A water supply and drainage monitoring and management system based on digital twin

Abstract

The present invention provides a water supply and drainage monitoring and management system based on digital twins, wherein the physical layer includes a distributed sensor network arranged in the water supply and drainage network, and the monitoring data of the water supply and drainage network is collected through the monitoring nodes inside the pipe and the monitoring nodes outside the pipe; the transmission layer is used to transmit the monitoring data collected by the physical layer to the digital twin platform in real time through the wireless communication network; the digital twin platform is used to integrate the received monitoring data into the digital twin model of the water supply and drainage network, and predict the potential risks of each position of the water supply and drainage network according to the monitoring data corresponding to the nodes inside the pipe and the nodes outside the pipe, and obtain the risk prediction results; the user layer is used for the user terminal to obtain the mirror data of the digital twin model of the water supply and drainage network, and to display it visually. The present invention is helpful to improve the management efficiency and intelligence level of the water supply and drainage system management.

Description

Technical Field

[0001] The present invention relates to the technical field of digital twins, and in particular to a water supply and drainage monitoring and management system based on digital twins. Background Art

[0002] At present, water supply and drainage systems play a vital role in urban infrastructure. Traditional monitoring methods are mostly based on the collection of data from decentralized sensors, and abnormal analysis is performed based on the data from a single sensor. This method usually lacks real-time and intelligence, and it is difficult to fully understand the operation status of the system. In addition, existing monitoring methods are difficult to accurately predict and quickly respond to potential risks (such as pipe network leakage). Therefore, it is urgent to propose a more intelligent monitoring and management system to improve the management efficiency and effectiveness of the water supply and drainage system. Summary of the invention

[0003] In response to the above problems, the present invention aims to provide a water supply and drainage monitoring and management system based on digital twins.

[0004] The purpose of the present invention is achieved by adopting the following technical solutions:

[0005] The present invention proposes a water supply and drainage monitoring and management system based on digital twins, comprising a physical layer, a transmission layer, a digital twin platform and a user layer; wherein:

[0006] The physical layer includes a distributed sensor network arranged in the water supply and drainage network, wherein the distributed sensor network includes an in-pipe monitoring node arranged inside the water supply and drainage network, and an out-pipe monitoring node arranged in the periphery of the water supply and drainage network, and the monitoring data of the water supply and drainage network is collected through the in-pipe monitoring node and the out-pipe monitoring node, wherein the monitoring data includes flow data, image data and vibration signal data;

[0007] The transport layer is used to transmit the monitoring data collected by the physical layer to the digital twin platform in real time through the wireless communication network;

[0008] The digital twin platform is used to integrate the received monitoring data into the digital twin model of the water supply and drainage network built based on GIS, and update the status of the corresponding elements in the digital twin model, where the elements include the in-pipe nodes and out-pipe nodes corresponding to the real monitoring nodes; the potential risks of each location of the water supply and drainage network are predicted based on the monitoring data corresponding to the in-pipe nodes and out-pipe nodes, and the risk prediction results are obtained, including:

[0009] Based on the flow data of the upstream and downstream nodes in the pipes of the selected location in the water supply and drainage network, the flow change at the selected location is obtained; image analysis and processing are performed based on the obtained image data to identify the crack characteristics of the inner wall of the network, and the length of the crack is further identified based on the crack characteristics; the impact force on the network location is analyzed based on the obtained vibration signal data; based on the obtained flow change, the identified crack length and the impact force, the potential risk factors of each location in the water supply and drainage network are comprehensively calculated; and the calculated potential risk factors are compared with the preset standard risk factors to obtain risk prediction results.

[0010] The user layer is used for user terminals to obtain mirror data of the digital twin model of the water supply and drainage network and to display it visually.

[0011] Preferably, in the distributed sensor network at the physical layer, the monitoring nodes inside the pipe include flow sensors and image sensors; the monitoring nodes outside the pipe include vibration sensors;

[0012] Flow sensors are used to collect flow data of water supply and drainage network segments;

[0013] The image sensor is used to collect image data inside the water supply and drainage network;

[0014] Vibration sensors are used to collect vibration signal data from the external environment of the water supply and drainage network.

[0015] Preferably, the transmission layer includes a gateway unit and a wireless transmission unit;

[0016] The gateway unit is used to collect monitoring data collected by various sensors in the physical layer;

[0017] The wireless transmission unit is used to transmit monitoring data to the digital twin platform via a wireless network.

[0018] Preferably, the digital twin platform includes a model building unit, a data integration unit, a risk analysis unit, and a visualization unit; wherein,

[0019] The model building unit is used to construct a three-dimensional digital twin model of the water supply and drainage network based on the real GIS data of the water supply and drainage network, wherein the digital twin model has corresponding in-pipe nodes and out-pipe nodes at corresponding positions in the model according to the distribution positions of the real monitoring nodes;

[0020] The data integration unit is used to complete the corresponding API interface settings for each in-pipe node and out-pipe node, and map the acquired monitoring data to the corresponding in-pipe node and out-pipe node;

[0021] The risk analysis unit is used to conduct joint potential risk prediction analysis for each location of the water supply and drainage network based on the monitoring data of the corresponding nodes inside and outside the pipe, obtain the corresponding risk prediction results, and further integrate the obtained risk prediction results into the digital twin model;

[0022] The visualization unit is used to generate visualization data based on the real-time monitoring data and risk analysis results of the digital twin model.

[0023] Preferably, the risk analysis unit includes a flow analysis unit, an image analysis unit, a vibration analysis unit and a quantitative analysis unit; wherein,

[0024] The flow analysis unit is used to obtain the flow change at the selected location in the water supply and drainage network according to the flow data of the upstream and downstream in-pipe nodes at the selected location;

[0025] The image analysis unit is used to perform image analysis processing based on the acquired image data, identify crack characteristics on the inner wall of the pipe network, and further identify the length of the crack based on the crack characteristics;

[0026] The vibration analysis unit is used to analyze the impact force on the pipe network location based on the acquired vibration signal data;

[0027] The quantitative analysis unit is used to comprehensively calculate the potential risk factors of each location in the water supply and drainage network based on the acquired flow change, the identified crack length and the impact force; and to compare the calculated potential risk factors with the preset standard risk factors to obtain the risk prediction results.

[0028] Preferably, the user layer includes a user management unit;

[0029] The user management unit is used to manage users who are allowed to access the water supply and drainage digital twin model. After the user terminal completes the permission / identity authentication, the user terminal is allowed to obtain the real-time mirror data of the water supply and drainage network digital twin model.

[0030] The beneficial effects of the present invention are as follows: The present invention proposes a water supply and drainage monitoring and management system based on digital twins, wherein a distributed sensor network is set for the physical layer, and corresponding sensor nodes are set in the inner and outer areas of the pipe network to collect the monitoring data of the pipe network in real time; the monitoring data is transmitted to the digital twin platform in real time based on the wireless transmission technology through the transmission layer, thereby improving the real-time level of monitoring data collection of the water supply and drainage system. Based on the constructed digital twin platform, a digital twin model corresponding to the real pipe network distribution can be built based on GIS technology, and the acquired monitoring data can be updated to the model in real time, and the real-time display of the real state of the water supply and drainage pipe network can be realized through the digital twin model. Based on the digital twin model, potential risk prediction is further performed based on the acquired monitoring data, and different locations in the pipe network can be quickly and accurately monitored for risks, thereby improving the reliability of risk monitoring. Finally, the mirror data of the digital twin model is provided to the user terminal through the user layer to assist the manager in visually displaying the data of the model, which is helpful to improve the efficiency and intelligence level of the comprehensive management of the water supply and drainage system.

[0031] Among them, for the potential risk prediction of the water supply and drainage pipeline network, the present invention particularly proposes a technical solution for potential risk prediction based on the pipeline flow data, image data and vibration signal data of the pipeline surrounding area. This solution, on the basis of the conventional risk analysis solution based on pipeline flow, further introduces crack identification data inside the pipeline network and vibration signal data of the pipeline surrounding area to comprehensively reflect the abnormal risks of the pipeline network in specific risk scenarios (such as large construction sites, railway surroundings, etc.), thereby comprehensively analyzing the current potential risk status of the pipeline network and intuitively displaying the analysis results, which helps managers to carry out targeted regulation and prevention of potential abnormal risk areas based on the obtained potential risk analysis results, thereby improving the management efficiency and targeted level of urban water supply and drainage systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.

[0033] Figure 1 A schematic diagram of the framework structure of a water supply and drainage monitoring and management system based on digital twins shown in an embodiment of the present invention;

[0034] Figure 2 For the present invention Figure 1 Schematic diagram of the framework structure of the risk analysis unit in the embodiment. DETAILED DESCRIPTION

[0035] The present invention is further described in conjunction with the following application scenarios.

[0036] See also Figure 1 As shown in the embodiment, a water supply and drainage monitoring and management system based on digital twin is shown, including a physical layer, a transmission layer, a digital twin platform and a user layer; wherein,

[0037] The physical layer includes a distributed sensor network arranged in the water supply and drainage network, wherein the distributed sensor network includes an in-pipe monitoring node arranged inside the water supply and drainage network, and an out-pipe monitoring node arranged in the periphery of the water supply and drainage network, and the monitoring data of the water supply and drainage network is collected through the in-pipe monitoring node and the out-pipe monitoring node, wherein the monitoring data includes flow data, image data and vibration signal data;

[0038] The transport layer is used to transmit the monitoring data collected by the physical layer to the digital twin platform in real time through the wireless communication network;

[0039] The digital twin platform is used to integrate the received monitoring data into the digital twin model of the water supply and drainage network built based on GIS, and update the status of the corresponding elements in the digital twin model, where the elements include the in-pipe nodes and out-pipe nodes corresponding to the real monitoring nodes; the potential risks of each location of the water supply and drainage network are predicted based on the monitoring data corresponding to the in-pipe nodes and out-pipe nodes, and the risk prediction results are obtained, including:

[0040] Based on the flow data of the upstream and downstream nodes in the pipes of the selected location in the water supply and drainage network, the flow change at the selected location is obtained; image analysis and processing are performed based on the obtained image data to identify the crack characteristics of the inner wall of the network, and the length of the crack is further identified based on the crack characteristics; the impact force on the network location is analyzed based on the obtained vibration signal data; based on the obtained flow change, the identified crack length and the impact force, the potential risk factors of each location in the water supply and drainage network are comprehensively calculated; and the calculated potential risk factors are compared with the preset standard risk factors to obtain risk prediction results.

[0041] The user layer is used for user terminals to obtain mirror data of the digital twin model of the water supply and drainage network and to display it visually.

[0042] The above-mentioned implementation mode of the present invention proposes a water supply and drainage monitoring and management system based on digital twins, wherein a distributed sensor network is set for the physical layer, and corresponding sensor nodes are set in the inner and outer areas of the pipe network to collect the monitoring data of the pipe network in real time; the monitoring data is transmitted to the digital twin platform in real time based on the wireless transmission technology through the transmission layer, thereby improving the real-time level of monitoring data collection of the water supply and drainage system. Based on the constructed digital twin platform, a digital twin model corresponding to the real pipe network distribution can be built based on GIS technology, and the acquired monitoring data can be updated to the model in real time, and the real-time display of the real state of the water supply and drainage pipe network can be realized through the digital twin model. Based on the digital twin model, potential risk prediction is further performed based on the acquired monitoring data, and different locations in the pipe network can be quickly and accurately monitored for risks, thereby improving the reliability of risk monitoring. Finally, the mirror data of the digital twin model is provided to the user terminal through the user layer to assist the manager in visualizing the data of the model, which is helpful to improve the efficiency and intelligence level of the comprehensive management of the water supply and drainage system.

[0043] Among them, for the potential risk prediction of the water supply and drainage pipeline network, the present invention particularly proposes a technical solution for potential risk prediction based on the pipeline flow data, image data and vibration signal data of the pipeline surrounding area. This solution, on the basis of the conventional risk analysis solution based on pipeline flow, further introduces crack identification data inside the pipeline network and vibration signal data of the pipeline surrounding area to comprehensively reflect the abnormal risks of the pipeline network in specific risk scenarios (such as large construction sites, railway surroundings, etc.), thereby comprehensively analyzing the current potential risk status of the pipeline network and intuitively displaying the analysis results, which helps managers to carry out targeted regulation and prevention of potential abnormal risk areas based on the obtained potential risk analysis results, thereby improving the management efficiency and targeted level of urban water supply and drainage systems.

[0044] Among them, the selected position in the water supply and drainage network mentioned above in the present invention can be understood as, for example, the position of the water supply and drainage network in a specific risk scenario (such as a large construction site, around a railway, etc.), or any position in the water supply and drainage network, or can be specifically selected based on the set in-pipe monitoring nodes and out-pipe monitoring nodes.

[0045] Preferably, in the distributed sensor network at the physical layer, the monitoring nodes inside the pipe include flow sensors and image sensors; the monitoring nodes outside the pipe include vibration sensors;

[0046] Flow sensors are used to collect flow data of water supply and drainage network segments;

[0047] The image sensor is used to collect image data inside the water supply and drainage network;

[0048] Vibration sensors are used to collect vibration signal data from the external environment of the water supply and drainage network.

[0049] By installing flow sensors and image sensors inside the pipeline network, the internal status of the pipeline network can be truly monitored so that risk prediction can be made for the pipeline network status in the future. In addition, vibration sensors are further installed in the surrounding area of ​​the pipeline network, especially in risk scenarios (such as large-scale construction projects, railway network areas, etc.) to collect vibration signal data in the surrounding area of ​​the pipeline network. This will help to further feedback the impact of external risk scenarios on the pipeline network through vibration signal data, and can meet the needs of pipeline network abnormality analysis under different risk scenarios. It has strong adaptability and flexible installation, which helps to improve the adaptability and pertinence of potential risk analysis and feedback of the pipeline network.

[0050] Considering the water supply and drainage pipelines near large-scale construction projects, although the existing construction standards set a minimum safety distance, in the actual construction site, the impact force caused by the construction will still have a non-negligible impact on the pipeline network. Therefore, when conducting a potential risk analysis of the pipeline network, it is necessary to take into account the environmental factors surrounding the pipeline network, which will help improve the adaptability and reliability of the pipeline network risk analysis.

[0051] Preferably, the transmission layer includes a gateway unit and a wireless transmission unit;

[0052] The gateway unit is used to collect monitoring data collected by various sensors in the physical layer;

[0053] The wireless transmission unit is used to transmit monitoring data to the digital twin platform via a wireless network.

[0054] Optionally, the wireless transmission unit can transmit the monitoring data to the digital twin platform through wireless transmission methods such as LoRa, 5G, and ad hoc networks.

[0055] Transmitting monitoring data to the digital twin platform through wireless data transmission helps to improve the real-time and reliability of data transmission.

[0056] Preferably, the digital twin platform includes a model building unit, a data integration unit, a risk analysis unit, and a visualization unit; wherein,

[0057] The model building unit is used to construct a three-dimensional digital twin model of the water supply and drainage network based on the real GIS data of the water supply and drainage network, wherein the digital twin model has corresponding in-pipe nodes and out-pipe nodes at corresponding positions in the model according to the distribution positions of the real monitoring nodes;

[0058] The data integration unit is used to complete the corresponding API interface settings for each in-pipe node and out-pipe node, and map the acquired monitoring data to the corresponding in-pipe node and out-pipe node;

[0059] The risk analysis unit is used to conduct joint potential risk prediction analysis for each location of the water supply and drainage network based on the monitoring data of the corresponding nodes inside and outside the pipe, obtain the corresponding risk prediction results, and further integrate the obtained risk prediction results into the digital twin model;

[0060] The visualization unit is used to generate visualization data based on the real-time monitoring data and risk analysis results of the digital twin model.

[0061] The above-mentioned implementation mode of the present invention can restore the real pipe network distribution data based on the GIS geographic information system through the digital twin platform built, and collect the data of the pipe network and the surrounding area of ​​the pipe network at the same time, so as to build a digital twin model with strong authenticity. Through the built digital twin model, based on the corresponding nodes set in the model, the monitoring data collected by the real sensor can be synchronously updated to the corresponding position of the digital twin model based on the API interface setting, so as to provide real feedback on the status of the pipe network site.

[0062] Based on the risk analysis unit, targeted analysis and potential risk prediction can be carried out on the monitoring data obtained to provide real feedback on the potential risk situation at each location in the pipeline network, which helps to provide intuitive risk warnings for managers (avoiding the situation in which a large number of analysis results appear in traditional single-scale data analysis but the results do not point to obvious impacts on management effectiveness), and helps to improve the comprehensive management level of the water supply and drainage system.

[0063] Preferably, see Figure 2 The risk analysis unit includes a flow analysis unit, an image analysis unit, a vibration analysis unit and a quantitative analysis unit; wherein,

[0064] The flow analysis unit is used to obtain the flow change at the selected location in the water supply and drainage network according to the flow data of the upstream and downstream in-pipe nodes at the selected location;

[0065] The image analysis unit is used to perform image analysis processing based on the acquired image data, identify crack characteristics on the inner wall of the pipe network, and further identify the length of the crack based on the crack characteristics;

[0066] The vibration analysis unit is used to analyze the impact force on the pipe network location based on the acquired vibration signal data;

[0067] The quantitative analysis unit is used to comprehensively calculate the potential risk factors of each location in the water supply and drainage network based on the acquired flow change, the identified crack length and the impact force; and to compare the calculated potential risk factors with the preset standard risk factors to obtain the risk prediction results.

[0068] The above-mentioned embodiment of the present invention proposes a technical solution for performing potential risk analysis on each location of a water supply and drainage network based on flow data, image data and vibration signal data, wherein firstly, feature extraction is performed separately according to the flow data, image data and vibration data corresponding to the network location, the required feature data is extracted, and the potential risk of the network location is quantitatively calculated based on the feature data, so as to provide real feedback on the potential risk level of the network based on the proposed potential risk factors. The potential risk analysis results fed back based on the potential risk factors can comprehensively reflect the internal and external risks of the network location, and intuitively display the risk results to managers, which is helpful to improve the management level of risk management of water supply and drainage systems.

[0069] Preferably, the traffic analysis unit comprises:

[0070] According to the selected location in the water supply and drainage network, one or more nodes in the pipe are retrieved upstream to obtain the total input flow Q of the selected location. j-in ;

[0071] Further obtain the total output flow Q of the selected location to the downstream j-out ;

[0072] Calculate the flow change at the selected location .

[0073] Based on the flow characteristics at each location in the pipeline network, the loss situation in the pipeline network is reflected based on the flow change, and the current leakage risk of the pipeline network is truly fed back.

[0074] Preferably, the image analysis unit comprises:

[0075] According to the image data Pic obtained at the pipe network location j j Perform enhancement processing to obtain an enhanced image of the inner wall of the pipe network;

[0076] According to the acquired pipeline inner wall image, the deep learning model based on YoloV5 is used to identify cracks in the pipeline inner wall image and identify the crack features existing in the pipeline inner wall;

[0077] According to the obtained crack features, combined with the mapping relationship between pixels and actual length, the length of the crack feature is calculated to obtain the length of the crack Cq j .

[0078] Based on image analysis technology, cracks inside the pipeline network can be identified and their lengths measured, which can help provide feedback and reflect current or potential leakage risks of the current pipeline network based on the crack identification results.

[0079] The image processing model based on YoloV5 is used to identify cracks in the acquired pipe network inner wall image, which can accurately identify cracks appearing in the image based on the existing crack recognition model, thereby improving the accuracy of crack recognition. However, specific crack recognition can also be implemented based on other trained crack recognition models or open source image processing engines in the prior art, and the present invention does not make specific limitations here.

[0080] Among them, considering that the image data of the inner wall of the pipe network obtained from the inside of the pipe network is relatively lacking in the natural light environment inside the pipe network, the collected images usually show insufficient brightness, and usually need to rely on light sources (the light source of the camera or image sensor, infrared, LED lights, etc.), and may also be affected by water ripples or dirt reflections, so that the part of the inner wall of the pipe network in the image will appear unclear and interfere with information, affecting the accuracy and reliability of subsequent recognition of cracks on the inner wall of the pipe network. Therefore, before further recognition of inner wall cracks based on the acquired internal image of the pipe network, the present invention first enhances the acquired image data, thereby improving the clarity of the image, eliminating noise interference, and indirectly improving the accuracy of subsequent recognition of cracks in the pipe network.

[0081] Preferably, the image analysis unit further comprises an image enhancement unit, wherein:

[0082] The image enhancement unit is used to perform enhancement processing based on the image data of the pipe network location, specifically including:

[0083] According to the obtained internal image of the pipe network, the gray value h(x,y) of the pixel is obtained according to the RGB information of each pixel in the image;

[0084] The image is divided into M sub-image blocks of the same size. For each sub-image block, the gray level co-occurrence matrix is ​​obtained based on the gray level information of each pixel in the sub-image block. The texture feature factor of the sub-image block is calculated according to the obtained gray level co-occurrence matrix. The texture feature factor calculation function used is:

[0085]

[0086] In the formula, TCP m represents the texture feature factor of the m-th sub-image block, Po(i,j) represents the probability of gray level i and gray level j appearing together in the gray level co-occurrence matrix based on the m-th sub-image block, where i, j are variables in the summation function, and their value ranges are i=1,2,…,k; j=1,2,…,k; k represents the total number of gray levels in the gray level co-occurrence matrix, Indicates the preset adjustment parameters;

[0087] The internal image of the pipe network is further converted from the RGB color space to the Lab color space, and the brightness layer TL, color layer Ta and color layer Tb of the image are obtained respectively; the brightness component value L(x, y) of each pixel is obtained according to the brightness layer TL;

[0088] Based on the brightness component value of each pixel, a Laplace filter with a size of 3×3 is used to perform convolution processing on each pixel to obtain the Laplace eigenvalue LH(x,y) of each pixel;

[0089] Based on the Laplace eigenvalue and brightness component value of each pixel point, the brightness characteristic factor of each sub-image block is calculated, wherein the brightness characteristic factor calculation function used is:

[0090]

[0091] Where LCI m represents the brightness characteristic factor of the mth sub-image block, Indicates that the pixel (a, b) is the pixel in the mth sub-image, LH(a, b) represents the Laplace eigenvalue of the pixel (a, b), L(a, b) represents the brightness component value of the pixel (a, b), meanL m Represents the average brightness component value of each pixel in the mth sub-image, σL m represents the standard deviation of the brightness component value of each pixel in the mth sub-image; θ represents the preset adjustment factor;

[0092] The brightness adjustment process is performed on each sub-image block, wherein the brightness adjustment process function used is:

[0093]

[0094] Where L'(x,y) represents the brightness component value of the pixel (x,y) after brightness adjustment, where the pixel (x,y) is the pixel in the mth sub-image block, TCP m represents the texture feature factor of the mth sub-image block, TCPth represents the preset texture feature threshold, It represents the average brightness component value of each pixel in the 3×3 range centered on the pixel point, L(x,y) represents the brightness component value of the pixel point (x,y), β represents the sensitivity adjustment factor, α represents the adjustment amplitude control factor, LCI m represents the brightness characteristic factor of the mth sub-image block, Lbs represents the preset standard brightness component value, and mean{∙} represents the averaging function;

[0095] After the brightness adjustment processing of each sub-image block is completed respectively, the brightness layer TL' after brightness adjustment is obtained, and reconstruction is performed based on the brightness layer TL' after brightness adjustment to obtain the enhanced pipe network inner wall image.

[0096] The Laplace filter used is .

[0097] The above-mentioned embodiment of the present invention proposes a targeted image enhancement scheme for the acquired internal image data of the pipe network. First, the internal image of the pipe network is divided into multiple sub-image blocks for separate processing, which can improve the effect of local image processing. Based on the grayscale information of each sub-image block, the proposed texture feature factor is used to reflect the consistency of the texture features in the sub-image block. The texture feature factor is used as a basis (considering that in the internal image of the pipe network, the greater the local consistency of the texture feature, the texture feature is a conventional image area part, and therefore the possibility of noise interference or key feature information contained in the part is lower), to perform targeted processing on the image; further, based on the brightness information of the sub-image block in the Lab color space, the brightness feature factor of each sub-image block is calculated in combination with the Laplace filter, and the brightness feature factor is used to provide real feedback on the concentration of the brightness information in the image. The brightness feature factor is used as a basis (considering that in the area where the brightness change information is concentrated, the probability of abnormal reflection is greater), to perform targeted processing on the image. Finally, the texture feature factors and brightness feature factors of each sub-image block in the image are combined to adaptively adjust the brightness features in the image. This can perform general brightness balance adjustment on the non-critical feature parts in the image. For the critical feature parts, the brightness feature factors are further combined to adaptively suppress the brightness of the parts with concentrated brightness change features (usually reflected as water ripple reflections or other reflective interference areas). This can adaptively identify the influence of noise factors and adaptively enhance the brightness of the real critical texture features (such as cracks, pipe edges, etc.), thereby improving the clarity of the key features in the image and laying the foundation for further crack identification based on the enhanced internal image of the pipeline.

[0098] Preferably, the vibration analysis unit comprises:

[0099] According to the vibration signal data sigZ obtained within a time period i (t), extract the vibration acceleration Az at each moment i (t) and vibration signal strength as the vibration characteristics of the corresponding nodes outside the tube;

[0100] The impact force on the pipe network location is calculated based on the obtained vibration characteristics, and the impact force calculation function used is:

[0101]

[0102] Among them, Fz j(t) represents the magnitude of the impact force on the pipe network location j, Do j represents the pipe diameter at location j in the pipe network, Ro j represents the pipe thickness at location j in the pipe network, Represents the material coefficient of the pipe network, Indicates the maximum vibration acceleration monitored by the node outside the pipe during the time period; represents the preset transfer coefficient, Indicates the average vibration signal strength in the current time period. Indicates the preset attenuation factor, d ij It represents the distance from the pipe network location j to the monitoring node i outside the pipe, which is obtained based on the real GIS data. Among them, the material coefficient is for different pipe network materials, and its material coefficient value is as follows: When the material of the pipe network is steel, ; When the material of the pipe network is PVC, , when the material of the pipe network is HDPE, ;

[0103] By analyzing the impact force on the pipeline network through the vibration signal data around the pipeline location, it is possible to provide real feedback on the potential risks of rupture, leakage, etc. in the pipeline network. The proposed impact force calculation function can accurately calculate the impact on the pipeline network based on the vibration signal data around the pipeline network, thereby quantifying and analyzing the impact of the surrounding abnormal environment and adapting to the potential risk characterization caused by the surrounding abnormal environment.

[0104] Preferably, the quantitative analysis unit specifically includes:

[0105] Based on the acquired flow change, identified crack length and impact force, the potential risk factors of each location in the water supply and drainage network are comprehensively calculated. The risk factor calculation function used is:

[0106]

[0107] In the formula, R j represents the potential risk factor of the pipeline network location j, ∆Q j represents the flow change at the pipe network location j, ptQ j represents the historical flow change at the pipe network location j, where , where ptQ j-min Indicates the historical minimum flow change at the pipe network location j within a period of time, ptQ j-meant represents the historical average flow change at the pipe network location j over a period of time, α represents the preset weight factor, and L j Indicates the length of the pipe section at the pipe network location j, Cq j represents the crack length at the pipe network location j, Fz jrepresents the impact force on the pipe network position j; β1, β2 and β3 represent the preset weight adjustment factors respectively;

[0108] According to the potential risk factor R j Compared with the preset standard risk factor RT, when R j >RT, the risk prediction result is abnormal. j When ≤RT, the risk prediction result is normal.

[0109] In an optional embodiment, the preset standard risk factor RT is obtained based on experience, wherein the flow change, the identified crack length and the impact force under standard (healthy, normal) conditions are obtained, and the corresponding theoretical standard risk factor is calculated based on the above parameters under standard conditions; and further based on (expert) experience analysis, the flow change, the identified crack length and the impact force obtained when there are obvious abnormal conditions (such as leakage, fracture, etc.) are used to calculate the corresponding theoretical abnormal risk factor under abnormal conditions, and the preset standard risk factor RT is set based on experience based on the theoretical standard risk factor and the theoretical abnormal risk factor.

[0110] In another optional implementation, the preset weight adjustment factors β1, β2 and β3 can be reasonably set based on experience to set the importance of the three parts of the potential risk factors to adapt to the effect of specific analysis. For example, taking the pipe section length of 4m or 6m as an example, the value range of β1 is set to 1 to 10; the value range of β2 is set to 1000 to 10000; and the value range of β3 is set to 1 / 100 to 1 / 10000.

[0111] Through the proposed risk factors, we can combine the real and potential risks of the above-mentioned pipeline network to perform multi-dimensional integrated quantitative calculations, and use the size of the risk factors to comprehensively reflect the potential risks of the pipeline network location. Different from the traditional abnormal data analysis based only on current data, the idea of ​​quantitative calculation of potential risks is added to the process of feedback on the potential risks of the pipeline network based on risk factors, which helps to improve the level of intelligent risk prediction.

[0112] Preferably, the user layer includes a user management unit;

[0113] The user management unit is used to manage users who are allowed to access the water supply and drainage digital twin model. After the user terminal completes the permission / identity authentication, the user terminal is allowed to obtain the real-time mirror data of the water supply and drainage network digital twin model.

[0114] Centralized management of system access rights through the user management unit helps improve system security.

[0115] It should be noted that each functional unit / module in each embodiment of the present invention may be integrated into one processing unit / module, or each unit / module may exist physically separately, or two or more units / modules may be integrated into one unit / module. The above-mentioned integrated unit / module may be implemented in the form of hardware or in the form of software functional unit / module.

[0116] Through the description of the above implementation modes, it can be clearly understood by those skilled in the art that the embodiments described herein can be implemented in hardware, software, firmware, middleware, code or any appropriate combination thereof. For hardware implementation, the processor can be implemented in one or more of the following units: an application specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field programmable gate array (FPGA), a processor, a controller, a microcontroller, a microprocessor, other electronic units designed to implement the functions described herein or a combination thereof. For software implementation, part or all of the processes of the embodiment can be completed by instructing the relevant hardware through a computer program. When implemented, the above program can be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein the communication media include any medium that facilitates the transmission of computer programs from one place to another. The storage medium can be any available medium that can be accessed by a computer. Computer-readable media can include, but are not limited to, RAM, ROM, EEPROM, CD-ROM or other optical disk storage, disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer.

[0117] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention, rather than to limit the scope of protection of the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should analyze that the technical solution of the present invention can be modified or replaced by equivalents without departing from the essence and scope of the technical solution of the present invention.

Claims

1. A water supply and drainage monitoring and management system based on digital twin, characterized in that: It includes physical layer, transmission layer, digital twin platform and user layer; among them, The physical layer includes a distributed sensor network arranged in the water supply and drainage network, wherein the distributed sensor network includes an in-pipe monitoring node arranged inside the water supply and drainage network, and an out-pipe monitoring node arranged in the periphery of the water supply and drainage network, and the monitoring data of the water supply and drainage network is collected through the in-pipe monitoring node and the out-pipe monitoring node, wherein the monitoring data includes flow data, image data and vibration signal data; The transport layer is used to transmit the monitoring data collected by the physical layer to the digital twin platform in real time through the wireless communication network; The digital twin platform is used to integrate the received monitoring data into the digital twin model of the water supply and drainage network built based on GIS, and update the status of the corresponding elements in the digital twin model, where the elements include the in-pipe nodes and out-pipe nodes corresponding to the real monitoring nodes; the potential risks of each location of the water supply and drainage network are predicted based on the monitoring data corresponding to the in-pipe nodes and out-pipe nodes, and the risk prediction results are obtained, including: According to the flow data of the upstream and downstream nodes in the pipes at the selected position in the water supply and drainage network, the flow change at the selected position is obtained; according to the obtained image data, image analysis and processing are performed to identify the crack characteristics of the inner wall of the pipe network, and further identify the length of the crack according to the crack characteristics; according to the obtained vibration signal data, the impact force on the pipe network position is analyzed; according to the obtained flow change, the identified crack length and the impact force, the potential risk factors of each position in the water supply and drainage network are comprehensively calculated; and the calculated potential risk factors are compared with the preset standard risk factors to obtain risk prediction results; The user layer is used for user terminals to obtain mirror data of the digital twin model of the water supply and drainage network and to display it visually; The digital twin platform includes a model building unit, a data integration unit, a risk analysis unit, and a visualization unit; The model building unit is used to construct a three-dimensional digital twin model of the water supply and drainage network based on the real GIS data of the water supply and drainage network, wherein the digital twin model has corresponding in-pipe nodes and out-pipe nodes at corresponding positions in the model according to the distribution positions of the real monitoring nodes; The data integration unit is used to complete the corresponding API interface settings for each in-pipe node and out-pipe node, and map the acquired monitoring data to the corresponding in-pipe node and out-pipe node; The risk analysis unit is used to conduct joint potential risk prediction analysis for each location of the water supply and drainage network based on the monitoring data of the corresponding nodes inside and outside the pipe, obtain the corresponding risk prediction results, and further integrate the obtained risk prediction results into the digital twin model; The visualization unit is used to generate visualization data based on the real-time monitoring data and risk analysis results of the digital twin model; The risk analysis unit includes a flow analysis unit, an image analysis unit, a vibration analysis unit and a quantitative analysis unit; The flow analysis unit is used to obtain the flow change at the selected location in the water supply and drainage network according to the flow data of the upstream and downstream in-pipe nodes at the selected location; The image analysis unit is used to perform image analysis processing based on the acquired image data, identify crack characteristics on the inner wall of the pipe network, and further identify the length of the crack based on the crack characteristics; The vibration analysis unit is used to analyze the impact force on the pipe network location based on the acquired vibration signal data; The quantitative analysis unit is used to comprehensively calculate the potential risk factors of each location of the water supply and drainage network based on the acquired flow change, the identified crack length and the impact force; and compare the calculated potential risk factors with the preset standard risk factors to obtain the risk prediction results; Wherein, the image analysis unit includes an image enhancement unit; The image enhancement unit is used to perform enhancement processing on the image data of the pipe network location before image analysis processing, specifically including: According to the obtained pipe network inner wall image, the gray value h(x,y) of the pixel is obtained according to the RGB information of each pixel in the image; The image is divided into M sub-image blocks of the same size. For each sub-image block, the gray level co-occurrence matrix is ​​obtained based on the gray level information of each pixel in the sub-image block. The texture feature factor of the sub-image block is calculated according to the obtained gray level co-occurrence matrix. The texture feature factor calculation function used is: In the formula, TCP m represents the texture feature factor of the m-th sub-image block, Po(i,j) represents the probability of gray level i and gray level j appearing together in the gray level co-occurrence matrix based on the m-th sub-image block, wherein i, j are variables in the summation function, and their value ranges are i=1,2,…,k; j=1,2,…,k; k represents the total number of gray levels in the gray level co-occurrence matrix, and γ represents a preset adjustment parameter; The inner wall image of the pipe network is further converted from the RGB color space to the Lab color space, and the brightness layer TL, color layer Ta and color layer Tb of the image are obtained respectively; the brightness component value L(x, y) of each pixel is obtained according to the brightness layer TL; Based on the brightness component value of each pixel, a Laplace filter with a size of 3×3 is used to perform convolution processing on each pixel to obtain the Laplace eigenvalue LH(x,y) of each pixel; Based on the Laplace eigenvalue and brightness component value of each pixel point, the brightness characteristic factor of each sub-image block is calculated, wherein the brightness characteristic factor calculation function used is: Where LCI m represents the brightness characteristic factor of the mth sub-image block, Indicates that the pixel (a, b) is the pixel in the mth sub-image, LH(a, b) represents the Laplace eigenvalue of the pixel (a, b), L(a, b) represents the brightness component value of the pixel (a, b), meanL m Represents the average brightness component value of each pixel in the mth sub-image, σL m represents the standard deviation of the brightness component value of each pixel in the mth sub-image; θ represents the preset adjustment factor; The brightness adjustment process is performed on each sub-image block, wherein the brightness adjustment process function used is: Where L'(x,y) represents the brightness component value of the pixel (x,y) after brightness adjustment, where the pixel (x,y) is the pixel in the mth sub-image block, TCP m represents the texture feature factor of the mth sub-image block, TCPth represents the preset texture feature threshold, meanL 3×3 (x, y) represents the average brightness component value of each pixel in the 3×3 range centered on the pixel, L(x, y) represents the brightness component value of the pixel (x, y), β represents the sensitivity adjustment factor, α represents the adjustment amplitude control factor, LCI m represents the brightness characteristic factor of the m-th sub-image block, Lbs represents the preset standard brightness component value, and mean{·} represents the averaging function; After the brightness adjustment processing of each sub-image block is completed respectively, the brightness layer TL' after brightness adjustment is obtained, and reconstruction is performed based on the brightness layer TL' after brightness adjustment to obtain the enhanced pipe network inner wall image.

2. According to the digital twin-based water supply and drainage monitoring and management system of claim 1, it is characterized in that: In the distributed sensor network at the physical layer, the monitoring nodes inside the pipe include flow sensors and image sensors; the monitoring nodes outside the pipe include vibration sensors; Flow sensors are used to collect flow data of water supply and drainage network segments; The image sensor is used to collect image data inside the water supply and drainage network; Vibration sensors are used to collect vibration signal data from the external environment of the water supply and drainage network.

3. According to the digital twin-based water supply and drainage monitoring and management system of claim 1, it is characterized in that: The transport layer includes a gateway unit and a wireless transmission unit; The gateway unit is used to collect monitoring data collected by various sensors in the physical layer; The wireless transmission unit is used to transmit monitoring data to the digital twin platform via a wireless network.

4. According to the digital twin-based water supply and drainage monitoring and management system of claim 1, it is characterized in that: The traffic analysis unit includes: According to the selected location in the water supply and drainage network, one or more nodes in the pipe are retrieved upstream to obtain the total input flow Q of the selected location. j-in ; Further obtain the total output flow Q of the selected location to the downstream j-out ; Calculate the flow change ΔQ at the selected location j =Q j-in -Q j-out .

5. The water supply and drainage monitoring and management system based on digital twin according to claim 1 is characterized in that: The image analysis unit includes: According to the image data Pic obtained at the pipe network location j j Perform enhancement processing to obtain an enhanced image of the inner wall of the pipe network; According to the acquired pipeline inner wall image, the deep learning model based on YoloV5 is used to identify cracks in the pipeline inner wall image and identify the crack features existing in the pipeline inner wall; According to the obtained crack features, combined with the mapping relationship between pixels and actual length, the length of the crack feature is calculated to obtain the length of the crack Cq j .

6. A water supply and drainage monitoring and management system based on digital twin according to claim 1, characterized in that: The vibration analysis unit includes: According to the vibration signal data sigZ obtained within a time period i (t), extract the vibration acceleration Az at each moment i (t) and vibration signal intensity Vz i (t)=|sigZ i (t)| 2 as the vibration characteristics of the corresponding nodes outside the tube; The impact force on the pipe network location is calculated based on the obtained vibration characteristics, and the impact force calculation function used is: Among them, Fz j (t) represents the magnitude of the impact force on the pipe network location j, Do j represents the pipe diameter at location j in the pipe network, Ro j represents the pipe thickness at location j in the pipe network, ρ o Indicates the material coefficient of the pipe network, Az i-max (t) represents the maximum vibration acceleration monitored by the node outside the pipe during the time period; γ represents the preset transfer coefficient, Vz i-mean (t) represents the average vibration signal strength in the current time period, λ represents the preset attenuation factor, and d ij It represents the distance from the pipe network location j to the monitoring node i outside the pipe, which is obtained based on the real GIS data.

7. A water supply and drainage monitoring and management system based on digital twin according to claim 1, characterized in that: The quantitative analysis unit specifically includes: Based on the acquired flow change, identified crack length and impact force, the potential risk factors of each location in the water supply and drainage network are comprehensively calculated. The risk factor calculation function used is: In the formula, R j represents the potential risk factor of the pipeline network location j, ΔQ j represents the flow change at the pipe network location j, ptQ j represents the historical flow change at the pipe network location j, where ptQ j =α×ptQ j-min +ptQ j-meant , where ptQ j-min Indicates the historical minimum flow change at the pipe network location j within a period of time, ptQ j-meant represents the historical average flow change at the pipe network location j within a period of time, α represents the preset weight factor, and L j Indicates the length of the pipe section at the pipe network location j, Cq j represents the crack length at the pipe network location j, Fz j represents the impact force on the pipe network position j; β1, β2 and β3 represent the preset weight adjustment factors respectively; According to the potential risk factor R j Compared with the preset standard risk factor RT, when R j >RT, the risk prediction result is abnormal. j When ≤RT, the risk prediction result is normal.

8. The water supply and drainage monitoring and management system based on digital twin according to claim 1 is characterized in that: The user layer includes a user management unit; The user management unit is used to manage users who are allowed to access the water supply and drainage digital twin model. After the user terminal completes the permission / identity authentication, the user terminal is allowed to obtain the real-time mirror data of the water supply and drainage network digital twin model.

Images