Underground sewage pipeline-oriented multi-parameter risk assessment method and collaborative detection system

Abstract

The invention discloses an underground sewage pipeline-oriented multi-parameter risk assessment method and a collaborative detection system, and relates to the technical field of underground pipeline detection. The method comprises the following steps: acquiring shooting data of a binocular camera and sensor data of water quality, gas quality and flow in a pipeline; performing three-dimensional mapping based on the shooting data of the binocular camera to obtain environment information; constructing a fuzzy comprehensive evaluation model; the fuzzy comprehensive evaluation model comprises an index system and a corresponding index weight; the index system is composed of one target layer, three criterion layers and nine index layers; the index weight is corrected by adopting an entropy weight correction optimization method; and based on the sensor data and the environment information, performing fuzzy comprehensive evaluation by using the fuzzy comprehensive evaluation model to obtain a final risk level. The problems of low detection efficiency, single parameter, poor environmental adaptability and the like in the prior art can be solved, and the intelligent level of pipeline operation and maintenance is improved.

Description

Technical Field

[0001] This invention relates to the field of underground pipeline inspection technology, and in particular to a multi-parameter risk assessment method and collaborative inspection system for underground sewage pipelines. Background Technology

[0002] Current urban underground sewage pipeline inspections face numerous technical bottlenecks: Over-reliance on manual inspections limits the inspection range and accuracy, while also posing safety risks to inspectors, such as poisoning from toxic gases and suffocation in confined spaces, making it difficult to effectively handle complex underground environments. Regarding equipment, traditional devices can only measure water quality or gas parameters at single points, failing to perform multi-parameter collaborative analysis and thus unable to comprehensively and accurately reflect the pipeline environment. In terms of robot applications, most are corded types, with cable lengths limiting the inspection range, prone to tangling in complex pipe sections, poor adaptability to high-flow-rate conditions, and difficulty in guaranteeing real-time data transmission. Furthermore, the harsh conditions inside sewage pipelines, such as high humidity, corrosion, and siltation, render existing equipment inadequately waterproof and dustproof, and sensors susceptible to interference. Summary of the Invention

[0003] The purpose of this invention is to provide a multi-parameter risk assessment method and collaborative detection system for underground sewage pipelines, aiming to solve or improve at least one of the above-mentioned technical problems.

[0004] To achieve the above objectives, the present invention provides the following solution:

[0005] A multi-parameter risk assessment method for underground sewage pipelines includes:

[0006] Acquire data captured by a binocular camera, as well as sensor data on water quality, gas quality, and flow rate within the pipeline;

[0007] Based on the data captured by the binocular camera, a 3D map is constructed to obtain environmental information;

[0008] A fuzzy comprehensive evaluation model is constructed; the fuzzy comprehensive evaluation model includes an indicator system and corresponding indicator weights; the indicator system consists of one indicator layer, three criteria layers, and nine indicator layers; the indicator weights are corrected using an entropy weight correction optimization method;

[0009] Based on the sensor data and the environmental information, the fuzzy comprehensive evaluation model is used to perform fuzzy comprehensive assessment to obtain the final risk level; the fuzzy comprehensive assessment adopts a two-level evaluation process.

[0010] Optionally, the target layer is the safety risk of underground pipeline network, the criterion layers are gas safety accident category, environmental pollution category and dredging category, and the indicator layers are methane, sulfur dioxide, hydrogen sulfide, carbon monoxide, carbon dioxide, copper ions in water, lead ions in water, cadmium ions in water and sewage flow rate.

[0011] Optionally, the process for determining the indicator weights of the indicator system is as follows:

[0012] Acquire historical indicator layer data and define the category weights of each criterion layer according to the standard document;

[0013] Based on the historical indicator layer data, normalization processing is performed, a single-factor evaluation matrix is ​​constructed based on the normalized data, and the corresponding single-factor risk level is determined based on the single-factor evaluation matrix.

[0014] Define the local weight of each risk factor according to the single-factor risk level described above;

[0015] The category weights and local weights are multiplied to obtain the overall weights of each indicator;

[0016] The overall weight of each indicator is adjusted by using the entropy weight correction optimization method to obtain the final indicator weight of each risk factor in the indicator system.

[0017] Optionally, the step of using the entropy weight correction optimization method to perform weight fusion correction on the overall weight of each indicator to obtain the final indicator weight of each risk factor in the indicator system specifically includes:

[0018] Construct an initial expert weighting scoring table; the initial expert weighting scoring table includes each risk factor and its corresponding expert weighting score;

[0019] Based on the expert initial weight scoring table, the entropy value of the indicator is calculated to obtain the entropy value calculation result corresponding to each risk factor.

[0020] Entropy weights are calculated based on the entropy values, and weights are fused based on the total weights and entropy weights to obtain a fused weight vector.

[0021] The fused weight vector is orthogonally rotated to maximize the weight variance, resulting in rotated weights, which are then determined as the final indicator weights for the risk factors.

[0022] Optionally, the formula for calculating the weight fusion is:

[0023]

[0024] Among them, W 主观 W represents the fused weight vector.AHp ω represents the total weight, and ω represents the entropy weight.

[0025] Optionally, the step of performing fuzzy comprehensive evaluation based on the sensor data and the environmental information using the fuzzy comprehensive evaluation model to obtain the final risk level specifically includes:

[0026] Based on the sensor data, the environmental information, and the fuzzy comprehensive evaluation model, a first-level fuzzy evaluation is performed on the categories of each criterion layer to obtain the first-level evaluation matrix corresponding to each category;

[0027] The various primary evaluation matrices are integrated into a comprehensive risk matrix, and the final risk level is determined by the principle of maximum membership.

[0028] This invention also provides a multi-parameter collaborative detection system for underground sewage pipelines, which applies the method described above, including:

[0029] The robot motion and navigation module is used to acquire odometry data and IMU data, estimate its own motion state, and output navigation commands based on the estimation results.

[0030] The environmental perception module is used to create a 3D map of the internal environment of the pipeline and obtain environmental information.

[0031] A multi-parameter collaborative detection module is used to detect sensor data on water quality, gas quality, and flow rate within the pipeline;

[0032] A remote data communication module is used to transmit the sensor data and the environmental information;

[0033] The power management module is used to provide power using a lithium battery.

[0034] The fuzzy risk assessment module is used to perform risk assessment on the environmental information and sensor data using a fuzzy comprehensive evaluation model to obtain the final risk level. The fuzzy comprehensive evaluation model includes an indicator system and corresponding indicator weights. The indicator system consists of one objective layer, three criteria layers, and nine indicator layers. The indicator weights are corrected using an entropy weight correction optimization method.

[0035] Optionally, the multi-parameter collaborative detection module specifically includes: an air parameter acquisition module, a heavy metal parameter acquisition module, and a wastewater flow acquisition module;

[0036] The air parameter acquisition module includes multiple gas sensors for detecting the concentration of harmful gases in the sewage pipe; the heavy metal parameter acquisition module includes multiple heavy metal ion sensors for monitoring the concentration of heavy metal ions in the sewage pipe, including lead, copper, and cadmium; the sewage flow acquisition module uses a Doppler flow sensor.

[0037] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0038] This invention discloses a multi-parameter risk assessment method and collaborative detection system for underground sewage pipelines. The method includes acquiring data from a binocular camera and sensor data on water quality, gas quality, and flow rate within the pipeline; constructing a 3D map based on the binocular camera data to obtain environmental information; building a fuzzy comprehensive evaluation model; the fuzzy comprehensive evaluation model includes an index system and corresponding index weights; the index system consists of one objective layer, three criteria layers, and nine index layers; the index weights are corrected using an entropy weight correction optimization method; based on the sensor data and the environmental information, a fuzzy comprehensive evaluation is performed using the fuzzy comprehensive evaluation model to obtain the final risk level; the fuzzy comprehensive evaluation adopts a two-stage evaluation process. This invention can solve the problems of low detection efficiency, single parameters, and poor environmental adaptability in existing technologies, enabling simultaneous acquisition and real-time transmission of water quality, gas quality, and flow rate parameters, and allowing wireless robots to operate stably in complex pipeline environments, achieving remote monitoring and risk assessment, and improving the intelligent level of pipeline operation and maintenance. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 This is a flowchart illustrating the multi-parameter risk assessment method in this embodiment;

[0041] Figure 2 This is a schematic diagram of the overall structure in this embodiment;

[0042] Figure 3 This is an external view of the wireless robot in this embodiment; wherein, (a) is the external view of the robot; (b) is a diagram showing the installation positions of the robot components; (c) is a front view of the robot moving through the large pipeline; and (d) is a rear view of the robot moving through the large pipeline.

[0043] Figure 4 These are test images of the image sensor in this embodiment; wherein, (a) is a diagram of the pipe opening location; and (b) is a diagram of the location inside the pipe.

[0044] Figure 5 This is a network circuit design diagram for the five air parameter acquisition modules in this embodiment;

[0045] Figure 6 This is a schematic diagram of the square transparent acrylic housing of the gas sensor assembly in this embodiment;

[0046] Figure 7 This is a graph showing the gas concentration measurement inside the pipeline in this embodiment;

[0047] Figure 8 This is a network design diagram of the three heavy metal ion sensors in this embodiment;

[0048] Figure 9 This is a schematic diagram of the underground pipeline remote communication scheme in this embodiment;

[0049] Figure 10 This is a schematic diagram of the power supply structure of the device power module in this embodiment. Detailed Implementation

[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] The purpose of this invention is to provide a multi-parameter risk assessment method and collaborative detection system for underground sewage pipelines, aiming to solve or improve at least one of the above-mentioned technical problems.

[0052] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0053] As a first aspect, the present invention provides, as follows Figure 1 The method for multi-parameter risk assessment of underground sewage pipelines, as shown, includes:

[0054] Step 100: Acquire data captured by the binocular camera and sensor data on water quality, gas quality, and flow rate within the pipeline.

[0055] Step 200: Perform 3D mapping based on the data captured by the binocular camera to obtain environmental information.

[0056] Step 300: Construct a fuzzy comprehensive evaluation model; the fuzzy comprehensive evaluation model includes an indicator system and corresponding indicator weights; the indicator system consists of one indicator layer, three criterion layers and nine indicator layers; the indicator weights are corrected using the entropy weight correction optimization method.

[0057] Step 400: Based on the sensor data and the environmental information, perform fuzzy comprehensive evaluation using the fuzzy comprehensive evaluation model to obtain the final risk level; the fuzzy comprehensive evaluation adopts a two-level evaluation process.

[0058] As a second aspect, the present invention also constructs, based on the above evaluation method, the following: Figure 2 and Figure 3 A multi-parameter collaborative detection system for underground sewage pipelines includes:

[0059] The robot motion and navigation module is used to acquire odometry data and IMU data, estimate its own motion state, and output navigation commands based on the estimation results.

[0060] The environmental perception module is used to create a 3D map of the internal environment of the pipeline and obtain environmental information.

[0061] The multi-parameter collaborative detection module is used to detect sensor data on water quality, gas quality, and flow rate within the pipeline.

[0062] A remote data communication module is used to transmit the sensor data and the environmental information.

[0063] The power management module is used to supply power using a lithium battery.

[0064] The fuzzy risk assessment module is used to perform risk assessment on the environmental information and sensor data using a fuzzy comprehensive evaluation model to obtain the final risk level. The fuzzy comprehensive evaluation model includes an indicator system and corresponding indicator weights. The indicator system consists of one objective layer, three criteria layers, and nine indicator layers. The indicator weights are corrected using an entropy weight correction optimization method.

[0065] As a specific implementation method, the air parameter acquisition module, water quality sensor module, and flow sensor group are modularly installed on the top of the wheeled robot, and the signals are converged through a 485 hub.

[0066] The robot can communicate wirelessly with the host computer and can independently complete multi-parameter detection of underground pipe networks. However, in actual use, a cable can be added to the back of the robot car to prevent the car from running out of power or other accidents and to pull the car out of the underground pipe network.

[0067] Based on the above evaluation method and detection system, in specific applications, the evaluation method is loaded into the detection system for calculation and processing. The following specific implementation process is used as an example for illustration.

[0068] Robot Motion and Navigation Module:

[0069] The robot's main body employs a four-wheel differential structure and a three-layer architecture. Placing the sensors in the first layer allows them to be closer to the target detection area, reducing interference during sensor signal transmission. The main circuitry is placed in the middle layer for better protection of the circuit boards, minimizing the impact of the external environment. The motors and wheels are placed in the bottom layer, lowering the robot's center of gravity and improving its stability and flexibility during movement.

[0070] The robot is equipped with odometry and inertial measurement unit (IMU) sensors to monitor and record its position and attitude information within the pipeline in real time. By accurately combining odometry and IMU data, the robot can accurately estimate its own motion state, thereby achieving precise positioning and efficient navigation in complex pipeline environments. IMU data, including accelerometer and gyroscope readings, provides information such as the robot's attitude, acceleration, and angular velocity. By analyzing IMU data, abnormal attitude changes, vibrations, or motion instability can be detected, thus assessing the robot's motion state. Odometry data estimates the robot's pose changes, including changes in position and orientation, by measuring wheel rotation. By comparing the actual distance traveled with the odometry-estimated distance, the robot's positioning accuracy and path planning accuracy can be evaluated. Integrating voltage data, IMU data, and odometry data for comprehensive analysis provides a more complete assessment of the robot's operational status.

[0071] The robot employs a combination of PID and PWM motor control. The PID controller calculates the control input and generates control signals based on the deviation between the current and desired positions. PWM technology converts these control signals into pulse signals, adjusting the motor's speed and position by controlling the duty cycle of these pulse signals, thus achieving precise motion control for the pipeline robot. The robot is adaptable to pipes with a diameter ≥600mm, features a waterproof (IP68) and dustproof design, and has an adjustable chassis height to handle water accumulation environments.

[0072] Environmental perception module:

[0073] Binocular cameras capture images of a scene simultaneously using two cameras, simulating the perspective of human binoculars to achieve stereoscopic vision. This design provides more realistic and depth-perceived image information, contributing to 3D visual effects and spatial awareness. Equipped with two high-resolution cameras, binocular cameras can capture clear and detailed images, which is beneficial for computer vision tasks such as image recognition and object detection.

[0074] Inside a completely dark pipe at a depth of 20 meters, the night vision capability of the robot's image sensor was tested. Figure 4As shown in section (a), at the pipe opening location, the lighting conditions are sufficient, and the robot's vision system performs excellently, accurately identifying obstacles and environmental changes within the pipe, thus effectively avoiding collisions. Figure 4 As shown in section (b), even in low-light environments, the sensor is still able to capture the necessary environmental information, and the images clearly show the fallen leaves on the ground inside the pipe and the texture of the pipe wall, demonstrating good adaptability.

[0075] LiDAR (LiDAR) uses the time-of-flight measurement method to measure distance. Its working principle is as follows: The LiDAR emits a laser pulse, which is scattered or reflected back to the LiDAR receiver upon encountering a target object. The receiver receives the reflected light signal. A timer (t1) is started when the laser pulse is emitted, and the receiver stops timing (t2) when the laser pulse returns. The round-trip time of the laser is calculated by measuring the time difference (t2-t1), thus estimating the distance between the target object and the LiDAR. Based on the principle of the constancy of the speed of light, the distance between the target object and the LiDAR can be calculated using the following formula:

[0076]

[0077] Where t1 is the timing of the laser pulse emission, t2 is the timing of the laser pulse return, and c is the speed of light in a vacuum (approximately 3 × 10⁸ m / s). Since lasers travel at the speed of light, dividing the round-trip time by 2 yields the one-way distance. Using this time-of-flight measurement method, lidar can accurately measure the distance between a target object and the lidar, thus achieving high-precision distance measurement and 3D environment mapping.

[0078] Multi-parameter collaborative detection module:

[0079] (1) Air Parameter Acquisition Module: Integrates CO2, CO, H2S, SO2, and CH4 gas sensors. The sensor measurement ranges are 0-10000ppm, 0-10000ppm, 0-100ppm, 0-1000ppm, and 0-5%VOL, respectively, with an accuracy of ±5%FS. Methane and carbon dioxide are detected using infrared detection principles; sulfur dioxide, hydrogen sulfide, and carbon monoxide are detected using electrochemical principles. Since individual sensor probes are more complex in terms of wiring and power supply, five gas sensor probes are integrated onto a single PCB board, and a 485 communication circuit and synchronous power supply circuit are designed to achieve synchronous data acquisition and power supply for multiple sensors. The network circuit design is as follows: Figure 5 As shown.

[0080] Considering the harsh internal environment of sewage pipes, the sensor probes must possess excellent waterproof performance to ensure long-term stable operation. A square transparent acrylic housing was designed for the gas sensor assembly, with glass glue used to seal the interfaces, thus improving the sensor assembly's waterproof performance. The transparent acrylic housing also offers advantages such as transparency, strong corrosion resistance, and ease of installation and maintenance, making it suitable for gas detection applications in various complex environments. The overall design and installation are as follows... Figure 6 As shown.

[0081] Air parameters were collected from a completed but not yet operational sewage pipeline. A small amount of fallen leaves and an average thickness of 10 cm of dried sludge were present at the bottom of the pipeline. A pipeline robot remained at a depth of 20 meters inside the pipeline for one hour to monitor the gas composition and concentration within the pipeline. Figure 7 As shown, the carbon dioxide concentration remained stable between 420 and 480 ppm, the carbon monoxide concentration fluctuated intermittently, ranging from 0 to 1 ppm, while the concentrations of sulfur dioxide, methane, and hydrogen sulfide were all 0 ppm.

[0082] (2) Water Quality Sensor Module: Includes copper, lead, and cadmium plasma sensors, with a measurement range of 0-200ppm and a resolution of 0.01ppm. This module uses a conventional multi-parameter water quality probe integrated device, acquiring water quality parameter data by installing multiple sensors. It is powered by a 12V DC power supply and transmits data in Modbus format. The overall circuit design fully considers waterproofing and electromagnetic interference issues, with shielded wires wrapped around the connecting cables to ensure signal transmission and stable and reliable system operation. The overall networking and design are as follows: Figure 8 As shown.

[0083] (3) Flow sensor module: Doppler flow velocity sensor (0.03-5.00 m / s) and liquid level sensor (0.02-3.00 m), with accuracies of ±5%FS and ±10%FS, respectively. The Doppler probe first emits a beam of ultrasonic waves. When the ultrasonic waves propagate in the fluid, they encounter impurities such as bubbles or particles, causing a frequency shift. When the ultrasonic waves pass through the fluid and interact with the impurities, the reflected wave generates a Doppler frequency shift Δf, which is proportional to the flow velocity. By measuring this frequency shift Δf, the actual flow velocity of the fluid can be calculated. The Doppler flow velocity sensor was fixed in a large pipe, and three flow velocity points were selected for measurement. The measured flow velocities of 0.57 m / s, 0.95 m / s, and 1.63 m / s were used as standard values ​​for the experiment. To prevent the measurement results from shifting or becoming erroneous, the measurement was repeated multiple times, and the average value was taken as the result. The relative error and repeatability of the flow velocity calculated by the Doppler flow velocity sensor are shown in Table 1. At lower speeds, the system's measurement stability is relatively poor. However, when the speed is increased to 1.63 m / s, the stability of the measurement results is significantly improved, the data reliability is enhanced, and the measurement accuracy remains within ±5% FS.

[0084] Table 1 Static Test Performance Indicators of Doppler Flow Sensor

[0085] Flow rate (standard value) Indication error Repeatability 0.57 8.39％ 15.40％ 0.95 6.87％ 11.50％ 1.63 7.43％ 4.90％

[0086] Remote data communication module:

[0087] Communication and data transmission modules, such as Figure 9 As shown, in order to reduce the storage burden on the robot and improve the overall response speed of the system, the intelligent sensor converts the collected data into voltage signals, which are then collected by the DTU device, converted into digital signals, and sent to the ground computer; the image sensor connects to the ground computer via Wi-Fi to share image and positioning data.

[0088] The ground computer is responsible for displaying environmental information and robot operating status data to the user, and integrating and analyzing this information. The signal source uses 2.4 & 5G, dual-polarized MIMO (Multiple-Input Multiple-Output) panel antennas, supporting 2.4-2.5 / 5.15-5.85GHz. Each antenna contains two independent antennas and two N-mother antenna ports, supporting 2X2 MIMO and improving wireless transmission bandwidth.

[0089] Power management module:

[0090] like Figure 10As shown, this module is powered by a 24V / 10Ah lithium battery. The robot is driven by a motor and requires 24V power. The multi-parameter gas sensor group, water quality heavy metal ion concentration sensor group, image sensor, DTU module, and 485 hub are powered by 12V. The Doppler flow velocity sensor requires a 5V DC power supply. Therefore, when designing the power module, the 24V DC voltage should first be converted to 12V to meet the power supply requirements of the image sensor and wireless communication module. Then, the 12V should be stepped down to 5V to provide a stable voltage for the Doppler flow velocity sensor.

[0091] Fuzzy Risk Assessment Module: Built-in fuzzy risk assessment process for the environmental safety of underground drainage pipe networks.

[0092] First, identify the risk factors and sources that affect the environmental safety of underground drainage pipe networks.

[0093] The safety risks of urban underground drainage pipe networks mainly stem from gas accumulation, heavy metal pollution, and abnormal flow velocity. Specific risk factors and corresponding standards are as follows:

[0094] (1) Risk of toxic and harmful gases (GBZ 2.1-2019 Occupational Exposure Limits for Hazardous Factors in the Workplace).

[0095] Methane (CH4): The lower explosive limit (LEL) is 5% VOL. A warning is required when the volume concentration exceeds 12.5% ​​LEL, and there is a high risk of explosion when it exceeds 50% LEL.

[0096] Hydrogen sulfide (H2S): Maximum permissible concentration (MAC) is 10 mg / m³ 3 Exceeding 50% of the occupational exposure limit (OEL) may cause poisoning, and exceeding 200% OEL directly threatens life.

[0097] Carbon monoxide (CO): The time-weighted average permissible concentration (PC-TWA) in non-high-altitude areas is 20 mg / m³. 3 The short-term exposure limit (PC-STEL) is 30 mg / m³. 3 .

[0098] Carbon dioxide (CO2): PC-TWA is 9000 mg / m³ 3 Excessive concentrations can lead to oxygen deficiency and suffocation; the PC-TWA concentration of sulfur dioxide (SO2) is 5 mg / m³. 3 Excessive levels can corrode pipes and damage the respiratory tract.

[0099] Sulfur dioxide (SO2): Not only is it corrosive, but concentrations exceeding the prescribed standards can cause respiratory damage. Its PC-TWA is 5 mg / m³. 3 PC-STEL is 10 mg / m² 3 .

[0100] (2) Risk of heavy metal ion pollution (GB / T 31962-2015 "Water Quality Standard for Wastewater Discharge into Urban Sewerage Systems"). Among them, copper ions (Cu... 2+ The emission limit is 2.0 mg / L; a pollution warning must be activated if the concentration exceeds 1.0 mg / L. Lead ions (Pb) 2+ The emission limit is 1.0 mg / L; levels exceeding 0.5 mg / L may cause harm to the nervous system. Cadmium ions (Cd) 2+ The emission limit is 0.1 mg / L; emissions exceeding 0.05 mg / L are considered highly toxic pollutants.

[0101] (3) Risk of abnormal flow velocity (GB 50014-2021 "Outdoor Drainage Design Standard").

[0102] The design flow velocity range for gravity flow sewage pipelines is 0.6 to 5.0 m / s. When the flow velocity is below 0.6 m / s, it is prone to siltation, and when it is above 5.0 m / s, it may cause pipeline erosion and damage.

[0103] Then, a fuzzy comprehensive evaluation model is established.

[0104] 1. Establish a set of risk factors

[0105] Based on the needs of pipeline network environmental monitoring, the core risk U is identified:

[0106] U = {U1, U2, U3, U4, U5, U6, U7, U8, U9} = {methane, sulfur dioxide, hydrogen sulfide, carbon monoxide, carbon dioxide, copper ions in water, lead ions in water, cadmium ions in water, wastewater flow rate}.

[0107] 2. Establish a risk level set

[0108] A four-level risk classification system is adopted, combining national standards and emergency response thresholds to define level V:

[0109] V = {1 (safe), 2 (relatively safe), 3 (dangerous), 4 (high risk)}, with specific classification thresholds shown in Table 2.

[0110] Table 2. Classification criteria for risk factors

[0111]

[0112] According to the "Workplace Ambient Gas Detection and Alarm Instruments," the setpoint for a first-level alarm for combustible gases should be less than or equal to 25% LEL, and the setpoint for a second-level alarm should be less than or equal to 50% LEL. For toxic gases, the setpoint for a first-level alarm should be less than or equal to 100% OEL, and the setpoint for a second-level alarm should be less than or equal to 200% OEL. When the measurement range of existing detectors cannot meet the measurement requirements, the setpoint for a first-level alarm for toxic gases should not exceed 5% IDLH, and the setpoint for a second-level alarm should not exceed 10% IDLH. Heavy metal ion concentrations in water are divided into four levels. According to the Integrated Wastewater Discharge Standard GB 8978-1996, cadmium and lead ions exceeding the maximum discharge standard are classified as level 4 (high risk), the maximum presence in sewage pipes is level 3 (dangerous), and exceeding the maximum presence in sewage pipes is level 2 (relatively safe). Regarding pipeline risks caused by flow velocity, level 4 (high risk) is defined as less than 0.6 m / s and greater than 5.0 m / s; intermediate levels determine safety.

[0113] 3. Determine the risk factor level

[0114] Data is collected in real time by sensors and normalized according to the above-mentioned grading standards to construct a single-factor evaluation matrix R. For example, when the methane concentration is 20% LEL, the normalized value is... Corresponding risk level 2 (relatively safe).

[0115] 4. Determine the weights of risk factors

[0116] Based on statistics of 12 types of underground pipeline accidents in China from October 2022 to September 2023 (a total of 814 typical accidents), the specific steps of the Analytic Hierarchy Process (AHP) are as follows:

[0117] (1) Construction of AHP subjective weights, as shown in Table 3.

[0118] Establish a hierarchical structure: divide risk factors into target layer (underground pipeline safety risks), criterion layer (gas safety accidents, environmental pollution, dredging risks), and indicator layer (9 specific risk factors).

[0119] Table 3 Weights of Initial Risk Factors

[0120]

[0121]

[0122] Constructing the judgment matrix: The system simulates 10 experts in the field of underground pipe networks (including 5 senior engineers, 3 university researchers, and 2 municipal management personnel), and scores the relative importance of indicators based on the Satty 1-9 scaling method to form an initial judgment matrix A.

[0123] ① Taking the five indicators (hydrogen sulfide, methane, carbon dioxide, carbon monoxide, and sulfur dioxide) in the "gas safety accident" category as an example, the judgment matrix is ​​as follows:

[0124]

[0125] Normalizing matrix A by its columns, we get:

[0126] Summing by rows yields:

[0127] Normalization yields the weight vector:

[0128] calculate Where (Aw)i represents the i-th element after multiplying matrix A by vector W.

[0129] The maximum eigenvalue can be obtained by calculating the above steps: λ max ≈5.21.

[0130] ② Calculate the consistency index CI

[0131]

[0132] Where n is the order of the judgment matrix, n = 5. When λ max When n = 5, approximately 5.21

[0133] ③ Calculate the random consistency ratio CR

[0134]

[0135] RI is the average random consistency index, which varies with n and is obtained through statistical analysis of a large number of random matrices. The standard values ​​are shown in Table 4.

[0136] Table 4 Average Random Consistency Index

[0137] n 1 2 3 4 5 6 7 8 9 RI 0 0 0.58 0.90 1.12 1.24 1.32 1.41 1.45

[0138] When CI = 0.0525 and n = 5, RI = 1.12, then The consistency of the matrix is ​​acceptable. For matrices with a CR > 0.1, experts need to re-evaluate them.

[0139] (2) Entropy weight correction and optimization

[0140] Based on the setting of 9 risk factors (n=9) and 10 experts (m=10), the calculation is carried out as follows:

[0141] ①Preparation of expert weighting data: The initial weighting scores of 9 risk factors by 10 experts are shown in Table 5:

[0142] Table 5 Initial Weight Scoring

[0143]

[0144]

[0145] ② Calculation of index entropy value

[0146] 1) Normalize the weight vector for each expert:

[0147] Taking expert 1 as an example, its weight vector sum is:

[0148]

[0149] Taking expert 3 as an example, its weight vector sum is:

[0150]

[0151] Similarly, all expert weight vectors have been normalized.

[0152] 2) Calculate the entropy value of the index:

[0153] Taking hydrogen sulfide (U3) as an example, the weights of 10 experts for it are: 0.60, 0.58, 0.65, 0.55, 0.62, 0.61, 0.59, 0.63, 0.57, 0.60.

[0154] For each expert k, calculate but: Substitute the data:

[0155]

[0156] but:

[0157]

[0158] Taking cadmium ions (U8) as an example, the weights of the 10 experts are: 0.03, 0.03, 0.03, 0.03, 0.03, 0.04, 0.03, 0.03, 0.03, 0.04. but:

[0159]

[0160] The calculation results of the complete entropy value are shown in Table 6 based on the above calculation method:

[0161] Table 6. Calculation results of complete entropy value

[0162]

[0163] 3) Entropy weight ωi calculate

[0164]

[0165] The denominator calculation results are shown in Table 7:

[0166]

[0167] Table 7 Molecular Calculation Results

[0168]

[0169] Entropy weight:

[0170]

[0171] Since ω3 is negative, while the entropy weight needs to be positive, the entropy weights are shown in Table 8 after the data is readjusted and corrected according to the steps above.

[0172] Table 8. Corrected Entropy Weights

[0173]

[0174] 4) Weight fusion: Calculation The results are shown in Table 9.

[0175] Table 9W AHp Initial weights

[0176]

[0177] Calculate the fusion weights:

[0178]

[0179] The final calculation result of the fusion weight according to the above formula is:

[0180] W=[0.10,0.03,0.2027,0.01,0.03,0.07,0.13,0.1897,0.12]

[0181] 5) Variance-maximizing orthogonal rotation

[0182] ① Perform principal component analysis on the fused weight vector, and maximize the weight variance through orthogonal rotation. The calculation process is as follows:

[0183] Original loads: ω1 = 0.10, ω3 = 0.2027

[0184]

[0185] Rotational backload:

[0186] ω'1=0.10cosθ-0.2027sinθ≈0.1125, ω'3=0.10sinθ+0.2027cosθ≈0.223

[0187] Similarly, the other index pairs are iteratively rotated to obtain the rotated weights:

[0188] w'=[0.1125,0.025,0.223,0.0075,0.025,0.06,0.13,0.21,0.1075]

[0189] ②The comparison and effect before and after rotation are shown in Table 10.

[0190] Table 10 Comparison and Effects Before and After Rotation

[0191]

[0192] Rotational forward deviation: σ 2 ≈0.125, Variance after rotation: σ '2 ≈0.132

[0193] The variance increased by about 5.6%, indicating that the weight distribution was more concentrated and the differentiation between hydrogen sulfide and cadmium ions was enhanced.

[0194] Table 11 Revised Risk Factor Weighting Table

[0195]

[0196] Table 12: Correspondence between the revised risk factor weight table and the initial risk factor weight data

[0197]

[0198] The following conclusions can be drawn from the comparison of data in Table 12:

[0199] The local weight of hydrogen sulfide increased from 68.50% to 72.30%, further highlighting its importance among risk factors in gas safety accidents and aligning with the high risk of hydrogen sulfide poisoning in underground pipeline accidents. The local weight of methane decreased from 25.00% to 22.50%, carbon dioxide from 5.00% to 3.20%, and carbon monoxide from 1.00% to 1.50%, reflecting a more refined adjustment of the importance of gas risk factors and making the allocation of local weights more closely reflect the actual level of risk.

[0200] The revised weights of each risk factor by the General Administration make the overall weight structure of underground pipeline risk assessment more accurate, more scientifically reflect the proportion of different risk factors in the overall risk, and improve the accuracy of subsequent fuzzy comprehensive evaluation models for risk identification and early warning.

[0201] (5) Fuzzy comprehensive evaluation

[0202] 1. Level 1 Fuzzy Evaluation:

[0203] Calculations are made separately for three categories: gas safety accidents, environmental pollution, and dredging risks.

[0204] B1 = W1oR1 (gas risk), B2 = W2oR2 (pollution risk), B3 = W3oR3 (dredging risk). Where B is the fuzzy comprehensive evaluation set.

[0205] W1 = [0.225, 0.005, 0.723, 0.015, 0.032] is the weight vector for gas factors, W2 is the weight vector for environmental pollution, and W3 is the weight vector for dredging; R1 is the gas parameter evaluation matrix, R2 is the environmental pollution evaluation matrix, and R3 is the dredging risk evaluation matrix. This indicates a fuzzy synthesis operation.

[0206] 2. Second-level fuzzy evaluation:

[0207] The primary assessment results are integrated into a comprehensive risk matrix:

[0208]

[0209] Combining the total risk category weights W = [0.5, 0.4, 0.1], calculate the final risk assessment vector:

[0210] B = WoR = (b1, b2, b3, b4)

[0211] Among them, b j This represents the degree of membership of the comprehensive risk to level j. The final risk level is determined by the principle of maximum membership. The calculation results are shown in Table 13.

[0212] Table 13 Risk Levels of Gas Safety Accident Data and Model Output

[0213]

[0214]

[0215] Finally, the model is validated and applied.

[0216] The model has been validated through field testing: the measured data and model output risk levels for a certain municipal pipeline, as shown in Table 13, are consistent with the on-site emergency response results. The model supports real-time data access and can dynamically generate risk warning reports, providing decision support for pipeline network operation and maintenance.

[0217] Data Explanation:

[0218] All gas concentrations did not exceed the occupational exposure limit (OEL), falling within the safe range. The concentrations of the three heavy metal ions were all far below the limits specified in the Integrated Wastewater Discharge Standard, posing no pollution risk. The measured flow velocities were all below 0.6 m / s; flow velocities <0.6 m / s are prone to sedimentation and fall within the "high risk" range. If subsequent data collection exceeds the safety threshold, the risk level will need to be recalculated and an alert triggered.

[0219] In summary, the present invention has the following beneficial effects:

[0220] 1. Multi-parameter collaborative detection: Breaking through the limitations of traditional single-parameter measurement, it realizes simultaneous monitoring of water quality, air quality and flow rate, improving data integrity by more than 80%.

[0221] 2. Remote wireless communication: Free from the constraints of cables, the detection range is extended to more than 200 meters, and the data transmission delay is ≤1 second.

[0222] 3. Environmental adaptability: The robot is adaptable to pipes with a diameter of ≥600mm and can operate stably in pipes with a water depth of ≤0.5m. The sensor maintains an accuracy of ±5%FS in high humidity environments.

[0223] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0224] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A multi-parameter risk assessment method for underground sewer pipes, characterized in that, The method comprises the following steps: acquiring binocular camera shooting data and sensor data of water quality, gas quality and flow rate in the pipeline; performing three-dimensional mapping based on the binocular camera shooting data to obtain environmental information; constructing a fuzzy comprehensive evaluation model; the fuzzy comprehensive evaluation model comprises an index system and corresponding index weights; the index system comprises one target layer, three rule layers and nine index layers; the index weights are corrected by an entropy weight correction optimization method; based on the sensor data and the environmental information, fuzzy comprehensive judgment is performed by using the fuzzy comprehensive evaluation model to obtain a final risk level; the fuzzy comprehensive judgment adopts a two-stage evaluation process.

2. The multi-parameter risk assessment method for underground sewer pipes according to claim 1, characterized in that, The target layer is the safety risk of underground pipe network, the rule layers are respectively gas safety accident category, environmental pollution category and dredging category, and the index layers are respectively methane, sulfur dioxide, hydrogen sulfide, carbon monoxide, carbon dioxide, copper ions in water, lead ions in water, cadmium ions in water and sewage flow rate.

3. The multi-parameter risk assessment method for underground sewer pipes according to claim 1, characterized in that, The index weight determination process of the index system comprises the following steps: acquire historical index layer data, and define the category weights of each rule layer according to a standard file; based on the historical index layer data, perform normalization processing, construct a single-factor evaluation matrix according to the normalized data, and determine the corresponding single-factor risk level according to the single-factor evaluation matrix; define the local weights of each risk factor according to the single-factor risk level; multiply the category weights and the local weights to obtain the total local weights of each index; use the entropy weight correction optimization method to fuse and correct the total local weights of each index to obtain the final index weights of each risk factor in the index system.

4. The multi-parameter risk assessment method for underground sewer pipes according to claim 3, characterized in that, The use of the entropy weight correction optimization method to fuse and correct the total local weights of each index to obtain the final index weights of each risk factor in the index system comprises the following steps: construct an expert initial weight score table; the expert initial weight score table comprises each risk factor and the corresponding expert weight score; based on the expert initial weight score table, perform index entropy value calculation to obtain the entropy value calculation results corresponding to each risk factor; perform entropy weight calculation according to the entropy value calculation results, and fuse the weights according to the total local weights and the entropy weight calculation results to obtain a fused weight vector; maximize the weight variance of the fused weight vector by orthogonal rotation to obtain rotated weights, and determine the rotated weights as the final index weights of the risk factors.

5. The multi-parameter risk assessment method for underground sewer pipes according to claim 4, characterized in that, The calculation formula of the weight fusion is: where W 主观 represents the fused weight vector, W AHp represents the total local weight, and ω represents the entropy weight.

6. The multi-parameter risk assessment method for underground sewer pipes according to claim 1, characterized in that, based on the sensor data and the environmental information, fuzzy comprehensive judgment is performed by using the fuzzy comprehensive evaluation model to obtain a final risk level, which comprises the following steps: perform one-stage fuzzy evaluation on the categories of each rule layer according to the sensor data, the environmental information and the fuzzy comprehensive evaluation model to obtain a one-stage evaluation matrix corresponding to each category; integrate each one-stage evaluation matrix into an overall risk matrix, and determine the final risk level by the maximum membership principle.

7. A multi-parameter coordinated detection system for underground sewage pipes, applying the method according to any one of claims 1-6, characterized in that, The method comprises the following steps: a robot motion and navigation module is used to acquire odometer data and IMU data, estimate the motion state of the robot, and output navigation instructions according to the estimation results; An environment perception module is configured to perform three-dimensional mapping on an internal environment of the pipeline to obtain environment information. A multi-parameter collaborative detection module is configured to detect sensor data of water quality, gas quality, and flow rate in the pipeline. A remote data communication module is configured to transmit the sensor data and the environment information. A power management module is configured to supply power by using a lithium battery. A fuzzy risk assessment module is configured to perform risk assessment on the environment information and the sensor data by using a fuzzy comprehensive evaluation model to obtain a final risk level. The fuzzy comprehensive evaluation model includes an index system and corresponding index weights. The index system includes one target layer, three criterion layers, and nine index layers. The index weights are corrected by using an entropy weight correction optimization method.

8. The multi-parameter synergic detection system for underground sewer pipes according to claim 7, characterized in that, The multi-parameter collaborative detection module specifically includes an air parameter acquisition module, a heavy metal parameter acquisition module, and a sewage flow acquisition module. The air parameter acquisition module includes multiple gas sensors configured to detect concentrations of harmful gases in the sewage pipeline. The heavy metal parameter acquisition module includes multiple heavy metal ion sensors configured to monitor concentrations of heavy metal ions in the sewage pipeline, the heavy metal ions including lead, copper, and cadmium. The sewage flow acquisition module uses a Doppler flow sensor.

Images