A method, system, and computer program product for monitoring anomalies in a fluid piping network

Abstract

The present disclosure provides a fluid pipe network anomaly monitoring method, system and computer program product, relating to the technical field of transportation and distribution. The fluid pipe network anomaly monitoring method disclosed by the present disclosure comprises: collecting, by a sensor, measured fluid data corresponding to one or more data collection points in a fluid pipe network in a preset geographical area; calculating, by a fluid pipe network simulation model, simulated fluid data corresponding to the one or more data collection points in the fluid pipe network; the fluid pipe network simulation model is constructed based on the pipe network structure of the fluid pipe network and the measured fluid data corresponding to the one or more data collection points; comparing the measured fluid data and the simulated fluid data corresponding to the same data collection point in the fluid pipe network to obtain a data comparison result; determining an abnormal area from the fluid pipe network based on the data comparison result.

Description

Technical Field

[0001] This specification relates to the field of transmission and distribution technology, and in particular to a method, system and computer program product for monitoring anomalies in fluid pipeline networks. Background Technology

[0002] In many regions, fluid pipeline networks have been in operation for a long time, leading to issues such as aging equipment and increased leakage risks. Currently, most methods for monitoring fluid pipeline anomalies rely on manual on-site inspections, which are not only costly in terms of manpower but also have poor timeliness in accident response. Therefore, mastering a more reliable and effective method for monitoring fluid pipeline anomalies is of great significance.

[0003] In view of this, some embodiments of this specification provide a method and system for monitoring anomalies in fluid pipeline networks. Summary of the Invention

[0004] This specification provides one or more embodiments of a fluid pipeline network anomaly monitoring method, executed by an edge server, comprising: collecting measured fluid data corresponding to one or more data collection points in a fluid pipeline network within a preset geographical area using sensors; calculating simulated fluid data corresponding to one or more data collection points in the fluid pipeline network using a fluid pipeline network simulation model; the fluid pipeline network simulation model being constructed based on the pipeline network structure and the measured fluid data corresponding to one or more data collection points; comparing the measured fluid data corresponding to the same data collection point in the fluid pipeline network with the simulated fluid data to obtain a data comparison result; and determining an abnormal area from the fluid pipeline network based on the data comparison result.

[0005] According to some embodiments of the fluid pipeline network anomaly monitoring method described in this specification, the fluid data includes multiple types; the step of comparing the measured fluid data corresponding to the same data acquisition point in the fluid pipeline network with the simulated fluid data to obtain a data comparison result includes: comparing the measured fluid data corresponding to the target type at the same data acquisition point in the fluid pipeline network with the simulated fluid data of the target type to obtain a data comparison error value corresponding to the target type at the data acquisition point; determining a preset error value corresponding to the target type at the data acquisition point; and generating a data comparison result based on the data comparison error value and the preset error value.

[0006] According to some embodiments of the fluid pipeline network anomaly monitoring method described in this specification, the multiple types include one or more of the following types: temperature type, pressure type, and flow type.

[0007] According to some embodiments of the fluid pipeline network anomaly monitoring method described in this specification, determining the preset error value corresponding to the target type at the data acquisition point includes: acquiring historical data comparison error values ​​corresponding to the target type at the data acquisition point within a historical time period; and determining the preset error value corresponding to the target type at the data acquisition point based on the historical data comparison error values.

[0008] According to some embodiments of the fluid pipeline network anomaly monitoring method described in this specification, the fluid pipeline network includes multiple valves; the method further includes: receiving valve opening / closing notification information of a target valve; calling the fluid pipeline network simulation model based on the valve opening / closing notification information; and determining whether the target valve is an abnormal valve based on the fluid pipeline network simulation model.

[0009] According to some embodiments of the fluid pipeline network anomaly monitoring method described in this specification, the step of determining whether the target valve is an abnormal valve based on the fluid pipeline network simulation model includes: calculating the simulated fluid data corresponding to the data acquisition point where the target valve is located using the fluid pipeline network simulation model; comparing the measured fluid data corresponding to the data acquisition point where the target valve is located with the simulated fluid data to obtain a data comparison result; and determining whether the target valve is an abnormal valve based on the data comparison result.

[0010] According to some embodiments of the fluid pipeline network anomaly monitoring method described in this specification, the method further includes: acquiring a leak detection network model; using the leak detection network model to process measured fluid data from one or more data acquisition points and / or one or more simulated fluid data from the fluid pipeline network to predict leak information in the fluid pipeline network.

[0011] According to some embodiments of the fluid pipeline network anomaly monitoring method described in this specification, the leakage information includes one or more of the following: the leakage point and the leakage amount corresponding to the leakage point.

[0012] According to some embodiments of the fluid pipeline network anomaly monitoring method described in this specification, the leakage detection network model is trained by a cloud computing center, and the training data of the leakage detection network model includes historical leakage information, as well as historical measured fluid data and / or historical simulated fluid data corresponding to the historical leakage information; the step of obtaining the leakage detection network model includes: obtaining the trained leakage detection network model from the cloud computing center.

[0013] According to some embodiments of the fluid pipeline network anomaly monitoring method described in this specification, the fluid pipeline network includes a gas pipeline network; the data acquisition point is located at one or more of the following nodes in the gas pipeline network: gate station, pressure regulating station, storage and distribution station, diversion pipeline connector, pipeline valve, and client.

[0014] According to some embodiments of the fluid pipeline network anomaly monitoring method described in this specification, the edge server is deployed at one of the one or more data acquisition points.

[0015] A fluid pipeline network anomaly monitoring system according to one or more embodiments of this specification is deployed on an edge server and includes: a data acquisition module for acquiring measured fluid data corresponding to one or more data acquisition points in a fluid pipeline network within a preset geographical area via sensors; a calculation module for calculating simulated fluid data corresponding to one or more data acquisition points in the fluid pipeline network using a fluid pipeline network simulation model; the fluid pipeline network simulation model is constructed based on the pipeline network structure and the measured fluid data corresponding to one or more data acquisition points; a comparison module for comparing the measured fluid data corresponding to the same data acquisition point in the fluid pipeline network with the simulated fluid data to obtain a data comparison result; and a determination module for determining an abnormal area from the fluid pipeline network based on the data comparison result.

[0016] According to some embodiments of the fluid pipeline network anomaly monitoring system described in this specification, the fluid data includes multiple types; the comparison module is further used to: compare the measured fluid data corresponding to the target type at the same data acquisition point in the fluid pipeline network with the simulated fluid data of the target type to obtain the data comparison error value corresponding to the target type at the data acquisition point; determine the preset error value corresponding to the target type at the data acquisition point; and generate a data comparison result based on the data comparison error value and the preset error value.

[0017] According to some embodiments of the fluid pipeline network anomaly monitoring system described in this specification, the fluid pipeline network includes multiple valves; the system also includes an abnormal valve determination module, which is used to: receive valve opening / closing notification information of a target valve; call the fluid pipeline network simulation model based on the valve opening / closing notification information; and determine whether the target valve is an abnormal valve based on the fluid pipeline network simulation model.

[0018] According to some embodiments of the fluid pipeline network anomaly monitoring system described in this specification, in order to determine whether a target valve is an abnormal valve based on the fluid pipeline network simulation model, the abnormal valve determination module is further configured to: calculate the simulated fluid data corresponding to the data acquisition point where the target valve is located using the fluid pipeline network simulation model; compare the measured fluid data corresponding to the data acquisition point where the target valve is located with the simulated fluid data to obtain a data comparison result; and determine whether the target valve is an abnormal valve based on the data comparison result.

[0019] According to some embodiments of this specification, the fluid pipeline network anomaly monitoring system further includes a leak detection module, which is used to: acquire a leak detection network model; process measured fluid data from one or more data acquisition points and / or one or more simulated fluid data from the fluid pipeline network using the leak detection network model, and predict leak information in the fluid pipeline network.

[0020] This specification provides one or more embodiments of a computer program product, including a computer program or instructions, characterized in that the computer program or instructions, when executed by a processor, implement the aforementioned monitoring method. Attached Figure Description

[0021] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. The same numbers in the drawings denote the same structures or steps.

[0022] Figure 1 This is a schematic diagram of an edge server network topology based on some embodiments of this specification.

[0023] Figure 2 This is a flowchart of a fluid pipeline network anomaly monitoring method according to some embodiments of this specification.

[0024] Figure 3 This is a schematic block diagram of a fluid pipeline network anomaly monitoring system according to some embodiments of this specification. Detailed Implementation

[0025] To more clearly illustrate the technical solutions of the embodiments in this specification, the embodiments will be described in detail below with reference to the accompanying drawings. Obviously, the content described below are some examples or embodiments of this specification. For those skilled in the art, without creative effort, the technical solutions or means disclosed in this specification can be applied to other scenarios based on this technical content.

[0026] It should be understood that the terms "system," "device," "unit," and / or "module" used in this specification are a method of distinguishing different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0027] Unless otherwise specified, the technical terms used to describe components, elements, etc. in this specification are not singular but may include plural. Generally speaking, terms such as "comprising" or "including" only indicate that explicitly identified steps, elements, or components are included, and these steps, elements, and components do not constitute an exclusive list, as the described method or apparatus may also include other steps or components.

[0028] This specification uses flowcharts to illustrate the operational steps performed by the apparatus or system of related embodiments. However, unless otherwise specified, the order in which these steps are described should not be construed as a limitation on the order of execution. Those skilled in the art can adjust the order of these steps based on the knowledge and information conveyed by the embodiments in this specification. Such adjustments include, but are not limited to, reversing the order of steps, merging multiple steps, and splitting a step.

[0029] Fluid, broadly speaking, refers to any substance capable of flowing, specifically including liquids and gases. For example, liquids transported and distributed through pipeline networks can include water needed for daily life and production, as well as hot water for heating; gases transported and distributed through pipeline networks can be fuel gases, such as natural gas, manufactured gas, liquefied petroleum gas, biogas, and coal gas. Correspondingly, the pipeline network for transporting and distributing fluids can be called a fluid pipeline network. The safe and stable transport and distribution of these fluid resources is crucial to people's daily lives and production, and is an important guarantee for a high-quality life and stable economic development. Therefore, more effective anomaly monitoring of fluid pipeline networks is of great significance.

[0030] Fluid piping networks may further include pipe segments and nodes, with different pipe segments connected by nodes. In some embodiments, nodes can alter the physical properties of the fluid, such as temperature, flow rate, pressure, and direction of movement. The following description primarily uses a gas piping network as an example. It should be understood that the following descriptions of gas piping networks also apply to other fluid piping networks and should not be construed as limiting the scope of this specification.

[0031] Specifically, for a gas pipeline network, its nodes can include gate stations, pressure regulating stations, storage and distribution stations, diversion pipeline connectors, pipeline valves, and user terminals. A gate station, such as a city gate station, is a receiving station where gas enters a local pipeline network (such as a city's pipeline network) from a long-distance pipeline. A pressure regulating station is a facility used to regulate or stabilize the gas pressure in the pipeline network. A storage and distribution station is a facility for storing and distributing gas. Diversion pipeline connectors can further include two-way diversion connectors, three-way connectors, four-way connectors, etc. User terminals refer to end users who use gas, specifically residential buildings, commercial buildings, and factory production workshops. In some embodiments, sensors can be installed at pipeline network nodes to collect information such as the physical parameters of the gas at the node. For example, sensors deployed at pipeline network nodes can include one or more combinations of the following: temperature sensors, pressure sensors, flow sensors, valve opening sensors, etc. Installing various sensors at nodes allows for data collection during the gas transmission and distribution process of the pipeline network, monitoring the network's operational status.

[0032] In practical applications, many regional gas pipeline networks have been in operation for a long time, leading to issues such as aging equipment and increased leakage risks. However, current methods for monitoring fluid network anomalies largely rely on manual on-site inspections, which are not only costly in terms of manpower but also have poor timeliness in accident response. Therefore, this specification provides a method and system for monitoring fluid network anomalies in some embodiments. This method involves collecting and measuring fluid data at one or more data acquisition points within the pipeline network, calculating simulated fluid data corresponding to one or more data acquisition points using a fluid network simulation model, and then identifying abnormal areas within the fluid network by comparing the measured fluid data with the simulated fluid data at the same data acquisition point.

[0033] Figure 1 This specification provides a schematic diagram of a network topology consisting of edge servers, as illustrated in some embodiments. Edge servers can be deployed at multiple nodes of the fluid pipeline network. The edge servers can acquire sensor data from their own node or nearby nodes (such as adjacent nodes), and then perform simulation calculations based on this data. The calculation results are then transmitted to the cloud computing center via the enterprise's internal LAN. For example, the edge server can be a lightweight system-on-a-chip, an electronic device assembled from multiple functional components, or a desktop computer, laptop computer, tablet computer, smartphone, etc. In some embodiments, sensors are installed at the edge server. The edge server may include a data preprocessing module, a data analysis and verification module, a leakage module, and a visualization interface. The sensors are used for data acquisition. The edge server is deployed near the sensors to process the data collected by the sensors in a timely manner, enabling staff to quickly understand the occurrence of anomalies in the fluid pipeline network. The data preprocessing module is a functional module that preprocesses the data acquired by the sensors to improve the convenience and accuracy of subsequent data processing. The data analysis and verification module performs simulation calculations on the preprocessed data and uses the simulated data for data analysis and verification to understand whether the fluid pipeline network is operating normally. For more information on simulation calculations, please refer to the relevant description in step S202. The leakage module is a functional module that determines whether there is a leak in the fluid pipeline network based on the preprocessed data. In some embodiments, the leakage module can detect whether there is a leak in the fluid pipeline network through a leakage detection network model. The visualization interface is a module used to display the current alarm data to staff so that they can understand the pipeline network anomalies in a timely manner.

[0034] Figure 2 This specification describes a method for monitoring fluid network anomalies via an edge server, as illustrated in some embodiments, which is an improvement over manual monitoring of fluid network anomalies. Figure 2 The method shown is more efficient and accurate. In some embodiments, Figure 2The process 200 shown can be implemented by an edge server, such as by... Figure 1 This can be implemented by an edge server on any node within the system, or further, it can be executed by a monitoring system 300 deployed on the edge server. For example... Figure 2 As shown, process 200 includes the following steps.

[0035] Step S201 involves acquiring measurement fluid data corresponding to one or more data acquisition points in the fluid pipeline network of a preset geographical area using sensors. In some embodiments, step S201 can be implemented by the acquisition module 310.

[0036] In this specification, data acquisition points are points distributed on the fluid pipeline network that can be used to collect data. The embodiments of this specification do not limit the specific location of the data acquisition points. Data acquisition points can be located at the location of the sensor in order to collect the data sent by the sensor; data acquisition points can also be located at the location of the node in order to collect the data corresponding to different nodes.

[0037] Among them, the measured fluid data is the data obtained through actual collection, such as the data collection point located at the sensor location, where the data collected by the sensor can be used as the measured fluid data.

[0038] In some embodiments, for example, a preset geographical area where data needs to be collected can be determined first. Then, one or more data collection points in the fluid pipeline network of the preset geographical area are used to collect data, and the measurement fluid data corresponding to each data collection point is obtained. The collection of measurement fluid data can be set to be continuous, or it can be set to collect measurement fluid data at regular intervals depending on the actual application. The preset geographical area can correspond to an edge server; that is, one edge server can be responsible for the simulation calculation and anomaly detection of a geographical area.

[0039] Specifically, at the same node, the deployed sensors can connect to the edge server via data cables, allowing the edge server to directly acquire the measured fluid data collected by the sensors. Alternatively, the edge server can connect to the sensors at its node via a wireless network, acquiring the measured fluid data collected by the sensors through the wireless network. In some embodiments, the sensors can be buried underground, closer to the pipe section, while the edge server can be installed above ground. In some embodiments, the edge server can also acquire measured fluid data from sensors at other nodes besides its own.

[0040] In some embodiments, gas pipeline failures can cause significant inconvenience to users and substantial losses to gas companies. Therefore, to enable gas companies to respond promptly, identify, and repair abnormal situations, edge servers can be deployed close to sensors. Specifically, edge servers can be deployed at data collection points, allowing edge computing to process and analyze data where it is generated. This enables information exchange between the cloud computing center server and the edge server, allowing the edge server to receive abnormal reports and other data from the edge server, rather than transmitting all data to the cloud computing center server for processing. This approach can significantly improve data processing efficiency and real-time performance, enabling the system to quickly detect and respond to gas pipeline failures.

[0041] The embodiments described in this specification enable close collaboration between edge servers and cloud computing center servers to complete information transmission, better meeting the needs of various scenarios and thus amplifying the application value of edge computing and cloud computing. From a communication perspective, edge servers and cloud computing center servers can communicate via an enterprise intranet or via an MPLS leased line.

[0042] In some embodiments, the measured fluid data can also be verified by a preprocessing algorithm module, and problematic data (such as outliers or null values) can be corrected or compensated. The verified and correct data can be stored in a historical database for future retrieval.

[0043] Step S202 involves calculating simulated fluid data corresponding to one or more data acquisition points in the fluid pipeline network using a fluid pipeline network simulation model. In some embodiments, step S202 can be implemented by the calculation module 320.

[0044] A fluid network simulation model is a model used to simulate the fluid flow state in a fluid network. For example, a fluid network model can be used to simulate data such as temperature, pressure, and flow rate at different locations within the fluid network. In the embodiments of this specification, the fluid network simulation model can be constructed based on the network structure and the measured fluid data corresponding to one or more data acquisition points. In some embodiments, the fluid network simulation model can be a fluid equation or a machine learning model; for example, it can be a neural network model, a linear model, a support vector machine, a decision tree, etc. The edge server can input the measured fluid data corresponding to the data acquisition points into the machine learning model to obtain simulated fluid data.

[0045] Among them, simulated fluid data is calculated by simulating the operating state of a fluid pipeline network through a fluid pipeline network simulation model. Since there are numerous nodes and pipes in a fluid pipeline network, but due to the difficulty in deploying sensors in most areas or other practical reasons, the available measured fluid data is limited. Therefore, data simulation technology can be introduced. Using data simulation technology, simulated fluid data at any location in the fluid pipeline network can be obtained from limited measured fluid data.

[0046] In some embodiments, such as gas pipeline networks in many regions that have been built for a long time, problems such as equipment aging and increased risk of leakage may occur. As a result, when sensors collect fluid data such as temperature, pressure, and flow rate, abnormalities may occur due to sensor aging or network communication issues during the acquisition or transmission process. Therefore, this embodiment can simulate the fluid flow state in the fluid pipeline network using a fluid pipeline network simulation model to obtain simulated fluid data, and compare it with the measured fluid data obtained from actual measurements to understand whether there are any abnormalities in the operation of the gas pipeline network.

[0047] Step S203 involves comparing the measured fluid data and simulated fluid data corresponding to the same data acquisition point in the fluid pipeline network to obtain a data comparison result. In some embodiments, step S203 can be implemented by the comparison module 330.

[0048] In some embodiments, in order to identify abnormal areas in the fluid pipeline network, the measured fluid data corresponding to the same data acquisition point in the fluid pipeline network can be compared with the simulated fluid data to obtain the data comparison results, and the abnormality and the area of ​​abnormality can be determined based on the data comparison results.

[0049] In practical applications, to improve the speed and timeliness of anomaly detection, sensors can be used to collect and measure fluid data in real time. This data is then transmitted from the sensors to an edge server, where a fluid network simulation model generates simulated fluid data in real time. In other words, fluid data is collected and simulated at every moment. Data comparisons are then performed, allowing staff to immediately identify and address any anomalies in the fluid network, minimizing inconvenience to users.

[0050] Optionally, in some embodiments, the step of "comparing the measured fluid data and simulated fluid data corresponding to the same data acquisition point in the fluid pipeline network to obtain the data comparison result" may include the following steps.

[0051] The measured fluid data corresponding to the target type at the same data acquisition point in the fluid pipeline network is compared with the simulated fluid data of the target type to obtain the data comparison error value corresponding to the target type at the data acquisition point.

[0052] Determine the preset error value corresponding to the target type at the data collection point.

[0053] Data comparison results are generated based on the data comparison error value and the preset error value.

[0054] Fluid data can include various types.

[0055] In some embodiments, for example, the measured fluid data corresponding to type 1 at data acquisition point A in the fluid pipeline network can be compared with the simulated fluid data corresponding to type 1 to obtain the data comparison error value corresponding to type 1 at data acquisition point A, and the preset error value corresponding to type 1 at data acquisition point A can be determined. Then, based on the obtained data comparison error value and the preset error value, the corresponding data comparison result can be generated.

[0056] Optionally, in some embodiments, the multiple types of fluid data may include one or more of the following types: temperature type, pressure type, and flow rate type.

[0057] In some embodiments, for example, the measured pressure value corresponding to the pressure type at data acquisition point A in the fluid pipeline network can be compared with the simulated pressure value corresponding to the pressure type to obtain the data comparison error value corresponding to the pressure type at data acquisition point A, and the preset error value corresponding to the pressure type at data acquisition point A can be determined. Then, based on the obtained data comparison error value and the preset error value, the corresponding data comparison result can be generated.

[0058] In some embodiments, during data comparison, the network structure and equipment status at different data collection points vary. For example, some meters may have been recently replaced, resulting in smaller errors, with a historical average error of only 1%. If the measured data comparison error reaches 5%, it indicates a serious anomaly. Conversely, some meters may have been used for a long time, resulting in larger errors, with a historical average error potentially reaching 10%. If the measured data comparison error is only 5%, it indicates no serious anomaly and no need for repair. However, if the measured data comparison error reaches a preset error value of 20%, it indicates a serious anomaly. Therefore, different preset error values ​​can be set for different data collection points to adapt to different actual situations and further improve the accuracy of anomaly monitoring.

[0059] Optionally, in some embodiments, the step "determining the preset error value corresponding to the target type at the data acquisition point" may include the following steps.

[0060] Obtain the historical data comparison error value corresponding to the target type at the data collection point within the historical time period.

[0061] Based on the error values ​​compared with historical data, the preset error values ​​corresponding to the target types at the data collection points are determined.

[0062] In some embodiments, multiple historical data comparison error values ​​corresponding to type 1 at data collection points within a historical time period can be obtained. Then, these historical data comparison error values ​​are processed and statistically analyzed to determine the preset error value corresponding to type 1 at the data collection point. For example, for pressure gauge 1, the average error of pressure gauge 1 over the previous 20 times can be obtained as 3%. The pressure gauge error can be obtained by comparing the readings of the pressure gauge under test with those of a precision pressure gauge. The closeness of the indicated value to the true value can be determined by comparing the absolute value of the maximum error with the true value as a percentage. Alternatively, the pressure gauge error can be obtained by dividing the absolute value of the error between the historical indicated value and the historical simulated value of the pressure gauge under test by the historical value. 3% can be used as the historical data comparison error value. After obtaining the historical data comparison error value, the corresponding preset error value for the pressure gauge can be obtained. That is, when the error of the pressure gauge is within the range of the preset error value, it can be inferred that the pressure gauge is working normally. Specifically, based on the historical data comparison error value of 3%, the preset error value corresponding to pressure gauge 1 can be determined to be 5%. If the data obtained at this moment has a comparison error value of 10%, it indicates that there may be abnormalities such as leakage, blockage, valve not closed, or pressure gauge failure, and staff can be notified in time to carry out corresponding maintenance.

[0063] Step S204: Based on the data comparison results, anomaly areas are identified from the fluid pipeline network. In some embodiments, step S204 can be implemented by the determination module 340.

[0064] An abnormal area refers to an area where abnormal situations have occurred. For example, blockages or leaks in the fluid pipeline network, abnormalities in data acquisition and transmission due to equipment aging or network communication, and theft are all considered abnormal situations.

[0065] In some embodiments, for example, a preset difference can be set first. If the portion of the data comparison error value that is larger than the preset error value reaches the preset difference, it indicates that the data acquisition point may be abnormal, such as data offset caused by aging acquisition equipment, or even human factors such as gas theft. Here, the error is the relative error between the two. After comparing the data comparison error value with the preset error value, if it is found that the data comparison error value is larger than the preset error value and the difference has reached the preset difference, the data acquisition point can be identified as an abnormal data acquisition point. Based on the areas where multiple abnormal data acquisition points are located, the abnormal area can be determined from the fluid pipeline network.

[0066] In some embodiments, if an abnormal area is detected in the fluid pipeline network, it can be reported first and staff can be asked to check whether it is an easily diagnosable problem such as a sensor malfunction. Furthermore, if it is determined that it is not a sensor malfunction, further investigation can be conducted to determine whether there is a valve malfunction or leakage.

[0067] In practical applications, when maintaining or repairing pipeline networks, workers may need to change the status of certain valves, such as opening valves that were originally closed or closing valves that were originally open. After maintenance, it is inevitable that valves may not be switched back to their proper state due to negligence or other reasons. Valve anomalies in fluid pipeline networks can lead to serious consequences; however, there are numerous valves in the network, and manually checking them one by one is time-consuming, labor-intensive, and prone to omissions. Optionally, in some embodiments, the fluid pipeline network anomaly monitoring method may further include an abnormal valve determination process, which may specifically include the following steps. In some embodiments, the abnormal valve determination process may be implemented by the abnormal valve determination module 350.

[0068] Receive valve opening / closing notification information from the target valve.

[0069] Based on the valve switching notification information, the fluid pipeline network simulation model is invoked.

[0070] Determine whether the target valve is an abnormal valve based on the fluid pipeline network simulation model.

[0071] The fluid pipeline network includes multiple valves.

[0072] In some embodiments, if a valve in the fluid pipeline is closed, the pressure downstream of the valve will be a fixed value, i.e., it will be in a pressure-holding state with no flow. If a valve in the fluid pipeline is open, the pressure downstream of the valve will change with the fluid flow, and the corresponding flow rate will be measured. By observing the difference in fluid pressure and flow rate between the closed and open states of the valves, the technical solutions of the embodiments in this specification can not only promptly detect abnormal areas in the fluid pipeline, but also determine whether a malfunction occurred when valve opening / closing notification information was pushed to the fluid pipeline, and whether the valves in the fluid pipeline are abnormal valves, etc.

[0073] For example, in practical applications, if a valve in a fluid pipeline changes state, such as from closed to open, the SCADA system receives the corresponding valve opening / closing notification and pushes it to the edge server. Upon receiving the notification, the edge server updates the valve's opening / closing information in the fluid pipeline simulation model and performs a simulation. If the simulation shows the valve is not actually open, it notifies on-site personnel whether the valve is indeed open. SCADA, or Data Acquisition and Monitoring Control System, is a computer-based DCS and power automation monitoring system. It has a wide range of applications, including data acquisition and monitoring control, and process control in fields such as power, metallurgy, petroleum, chemical, gas, and railway.

[0074] Optionally, in some embodiments, the step "determine whether the target valve is an abnormal valve based on the fluid network simulation model" may include the following steps.

[0075] The simulated fluid data corresponding to the data acquisition point where the target valve is located is calculated using a fluid pipeline network simulation model.

[0076] The measured fluid data corresponding to the data acquisition point where the target valve is located is compared with the simulated fluid data to obtain the data comparison results.

[0077] Based on the data comparison results, determine whether the target valve is an abnormal valve.

[0078] In some embodiments, for example, if a valve in a fluid pipeline is closed, the pressure downstream of the valve will be a fixed value, i.e., it will be in a pressure-holding state with no flow. If a valve in a fluid pipeline is open, the pressure downstream of the valve will change with the fluid flow, and the corresponding flow rate will be measured. Therefore, if the valve used in the fluid pipeline simulation model is closed, but the simulated pressure or flow rate downstream of the valve should be 0, but the actual measured fluid data shows that the pressure or flow rate is not 0, it can be determined that the valve information pushed by the SCADA system is incorrect. Therefore, the simulated fluid data corresponding to the data acquisition point of the target valve can be calculated using the fluid pipeline simulation model. Then, the measured fluid data corresponding to the data acquisition point of the target valve is compared with the simulated fluid data to obtain the data comparison result. Based on the data comparison result, it can be determined whether the target valve is an abnormal valve.

[0079] In some embodiments, the fluid network simulation model can be constructed based on the network structure and the measured fluid data corresponding to one or more data acquisition points in the network. The network structure data can come from a GIS (Geographic Information System) and includes node location information, node type, pipe length, and pipe endpoint location information, as well as valve status information. As mentioned earlier, in practical applications, due to human error or other reasons, some valves in the fluid network may fail to switch to their proper state, leading to a discrepancy between the actual situation of some valves and the valve status information used in the modeling. Therefore, in some embodiments, the fluid network simulation model can also periodically calculate the simulated fluid data of pipe segments and / or nodes related to the target valve. The target valve can be a valve in an important area or segment of the network, or a valve in an area that has recently undergone maintenance. The pipe segment related to the target valve can be the segment where the target valve is located, or a segment directly connected to the target valve. The node related to the target valve can be a node within the aforementioned pipe segment. The simulated pressure and flow data of these pipe segments or nodes can be calculated using the fluid network simulation model. The system acquires measured fluid data for pipe sections and / or nodes related to the target valve, compares the aforementioned simulation results with the corresponding measured fluid data, and if the difference exceeds a set threshold, it can be determined that the target valve may be in an abnormal state, and relevant personnel are notified to investigate. This embodiment can proactively and periodically investigate abnormal valve states in the fluid pipeline network, improving the reliability of the fluid pipeline network while reducing the manpower costs required for valve anomaly investigation.

[0080] Optionally, in some embodiments, the embodiments of this specification can not only monitor anomalies in fluid data such as temperature, pressure, and flow rate in real time, but also detect leaks in the fluid pipeline network. That is, the fluid pipeline network anomaly monitoring method can also include a leak detection process, which may specifically include the following steps. In some embodiments, the leak detection process can be implemented by the leak detection module 360.

[0081] Determine the leak detection network model.

[0082] Leak detection network models are used to process measured fluid data from one or more data acquisition points and / or simulated fluid data from one or more fluid networks to predict leak information in the fluid network.

[0083] In some embodiments, a leak detection network model can be introduced. This leak detection network model is a neural network that deeply integrates big data and AI algorithms to identify leaks in the overall fluid pipeline network. It is used to simulate the performance of different measuring instruments in different locations and under different leak conditions in the fluid pipeline network. By applying the leak detection network model, it is possible to infer whether the on-site situation is similar to the preset leak situation, thereby predicting the leak information in the fluid pipeline network. That is, by inputting the measured fluid data such as temperature, pressure, and flow rate of multiple data sampling points at the current moment into the leak detection network model, the leak detection network can output the corresponding leak information at the current moment.

[0084] Optionally, in some embodiments, the leakage information includes one or more of the following: the leakage point and the leakage amount corresponding to the leakage point.

[0085] In some embodiments, such as leak detection network models, it is possible to identify which part of the fluid pipeline network is leaking, that is, the specific location of the leak point and the amount of leakage at that point. This allows staff to accurately locate the leak and understand the severity of the leak in a timely manner so as to make appropriate responses.

[0086] Optionally, in some embodiments, the leak detection network model is trained by a cloud computing center, and the training data of the leak detection network model includes historical leak information, as well as historical measured fluid data and / or historical simulated fluid data corresponding to the historical leak information.

[0087] Determine the leak detection network model, including: obtaining the trained leak detection network model from the cloud computing center.

[0088] In some embodiments, a leak detection network model can be constructed in a cloud computing center server to obtain historical leak information, as well as historical measured fluid data and / or historical simulated fluid data corresponding to the historical leak information, as training data for the leak detection network model. This data is then stored in a leak sample library on the cloud computing center server. The leak sample library includes the correspondence between historical measured fluid data and / or historical simulated fluid data and historical leak information. For example, historical measured fluid data 1 represents a leak in the fluid pipeline network, with the leak point being point A and the leak amount being N, and so on.

[0089] Then, the training data can be used in the cloud computing center server to perform supervised model training on the leak detection network model, resulting in a trained leak detection network model. During the training process, leak information stored in the leak sample library can be used as corresponding labels for training the network model. Historical measured fluid data and / or historical simulated fluid data can be used as training data to input into the leak detection network model to predict the leak information. Then, the corresponding loss function value is calculated using the historical leak information and the predicted leak information, and the model parameters of the leak detection network model are updated using the loss function value to reduce the difference between the prediction result and the label, thereby completing the training of the network model.

[0090] Then, the edge server acquires the collected measured fluid data and the simulated fluid data obtained from the simulation, and uses the leak detection network model to predict leak information, thereby promptly detecting leaks. That is, the embodiments in this specification can use a cloud computing center server for the network model training step and an edge server for the network model detection step. Furthermore, if computing power allows, the training process can also be implemented on the edge server.

[0091] Optionally, in some embodiments, the fluid pipeline network anomaly monitoring method further includes pushing alarm information to a cloud computing center, the alarm information including abnormal areas and / or leakage information in the fluid pipeline network.

[0092] In some embodiments, if an abnormal area or leak is detected in the fluid pipeline network, an alarm message can be pushed to the cloud computing center and displayed on the system page to notify staff to investigate.

[0093] Figure 3 These are schematic block diagrams of the monitoring system shown in some embodiments of this specification, such as... Figure 3As shown, the monitoring system 300 includes a data acquisition module 310, a calculation module 320, a comparison module 330, and a determination module 340. The data acquisition module 310 is used to acquire measured fluid data corresponding to one or more data acquisition points in a fluid pipeline network within a preset geographical area using sensors. The calculation module 320 is used to calculate simulated fluid data corresponding to one or more data acquisition points in the fluid pipeline network using a fluid pipeline network simulation model, which is constructed based on the pipeline network structure and the measured fluid data corresponding to one or more data acquisition points. The comparison module 330 is used to compare the measured fluid data corresponding to the same data acquisition point in the fluid pipeline network with the simulated fluid data to obtain a data comparison result. The determination module 340 is used to determine abnormal areas from the fluid pipeline network based on the data comparison result.

[0094] In some optional embodiments, the fluid data includes multiple types; the comparison module 330 can be further used to: compare the measured fluid data corresponding to the target type at the same data acquisition point in the fluid pipeline network with the simulated fluid data of the target type to obtain the data comparison error value corresponding to the target type at the data acquisition point; determine the preset error value corresponding to the target type at the data acquisition point; and generate a data comparison result based on the data comparison error value and the preset error value.

[0095] In some alternative embodiments, the fluid network includes multiple valves; the system 300 may further include an abnormal valve determination module 350, which is used to: receive valve opening / closing notification information of a target valve; invoke the fluid network simulation model based on the valve opening / closing notification information; and determine whether the target valve is an abnormal valve based on the fluid network simulation model. To determine whether the target valve is an abnormal valve based on the fluid network simulation model, the abnormal valve determination module 350 may be further used to: calculate the simulated fluid data corresponding to the data acquisition point where the target valve is located using the fluid network simulation model; compare the measured fluid data corresponding to the data acquisition point where the target valve is located with the simulated fluid data to obtain a data comparison result; and determine whether the target valve is an abnormal valve based on the data comparison result.

[0096] In some alternative embodiments, system 300 may further include a leak detection module 360, which is used to: acquire a leak detection network model; process measurement fluid data from one or more data acquisition points and / or one or more simulated fluid data from the fluid network using the leak detection network model, and predict leak information in the fluid network.

[0097] For more information on each module, please refer to [link / reference]. Figure 2 The relevant explanations will not be repeated here. It should be understood that... Figure 3 The systems and modules shown can be implemented in various ways. For example, in some embodiments, the systems and modules can be implemented by hardware, software, or a combination of both. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated-design hardware. Those skilled in the art will understand that the methods and systems described above can be implemented using computer-executable instructions and / or included in the control code of a processor, such as on a media such as a disk, CD, or DVD-ROM, or in the memory of a programmable device. The systems and modules of this specification can be implemented not only by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips and transistors, or programmable hardware devices such as field-programmable gate arrays and programmable logic devices, but also by software, for example, executed by various types of processors, or by a combination of the aforementioned hardware circuits and software (e.g., firmware).

[0098] It should be noted that the above description of the system and its modules is for convenience only and should not be construed as limiting this specification to the embodiments described. It is understood that those skilled in the art, after understanding the principles of this system, may arbitrarily combine the various modules without departing from these principles to form subsystems connected to other modules. Alternatively, some modules may be split to obtain more modules or multiple units under a single module. Such modifications are all within the scope of this specification.

[0099] The beneficial effects that the embodiments of this specification may bring include, but are not limited to: (1) Real-time performance: Edge computing allows processing to be performed where data is generated, rather than transmitting all data to a central server. This improves real-time performance, enabling the system to quickly detect and respond to faults in the gas pipeline network; (2) Reduced latency: Edge computing reduces the latency of data transmission to the cloud server, making the system more agile and helping to make decisions and perform fault repair operations faster; (3) Bandwidth saving: By processing data at the edge, the system can reduce the amount of data transmitted to the cloud, thereby reducing bandwidth requirements and improving system efficiency; (4) Improved security: Edge computing can process and store data locally, reducing the risk of sensitive data being transmitted to the cloud and improving system security; (5) Autonomous decision-making: Edge computing enables the gas pipeline network system to make autonomous decisions locally, and can still perform critical fault diagnosis and repair even if the connection with the cloud is lost; (6) Flexibility: Edge computing systems are generally more flexible and adaptable to different network conditions and environments, thereby improving the robustness of the system; (7) Cost-effectiveness: Edge computing can reduce the dependence on large cloud servers, thereby saving costs, and can improve energy efficiency due to more efficient data transmission and processing. It should be noted that different embodiments may produce different beneficial effects. In different embodiments, the beneficial effects may be any one or a combination of the above, or any other possible beneficial effects.

[0100] The basic concepts have been described above. It is obvious that the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, various modifications, improvements, and corrections may be made to this specification by those skilled in the art. Such modifications, improvements, and corrections are taught in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

Claims

1. A method for monitoring anomalies in a fluid pipeline network, characterized in that, The method is executed by an edge server and includes: The system collects measurement fluid data from one or more data acquisition points in the fluid pipeline network of a preset geographical area using sensors. The simulation model of the fluid network is used to calculate the simulated fluid data corresponding to one or more data acquisition points in the fluid network. The simulation model of the fluid network is constructed based on the network structure of the fluid network and the measured fluid data corresponding to one or more data acquisition points. The measured fluid data and simulated fluid data corresponding to the same data acquisition point in the fluid pipeline network are compared to obtain the data comparison results; Based on the data comparison results, abnormal areas were identified from the fluid pipeline network; Fluid data includes various types; the comparison of measured fluid data and simulated fluid data corresponding to the same data acquisition point in the fluid pipeline network to obtain data comparison results includes: The measured fluid data corresponding to the target type at the same data acquisition point in the fluid pipeline network is compared with the simulated fluid data of the target type to obtain the data comparison error value corresponding to the target type at the data acquisition point; Determine the preset error value corresponding to the target type at the data acquisition point; Based on the data comparison error value and the preset error value, a data comparison result is generated; The fluid pipeline network includes multiple valves; the method further includes: Receive valve opening / closing notification information from the target valve; Based on the valve switch notification information, the fluid pipeline simulation model is invoked; Based on the fluid pipeline network simulation model, determine whether the target valve is an abnormal valve; The step of determining whether the target valve is an abnormal valve based on the fluid pipeline network simulation model includes: Using the fluid network simulation model, the simulated fluid data corresponding to the data acquisition point where the target valve is located is calculated; The measured fluid data corresponding to the data acquisition point where the target valve is located is compared with the simulated fluid data to obtain the data comparison results; Based on the data comparison results, it is determined whether the target valve is an abnormal valve.

2. The fluid pipeline network anomaly monitoring method according to claim 1, characterized in that, The various types include one or more of the following types: temperature type, pressure type, and flow type.

3. The fluid pipeline network anomaly monitoring method according to claim 1, characterized in that, Determining the preset error value corresponding to the target type at the data collection point includes: Obtain the historical data comparison error value corresponding to the target type at the data collection point within the historical time period; Based on the historical data comparison error value, a preset error value corresponding to the target type at the data collection point is determined.

4. The fluid pipeline network anomaly monitoring method according to claim 1, characterized in that, The method further includes: Obtain the leak detection network model; The leak detection network model is used to process measured fluid data from one or more data acquisition points and / or one or more simulated fluid data from the fluid network to predict leak information in the fluid network.

5. The fluid pipeline network anomaly monitoring method according to claim 4, characterized in that, The leakage information includes one or more of the following: the leakage point and the leakage amount corresponding to the leakage point.

6. The fluid pipeline network anomaly monitoring method according to claim 4, characterized in that, The leak detection network model is trained by a cloud computing center. The training data of the leak detection network model includes historical leak information, as well as historical measured fluid data and / or historical simulated fluid data corresponding to the historical leak information. The process of obtaining the leak detection network model includes: obtaining a trained leak detection network model from the cloud computing center.

7. The fluid pipeline network anomaly monitoring method according to any one of claims 1 to 6, characterized in that, It also includes pushing alarm information to the cloud computing center, the alarm information including abnormal areas and / or leakage information in the fluid pipeline network.

8. The fluid pipeline network anomaly monitoring method according to claim 1, characterized in that, The fluid pipeline network includes a gas pipeline network; the data acquisition points are located at one or more of the following nodes in the gas pipeline network: gate station, pressure regulating station, storage and distribution station, diversion pipeline connector, pipeline valve, and client.

9. The fluid pipeline network anomaly monitoring method according to claim 1, characterized in that, The edge server is deployed at one of the one or more data collection points.

10. A fluid pipeline network anomaly monitoring system, wherein the fluid pipeline network anomaly monitoring system is deployed on an edge server, comprising: The acquisition module is used to acquire measurement fluid data corresponding to one or more data acquisition points in the fluid pipeline network of a preset geographical area through sensors; The calculation module is used to calculate the simulated fluid data corresponding to one or more data acquisition points in the fluid pipeline network through a fluid pipeline network simulation model; the fluid pipeline network simulation model is constructed based on the pipeline network structure of the fluid pipeline network and the measured fluid data corresponding to one or more data acquisition points therein; The comparison module is used to compare the measured fluid data and simulated fluid data corresponding to the same data acquisition point in the fluid pipeline network to obtain the data comparison results; The determination module is used to identify abnormal areas from the fluid pipeline network based on the data comparison results; Fluid data includes various types; the comparison module is further used for: The measured fluid data corresponding to the target type at the same data acquisition point in the fluid pipeline network is compared with the simulated fluid data of the target type to obtain the data comparison error value corresponding to the target type at the data acquisition point; Determine the preset error value corresponding to the target type at the data acquisition point; Based on the data comparison error value and the preset error value, a data comparison result is generated; The fluid pipeline network includes multiple valves; the system also includes an abnormal valve detection module, which is used for: Receive valve opening / closing notification information from the target valve; Based on the valve switch notification information, the fluid pipeline simulation model is invoked; Based on the fluid pipeline network simulation model, determine whether the target valve is an abnormal valve; To determine whether the target valve is an abnormal valve based on the fluid network simulation model, the abnormal valve determination module is further configured to: Using the fluid network simulation model, the simulated fluid data corresponding to the data acquisition point where the target valve is located is calculated; The measured fluid data corresponding to the data acquisition point where the target valve is located is compared with the simulated fluid data to obtain the data comparison results; Based on the data comparison results, it is determined whether the target valve is an abnormal valve.

11. The fluid pipeline network anomaly monitoring system according to claim 10, characterized in that, The system also includes a leak detection module, which is used for: Obtain the leak detection network model; The leak detection network model is used to process measured fluid data from one or more data acquisition points and / or one or more simulated fluid data from the fluid network to predict leak information in the fluid network.

12. A computer program product, comprising a computer program or instructions, characterized in that, When the computer program or instructions are executed by a processor, they implement the method described in any one of claims 1-9.

Images