Intelligent monitoring and data analysis management platform for municipal pipe network

Abstract

The invention discloses an intelligent monitoring and data analysis management platform for a municipal pipe network. The system comprises a plurality of monitoring areas, an environment sensor group, an environment data processing unit, a corrosion risk assessment module, an intelligent inspection unit, a data transmission unit and a cloud platform. The environment sensor group collects environment data in a monitoring area in real time, and the environment data processing unit analyzes the data in real time through an edge computing technology and generates historical data. And the corrosion risk assessment module assesses the corrosion risk according to the real-time data and the historical data, determines a high-risk area and generates a detection task. The intelligent inspection unit automatically navigates to a target area for light spot projection detection, and the deformation analysis module analyzes image data through a computer vision algorithm and identifies a corrosion area and information thereof. And the data transmission unit uploads the detection information to the cloud platform. Through automatic inspection and light spot projection, the corrosion detection efficiency and precision are improved, and safe operation of pipe gallery equipment is ensured.

Description

Technical Field

[0001] The present invention relates to the field of intelligent monitoring, and particularly to an intelligent monitoring and data analysis management platform for municipal pipe networks. Background Art

[0002] The underground utility tunnel is an important part of modern urban infrastructure, mainly used to carry and maintain various pipeline equipment in the city, such as water pipes, cables, gas pipelines, etc. Due to the special environment of the underground utility tunnel, its pipeline equipment is in complex conditions such as humidity, low temperature, and low light for a long time, and is extremely prone to problems such as corrosion and aging. In severe cases, it may lead to equipment damage and even safety accidents. Therefore, timely and effective corrosion detection and maintenance are crucial for ensuring the normal operation of the underground utility tunnel.

[0003] Currently, most traditional underground utility tunnel corrosion detection methods rely on manual inspections or simple automated monitoring systems. Manual inspections not only have a high work intensity and low efficiency, but are also easily affected by environmental factors, with the risks of missed inspections and false inspections. And some automated monitoring systems usually check the pipeline surface by installing visual monitoring devices such as cameras. However, this camera-dependent monitoring method faces many limitations in scenarios with insufficient light and complex environments like underground utility tunnels.

[0004] Firstly, the underground utility tunnel is often in an environment with extremely weak light, and the camera cannot obtain clear images. Especially when there are dirt, rust and other obstructions on the pipeline surface, the image quality of the camera is greatly affected. In addition, the camera has high requirements for image resolution and contrast, and is prone to problems such as blurred images and excessive color differences in low-light environments, thus affecting the accuracy of the detection results.

[0005] Secondly, the environment in the underground utility tunnel is complex, and the pipelines and equipment are often in narrow spaces. The camera may not be able to cover all areas that need to be inspected, resulting in missed inspections or the inability to comprehensively detect different angles of the pipeline. Especially at pipeline connection parts and hidden positions, it is very difficult for the camera to conduct effective monitoring and inspection.

[0006] Furthermore, traditional camera monitoring systems require a large amount of data storage and transmission support. In the case of large amounts of data and high processing requirements, the storage and calculation pressures are relatively large, which poses high requirements for the real-time performance and scalability of the monitoring system. Moreover, camera monitoring usually can only provide static images or video streams, and it is impossible to obtain the deformation information of pipeline equipment in a timely manner, making it difficult to accurately determine whether the pipeline has corroded or been damaged.

[0007] Therefore, the existing detection method of monitoring through cameras has significant limitations in the low-light and complex environment of underground pipe corridors, and cannot meet the needs of efficient, accurate and real-time detection of corrosion risks. More intelligent and precise technical means are urgently needed to improve the effect of underground pipe corridor corrosion detection. Summary of the invention

[0008] The purpose of the present invention is to provide a municipal pipe network intelligent monitoring and data analysis management platform, which has the advantages of high corrosion detection accuracy and reliability and can detect early corrosion and deal with it in time.

[0009] The above technical objectives of the present invention are achieved through the following technical solutions:

[0010] A municipal pipe network intelligent monitoring and data analysis management platform, including:

[0011] A plurality of monitoring areas, each of which includes at least one environmental sensor group, and the environmental sensor group is used to collect environmental data in the area in real time;

[0012] An environmental data processing unit, electrically connected to the environmental sensor group, for receiving and processing in real time environmental data from each monitoring area, wherein the environmental data processing unit processes and analyzes the environmental data in real time through edge computing technology and stores the data as historical data;

[0013] A corrosion risk assessment module is connected to the environmental data processing unit and is used to rate the corrosion risk of the monitoring area based on the real-time collected environmental data and historical data, determine the corrosion risk area, and sort the detection order of the monitoring area according to the corrosion risk level to generate a detection task;

[0014] An intelligent inspection unit, which can automatically navigate in the underground pipe gallery and go to the monitoring area to perform light spot projection detection according to the detection task. The intelligent inspection unit includes a light spot projection module and a deformation analysis module; the light spot projection module is used to project a light spot of a specific shape onto the surface of the pipeline or equipment; the deformation analysis module is used to obtain image data after the light spot is projected onto the surface of the equipment or pipeline, and perform real-time image processing and deformation analysis based on the computer vision algorithm image data to determine the corrosion area and the corresponding corrosion information;

[0015] A data transmission unit, used to transmit the corrosion information determined by the deformation analysis module to the environmental data processing unit and the cloud platform, wherein the data transmission unit performs two-way data transmission with the cloud platform through the Internet of Things technology;

[0016] The cloud platform is connected to the data transmission unit and is used to store the uploaded corrosion information, generate corrosion monitoring reports and provide real-time warnings.

[0017] Further setting: The environmental data processing unit includes a processing module and a storage module. The processing module is used to preprocess the received environmental data and output the preprocessed environmental data to the corrosion risk assessment module and the storage module. The storage module stores the corrosion information and the corresponding environmental data as historical data and stores them using a time series database.

[0018] Further setting: The preprocessing of the received environmental data specifically includes noise filtering, outlier detection, and data standardization, where the outlier detection uses the Z-Score method.

[0019] Further setting: The steps for rating the corrosion risk of the monitored area, determining the corrosion risk area, and sorting the detection order of the monitored area according to the corrosion risk level to generate a detection task specifically include the following steps:

[0020] Calculate the corrosion risk score of each monitored area using a linear regression model based on historical data and real-time collected environmental data;

[0021] Classify the corrosion risk scores according to the preset classification rules to obtain the corrosion risk levels, and the corrosion risk levels specifically include high risk, medium risk, and no risk;

[0022] Determine the high-risk and medium-risk monitored areas as corrosion risk areas;

[0023] Generate a detection task after sorting according to the specific corrosion risk scores of the corrosion risk areas. The priority of the corrosion risk areas in the detection task is in a direct proportional relationship with the corrosion risk scores.

[0024] Further setting: The spot projection module includes a zoning sub-module, a shape selection sub-module, and a projection sub-module. The zoning sub-module obtains the structure information corresponding to the preset corrosion risk area and divides each device and pipeline in the corrosion risk area into multiple projection areas according to the structure information and the preset zoning rules. The shape selection sub-module calculates the specific shape used for each projection area according to the structure information and environmental data, and the specific shape specifically includes a dot spot, a grid spot, and a stripe spot. The projection sub-module is used to project the selected specific-shaped spot onto the corresponding projection area.

[0025] Further setting: The steps for the shape selection sub-module to calculate the specific shape used for each projection area according to the structure information and environmental data specifically include the following steps:

[0026] Obtain the structure parameters from the structure information;

[0027] Calculate the judgment index for each projection area based on the structural parameters and environmental data;

[0028] Select the specific shape of each projection area according to the judgment index.

[0029] Further set: The deformation analysis module is used to obtain the image data after the light spot is projected onto the surface of the device or pipeline, and perform real-time image processing and deformation analysis on the image data based on computer vision algorithms to determine the corrosion occurrence area and the corresponding corrosion information. The specific steps are as follows:

[0030] Preprocess the image data, and the preprocessing includes denoising and enhancing the contrast;

[0031] Perform illumination normalization on the preprocessed image data;

[0032] Select different deformation analysis methods according to the specific shape of the light spot in the image to analyze and calculate the image data to obtain the corrosion information;

[0033] For a circular dot light spot, analyze and calculate based on the displacement of the circular dot and the change in the distance between circular dots;

[0034] For a grid light spot, analyze and calculate based on the displacement of the intersection points and geometric fitting;

[0035] For a stripe light spot, analyze and calculate based on the change in the stripe angle and the change in the spacing.

[0036] In summary, the present invention has the following beneficial effects:

[0037] 1. Traditional camera monitoring systems are prone to problems such as blurred images and excessive color differences in the low-light and complex environments of underground pipe galleries, resulting in poor detection effects. In contrast, the present invention adopts a light spot projection technology. By projecting a light spot with a specific shape onto the surface of the pipeline or device and combining a deformation analysis module for real-time image processing, it does not rely on environmental light, and thus can provide clear detection results in low-light or even dark environments, effectively overcoming the influence of low-light environments on image quality.

[0038] 2. Through the intelligent inspection unit, the system can automatically navigate to the target monitoring area for detection, no longer relying on manual inspections, greatly reducing labor costs and work intensity, and avoiding omissions or misjudgments that may occur during manual inspections. The intelligent inspection system not only improves the detection efficiency but also ensures that all monitoring areas of the underground pipe gallery are comprehensively inspected, effectively improving the overall working efficiency of the system.

[0039] 3. Through the corrosion risk assessment module, the system can accurately calculate the corrosion risk score for each monitored area based on the real-time collected environmental data and historical data using a linear regression model, and generate corresponding detection tasks according to the score. By sorting and prioritizing the monitored areas, the system can efficiently schedule inspection tasks to ensure that high-risk areas are detected first, effectively avoiding the risks of missed and misdetected inspections.

[0040] 4. The deformation analysis module combines spot projection and computer vision algorithms to accurately detect the deformation on the surface of pipelines or equipment, and determine the area and degree of corrosion occurrence based on the deformation information. This technical solution can not only detect surface corrosion, but also analyze the specific morphology and distribution of corrosion through image data analysis to ensure the accurate extraction of corrosion information, improving the accuracy and reliability of corrosion detection.

[0041] 5. By using a time series database to store historical data, the system can long-term track the corrosion status of pipeline equipment, conduct trend analysis and historical data backtracking. This not only provides data support for the long-term monitoring of pipe gallery equipment, but also helps analyze the periodic laws of corrosion occurrence, providing a basis for future maintenance decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is the overall flow block diagram of the embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] The present invention will be further described in detail below with reference to the accompanying drawings.

[0044] Embodiment:

[0045] As Figure 1 shown, an intelligent monitoring and data analysis management platform for municipal pipe networks includes:

[0046] A number of monitored areas, each of which includes at least one environmental sensor group for real-time collecting environmental data within the area;

[0047] An environmental data processing unit electrically connected to the environmental sensor group for receiving and real-time processing environmental data from each monitored area. The environmental data processing unit performs real-time processing and analysis on the environmental data through edge computing technology and stores it as historical data;

[0048] A corrosion risk assessment module connected to the environmental data processing unit for rating the corrosion risk of the monitored areas based on the real-time collected environmental data and historical data, determining the corrosion risk areas, and sorting the detection order of the monitored areas according to the corrosion risk level to generate detection tasks;

[0049] Intelligent inspection unit, which can automatically navigate in the underground pipe gallery and go to the monitoring area according to the detection task for spot projection detection. The intelligent inspection unit includes a spot projection module and a deformation analysis module; the spot projection module is used to project a spot with a specific shape onto the surface of the pipeline or equipment; the deformation analysis module is used to obtain the image data after the spot is projected onto the surface of the equipment or pipeline, and perform real-time image processing and deformation analysis on the image data based on computer vision algorithms to determine the corrosion occurrence area and the corresponding corrosion information;

[0050] Data transmission unit, which is used to transmit the corrosion information determined by the deformation analysis module to the environmental data processing unit and the cloud platform. The data transmission unit performs two-way data transmission with the cloud platform through Internet of Things technology;

[0051] Cloud platform, which is connected to the data transmission unit, is used to store the uploaded corrosion information, generate a corrosion monitoring report and provide real-time early warning.

[0052] The environmental data processing unit includes a processing module and a storage module. The processing module is used to preprocess the received environmental data and output the preprocessed environmental data to the corrosion risk assessment module and the storage module; the storage module stores the corrosion information and the corresponding environmental data as historical data and stores them using a time series database. Environmental data includes environmental parameters related to corrosion such as humidity, temperature, oxygen concentration, and harmful gas concentration. The environmental data (humidity H, temperature T, oxygen concentration 0, harmful gas concentration G, etc.) collected by the sensor group in each monitoring area is transmitted to the processing module through a communication network (such as LoRa, Wi-Fi, etc.). The data is first subjected to noise filtering, outlier detection, and normalization processing to ensure the accuracy of the data. Among them, the Z-Score method is used to normalize the collected environmental data to detect the outlier points exceeding the threshold in the data. The Z-Score calculation formula is:

[0053]

[0054] where X is the current data value, μ is the mean value of the environmental data, and σ is the standard deviation of the environmental data. If |Z| > 3, the data is considered an outlier and is processed or removed.

[0055] The processed environmental data will be corresponded one by one with the corrosion information calculated by the subsequent deformation analysis module. The environmental data and the specific corrosion conditions of the monitored areas where corrosion occurs will be stored in the storage module as historical data, which serves as one of the bases for future corrosion risk assessment to improve the accuracy of corrosion assessment according to historical situations. The data storage uses a time series database (such as InfluxDB) for efficient storage to ensure the scalability of large-scale data processing. The data compression technology, such as the Zlib compression algorithm, is adopted in the storage process to reduce the data storage space and improve the data access efficiency.

[0056] Whenever new environmental data is collected, the environmental data processing unit will dynamically update the historical data using the Weighted Moving Average (WMA). The environmental data processing unit regularly checks the data consistency through the hash verification method to prevent data corruption or tampering, improving the data stability and prediction accuracy. At the same time, it optimizes the storage and query of large-scale data in combination with the time series database. By generating the data hash value and comparing it with the previously stored hash value, the integrity and accuracy of the data are ensured.

[0057] The corrosion risk assessment module is the key part of the whole system, which is used to evaluate the corrosion risk based on the corrosion prediction model according to the real-time environmental data and historical data, and generate detection tasks. After processing the input data, this module outputs the corrosion risk ratings (none, medium, high), and provides priority sorting for the subsequent intelligent inspection unit. The specific steps are as follows:

[0058] Calculate the corrosion risk score of each monitored area using the linear regression model based on the historical data and the real-time collected environmental data;

[0059] Classify the corrosion risk score according to the preset classification rules to obtain the corrosion risk level, and the corrosion risk level specifically includes high risk, medium risk and no risk;

[0060] Determine the monitored areas with high risk and medium risk as corrosion risk areas;

[0061] Generate detection tasks after sorting according to the specific corrosion risk scores of the corrosion risk areas. The priority of the corrosion risk areas in the detection tasks is in a direct proportional relationship with the corrosion risk scores.

[0062] To accurately evaluate the corrosion risk, multiple regression analysis or support vector machine (SVM) is used as a mathematical model for corrosion risk prediction. The corrosion risk assessment model is built based on the relationship between environmental parameters (humidity, temperature, oxygen concentration, and harmful gas concentration) and the corrosion rate. In this embodiment, a linear regression model in the regression analysis model is adopted to model the relationship between corrosion risk and environmental factors. The evaluation formula of the linear regression model is:

[0063] R(t) = β 0 + β 1 ·H(t) + β 2 ·T(t) + β 3 ·0(t) + β 4 ·G(t)

[0064] Wherein, R(t) is the corrosion risk score, β 0 : the intercept term of the regression model, representing the baseline impact of environmental parameters on corrosion risk, H(t), T(t), 0(t), G(t) are humidity, temperature, oxygen concentration, and harmful gas concentration respectively, and β 1 , β 2 , β 3 , β 4 are regression coefficients, representing the degree of influence of each environmental parameter on corrosion risk, and are obtained through training with historical data.

[0065] According to the evaluated corrosion risk value R(t), it is divided into different corrosion risk levels. The corrosion risk classification criteria are as follows: High risk (R(t) ≥ 0.8): This area requires immediate comprehensive inspection; Medium risk (0.5 ≤ R(t) < 0.8): This area also needs to be inspected but the risk is not that high and the priority is behind the high risk; No risk (R(t) < 0.5): The risk of corrosion in this area is relatively low, and there is no need to conduct an inspection for the time being. The risk can be excluded during the subsequent overall inspection. For corrosion risk areas belonging to the high-risk category, they are sorted according to the size of their corrosion risk scores. The higher the corrosion risk score, the higher the priority.

[0066] For the intelligent inspection unit, it is required to be able to navigate automatically in the underground pipe gallery and go to the monitoring area according to the detection task for spot projection detection. The intelligent inspection unit can be a robot, an intelligent trolley or a drone for inspection. In this embodiment, a drone is selected as the carrier of the intelligent inspection unit, which can better project spots from various angles. The spot projection module includes a zoning sub-module, a shape selection sub-module and a projection sub-module. The zoning sub-module obtains the structural information corresponding to the preset corrosion risk area, and divides each device and pipeline in the corrosion risk area into multiple projection areas according to the structural information and the preset zoning rules; the shape selection sub-module calculates the specific shape used for each projection area according to the structural information and environmental data analysis. The specific shape specifically includes a dot spot, a grid spot and a stripe spot; the projection sub-module is used to project the selected spot of a specific shape onto the corresponding projection area. The preset structural information specifically includes position information and structural parameters. The position information represents the specific position distribution of each pipeline and each device. The structural parameters include the three-dimensional structural information of the pipeline and the device, the type of material used for the outer layer, and the shape. The preset zoning rules are based on the three-dimensional structural information and shape of the pipeline and the device. For interfaces and connection points: for example, at the interface of the pipeline, both sides of the connection should be separately divided into different areas, and the interface itself is used as a separate area for detection. This can avoid the interference of complex shapes at the interface on the analysis of other parts; for devices with complex shapes (such as pipelines connected to devices, cabinets externally connected to devices, etc.), they should be segmented according to the geometric shape of the device. For example, the area around the externally connected cabinet is treated as a separate area, different surfaces of the cabinet are used as independent areas, and the places where the pipeline has bends or transitions should be treated as a separate area to reduce the interference of the pipeline bending part on the deformation analysis. The pipelines and devices are divided into multiple projection areas according to the shape and three-dimensional structure as a whole according to the complexity of each part. Through the segmentation of the projection areas, the structure of the projection surface of each independent projection area is made as simple and smooth as possible, reducing complex situations and reducing the interference of complex structures on the analysis of spot projection corrosion conditions, ensuring that each part is independently and accurately detected.

[0067] The steps for the shape selection sub-module to calculate the specific shape used for each projection area according to the structural information and environmental data analysis are as follows:

[0068] Obtain the structural parameters from the structural information;

[0069] Calculate the judgment index for each projection area according to the structural parameters and environmental data;

[0070] Select the specific shape for each projection area according to the judgment index.

[0071] Humidity, temperature, and oxygen concentration are selected from the environmental parameters and added to the calculation of the judgment index. Humidity is usually one of the most important factors affecting the corrosion rate, especially in humid environments. When the humidity is high, the corrosion accelerates. Temperature directly affects the oxidation reaction rate on the metal surface. In an environment with a higher temperature, the corrosion rate is usually faster. The oxygen concentration determines the intensity of the oxidation reaction. In a low-oxygen environment, the corrosion is slower, while in an environment with sufficient oxygen, the corrosion is faster. The corrosion rate affects the specific degree and scope of corrosion if it occurs. For different degrees of corrosion, different-shaped light spots need to be used for projection. For a more complex corrosion situation, using a grid light spot can obtain more detailed data at each location. For a large corrosion area with a low degree, using a stripe light spot can obtain results faster. Therefore, these parameters are added to the calculation of the judgment index. From the type of material used in the outer layer of the structural parameters, a smoothness index is obtained, and the smoothness index and the area of the projection region are added to the calculation of the judgment index. The calculation formula of the judgment index is as follows:

[0072] CRS = w 1 ·H + w 2 ·T + w 3 ·O 2 + w 4 ·S + w 5 ·D

[0073] CRS: Judgment index, w 1 、w 2 、w 3 、w 4 、w 5 are the weights of each parameter, calculated according to their influence degree on the accuracy of the light spot shape. H is humidity, T is temperature, O 2 is the oxygen concentration, S is the area of the projection region, and D is the smoothness index.

[0074] CRS < low threshold: Select stripe light spot or dot light spot; low threshold ≤ CRS < high threshold: Select grid light spot; CRS ≥ high threshold: Select dense grid light spot or adaptive light spot. Adjust the size of the light spot projection according to the area size and shape of the projection region. The density of the light spot: Adjust the density of the light spot according to the level of the judgment index. The light spot density is high in the area with a high judgment index, and the light spot density is low in the area with a low judgment index.

[0075] The deformation analysis module is used to obtain the image data after the light spot is projected onto the surface of the device or pipeline, and perform real-time image processing and deformation analysis on the image data based on computer vision algorithms to determine the corrosion occurrence area and the corresponding corrosion information. The specific steps are as follows:

[0076] Preprocess the image data, and the preprocessing includes denoising and enhancing the contrast;

[0077] Perform illumination normalization on the preprocessed image data;

[0078] Select different deformation analysis methods according to the specific shape of the light spot in the image to analyze and calculate the image data to obtain corrosion information;

[0079] For circular dot light spots, the deformation analysis of circular dot light spots is usually based on the displacement and shape change of the circular dots. The specific steps are as follows:

[0080] Step 1: Circular dot detection and fitting Use the Hough transform or circular fitting algorithm to detect the circular dots in the image and fit the center position and radius of the circular dots.

[0081] Step 2: Calculate the displacement of the circular dots Compare the ideal position and the actual position of the circular dots, and calculate the displacement Δx, Δy of each circular dot,

[0082]

[0083] Through the change of displacement, the area where corrosion occurs can be judged. Displacement is the displacement of the circular dot, (x i ,y i ) is the position of the circular dot in the actual image, (x ideal ,y ideal ) is the ideal circular dot position.

[0084] Step 3: Calculate the change in the distance between circular dots The distance between circular dots can also reflect the deformation of the surface. If the circular dots shift, the distance between circular dots will change, and further calculate the area of corrosion.

[0085] Δd = d t - d 0

[0086] where, d t and d 0 are the distances between circular dots after deformation and in the ideal case respectively.

[0087] Step 4: Output the type and area of corrosion According to the pattern of circular dot offset, judge the type of corrosion (such as pitting corrosion, local corrosion, etc.), and calculate the area of the corrosion area.

[0088] A corrosion = ∑Δd i

[0089] where, A corrosion is the area of the corrosion area, and Δd i is the change in the distance between each circular dot.

[0090] Step 5: Judge the degree of corrosion Determine the degree of corrosion (mild, moderate, severe) according to the change of the offset and the distance between circular dots.

[0091] For a grid light spot, the grid light spot consists of multiple intersecting horizontal and vertical lines, forming a rectangular or square grid structure. When the surface of a pipeline or device deforms, the intersections of the grid will undergo displacement or distortion. The deformation analysis of the grid light spot is mainly based on the displacement of the grid intersections, and is usually carried out through feature point detection and geometric fitting methods. The specific steps are as follows:

[0092] Step 1: Detect grid intersections. Use a feature point detection algorithm (such as SIFT, SURF, or ORB algorithm) to extract the intersections in the grid light spot. Each intersection represents the intersection position of the grid, and the deformation analysis is based on the displacement of these intersections.

[0093] Step 2: Calculate the intersection displacement. Compare the positions of the intersections in the original (ideal) grid and the actual image, and calculate the displacement amounts Δx and Δy of each intersection.

[0094]

[0095] where Displacement is the displacement of the grid intersection, (x i , y i ) is the intersection position in the actual image, and (x ideal , y ideal ) is the intersection position in the reference grid.

[0096] Step 3: Geometric fitting and deformation model. Use the least squares method to fit the ideal grid and the actual grid, and calculate the offset and deformation of the intersections. Through the fitting method, the deformation mode of the entire grid can be obtained, and the position and area of corrosion can be further deduced. Least squares fitting:

[0097] Residual = ∑((x i - x ideal ) 2 + (y i - y ideal ) 2 )

[0098] Residual: The residual of the least squares method, used to measure the fitting error.

[0099] Calculate the corrosion area: According to the displacement and offset of the intersections, combined with the area of the grid region, deduce the actual area of corrosion:

[0100] A corrosion = ∑Δx·Δy

[0101] where A corrosion is the area of the corrosion region, and Δx·Δy is the displacement amount of the intersection position.

[0102] Step 4: Output the type and degree of corrosion Based on the deformation pattern of the grid intersections, determine the type of corrosion (such as uniform corrosion, crevice corrosion, etc.) and output the degree of corrosion (mild, moderate, severe).

[0103] For stripe-shaped light spots, stripe-shaped light spots are usually composed of a series of parallel bright or dark lines with a fixed spacing. When the stripe-shaped light spot is projected onto the surface of the pipeline, if the surface deforms (such as corrosion, crack, etc.), these stripes will bend or stretch. The deformation analysis of the stripe-shaped light spot is mainly based on the changes of the stripes, including the curvature, spacing change and angle change of the stripes. The specific steps are as follows:

[0104] Step 1: Extract the stripe edges Use an edge detection algorithm (such as Canny edge detection) to extract the stripe edges. Edge detection can extract the contour of the stripes, which is convenient for subsequent analysis.

[0105] Step 2: Calculate the geometric deformation of the stripes:

[0106] Calculation of angle change: For the projected stripe-shaped light spot, calculate the inclination angle of each stripe and compare it with the angle of the reference image. Calculate the offset angle Δθ of each stripe,

[0107] Δθ = θ observed - θ ideal

[0108] where θ observed is the angle of the actual stripe, and θ ideal is the angle of the reference stripe. The change of the offset angle reflects the degree of surface deformation.

[0109] Calculation of stripe spacing change: Calculate the stripe spacing change Δd and judge the compression or expansion of the stripes. The spacing change directly reflects the expansion or contraction of the surface,

[0110] Δd = d observed - d ideal

[0111] where d observed is the actual stripe spacing, and d ideal is the spacing of the reference stripe.

[0112] Step 3: Deduce the corrosion location and area Based on the stripe angle change and spacing change, deduce the corrosion location and area. Especially in the area where the corrosion is relatively uniform, the deformation of the stripes is relatively regular and the area calculation is more intuitive,

[0113] A corrosion = ∫Δd(x)dx

[0114] where Δd(x) is the change amount of the stripe spacing, and A corrosion is the area of the corrosion area.

[0115] The deformation analysis results of each light spot will output corrosion information, specifically including: corrosion type (such as uniform corrosion, pitting corrosion, crevice corrosion, etc.); corrosion area (the area of the region calculated based on deformation); corrosion degree (such as mild, moderate, severe); and corrosion location.

[0116] An intelligent monitoring method for corrosion detection of pipeline equipment in an underground utility tunnel, which is applied to a municipal pipe network intelligent monitoring and data analysis management platform described above, specifically includes the following steps:

[0117] Through an environmental sensor group set in multiple monitoring areas, the environmental data in the monitoring areas is collected in real time, and the environmental data includes temperature, humidity, and gas concentration;

[0118] The collected environmental data is preprocessed by an environmental data processing unit, and the preprocessing includes noise filtering, outlier detection, and data standardization, where the outlier detection uses the Z-Score method;

[0119] Through edge computing technology, the processed environmental data is analyzed in real time and historical data records are generated;

[0120] Based on the real-time collected environmental data and historical data, a corrosion risk score for each monitoring area is calculated using a linear regression model;

[0121] According to the corrosion risk score, the corrosion risk levels of the monitoring areas are divided, specifically including three levels: high risk, medium risk, and no risk, and the high-risk and medium-risk areas are determined as corrosion risk areas;

[0122] The detection order of the monitoring areas is sorted according to the corrosion risk scores of the corrosion risk areas, and a detection task is generated according to the sorting. The priority of the corrosion risk area is directly proportional to its corrosion risk score;

[0123] Through an intelligent inspection unit, it automatically navigates to the target monitoring area according to the detection task for light spot projection detection. The intelligent inspection unit includes a light spot projection module and a deformation analysis module. The light spot projection module projects a light spot with a specific shape onto the surface of the pipeline or equipment;

[0124] The image data after the light spot is projected onto the surface of the pipeline or equipment is obtained through the deformation analysis module, and image processing is performed based on computer vision algorithms to analyze the deformation information in the image data to determine the corrosion occurrence area and its corrosion degree;

[0125] The corrosion information analyzed by the deformation analysis module is uploaded to the environmental data processing unit and the cloud platform through the data transmission unit, and the cloud platform generates a corrosion monitoring report and provides real-time warnings.

[0126] In summary, the present invention has the following beneficial effects:

[0127] In traditional camera monitoring systems, problems such as blurred images and excessive color differences are likely to occur in the low-light and complex environments of underground pipe galleries, resulting in poor detection effects. In contrast, the present invention adopts a spot projection technique. By projecting spots of a specific shape onto the surface of pipes or equipment and combining with a deformation analysis module for real-time image processing, it does not rely on ambient light, and thus can provide clear detection results in low-light or even dark environments, effectively overcoming the impact of low-light environments on image quality.

[0128] Through the intelligent inspection unit, the system can automatically navigate to the target monitoring area for detection, no longer relying on manual inspections, greatly reducing labor costs and work intensity, and avoiding possible omissions or misjudgments during manual inspections. The intelligent inspection system not only improves the detection efficiency but also ensures that all monitoring areas of the underground pipe gallery are comprehensively inspected, effectively enhancing the overall working efficiency of the system.

[0129] Through the corrosion risk assessment module, the system can accurately calculate the corrosion risk score of each monitoring area based on the real-time collected environmental data and historical data using a linear regression model, and generate corresponding detection tasks according to the score. By sorting and prioritizing the monitoring areas, the system can efficiently schedule inspection tasks to ensure that high-risk areas are detected first, effectively avoiding the risks of missed inspections and misinspections.

[0130] The deformation analysis module combines spot projection and computer vision algorithms to accurately detect the deformation of the surface of pipes or equipment, and determine the area and degree of corrosion occurrence based on the deformation information. This technical solution can not only detect surface corrosion but also analyze the specific form and distribution of corrosion through image data analysis, ensuring the accurate extraction of corrosion information and improving the accuracy and reliability of corrosion detection.

[0131] By using a time series database to store historical data, the system can long-term track the corrosion status of pipeline equipment, conduct trend analysis and historical data backtracking. This not only provides data support for the long-term monitoring of pipe gallery equipment but also helps analyze the periodic laws of corrosion occurrence, providing a basis for future maintenance decisions.

[0132] The above-described embodiments do not constitute a limitation on the protection scope of this technical solution. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the above embodiments shall be included within the protection scope of this technical solution.

Claims

1. A municipal pipe network intelligent monitoring and data analysis management platform, characterized in that: include: A plurality of monitoring areas, each of which includes at least one environmental sensor group, and the environmental sensor group is used to collect environmental data in the area in real time; An environmental data processing unit, electrically connected to the environmental sensor group, for receiving and processing in real time environmental data from each monitoring area, wherein the environmental data processing unit processes and analyzes the environmental data in real time through edge computing technology and stores the data as historical data; A corrosion risk assessment module is connected to the environmental data processing unit and is used to rate the corrosion risk of the monitoring area based on the real-time collected environmental data and historical data, determine the corrosion risk area, and sort the detection order of the monitoring area according to the corrosion risk level to generate a detection task; An intelligent inspection unit, which can automatically navigate in the underground pipe gallery and go to the monitoring area to perform light spot projection detection according to the detection task. The intelligent inspection unit includes a light spot projection module and a deformation analysis module; The light spot projection module is used to project a light spot of a specific shape onto the surface of a pipe or equipment; the deformation analysis module is used to obtain image data after the light spot is projected onto the surface of the equipment or pipe, and perform real-time image processing and deformation analysis based on the computer vision algorithm image data to determine the corrosion occurrence area and the corresponding corrosion information; A data transmission unit, used to transmit the corrosion information determined by the deformation analysis module to the environmental data processing unit and the cloud platform, wherein the data transmission unit performs two-way data transmission with the cloud platform through the Internet of Things technology; The cloud platform is connected to the data transmission unit and is used to store the uploaded corrosion information, generate corrosion monitoring reports and provide real-time warnings.

2. According to claim 1, an intelligent system for detecting corrosion of underground pipe gallery pipeline equipment is characterized in that: The environmental data processing unit includes a processing module and a storage module. The processing module is used to preprocess the received environmental data and output the preprocessed environmental data to the corrosion risk assessment module and the storage module; the storage module stores the corrosion information and the corresponding environmental data as historical data and stores them in a time series database.

3. A municipal pipe network intelligent monitoring and data analysis management platform according to claim 2, characterized in that: The preprocessing of the received environmental data specifically includes noise filtering, outlier detection and data standardization, wherein the outlier detection adopts the Z-Score method.

4. A municipal pipe network intelligent monitoring and data analysis management platform according to claim 2, characterized in that: The step of rating the corrosion risk of the monitoring area, determining the corrosion risk area, and sorting the detection order of the monitoring area according to the corrosion risk level to generate the detection task specifically includes the following steps: The corrosion risk score of each monitoring area is calculated using a linear regression model based on historical data and real-time environmental data; The corrosion risk scores are classified into corrosion risk levels according to preset classification rules, wherein the corrosion risk levels specifically include high risk, medium risk and no risk; Identify high-risk and medium-risk monitoring areas as corrosion risk areas; The detection tasks are generated after sorting the corrosion risk areas according to their specific corrosion risk scores, and the priority of the corrosion risk areas in the detection tasks is in direct proportion to the corrosion risk scores.

5. According to claim 4, a municipal network intelligent monitoring and data analysis management platform is characterized in that: The environmental data specifically include humidity, temperature, oxygen concentration and harmful gas concentration. The calculation formula of the corrosion risk score is specifically: R(t)=β0+β1·H(t)+β2·T(t)+β3·O(t)+β4·G(t) Among them, R(t) is the corrosion risk score, β0 is the intercept term of the linear regression model, which represents the baseline impact of environmental parameters on corrosion risk, H(t), T(t), O(t), and G(t) are humidity, temperature, oxygen concentration, and harmful gas concentration, respectively. β1, β2, β3, and β4 are regression coefficients, which represent the degree of influence of each environmental data on corrosion risk and are obtained through training with historical data.

6. The intelligent monitoring and data analysis management platform for municipal pipe networks according to claim 1 is characterized in that: The light spot projection module includes a partitioning submodule, a shape selection submodule and a projection submodule. The partitioning submodule obtains the structural information corresponding to the preset corrosion risk area, and divides each equipment and pipeline in the corrosion risk area into multiple projection areas according to the structural information and the preset partitioning rules; the shape selection submodule calculates the specific shape used for each projection area according to the structural information and environmental data analysis, and the specific shapes specifically include dot light spots, grid light spots and stripe light spots; the projection submodule is used to project the selected light spot of the specific shape to the corresponding projection area.

7. The intelligent monitoring and data analysis management platform for municipal pipe networks according to claim 6 is characterized in that: The shape selection submodule analyzes and calculates the specific shape used in each projection area according to the structural information and environmental data, and specifically includes the following steps: Get structure parameters from structure information; Calculate the judgment index of each projection area based on structural parameters and environmental data; A specific shape for each projection area is selected based on a judgment index.

8. The intelligent monitoring and data analysis management platform for municipal pipe networks according to claim 7 is characterized in that: The deformation analysis module is used to obtain image data after the light spot is projected onto the surface of the equipment or pipeline, and to perform real-time image processing and deformation analysis based on the computer vision algorithm image data to determine the corrosion occurrence area and the corresponding corrosion information. Specifically, the steps include: Preprocessing the image data, wherein the preprocessing includes denoising and contrast enhancement; Perform illumination normalization on the preprocessed image data; According to the specific shape of the light spot in the image, different deformation analysis methods are selected to analyze and calculate the image data to obtain corrosion information; For the dot light spot, analysis and calculation are performed based on the displacement of the dots and the change of the dot spacing; For the grid spot, the analysis and calculation are performed based on the displacement and geometric fitting of the intersection points; For the stripe light spot, analysis and calculation are performed based on the change of stripe angle and spacing.

9. The intelligent monitoring and data analysis management platform for municipal pipe networks according to claim 1 is characterized in that: The corrosion information specifically includes the type of corrosion, the area of ​​corrosion, and the degree of corrosion.

Images