position where the construction personnel enter the pipe gallery; Start the door control module on the optimal path according to the starting position and the ending position so that the pipe gallery chambers on the optimal path are connected to each other; Determine the first ventilation power of the first ventilation module and the second ventilation power of the second ventilation module according to the optimal path, the sensor information, the first volume, and the second volume, where the first ventilation module represents the fan module corresponding to the starting position, 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 towards the starting position at the first ventilation power, and control the second ventilation module to extract air along the optimal path at the second ventilation power, so that the air quality in the area to be constructed is greater than the first threshold and the air quality in the area passed through during construction is greater than the second threshold.

7. The method for controlling an underground utility tunnel based on an underground utility tunnel perception network according to claim 1, characterized in that 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, and the third water pump unit is arranged outside the second target chamber. The third water pump unit performs water extraction treatment on the second target chamber through a ventilation opening on the second target chamber; controlling the water pump module to perform water extraction treatment on the second target chamber includes: When the water level information indicates that the accumulated water depth in the second target chamber is greater than or equal to the first depth, start the first water pump unit to perform water extraction treatment on the second target chamber; When the water level information indicates that the accumulated water depth in the second target chamber is greater than or equal to the second depth, start the first water pump unit and the second water pump unit to perform water extraction treatment on the second target chamber; When the water level information indicates that the accumulated water depth in the second target chamber is greater than or equal to the third depth, start the first water pump unit, the second water pump unit, and the third water pump unit to perform water extraction treatment on the second target chamber.

8. The method for controlling an utility tunnel based on an utility tunnel perception network according to claim 7, characterized in that, The method further includes: When the water level information indicates that the accumulated water depth in the second target chamber is greater than or equal to the first depth, obtain a first water depth change rate according to the water level information, determine a first water extraction power of the first water pump unit according to the first water depth change rate, and control the first water pump unit to perform water extraction treatment on the second target chamber at the first water extraction power; When the water level information indicates that the accumulated water depth in the second target chamber is greater than or equal to the second depth, obtain a second water depth change rate according to the water level information, determine a second water extraction power of the first water pump unit and a third water extraction power of the second water pump unit according to the second water depth change rate, and control the first water pump unit to perform water extraction treatment on the second target chamber at the second water extraction power and the second water pump unit to perform water extraction treatment on the second target chamber at the third water extraction power; When the water level information indicates that the accumulated water depth in the second target chamber is greater than or equal to the third depth, obtain a third water depth change rate according to the water level information, determine a fourth water extraction power of the first water pump unit, a fifth water extraction power of the second water pump unit, and a sixth water extraction power of the third water pump unit according to the third water depth change rate, and control the first water pump unit to perform water extraction treatment on the second target chamber at the fourth water extraction power, the second water pump unit to perform water extraction treatment on the second target chamber at the fifth water extraction power, and the third water pump unit to perform water extraction treatment on the second target chamber at the sixth water extraction power.

9. An integrated corridor control device based on an integrated corridor perception network, characterized in that, Including: A memory, a processor, and a computer program stored on the memory and executable on the processor, wherein when the processor executes the computer program, the method for controlling a pipe gallery based on a pipe gallery sensing network according to any one of claims 1-8 is implemented.

10. A computer storage medium, characterized in that, The computer storage medium stores computer-executable instructions for executing the method for controlling a pipe gallery based on a pipe gallery sensing network according to any one of claims 1-8.

Images