Remote condition monitoring method and system based on smart water meter

Abstract

The embodiment of the application discloses a remote state monitoring method and system based on a smart water meter, which comprises the following steps: obtaining a pipeline BIM model data of a smart water meter pipeline network of a target building, and synchronously collecting a real-time monitoring data set of a plurality of sensing nodes in the smart water meter pipeline network; performing abnormal state identification processing on the real-time monitoring data set to generate an abnormal state distribution atlas matched with the spatial position of the pipeline BIM model data; based on the abnormal state distribution atlas and a preset pipeline topology constraint condition, performing dynamic visual reconstruction processing on the pipeline BIM model data to generate a visual output interface with an abnormal state mark; associating and mapping the visual output interface with the real-time monitoring data set to generate a state monitoring feedback strategy, and triggering a maintenance response operation of the smart water meter pipeline network through a remote service port.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the technical field of data processing, in particular to a remote state monitoring method and system based on a smart water meter. BACKGROUND

[0002] With the development of intelligent buildings, smart water meters are increasingly widely used in building water supply pipe networks. The state monitoring of smart water meters has become a key link to ensure the normal and efficient management of water resources in buildings.

[0003] In actual application, the existing smart water meter state monitoring technology has many shortcomings. On the one hand, although the BIM model can display the layout of the smart water meter pipe network, it cannot effectively combine real-time monitoring data and quickly find abnormal conditions of the smart water meter and its pipe. On the other hand, for the collected monitoring data, there is a lack of in-depth analysis and abnormal precise positioning capability, and abnormal information cannot be intuitively and accurately presented on the BIM model. Therefore, the existing smart water meter state monitoring technology has the problem of low intelligence and automation level. SUMMARY

[0004] Embodiments of the present application provide a remote state monitoring method and system based on a smart water meter, which is used to realize the deep fusion of smart water meter monitoring data and BIM model, thereby accurately positioning abnormalities and realizing dynamic visual display.

[0005] In a first aspect, embodiments of the present application provide a remote state monitoring method based on a smart water meter, applied to a remote state monitoring system, the method comprising: acquiring pipe BIM model data of a smart water meter pipe network of a target building, and synchronously collecting a real-time monitoring data set of a plurality of sensing nodes in the smart water meter pipe network; performing abnormal state identification processing on the real-time monitoring data set to generate an abnormal state distribution map matching the spatial position of the pipe BIM model data; based on the abnormal state distribution map and a preset pipe topology constraint condition, performing dynamic visual reconstruction processing on the pipe BIM model data to generate a visual output interface with an abnormal state marker; associating and mapping the visual output interface with the real-time monitoring data set to generate a state monitoring feedback strategy, and triggering a maintenance response operation of the smart water meter pipe network through a remote service port.

[0006] In a second aspect, embodiments of the present application provide a remote state monitoring system, comprising:

[0007] a processor;

[0008] a storage device having a computer program stored thereon,

[0009] When the computer program is executed by the processor, the processor implements any of the remote state monitoring methods based on a smart water meter.

[0010] The embodiment of the present application provides a readable storage medium, and the readable storage medium stores programs or instructions, and the programs or instructions are executed by a processor to implement the steps of the remote state monitoring method based on a smart water meter.

[0011] The embodiment of the present application can realize deep fusion of smart water meter monitoring data and a BIM model, thereby accurately positioning an anomaly and realizing dynamic visual display. In detail, the embodiment of the present application can accurately grasp a pipeline state by acquiring pipeline BIM model data and a real-time monitoring data set of a target building smart water meter pipeline network; an anomaly state distribution atlas is generated by performing anomaly state identification processing on the monitoring data, and an abnormal position can be clearly presented; a visual output interface with an anomaly mark is generated by performing dynamic visual reconstruction on the BIM model data based on the atlas and a topological constraint condition, so that an operation and maintenance personnel can intuitively understand the anomaly; the visual interface is associated and mapped with the monitoring data, a feedback strategy is generated and a maintenance response operation is triggered, efficient monitoring and timely processing of the smart water meter pipeline network state can be realized, and the intelligentization and automation level of the entire monitoring process is greatly improved, and operation and maintenance efficiency and accuracy are improved. BRIEF DESCRIPTION OF DRAWINGS

[0012] Figure 1 A flowchart of a remote state monitoring method based on a smart water meter provided by the embodiment of the present application.

[0013] Figure 2 A schematic diagram of the basic structure of a remote state monitoring system provided by the embodiment of the present application. DETAILED DESCRIPTION

[0014] In order to make the above-mentioned purposes, characteristics and advantages of the present application more obvious and easy to understand, the embodiment of the present application will be further described in detail below in combination with the drawings and specific embodiments.

[0015] Referring to Figure 1 The figure is a flowchart of a remote state monitoring method based on a smart water meter provided by the embodiment of the present application, and the method can be applied to a remote state monitoring system. As shown in Figure 1 The method can include steps 110-140.

[0016] Step 110: acquiring pipeline BIM model data of a smart water meter pipeline network of a target building, and synchronously collecting a real-time monitoring data set of a plurality of sensing nodes in the smart water meter pipeline network.

[0017] In this embodiment, first, the pipe BIM model data of the intelligent water meter pipeline network of the target building is acquired, which contains detailed information of the entire pipeline system, such as the connection mode of the pipeline, the spatial position, and the attributes of each device, etc. At the same time, the real-time monitoring data set of multiple sensing nodes is synchronously collected, which are distributed at different positions of the pipeline network and monitor various data in real time.

[0018] In an optional embodiment, the pipe BIM model data of the intelligent water meter pipeline network of the target building is acquired, and the real-time monitoring data set of multiple sensing nodes in the intelligent water meter pipeline network is synchronously collected, comprising:

[0019] Step 111: calling the BIM model database of the target building, extracting the spatial topology structure data and device attribute data constituting the pipe BIM model data; the spatial topology structure data is used to describe the connection relationship and spatial geometric layout of the pipeline, and the device attribute data contains the water meter installation coordinates and configuration parameters associated with the sensing nodes.

[0020] In this scenario, when the pipe BIM model data needs to be acquired, the system calls the BIM model database, and extracts the spatial topology structure data from the database. The spatial topology structure data shows how each pipeline is connected and its layout in space. For example, pipeline A and pipeline B are connected through a tee pipe at a certain spatial position, and their spatial coordinates are also accurately recorded. The device attribute data contains the water meter installation coordinates, such as the water meter M installed at coordinates (X1, Y1, Z1). At the same time, the configuration parameters are also extracted, which determine the working mode of the sensing nodes, such as the sampling frequency of the sensing nodes, the data transmission mode, etc.

[0021] Step 112: sending an activation instruction to each sensing node according to the configuration parameters, and collecting the water pressure monitoring data, flow fluctuation data and device running state data of each sensing node at a set period.

[0022] Based on the above-mentioned configuration parameters, the system sends an activation instruction to each sensing node. For example, there are N sensing nodes, and each sensing node has corresponding configuration parameters. For example, the configuration parameters of sensing node S1 stipulate that its sampling frequency is to collect data once every 5 minutes. When the activation instruction is sent, the sensing node starts to work according to the set period. In this process, the sensing node will collect water pressure monitoring data, such as the water pressure value P1 collected by sensing node S2 at a time, and flow fluctuation data, such as the flow value Q1 collected by the sensing node at the same time. In addition, device running state data can also be collected, such as whether the device is running normally, whether there is a fault alarm, etc. These data will be updated over time, providing dynamic data support for subsequent analysis.

[0023] Step 113: Perform spatio-temporal alignment processing on the water pressure monitoring data, the flow fluctuation data, and the equipment operation state data based on the water meter installation coordinates and the spatial topology level of the pipeline BIM model to generate a hierarchical real-time monitoring data set; wherein the topology structure of the hierarchical real-time monitoring data set and the pipe network level of the pipeline BIM model form a mapping relationship, the spatial topology level is a BIM model hierarchy structure divided according to building function areas, and the pipe network level is a physical hierarchy structure of pipelines divided based on the main and secondary relationship of water supply.

[0024] In order to better analyze and utilize the collected data, spatio-temporal alignment processing is needed. Taking the water meter installation coordinates as the reference, the water pressure, flow, and equipment operation state data collected by different sensor nodes are corresponded to the spatial topology level of the pipeline BIM model. For example, the data collected by the sensor nodes located in a function area (such as the kitchen area) of the target building will correspond to the spatial position of the function area in the BIM model. At the same time, considering the pipe network level, the data is arranged according to the physical hierarchy structure of the pipelines according to the main and secondary relationship of water supply. For example, the pipe network is divided into different levels such as the main water supply pipeline layer and the branch water supply pipeline layer, and then the data is arranged according to this level relationship. After the above processing, the generated hierarchical real-time monitoring data set has a clear structure, and its topology structure and the pipe network level of the pipeline BIM model form a one-to-one mapping relationship.

[0025] Step 120: Perform abnormal state identification processing on the real-time monitoring data set to generate an abnormal state distribution atlas matched with the spatial position of the pipeline BIM model data.

[0026] After obtaining the hierarchical real-time monitoring data set, the next step is to analyze these data, identify the abnormal states therein, and generate an abnormal state distribution atlas matched with the spatial position of the pipeline BIM model data. This atlas can intuitively show which positions in the pipeline network have abnormal conditions, as well as the type and degree of the abnormality.

[0027] In an optional embodiment, the abnormal state identification processing on the real-time monitoring data set to generate an abnormal state distribution atlas matched with the spatial position of the pipeline BIM model data includes:

[0028] Step 121: Perform mutation point detection processing on the water pressure monitoring data in the hierarchical real-time monitoring data set to identify water pressure abnormal areas exceeding a preset fluctuation threshold, and extract the duration and fluctuation amplitude of the water pressure abnormal areas.

[0029] For water pressure monitoring data, first, the mutation point detection processing is performed. For example, the water pressure data collected within a period of time is divided according to a certain time interval, for example, water pressure data is collected every 1 minute, and these data are divided into a plurality of data segments. Then, first-order difference calculation is performed on each data segment, that is, the difference value of water pressure values of adjacent two time points is calculated to obtain a water pressure change rate sequence. Then, sliding window standard deviation analysis is performed on the change rate sequence, for example, a sliding window size of 5 minutes is set, and the standard deviation of the data in the window is calculated. When the standard deviation exceeds the dynamic threshold, a mutation time point is found. According to the mutation time point, the corresponding water pressure monitoring sub-data is extracted. For example, the water pressure data within 10 minutes before and after the mutation time point is taken as the sub-data. Then, the offset percentage of the sub-data and the historical average data is calculated, for example, the historical average water pressure is P0, the average water pressure of the sub-data is P2, and the offset percentage = (P2-P0) / P0x100%. If the offset percentage exceeds the preset fluctuation threshold, the region corresponding to the sub-data is marked as a water pressure abnormal region, and the duration and fluctuation amplitude of the abnormal region are recorded.

[0030] In a preferred embodiment, the mutation point detection processing of the water pressure monitoring data in the hierarchical real-time monitoring data set, identifying the water pressure abnormal region exceeding the preset fluctuation threshold, comprises:

[0031] Step 1210: dividing the water pressure monitoring data into continuous water pressure data segments according to a preset time window, and performing first-order difference calculation on each water pressure data segment to obtain a water pressure change rate sequence; performing sliding window standard deviation analysis on the water pressure change rate sequence to identify a mutation time point whose standard deviation exceeds a dynamic threshold; extracting corresponding water pressure monitoring sub-data according to the mutation time point, and calculating the offset percentage of the water pressure monitoring sub-data and the historical average data; if the offset percentage exceeds the preset fluctuation threshold, marking the corresponding region of the water pressure monitoring sub-data as a water pressure abnormal region.

[0032] In actual operation, the preset time window can be adjusted according to actual conditions, such as being set to 3 minutes. The water pressure monitoring data is divided into continuous data segments according to the window. First-order difference calculation is performed on each data segment to obtain a water pressure change rate sequence. For example, for the data segment [P3, P4, P5], first-order difference calculation obtains the change rate sequence [P4-P3, P5-P4]. Then, sliding window standard deviation analysis is performed on the change rate sequence, for example, the sliding window size is 4 data points. When the calculated standard deviation exceeds the dynamic threshold, a mutation time point is determined. According to the mutation time point, the corresponding water pressure monitoring sub-data is extracted, such as the water pressure data within 15 minutes before and after the mutation time point. The offset percentage of the sub-data from the historical average data is calculated. If the offset percentage exceeds the preset fluctuation threshold, such as the preset fluctuation threshold is 15%, when the calculated offset percentage is greater than 15%, the region corresponding to the sub-data is marked as a water pressure abnormal region.

[0033] Step 122: Perform trend deviation analysis processing on the flow fluctuation data to determine the deviation degree of the flow trend from the preset flow model, and mark the flow abnormal region whose deviation degree exceeds the dynamic threshold.

[0034] In order to analyze the flow fluctuation data, the preset flow reference curve of each branch pipeline in the intelligent water meter pipeline network needs to be obtained first. These curves are generated based on historical data and contain the expected flow range at different times. For example, the expected flow range of a branch pipeline from 8 am to 10 am is [Q3, Q4]. Then, the flow fluctuation data is identified and classified by branch pipeline, and the flow data of each branch pipeline is processed separately. Based on the preset flow reference curve, the classified flow fluctuation data is analyzed for trend deviation, and a real-time flow trend line for each branch pipeline is generated through some data analysis methods, such as curve fitting. For example, for branch pipeline L1, a real-time flow trend line is obtained by fitting the collected flow data. Then, the deviation degree of this real-time flow trend line from the corresponding preset flow reference curve is calculated, including the deviation duration and the deviation cumulative amount. For example, from a certain time point, the real-time flow trend line is always higher than the preset flow reference curve, lasting for 30 minutes, which is the deviation duration; within these 30 minutes, the cumulative difference between the real-time flow and the preset flow is the deviation cumulative amount. If the deviation duration exceeds the first dynamic threshold, such as the first dynamic threshold is 20 minutes, and the deviation cumulative amount exceeds the second dynamic threshold, such as the second dynamic threshold is a certain total flow difference, then the spatial coordinates of the corresponding branch pipeline can be mapped to the pipeline BIM model data, and the flow abnormal region can be marked according to the mapped spatial coordinates.

[0035] In a preferred embodiment, the trend deviation analysis is performed on the flow fluctuation data to determine the deviation degree of the flow trend from the preset flow model, and the flow anomaly region with a deviation degree exceeding a dynamic threshold is marked.

[0036] Step 1221: Obtain the preset flow reference curve of each branch pipeline in the intelligent water meter pipeline network, wherein the preset flow reference curve comprises a time-periodic expected flow range generated based on historical data.

[0037] Optionally, the system extracts the preset flow reference curve of each branch pipeline from the historical data repository, which is obtained by statistical analysis of the flow data of each time period in the past. For example, by statistical analysis of the flow data from 9:00 to 11:00 am every day in the past month, the expected flow range of a branch pipeline in this time period is obtained, and the preset flow reference curve of this time period is generated. Different branch pipelines have different preset flow reference curves because their water supply demands and usage may be different.

[0038] Step 1222: Identify and classify the flow fluctuation data by branch pipeline, and perform trend deviation analysis on the classified flow fluctuation data based on the preset flow reference curve to generate a real-time flow trend line of each branch pipeline.

[0039] After collecting the flow fluctuation data, first, classify the data according to the branch pipeline. For example, all the flow data related to branch pipeline L2 are classified into one category. Then, for each category of data, perform trend deviation analysis based on the preset flow reference curve. By using some data analysis algorithms, such as the least square method to fit the curve, the classified flow data is fitted to generate a real-time flow trend line of each branch pipeline, which can reflect the current flow trend. By comparing with the preset flow reference curve, it can be seen whether the flow deviates abnormally.

[0040] Step 1223: Calculate the deviation degree of the real-time flow trend line from the corresponding preset flow reference curve, wherein the deviation degree comprises a deviation duration and a deviation cumulative amount.

[0041] In the embodiments of the present application, calculating the deviation degree of the real-time flow trend line from the preset flow reference curve is a key step. For the deviation duration, the time point at which the real-time flow trend line deviates from the preset flow reference curve is counted, and the time when the real-time flow trend line returns to the preset flow reference curve range or exceeds the set observation time is the end of the counting. The deviation duration is the time from the start point to the end point. For the deviation cumulative amount, the sum of the flow difference between the real-time flow trend line and the preset flow reference curve at each time point within the deviation duration is the deviation cumulative amount. For example, in a certain period of time, the flow values corresponding to the real-time flow trend line are Q5, Q6, and Q7, and the flow values corresponding to the preset flow reference curve are Q8, Q9, and Q10. Then the deviation cumulative amount is (Q5-Q8) + (Q6-Q9) + (Q7-Q10).

[0042] Step 1224: If the deviation duration exceeds the first dynamic threshold and the deviation cumulative amount exceeds the second dynamic threshold, the spatial coordinates of the corresponding branch pipeline are mapped to the pipeline BIM model data, and the flow anomaly area is marked according to the mapped spatial coordinates.

[0043] When the calculated deviation duration exceeds the first dynamic threshold, such as the first dynamic threshold being set to 15 minutes, and the deviation cumulative amount exceeds the second dynamic threshold, for example, the second dynamic threshold being a specific total flow difference, it can be judged that the branch pipeline has a flow anomaly. At this time, the spatial coordinates of the branch pipeline are mapped to the pipeline BIM model data, the corresponding position in the BIM model is found, and the flow anomaly area is marked according to the mapped spatial coordinates, so that it can be directly seen in the model which positions have flow anomalies.

[0044] Step 123: Map the spatial coordinates of the water pressure anomaly area and the flow anomaly area to the corresponding positions of the pipeline BIM model data to generate an initial abnormal state distribution atlas containing abnormal type labels and abnormal levels.

[0045] After determining the water pressure anomaly area and the flow anomaly area, the spatial coordinates of these areas are mapped to the corresponding positions in the pipeline BIM model data. For example, the spatial coordinates of the water pressure anomaly area are (X5, Y5, Z5), and the corresponding position in the pipeline BIM model is found, and the abnormal type label such as water pressure anomaly is marked, and the abnormal level is determined according to the severity of the anomaly, such as mild, moderate, and severe. The same operation is performed for the flow anomaly area, and the spatial coordinates are mapped to the model, and the flow anomaly type label and the corresponding abnormal level are marked. In this way, an initial abnormal state distribution atlas containing all abnormal information is generated, which can directly show the abnormal conditions of different positions in the pipeline network.

[0046] Step 124: Cross-validation processing is performed on the initial abnormal state distribution map based on fault alarm information in the device operation state data to eliminate false alarm abnormal areas and correct abnormal levels, and the abnormal state distribution map is generated.

[0047] Among them, the device operation state data may contain some fault alarm information, which can be used to further verify the accuracy of the initial abnormal state distribution map. For example, the device operation state data shows that a certain sensor node has failed, causing the data collected by it to be inaccurate. Therefore, in the initial abnormal state distribution map, if the area where the sensor node is located is marked as an abnormal area, it needs to be judged according to the fault alarm information. If it is determined that it is a false alarm caused by the failure of the sensor node, the abnormal area will be eliminated. At the same time, for other abnormal areas, according to more information provided by the device operation state data, such as the fault type of the device, the influence range, etc., the abnormal level is corrected. After the above cross-validation processing, a more accurate abnormal state distribution map is generated.

[0048] Step 130: Based on the abnormal state distribution map and the preset pipeline topology constraint condition, dynamic visual reconstruction processing is performed on the pipeline BIM model data to generate a visual output interface with abnormal state markers.

[0049] In order to more intuitively show the abnormal situation of the pipeline network, dynamic visual reconstruction processing needs to be performed on the pipeline BIM model data to generate a visual output interface with abnormal state markers, which can allow users to clearly see which positions have abnormalities and the specific situation of the abnormalities.

[0050] In an optional embodiment, the dynamic visual reconstruction processing of the pipeline BIM model data based on the abnormal state distribution map and the preset pipeline topology constraint condition to generate a visual output interface with abnormal state markers includes:

[0051] Step 131: Extracting pipeline connection relationship data and valve control node data from the pipeline BIM model data to construct a topological constraint relationship network of the intelligent water meter pipeline network.

[0052] From the pipeline BIM model data, the system will extract pipeline connection relationship data, which describes in detail the connection method of each pipeline with other pipelines, such as pipeline A and pipeline B connected through an elbow. At the same time, valve control node data is extracted, and the valve plays an important control role in the pipeline system. The valve control node data contains information such as the position and switch state of the valve. Using these data, a topological constraint relationship network of the intelligent water meter pipeline network is constructed, which shows the structure of the entire pipeline system and the relationship between each component, providing a basis for subsequent analysis and visualization processing.

[0053] Step 132: Based on the abnormality type label in the abnormality distribution map, assign a corresponding color code and dynamic flashing frequency to each abnormal region in the abnormality distribution map.

[0054] Optionally, based on the different anomaly type labels in the anomaly distribution map, a unique color code and dynamic flashing frequency are assigned to each anomaly region. For example, for anomaly regions in water pressure, red is assigned as the color code, and a dynamic flashing frequency of 2 flashes per second is set; for anomaly regions in flow rate, yellow is assigned as the color code, and it flashes once per second. In this way, different colors and flashing frequencies allow users to quickly distinguish different types of anomaly regions on the visualization interface.

[0055] Step 133: Superimpose the color code and dynamic flashing frequency onto the corresponding spatial location of the pipeline BIM model data to generate the first visual topology structure.

[0056] For example, in a pipeline BIM model, a location is marked as an area of ​​abnormal water pressure, and the attributes of red color and flashing twice per second are overlaid on that location. In this way, a first visual topology structure is generated, which can intuitively show the location of the abnormal area in the pipeline network and the type of abnormality.

[0057] Step 134: Based on the topological constraint relationship network, perform influence range diffusion simulation processing on the abnormal regions in the first visualized topological structure to generate a second visualized topological structure containing potential influence region markers.

[0058] Using the previously constructed topological constraint network, the impact range diffusion of anomalous regions in the first visualized topology is simulated. For example, when a pipe experiences abnormal water pressure, the potential impact of this anomaly on other connected pipes is analyzed based on pipe connections. By simulating the propagation path of water flow and pressure changes, potential affected areas are identified and marked in the visualized structure. The resulting second visualized topology not only shows the directly affected areas but also the potentially affected areas.

[0059] Step 135: Perform layer fusion processing on the second visualization topology and the device operation status data in the hierarchical real-time monitoring data set to generate the visualization output interface.

[0060] For example, the device running state data shows that a certain valve is in a closed state. By fusing this information with the second visualized topology, the abnormal area, the potential impact area, and the running state of the device are displayed on the visualized output interface at the same time. The visualized output interface generated in this way can provide more comprehensive information and help users better understand the running state of the pipeline network.

[0061] Step 140: associating the visualized output interface with the real-time monitoring data set to generate a state monitoring feedback strategy and triggering a maintenance response operation of the intelligent water meter pipeline network through a remote service port.

[0062] In order to take corresponding measures according to the monitored abnormal situation, it is necessary to associate the visualized output interface with the real-time monitoring data set to generate a state monitoring feedback strategy and trigger a maintenance response operation.

[0063] In an optional embodiment, the association of the visualized output interface with the real-time monitoring data set to generate a state monitoring feedback strategy comprises:

[0064] Step 141: extracting a spatial coordinate set and an abnormal type label of the abnormal area from the visualized output interface to generate an abnormal state feature vector.

[0065] After generating the visualized output interface, the system will extract key information from the interface. For each abnormal area, obtain its spatial coordinate set, for example, the spatial coordinates of a certain water pressure abnormal area are { (Xa, Ya, Za), (Xb, Yb, Zb), …}. This set of coordinates accurately locates the position of the abnormal area in the pipeline BIM model. At the same time, the abnormal type label corresponding to the abnormal area is extracted, such as water pressure abnormality, flow abnormality, etc. These spatial coordinate sets and abnormal type labels are combined together to form a multi-dimensional abnormal state feature vector, which can comprehensively describe the characteristics of the abnormal area and provide basic data for subsequent analysis and decision-making. For example, the abnormal state feature vector can be represented as [water pressure abnormality, (Xa, Ya, Za), (Xb, Yb, Zb), …], which includes the abnormal type and the location information of the abnormal occurrence.

[0066] Step 142: according to the abnormal state feature vector, matching the corresponding maintenance priority scoring rule from the pre-set strategy library, and calculating the maintenance priority score of each abnormal area extracted from the visualized output interface.

[0067] The preset strategy library stores various maintenance priority scoring rules for different abnormal situations. After obtaining the abnormal state feature vector, the system matches it with the rules in the strategy library. For example, if the abnormal type is water pressure abnormality and the abnormal area is near the main water supply pipeline, according to the rules in the strategy library, a set of set scoring standards will be corresponded. This set of standards may consider factors such as the severity of the abnormality (such as abnormality level), the influence range of the abnormal area on the entire pipeline system, potential risks, etc. For example, for a water pressure abnormal area, according to its abnormal level is severe, and it is located near the main water supply pipeline, it has a greater impact on the system. According to the scoring rules, through the set calculation method (such as assigning different weights to different factors and then performing weighted calculation), the maintenance priority score of the abnormal area is obtained. The exemplary calculation formula is: maintenance priority score = abnormal level weight × abnormal level + position weight × position influence + potential risk weight × potential risk degree. Among them, the abnormal level weight, the position weight, and the potential risk weight are pre-set according to the actual situation. Through this calculation method, the maintenance priority score of each abnormal area can be accurately calculated.

[0068] Step 143: Based on the maintenance priority score and the pipeline connection tightness in the topological constraint relationship network, a state monitoring feedback strategy containing maintenance sequence and resource allocation scheme is generated; the maintenance sequence is a sequence sorted according to the priority score and the pipeline connection tightness.

[0069] After obtaining the maintenance priority score of each abnormal area, the state monitoring feedback strategy is generated in combination with the pipeline connection tightness information in the topological constraint relationship network constructed before. The pipeline connection tightness represents the tightness of the connection between different pipelines, for example, the connection tightness of directly connected pipelines is higher, and the connection tightness of pipelines connected through multiple pipe fittings is relatively lower. For the determination of the maintenance sequence, the maintenance priority score and the pipeline connection tightness are considered comprehensively. For example, the maintenance priority score of abnormal area ErrA is higher, but the connection tightness with other important pipelines is lower; while the maintenance priority score of abnormal area ErrB is slightly lower than that of abnormal area ErrA, but it is tightly connected with multiple key pipelines. In this case, the priority and connection tightness of the two are evaluated comprehensively according to a certain algorithm, and abnormal area ErrB may be processed first because it has a greater impact on the entire pipeline system. The resource allocation scheme is also formulated according to the maintenance priority score and the pipeline connection tightness. For areas with high priority and tight connection, more maintenance resources such as more maintenance personnel and more advanced detection equipment are allocated; for areas with lower priority and relatively loose connection, relatively less resources are allocated. Through the above method, a reasonable state monitoring feedback strategy containing maintenance sequence and resource allocation scheme can be generated, ensuring that the maintenance work can be carried out efficiently and orderly.

[0070] In yet another optional embodiment, the triggering of the maintenance response operation of the smart water pipeline network through the remote service port comprises:

[0071] Step 144: converting the state monitoring feedback strategy into a remote control instruction set, which includes valve opening and closing instructions, data acquisition frequency adjustment instructions, and maintenance work order generation instructions.

[0072] Optionally, after the state monitoring feedback strategy is determined, it needs to be converted into a remote control instruction set that can be recognized and executed by the control devices of the smart water pipeline network. For valve opening and closing instructions, according to the maintenance strategy, if the abnormality of a certain area needs to be solved by closing or opening a specified valve, the corresponding instruction will be generated. For example, if there is an abnormal water pressure in a certain section of pipeline, the valve connected to it needs to be closed to prevent the problem from getting worse, and the instruction will clearly indicate the identification of the valve and the closing operation requirements. The data acquisition frequency adjustment instruction is to adjust the data acquisition frequency of the sensing nodes according to the maintenance requirements. For example, for an area with abnormality, in order to more closely monitor the situation, the data acquisition frequency of the sensing nodes near the area may be increased, and the instruction will specify the identification of the sensing nodes to be adjusted and the new data acquisition frequency. The maintenance work order generation instruction will generate detailed maintenance work order information according to the maintenance sequence and resource allocation scheme, including the specific content of the maintenance task, the required resources, the responsible maintenance personnel, etc., so as to accurately convey these instructions to the relevant execution devices and personnel.

[0073] Step 145: sending the remote control instruction set to the edge control node of the smart water pipeline network through an encrypted communication link, instructing the edge control node to perform valve state adjustment operation.

[0074] To ensure the security and accuracy of the remote control instruction set during transmission, the encrypted communication link is used for sending. First, a unique identity identifier and a dynamic encryption key are assigned to each edge control node. Before sending the remote control instruction set, the instruction header verification code is generated according to the identity identifier, which is used to verify the source and integrity of the instruction. Then the instruction content is segmented and encrypted using the dynamic encryption key, which divides the instruction content into multiple parts and encrypts each part to increase the security of the data. For example, the instruction content is divided into three segments, which are encrypted using the dynamic encryption key to generate encrypted instruction segments. The instruction content after segmented encryption is combined with the instruction header verification code to generate an encrypted data packet, which contains complete instruction information and verification information. The encrypted data packet is sent to each edge control node through the multi-path transmission protocol, which can improve the reliability of data transmission. Even if a path fails, the data packet can reach the target node through other paths. When the edge control node receives the encrypted data packet, it will first verify the correctness of the instruction header verification code, and then use its own dynamic encryption key to decrypt the encrypted instruction content to obtain the remote control instruction set, and execute the valve state adjustment operation according to the valve opening and closing instruction in the instruction set.

[0075] In an exemplary embodiment, the establishment process of the encrypted communication link includes:

[0076] Step 1450: Assign a unique identity identifier and a dynamic encryption key to each edge control node; before sending the remote control instruction set, generate an instruction header verification code according to the identity identifier, and use the dynamic encryption key to segment and encrypt the instruction content; combine the instruction content after segmented encryption with the instruction header verification code to generate an encrypted data packet, and send it to each edge control node through the multi-path transmission protocol; receive the instruction confirmation signal returned by each target edge control node, and update the validity status of the dynamic encryption key according to the instruction confirmation signal.

[0077] In the system initialization phase, each edge control node is assigned a unique identity identifier, for example, edge control node E1 is assigned identifier ID1, E2 is assigned ID2, etc. At the same time, a dynamic encryption key is generated for each node, which is updated regularly to improve security. When a remote control instruction set needs to be sent, a command header verification code is generated according to the identity identifier of each edge control node. For example, for the instruction to be sent to edge control node E1, a verification code is generated by combining ID1 through a set algorithm (such as a hash algorithm). Then, the content of the remote control instruction set is segmented according to certain rules, for example, the instruction set content is a piece of text information, which is divided into three parts: the first part is the valve opening and closing instruction part, the second part is the data acquisition frequency adjustment instruction part, and the third part is the maintenance work order generation instruction part. Use the dynamic encryption key assigned to E1 to encrypt the three parts respectively to generate encrypted three-part content. Combine the three encrypted contents with the command header verification code to form an encrypted data packet. Through the multi-path transmission protocol, the encrypted data packet is sent to E1. After E1 receives the data packet, it first verifies the correctness of the command header verification code, and if the verification is passed, it uses its own dynamic encryption key to decrypt the encrypted content. After completing the operation, E1 will return an instruction confirmation signal, and the system will receive the signal and update the dynamic encryption key according to the signal content. For example, extend the use time of the key or mark it as used and prepare to update a new key.

[0078] Step 146: Synchronize pushing the maintenance work order generation instruction to the preset maintenance management platform, and receive the work order execution state data returned by the maintenance management platform; update the abnormal state marker in the visual output interface according to the work order execution state data, and generate a maintenance progress tracking layer.

[0079] While sending the remote control instruction set, the system will push maintenance work order generation instructions to the preset maintenance management platform, which is responsible for managing and allocating maintenance tasks. After receiving the instructions, it will generate detailed maintenance work orders according to the instruction content and assign them to the corresponding maintenance personnel. For example, the maintenance work order will clearly indicate that the maintenance personnel's task is to go to a specific location to repair the abnormal pipeline, and the required tools and materials will be listed in detail. The maintenance management platform will track the execution status of the work order in real time, and when the maintenance personnel starts the task, completes part of the task, or completes the task, the platform will update the execution status data of the work order. The system will receive these returned work order execution status data and update the abnormal state markers in the visual output interface according to the data content. If the maintenance task of a certain abnormal area has started, the abnormal state marker of that area will be updated to "in maintenance" in the visual output interface, and the progress of the completed part will be displayed; when the maintenance task is completed, the abnormal state marker will be updated to "repaired", and whether there are other potential abnormalities will be determined according to the new monitoring data. In this way, a maintenance progress tracking layer is generated, and users can intuitively see the maintenance progress of each abnormal area and understand the maintenance situation of the entire pipeline network on the visual output interface.

[0080] In a non-limiting embodiment, after triggering the maintenance response operation of the intelligent water meter pipeline network through the remote service port, it further includes: receiving a set of maintenance verification data returned by the edge control node, the set of maintenance verification data including water pressure calibration data and flow recovery data after the maintenance operation; comparing and analyzing the set of maintenance verification data with the abnormal state distribution atlas to determine the abnormal elimination state and the residual abnormal area; updating the abnormal state distribution atlas according to the abnormal elimination state, and adjusting the collection frequency of the set of hierarchical real-time monitoring data based on the spatial coordinates of the residual abnormal area; synchronizing the updated abnormal state distribution atlas and the adjusted collection frequency to the visual output interface to generate a maintenance effect feedback layer, and triggering an alarm state release instruction of the user terminal through the remote service port.

[0081] It can be understood that after the maintenance operation is completed, the edge control node will return a set of maintenance verification data. Among them, the water pressure calibration data reflects the recovery of the water pressure in the pipeline system after the maintenance operation, for example, the water pressure value at a certain location is restored from the abnormal ErrP1 to the normal range NorP2 after maintenance; the flow recovery data shows whether the flow has returned to normal, such as the flow is restored from the abnormal ErrQ1 to the expected NorQ2. Compare and analyze these sets of maintenance verification data with the abnormal state distribution atlas to see if the previously marked abnormal areas have returned to normal.

[0082] If both water pressure and flow rate return to normal range, the corresponding abnormal area is considered as abnormality eliminated; if there are still some areas with abnormal water pressure or flow rate, these areas are residual abnormal areas. According to the abnormality elimination state, the abnormal state distribution map is updated, and the areas where the abnormality has been eliminated are marked as normal state. For the residual abnormal areas, based on their spatial coordinates, the collection frequency of the hierarchical real-time monitoring data set is adjusted.

[0083] For example, if a residual abnormal area is located on a key branch pipeline, in order to more closely monitor its situation, the data collection frequency of the sensor nodes near this area is doubled. The updated abnormal state distribution map and the adjusted collection frequency are synchronized to the visualization output interface to generate a maintenance effect feedback layer, which can allow the user to clearly see the effect of the maintenance operation, which abnormalities have been solved, and which still need to be paid attention to. At the same time, the alarm state release instruction of the user terminal is triggered through the remote service port, and if a certain abnormal area has been repaired, the corresponding alarm information will be released on the user terminal, allowing the user to timely understand the running state change of the pipeline network.

[0084] In a non-limiting embodiment, after the maintenance response operation of the intelligent water meter pipeline network is triggered through the remote service port, it further includes: collecting valve adjustment logs and sensor response data during the maintenance operation to generate a maintenance operation tracking data set; based on the maintenance operation tracking data set and the abnormal state distribution map, identifying the association relationship between the abnormal area and the valve control node, and constructing an abnormal control knowledge graph; superimposing and rendering the abnormal control knowledge graph and the visualization output interface to generate a dynamic interaction layer containing control node recommendation marks; according to the operation instruction of the user to the dynamic interaction layer, real-time adjusting the visualization focus area of the pipeline BIM model data, and updating the maintenance sequence in the state monitoring feedback strategy.

[0085] During the maintenance operation, the system will collect valve adjustment logs to record the adjustment time, adjustment direction (open or close) and adjustment degree of each valve, etc. At the same time, the sensor response data is collected to understand the data change of the sensor nodes during the maintenance operation. These data are combined together to generate a maintenance operation tracking data set. Based on this data set and the abnormal state distribution map, the association relationship between the abnormal area and the valve control node is analyzed.

[0086] For example, if the water pressure anomaly of a certain abnormal area is improved after a certain valve is closed, it can be determined that the abnormal area is associated with the valve control node. By organizing and analyzing a plurality of above-mentioned association relationships, an abnormal control knowledge graph is constructed, which shows the relationship between different abnormal conditions and valve control nodes, providing knowledge support for subsequent maintenance and control. The abnormal control knowledge graph is superimposed and rendered with the visual output interface to generate a dynamic interactive layer containing control node recommendation markers. When a user views a certain abnormal area on this layer, the valve control node recommendation information related to it will be displayed, helping the user understand how to better control the abnormal condition. The user can operate the dynamic interactive layer through operation instructions, for example, the user clicks on a certain valve control node recommendation marker, and the system will adjust the visual focus area of the pipeline BIM model data in real time to focus on the valve control node and its related pipeline part.

[0087] Meanwhile, according to the user's operation and newly acquired information, the maintenance order in the state monitoring feedback strategy is updated. For example, if it is found that a certain abnormal area with a lower priority is associated with a key valve control node, and the problem can be effectively solved by operating the valve, the maintenance priority of the abnormal area will be increased, and the maintenance order will be adjusted to more efficiently maintain and manage the pipeline network.

[0088] The embodiments of the present application can realize the deep fusion of intelligent water meter monitoring data and BIM model, thereby accurately positioning the abnormality and realizing dynamic visual display. In detail, the embodiments of the present application can accurately grasp the pipeline state by acquiring the pipeline BIM model data and real-time monitoring data set of the target building intelligent water meter pipeline network; the matching abnormal state distribution graph can be generated by performing abnormal state identification processing on the monitoring data, and the abnormal position can be clearly presented; the visual output interface with abnormal markers can be generated by dynamically visualizing the BIM model data based on the graph and topological constraint conditions, so that the operation and maintenance personnel can intuitively understand the abnormal condition; the feedback strategy can be generated by associating and mapping the visual interface with the monitoring data, and the maintenance response operation can be triggered, which can realize efficient monitoring and timely processing of the intelligent water meter pipeline network state, greatly improving the intelligent and automated level of the entire monitoring process, and improving the operation and maintenance efficiency and accuracy.

[0089] Referring to Figure 2 As shown in the figure, the figure is a schematic diagram of the basic structure of a remote state monitoring system 200 provided by the embodiments of the present application, and the remote state monitoring system 200 comprises:

[0090] a processor 201;

[0091] a storage device 202, which stores a computer program 2020 thereon;

[0092] The computer program 2020, when executed by the processor 201, causes the processor 201 to implement any of the above-mentioned intelligent water meter based remote condition monitoring methods.

[0093] Based on the above, a readable storage medium is provided, and the readable storage medium stores a program or instructions, and the program or instructions are executed by a processor to implement the steps of the above-mentioned method.

[0094] In the technical solutions related to the above-mentioned embodiments of the present application, whether it is a multi-dimensional feature comparison calculation or a composite parameter construction, if there are problems caused by significant differences in the number of dimensions, dimensional units and semantic meanings of different features, a person skilled in the art can fully understand that these differences need to be properly handled based on their professional knowledge and past practical experience, so that the calculation result is accurate and has comparability, and logical confusion, unclear mathematical meaning and other conditions are avoided.

[0095] In detail, when facing features with different numbers of dimensions, a person skilled in the art can use various strategies to accurately calculate the similarity, matching degree or feature distance between different features.

[0096] Feature selection is a commonly used method. For a high-dimensional feature set, a feature subset with the most representative features can be selected from the high-dimensional features according to the importance, correlation and other indicators of the features, so as to match the number of low-dimensional features. By using chi-square test, information gain and other methods for feature selection, the most valuable features of the technical solution are screened out, so as to reduce the high-dimensional features to a dimension comparable to the low-dimensional features, and then the similarity or distance calculation is performed.

[0097] Feature extraction is also an effective means. By constructing a suitable feature extraction model, features with different dimensions are mapped to a common low-dimensional feature space. Principal component analysis (PCA) can not only be used to process dimensional differences, but also project high-dimensional features to a low-dimensional space composed of principal components, so that features with different dimensions have comparability in this low-dimensional space. In addition, deep learning models such as autoencoders can also be used for feature extraction, which can automatically learn the latent representation of input features and convert features with different dimensions into feature vectors with the same dimension, so as to perform subsequent similarity, matching degree or feature distance calculation.

[0098] In addition, a kernel method can also be used. The kernel function can calculate the similarity between features in a high-dimensional space without explicitly mapping the features to a high-dimensional space. For features with different numbers of dimensions, a suitable kernel function such as a Gaussian kernel function or a polynomial kernel function can be selected to directly calculate the similarity between them. This method avoids the difficulty of direct calculation caused by different feature dimensions, and can effectively measure the relationship between features in the original feature space or an implicit high-dimensional space.

[0099] In the comparison of multi-dimensional features, in order to achieve the comparable alignment of the feature space, the skilled person can use a variety of existing common technical means.

[0100] Standardization preprocessing is a widely used and effective method, which converts the data into a standard normal distribution with a mean of 0 and a standard deviation of 1 by performing a specific linear transformation on the original feature data. This processing method can essentially eliminate the influence of the dimensions of different features, allowing all features to be compared on the same scale. For example, in a dataset containing features of different dimensions, after standardization preprocessing, these features can be calculated for similarity or distance on the same scale, avoiding calculation bias caused by different dimensions.

[0101] Mapping conversion is also an effective way to solve the problem of dimension difference, which can map the original features to a completely new space according to the specific properties of the features and actual business needs. In this new space, features of different dimensions can have better comparability. For features with nonlinear relationships, the skilled person can use logarithmic transformation, power transformation, etc. to convert them to linear relationships, making it easier to calculate similarity or distance. For example, when dealing with some features with exponential growth trends, logarithmic transformation can convert them to linear relationships, making subsequent calculations more accurate and convenient.

[0102] Space projection is also an important technical means, which projects high-dimensional feature space into low-dimensional space while preserving important information between features as much as possible. By carefully selecting the projection direction and projection dimension, the skilled person can reduce the influence of dimension difference on the calculation results while reducing the data dimension. Common space projection methods include principal component analysis (PCA), linear discriminant analysis (LDA), etc. Taking principal component analysis as an example, it projects high-dimensional data into a low-dimensional space composed of principal components by finding the principal component direction, simplifying the data structure while reducing the interference of dimension difference on feature comparison.

[0103] In the construction process of composite parameters (such as loss function values), different parameter terms often have different dimensions, and the skilled person can use normalization processing or adaptive weight distribution mechanism based on distribution characteristics.

[0104] Normalization is to unify the value range of different parameter items to a fixed interval, for example, [0, 1]. This processing method can eliminate the influence of dimensional differences and ensure that each parameter item has the same importance when weighted fusion. Common normalization methods include minimum-maximum normalization, Z-score normalization, etc. Taking minimum-maximum normalization as an example, it scales the value range of the parameter item to the [0, 1] interval through linear transformation, so that parameter items of different dimensions can be weighted and fused under the same standard.

[0105] The adaptive weight allocation mechanism based on distribution characteristics dynamically adjusts the weights of different parameter items according to their distribution characteristics. For parameter items with large variance, the skilled person in the art can appropriately reduce their weights; for parameter items with small variance, the skilled person in the art can appropriately increase their weights. In this way, the composite loss function can pay more attention to parameter items with small variance, thereby improving the stability and generalization ability of the model. For example, in a composite loss function containing multiple parameter items, if the variance of a certain parameter item is large, it means that its fluctuation is violent, which may adversely affect the stability of the model. At this time, reducing its weight can reduce this adverse effect; for parameter items with small variance, increasing their weight can make the model pay more attention to the information reflected by the parameter item, thereby improving the overall performance of the model.

[0106] The above-mentioned general technical means for solving the problems of feature matching and loss balancing belong to the common knowledge in the art. These technical means have been fully verified and widely used in a large number of practical applications, and the skilled person in the art can skillfully and flexibly use these methods to solve similar dimensional difference problems.

[0107] The formulas and calculation processes involved in the embodiments of the present application, whether for multi-dimensional feature comparison or composite loss function construction, strictly follow the dimensional correspondence principle. The variables in each formula have a clear and explicit physical meaning, and their operation logic completely conforms to the basic mathematical and physical logic, and the operation result is necessarily the reasonable result expected by the present application. The skilled person in the art has the ability to comprehensively use the above-mentioned general technical means according to the specific data situation and business requirements to effectively solve various problems caused by the number of dimensions, dimensional differences, etc. in the multi-dimensional feature comparison calculation and the composite loss function construction in the embodiments, and to ensure the accuracy, reliability and implementability of the technical solutions of the present application.

[0108] It should be noted that the various embodiments described in the specification are intended to be exemplary only and that the scope of the application is not intended to be limited to the embodiments described in the specification.

Claims

1. A method for remote condition monitoring based on smart water meter, characterized in that, The method comprises the following steps: acquiring the pipeline BIM model data of the intelligent water meter pipeline network of the target building, and synchronously collecting a real-time monitoring data set of a plurality of sensing nodes in the intelligent water meter pipeline network; performing abnormal state identification processing on the real-time monitoring data set to generate an abnormal state distribution map matching the spatial position of the pipeline BIM model data; based on the abnormal state distribution map and a preset pipeline topology constraint condition, performing dynamic visual reconstruction processing on the pipeline BIM model data to generate a visual output interface with an abnormal state marker; associating and mapping the visual output interface with the real-time monitoring data set to generate a state monitoring feedback strategy, and triggering a maintenance response operation of the intelligent water meter pipeline network through a remote service port; collecting valve adjustment logs and sensor response data during the maintenance operation to generate a maintenance operation tracking data set; based on the maintenance operation tracking data set and the abnormal state distribution map, identifying the association relationship between the abnormal area and the valve control node, and constructing an abnormal control knowledge graph; superimposing and rendering the abnormal control knowledge graph and the visual output interface to generate a dynamic interactive layer containing a control node recommendation marker; according to the operation instruction of the user to the dynamic interactive layer, real-time adjusting the visual focus area of the pipeline BIM model data, and updating the maintenance sequence in the state monitoring feedback strategy.

2. The method of claim 1, wherein, The method comprises the following steps: calling the BIM model database of the target building, extracting the spatial topology structure data and equipment attribute data constituting the pipeline BIM model data; the spatial topology structure data is used to describe the pipe connection relationship and spatial geometric layout, and the equipment attribute data includes the water meter installation coordinates and configuration parameters associated with the sensing nodes; sending an activation instruction to each sensing node according to the configuration parameters, and collecting water pressure monitoring data, flow fluctuation data and equipment running state data of each sensing node at a set period; performing space-time alignment processing on the water pressure monitoring data, the flow fluctuation data and the equipment running state data based on the water meter installation coordinates and the spatial topology level of the pipeline BIM model to generate a hierarchical real-time monitoring data set; wherein the topology structure of the hierarchical real-time monitoring data set and the pipeline network level of the pipeline BIM model form a mapping relationship, the spatial topology level is a BIM model hierarchy divided according to building function areas, and the pipeline network level is a physical pipeline hierarchy divided based on the primary and secondary relationship of water supply.

3. The method of claim 2, wherein, The method comprises the following steps: performing mutation point detection processing on the water pressure monitoring data in the hierarchical real-time monitoring data set to identify water pressure abnormal areas exceeding a preset fluctuation threshold, and extracting the duration and fluctuation amplitude of the water pressure abnormal areas; The flow fluctuation data is subjected to trend deviation analysis processing to determine the deviation degree of flow trend from the preset flow model, and the flow anomaly area with a deviation degree exceeding a dynamic threshold is marked; The spatial coordinates of the water pressure anomaly area and the flow anomaly area are mapped to the corresponding positions of the pipeline BIM model data to generate an initial anomaly state distribution map containing anomaly type labels and anomaly levels; According to the fault alarm information in the equipment operation state data, the initial anomaly state distribution map is subjected to cross-validation processing to eliminate false alarm anomaly areas and correct anomaly levels, thereby generating the anomaly state distribution map.

4. The method of claim 3, wherein, The anomaly state distribution map and the preset pipeline topology constraint condition are used to perform dynamic visual reconstruction processing on the pipeline BIM model data to generate a visual output interface with an anomaly state label, including: The pipeline connection relationship data and valve control node data are extracted from the pipeline BIM model data to construct a topology constraint relationship network of the intelligent water meter pipeline network; According to the anomaly type label in the anomaly state distribution map, each anomaly area in the anomaly state distribution map is assigned a corresponding color coding and dynamic flashing frequency; The color coding and dynamic flashing frequency are superimposed on the corresponding spatial positions of the pipeline BIM model data to generate a first visual topology structure; Based on the topology constraint relationship network, the anomaly areas in the first visual topology structure are subjected to influence range diffusion simulation processing to generate a second visual topology structure containing a potential influence area label; The second visual topology structure and the equipment operation state data in the hierarchical real-time monitoring data set are subjected to layer fusion processing to generate the visual output interface.

5. The method of claim 4, wherein, The visual output interface and the real-time monitoring data set are associated and mapped to generate a state monitoring feedback strategy, including: The spatial coordinate set of the anomaly area and the anomaly type label are extracted from the visual output interface to generate an anomaly state feature vector; According to the anomaly state feature vector, a corresponding maintenance priority scoring rule is matched from a preset strategy library to calculate the maintenance priority score of each anomaly area extracted from the visual output interface; Based on the maintenance priority score and the pipeline connection tightness in the topology constraint relationship network, a state monitoring feedback strategy containing a maintenance sequence and a resource allocation scheme is generated; the maintenance sequence is a sequence sorted according to the priority score and the pipeline connection tightness.

6. The method of claim 5, wherein, The maintenance response operation of the intelligent water meter pipeline network is triggered through a remote service port, including: The state monitoring feedback strategy is converted into a remote control instruction set, including valve opening and closing instructions, data acquisition frequency adjustment instructions, and maintenance work order generation instructions; The remote control instruction set is sent to the edge control node of the intelligent water meter pipeline network through an encrypted communication link to instruct the edge control node to perform valve state adjustment operations; The maintenance work order generation instruction is pushed to a preset maintenance management platform synchronously, and work order execution state data returned by the maintenance management platform is received; According to the work order execution state data, an abnormal state mark in the visualization output interface is updated, and a maintenance progress tracking layer is generated.

7. The method of claim 3, wherein, The mutation point detection processing on the water pressure monitoring data in the hierarchical real-time monitoring data set includes: The water pressure monitoring data is divided into continuous water pressure data segments according to a preset time window, and first-order difference calculation is performed on each water pressure data segment to obtain a water pressure change rate sequence; The sliding window standard deviation analysis is performed on the water pressure change rate sequence to identify a mutation time point with a standard deviation exceeding a dynamic threshold; According to the mutation time point, corresponding water pressure monitoring sub-data is extracted, and an offset percentage of the water pressure monitoring sub-data from historical mean data is calculated; If the offset percentage exceeds the preset fluctuation threshold, the corresponding region of the water pressure monitoring sub-data is marked as a water pressure abnormal region.

8. The method of claim 7, wherein, The trend deviation analysis processing on the flow fluctuation data includes: Obtaining a preset flow reference curve of each branch pipeline in the intelligent water meter pipeline network, the preset flow reference curve including a time-periodized expected flow range generated based on historical data; Identifying and classifying the flow fluctuation data according to branch pipelines, and performing trend deviation analysis processing on the classified flow fluctuation data based on the preset flow reference curve to generate a real-time flow trend line of each branch pipeline; Calculating the deviation degree of the real-time flow trend line from the corresponding preset flow reference curve, the deviation degree including a deviation duration and a deviation cumulative amount; If the deviation duration exceeds a first dynamic threshold and the deviation cumulative amount exceeds a second dynamic threshold, the spatial coordinates of the corresponding branch pipeline are mapped to the pipeline BIM model data, and a flow abnormal region is marked according to the mapped spatial coordinates.

9. The method of claim 6, wherein, The establishment process of the encrypted communication link includes: Assigning a unique identity identifier and a dynamic encryption key to each edge control node; Before sending the remote control instruction set, generating an instruction header verification code according to the identity identifier, and performing segmented encryption processing on the instruction content using the dynamic encryption key; Combining the segmented encrypted instruction content and the instruction header verification code to generate an encrypted data packet, and sending it to each edge control node through a multi-path transmission protocol; Receiving an instruction confirmation signal returned by each target edge control node, and updating the effective state of the dynamic encryption key according to the instruction confirmation signal.

10. A remote condition monitoring system characterised in that, It includes: A processor; A storage device having a computer program stored thereon, when the computer program is executed by the processor, the processor implements the intelligent water meter based remote state monitoring method according to any one of claims 1-9.

Images