Intelligent interconnected detection system based on pressure valve gas transmission characteristics

Abstract

The present application provides an intelligent interconnected detection system based on the gas transmission characteristics of pressure valves, which relates to the field of valve detection technology, including: a gas supply network acquisition module acquires and identifies N pressure valves on the gas supply pipeline network; a gas transmission monitoring module performs gas transmission monitoring on the pressure valves and outputs gas transmission monitoring data; an abnormality detection module performs abnormality detection based on the gas transmission characteristics and outputs the first abnormal pressure valve; a matrix construction module constructs an adjacency matrix based on the first abnormal pressure valve; a fault diffusion identification module constructs a fault diffusion model to perform fault diffusion identification on the adjacency matrix and outputs a predicted diffusion rate; an interconnected pressure valve output module interconnects pressure valves with a diffusion rate greater than or equal to a preset value and outputs an interconnected pressure valve; a fault detection module performs fault detection on the gas supply pipeline network based on the interconnected pressure valves. The present application solves the technical problem that traditional fault detection technology lacks fault diffusion prediction and improves the efficiency and accuracy of pressure valve fault detection.

Description

Technical Field

[0001] The present application relates to the field of valve detection technology, and in particular to an intelligent interconnected detection system based on the gas transmission characteristics of a pressure valve. Background Art

[0002] In modern industrial production, the gas supply pipeline network is one of the key infrastructures supporting various process flows. In the gas supply pipeline network, the pressure valve is a key component. When it fails, it may trigger a series of chain reactions, leading to greater production interruptions or safety accidents. When a pressure valve fails, this failure is not always an isolated incident. It may spread to other pressure valves through the connections in the network, thereby triggering a chain reaction, causing the performance of the entire system to deteriorate or even paralyze. The risk of this fault spreading makes the monitoring and fault detection of pressure valves particularly important.

[0003] In order to effectively manage and maintain the gas supply pipeline network, it is necessary to be able to predict and control the spread of faults. Traditional fault detection and control methods usually only focus on the performance of a single pressure valve, ignoring the mutual influence between pressure valves and the systematic nature of the entire gas supply pipeline network. Secondly, existing fault detection methods usually rely on simple threshold judgments, lack a deep understanding of fault spread and system dynamic behavior, and are unable to predict and prevent potential fault spread. In addition, existing valve fault detection focuses on post-processing rather than prevention, at which time the fault may have spread to other components, easily leading to production interruptions or safety hazards. Summary of the invention

[0004] The present application provides an intelligent interconnected detection system based on the gas transmission characteristics of pressure valves, which solves the technical problems that traditional fault detection technology cannot effectively predict the mutual influence between pressure valves and lacks fault diffusion prediction, resulting in low fault detection accuracy and efficiency, and achieves the technical effect of improving the efficiency and accuracy of pressure valve fault detection in the gas supply pipeline network.

[0005] In view of the above problems, the present application provides an intelligent interconnected detection system based on the gas transmission characteristics of pressure valves, the system comprising: a gas supply network acquisition module, the gas supply network acquisition module is used to acquire the gas supply pipeline network and identify N pressure valves on the gas supply pipeline network; a gas transmission monitoring module, the gas transmission monitoring module is used to perform gas transmission monitoring on the N pressure valves and output N groups of gas transmission monitoring data; an abnormality detection module, the abnormality detection module is used to perform abnormality detection according to the gas transmission characteristics of the N groups of gas transmission monitoring data, and output a first abnormal pressure valve; a matrix construction module, the matrix construction module is used to construct an adjacency matrix based on the first abnormal pressure valve; a fault diffusion identification module, the fault diffusion identification module is used to construct a fault diffusion model, perform fault diffusion identification on the adjacency matrix based on the fault diffusion model, and output a predicted diffusion rate corresponding to the adjacency matrix; an interconnected pressure valve output module, the interconnected pressure valve output module is used to interconnect the pressure valves whose predicted diffusion rates corresponding to the adjacency matrix are greater than or equal to a preset diffusion rate, and output an interconnected pressure valve; a fault detection module, the fault detection module is used to perform fault detection on the gas supply pipeline network according to the interconnected pressure valves.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The gas supply network acquisition module acquires the gas supply pipeline network, identifies the N pressure valves on the gas supply pipeline network, and provides basic data for subsequent monitoring and analysis, ensuring that the system can cover all key nodes and achieve comprehensive monitoring. The gas transmission monitoring module monitors the gas transmission of the N pressure valves, outputs N groups of gas transmission monitoring data, and provides real-time data streams so that the system can continuously track the working status of the pressure valves and capture any abnormal changes in time. The anomaly detection module performs anomaly detection based on the gas transmission characteristics of the N groups of gas transmission monitoring data, outputs the first abnormal pressure valve, quickly locates the source of the problem, provides guidance for troubleshooting, and reduces the troubleshooting time. The matrix construction module constructs an adjacency matrix based on the first abnormal pressure valve, creates an intuitive network relationship map, and facilitates understanding of the mutual influence and fault propagation path between the pressure valves. The fault diffusion identification module constructs a fault diffusion model, performs fault diffusion identification on the adjacency matrix based on the fault diffusion model, outputs the predicted diffusion rate corresponding to the adjacency matrix, predicts and quantifies the impact of the fault, and provides a basis for preventive maintenance. The interconnected pressure valve output module interconnects the pressure valves whose predicted diffusion rates are greater than or equal to the preset diffusion rates corresponding to the adjacency matrix, outputs the interconnected pressure valves, and determines the pressure valves that need special attention to prevent the spread of faults. The fault detection module performs fault detection on the gas supply pipeline network according to the interconnected pressure valves, and promptly discovers and handles potential problems.

[0008] In summary, this application integrates advanced sensor technology, data analysis and artificial intelligence algorithms to build a fault diffusion model, realize real-time monitoring and intelligent analysis of the pressure valve status, and conduct interconnection analysis on all pressure valves based on the preset diffusion detection rate, thereby defining pressure valves with potential interconnection risks. In this way, the system can accurately complete fault detection and early warning, improve the efficiency and accuracy of pressure valve fault detection in the gas supply pipeline network, predict and prevent potential fault diffusion, and thus improve the reliability and safety of the entire gas supply system.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A schematic diagram of the structure of an intelligent interconnected detection system based on the gas transmission characteristics of a pressure valve provided in an embodiment of the present application;

[0011] Figure 2 A schematic diagram of a process for constructing a fault diffusion model in an intelligent interconnected detection system based on the gas transmission characteristics of a pressure valve provided in an embodiment of the present application;

[0012] Figure 3 A schematic diagram of a flow chart for fault diffusion identification in an intelligent interconnected detection system based on pressure valve gas transmission characteristics provided in an embodiment of the present application.

[0013] Explanation of the reference numerals: gas supply network acquisition module 10 , gas transmission monitoring module 20 , anomaly detection module 30 , matrix construction module 40 , fault diffusion identification module 50 , interconnected pressure valve output module 60 , fault detection module 70 . DETAILED DESCRIPTION

[0014] The embodiment of the present application provides an intelligent interconnected detection system based on the gas transmission characteristics of pressure valves. By constructing an adjacency matrix and a fault diffusion model, the system clearly displays the relationship between the faulty valve and other valves, quickly identifies abnormal gas transmission characteristics, and predicts the diffusion trend that may be caused by the fault. This solves the technical problems that traditional fault detection technology cannot effectively predict the mutual influence between pressure valves and lacks fault diffusion prediction, resulting in low fault detection accuracy and efficiency, and achieves the technical effect of improving the efficiency and accuracy of pressure valve fault detection in the gas supply pipeline network.

[0015] like Figure 1 As shown, the embodiment of the present application provides an intelligent interconnection detection system based on the gas transmission characteristics of the pressure valve, and the system includes:

[0016] The gas supply network acquisition module 10 is used to acquire a gas supply pipeline network and identify N pressure valves on the gas supply pipeline network.

[0017] Specifically, the gas supply pipeline network refers to a pipeline system connected to multiple pressure valves, which constitute the transmission path of gas from the supply source to the consumer. The pressure valve is a key component for controlling and regulating the flow of gas. It can open, close or partially block the flow of gas to maintain the pressure stability in the system. The gas supply pipeline network can be any gas supply pipeline network that requires valve detection, including N pressure valves, where N is a positive integer and is the number of pressure valves.

[0018] The gas supply network acquisition module 10 uses GIS (Geographic Information System) and field survey data to build a detailed pipeline network map and obtain the structure and layout information of the gas supply pipeline network. Then, by using image recognition, data analysis or other algorithmic technologies, the N pressure valves on the network are automatically identified and their geographical locations and connection relationships are recorded to provide accurate location information for subsequent monitoring work.

[0019] The gas transmission monitoring module 20 is used to perform gas transmission monitoring on the N pressure valves and output N groups of gas transmission monitoring data.

[0020] Specifically, gas transmission monitoring data refers to a series of data sets collected for each pressure valve, including readings of parameters such as pressure, flow, and temperature, which reflect the operating status of each pressure valve. N pressure valves correspond to N sets of gas transmission monitoring data.

[0021] The gas transmission monitoring module 20 continuously monitors the gas transmission of N pressure valves and tracks the working status of the pressure valves in real time. First, a series of sensors, including pressure sensors, flow meters, and temperature sensors, are installed on each pressure valve. These sensors are responsible for collecting data in real time, and then transmit the collected data to the gas transmission monitoring module 20 through wireless communication technologies such as Wi-Fi or LoRaWAN for aggregation to generate N groups of gas transmission monitoring data. The gas transmission monitoring module 20 transmits these data to the anomaly detection module 30 to achieve real-time monitoring of each pressure valve and provide detailed data support for subsequent anomaly detection and fault analysis.

[0022] The abnormality detection module 30 is used to perform abnormality detection according to the gas transmission characteristics of the N groups of gas transmission monitoring data and output a first abnormal pressure valve.

[0023] Specifically, the gas transmission characteristics refer to the various parameters and characteristics exhibited by the pressure valve during the gas transmission process under normal working conditions. The first abnormal pressure valve refers to the first pressure valve detected to exhibit abnormal behavior. The abnormal detection module 30 receives data from the gas transmission monitoring module 20 and analyzes the data using statistical methods or machine learning algorithms. The module establishes a data baseline under normal operating conditions, which is a normal working mode obtained by analyzing historical data. First, the module collects normal operating data over a period of time, including historical records of parameters such as pressure, flow, and temperature of the pressure valve. For example, the module may collect data for a year to cover various seasons and operating conditions. The collected data is cleaned to remove obvious errors or missing values ​​to ensure the accuracy of the analysis. This may include removing obvious errors or missing values. Next, the module extracts useful features from these historical data. For example, statistical features such as maximum, minimum, average, and variance can be extracted from the pressure curve. Using these features, these features are analyzed by statistical methods, such as calculating the mean and standard deviation, or machine learning methods, such as K-means clustering, can be used to determine the typical characteristics of normal operation. Establish a baseline for normal operation. Then, the real-time data is compared with this baseline to identify any deviations from the normal range. If the data from a particular pressure valve is significantly different than expected, it is flagged as an anomaly. This process can use specific algorithms to identify anomalies. For example, the Isolation Forest algorithm can identify data points that are significantly different from the majority of data points, while the Adaptive Neuro-Fuzzy Inference Engine can take into account multiple factors such as pressure, flow, and temperature to determine whether an anomaly has occurred.

[0024] Through the above process, the abnormality detection module 30 can realize comprehensive monitoring of the pressure valves on the gas supply pipeline network, timely discover and output abnormal pressure valves, and provide important information support for subsequent fault handling and maintenance.

[0025] A matrix construction module 40 is used to construct an adjacency matrix based on the first abnormal pressure valve.

[0026] Specifically, the adjacency matrix is ​​a mathematical structure used to represent the relationship between each pressure valve in the gas supply pipeline network. In the gas supply network, the adjacency matrix can show which pressure valves are directly connected and the strength of the connection between them. The matrix construction module 40 uses the network topology diagram to determine the connection mode between the pressure valves. First, the module uses GIS software and field measurement data to determine the physical location and connection relationship information of the first abnormal pressure valve output by the abnormality detection module 30. Then, a matrix is ​​constructed based on these connection relationship information, in which each element represents the direct connectivity between the two pressure valves. If there is a direct connection between the two pressure valves, the corresponding position in the matrix will be marked as 1; if there is no direct connection, it will be marked as 0. In this way, the matrix construction module 40 can clearly show the relationship between the pressure valves and provide an accurate information basis for subsequent fault diffusion identification and fault detection.

[0027] Furthermore, the matrix construction module 40 described in the embodiment of the present application is also used to perform the following steps:

[0028] Obtain a sub-gas supply pipeline network corresponding to the first abnormal pressure valve, wherein the sub-gas supply pipeline network is a sub-network of the gas supply pipeline network, and the gas supply pipeline network includes multiple sub-gas supply pipeline networks, and the supply properties of each sub-gas supply pipeline network are different; according to the sub-gas supply pipeline network corresponding to the first abnormal pressure valve, construct an adjacency matrix based on the first abnormal pressure valve, and the elements of the adjacency matrix represent the connection strength between the two pressure valves.

[0029] Specifically, a sub-gas supply pipeline network refers to a part of the gas supply pipeline network, which contains specific pressure valves and the connections between them. Each sub-network has unique supply properties. In the gas supply network, these sub-networks may represent an area or a specific application scenario, such as a residential area or an industrial area. Supply properties refer to the characteristics of the sub-gas supply pipeline network, which determine how the network distributes and controls gas flow. Supply properties include but are not limited to flow capacity, pipeline diameter, pipeline material, pipeline length, connection type, geographical location, and service objects.

[0030] The matrix construction module 40 first determines the sub-gas supply pipeline network associated with the first abnormal pressure valve by analyzing the location of the first abnormal pressure valve and its upstream and downstream pressure valves. During implementation, the module can use GIS software and field measurement data to determine the exact locations of these pressure valves and the connection paths between them. Then, an adjacency matrix is ​​constructed based on these connection paths, and the rows and columns of the matrix represent the pressure valves in the sub-gas supply pipeline network, respectively, and each element in the matrix represents the connection strength between two pressure valves. If there is a direct connection between the two pressure valves, and this connection is very important for gas supply, then the corresponding position in the matrix may be assigned a higher value. If the connection is not important, the value may be lower.

[0031] Through the above process, the matrix construction module 40 can realize the comprehensive construction of the connection relationship between the pressure valves in the gas supply pipeline network, and provide an accurate information basis for subsequent fault diffusion identification and fault detection.

[0032] The fault diffusion identification module 50 is used to construct a fault diffusion model, perform fault diffusion identification on the adjacency matrix based on the fault diffusion model, and output a predicted diffusion rate corresponding to the adjacency matrix.

[0033] Specifically, the fault diffusion model is a mathematical model used to simulate and predict the diffusion process of faults in a pressure valve network. The predicted diffusion rate refers to the speed and scale at which the model estimates that the fault may spread.

[0034] The fault diffusion identification module 50 first constructs a fault diffusion model, such as the SIR model (susceptible-infected-removed model). Then, the model uses the adjacency matrix generated by the matrix construction module 40 as input to analyze the impact of the abnormal pressure valve on the pressure valve directly connected to it, and predict how this impact will further spread to the entire sub-gas supply pipeline network. The model will consider the importance, connection strength and other supply attributes of each pressure valve to estimate the possibility and speed of fault diffusion, and output the predicted diffusion rate corresponding to the adjacency matrix. Through the above process, the fault diffusion identification module 50 can provide a prediction framework to simulate and predict the fault diffusion process in the gas supply pipeline network, and provide support for the subsequent determination of interconnected pressure valves.

[0035] Further, such as Figure 2 As shown, the fault diffusion identification module 50 described in the embodiment of the present application is also used to perform the following steps:

[0036] Initialize the SIR model architecture; define model parameters according to the initialized SIR model architecture, wherein the model parameters include the number of pressure valves in the first state, the number of pressure valves in the second state, and the number of pressure valves in the third state, as well as a fault transfer coefficient and a fault recovery coefficient; obtain pressure valve fault history data, use the pressure valve fault history data as a training data set to learn the model parameters of the initialized SIR model architecture, and obtain a fault diffusion model.

[0037] Preferably, in the fault diffusion identification module 50 described in the embodiment of the present application, the number of pressure valves in the first state is the number of pressure valves in a fault-susceptible state, the number of pressure valves in the second state is the number of pressure valves in a fault-infected state, and the number of pressure valves in the third state is the number of pressure valves in a fault-removed state.

[0038] Specifically, the SIR model is an epidemiological model that is often used to simulate the spread of diseases and is also used to simulate the spread of faults in a system. The SIR model divides individuals in the system into three categories: susceptible, infected, and removed. In the context of fault spread, the susceptible state means that the pressure valve has not been affected but may be affected, the infected state means that the pressure valve has failed, and the removed state means that the pressure valve has recovered from the failure or has been replaced.

[0039] The fault diffusion identification module 50 will first initialize a SIR model architecture and define the SIR model parameters, namely the number of pressure valves in the first state, the number of pressure valves in the second state, the number of pressure valves in the third state, the fault transfer coefficient and the fault recovery coefficient. Among them, the number of pressure valves in the first state is the number of pressure valves in the fault-susceptible state, the number of pressure valves in the second state is the number of pressure valves in the fault-infected state, and the number of pressure valves in the third state is the number of pressure valves in the fault removal state. The fault transfer coefficient is a parameter used in the model to measure the rate at which a fault propagates from one pressure valve to another. This coefficient can be calculated from historical data, such as by analyzing the average propagation time after the fault occurs and the number of pressure valves affected. The fault recovery coefficient is a parameter used in the model to measure the rate at which a pressure valve recovers from a fault state to a normal state. This coefficient can also be calculated from historical data, such as by analyzing the time required for fault repair.

[0040] The module then uses historical fault data to train the model so that it can predict how faults spread in the network. The specific process includes: First, collect historical fault data of the gas pipeline network, including the time, location, scope of impact, and type and severity of each fault. This data can be extracted from maintenance logs, sensor records, and other relevant records. For example, to analyze a city's gas supply network, data can be collected from past maintenance reports to record which pressure valves have failed before, as well as the frequency and duration of these failures. Then, the historical data is input into the model, and the historical data of pressure valve failures is used as a training data set to adjust the model parameters until the model can accurately reflect the actual fault propagation. For example, if historical records show that it takes an average of two days for a pressure valve to propagate to adjacent pressure valves after a pressure valve fails, the fault transfer coefficient can be set to 0.5, that is, about half of the related pressure valves will be affected every day. Once the model is trained, it can be used to predict future fault propagation. If a new fault occurs, the model can predict which pressure valves may be affected and give a diffusion rate.

[0041] Further, such as Figure 3 As shown, the fault diffusion identification module 50 described in the embodiment of the present application is also used to perform the following steps:

[0042] Obtain a current state vector of each pressure valve in the adjacency matrix, wherein the state vector is used to represent the current state of each pressure valve between the interval [0,1], where 0 represents normal and 1 represents fault; analyze each pressure valve in the adjacency matrix based on the fault diffusion model to obtain an updated state vector; and output a predicted diffusion rate corresponding to each pressure valve in the adjacency matrix according to the updated state vector.

[0043] Specifically, the state vector is a numerical vector used to represent the current state of each pressure valve, where 0 represents normal and 1 represents fault. The updated state vector refers to the updated state vector of the pressure valve after the fault diffusion model analysis. The predicted diffusion rate refers to the predicted diffusion rate of the fault in the gas supply pipeline network based on the updated state vector.

[0044] The fault diffusion identification module 50 first obtains the state vector of each pressure valve from the adjacency matrix. These state vectors represent the current state of each pressure valve between the interval [0,1]. Then, the fault diffusion model is used to analyze these state vectors. The model considers the relationship between each pressure valve and other pressure valves and factors such as the fault transfer coefficient to predict how the fault propagates from one pressure valve to another, and updates the state vector by iterative calculation output. For example, if we know that a pressure valve has failed, the model may predict that its neighboring pressure valve has a 50% probability of also failing, depending on the strength of the connection between them. Finally, based on the updated state vector, the predicted diffusion rate is output.

[0045] Preferably, the fault diffusion identification module 50 described in the embodiment of the present application is further configured to perform the following steps:

[0046] The time step is set, and the values ​​of the number of pressure valves in the first state, the number of pressure valves in the second state, and the number of pressure valves in the third state at each time step are iteratively calculated based on the Euler algorithm; the state of each pressure valve in the adjacency matrix is ​​updated according to the values ​​of the number of pressure valves in the first state, the number of pressure valves in the second state, and the number of pressure valves in the third state at each time step, and an updated state vector is obtained.

[0047] Preferably, the expression based on the iterative calculation of the Euler algorithm is as follows:

[0048]

[0049] Among them, S(t), I(t), and R(t) are the number of pressure valves in the first state, the number of pressure valves in the second state, and the number of pressure valves in the third state corresponding to time t, respectively, β is the fault transfer coefficient, γ is the fault recovery coefficient, S(t+Δt), I(t+Δt), and R(t+Δt) are the number of pressure valves in the first state, the number of pressure valves in the second state, and the number of pressure valves in the third state corresponding to time t+Δt, respectively, and Δt is the time step.

[0050] Specifically, the time step is a unit of time interval used to track the change of fault status over time. In each time step, the model updates the state of the pressure valve. The Euler algorithm is an iterative algorithm for numerically solving differential equations and is used to simulate the propagation of faults in the pressure valve network.

[0051] Based on the fault diffusion model, each pressure valve in the adjacency matrix is ​​analyzed, specifically including: first, a time step Δt is set according to the prediction accuracy requirement. Then, the Euler algorithm is used to simulate the propagation of the fault in each time step. For each time step, the number of pressure valves in the susceptible, infected, and removed states is updated, that is, the number of pressure valves in the first state, the number of pressure valves in the second state, and the number of pressure valves in the third state. For example, if a pressure valve fails in the current time step, it will transfer from the susceptible state to the infected state. The iterative calculation expression is as follows:

[0052]

[0053] Among them, S(t), I(t), and R(t) are the number of pressure valves in the first state, the number of pressure valves in the second state, and the number of pressure valves in the third state corresponding to time t, respectively, β is the fault transfer coefficient, γ is the fault recovery coefficient, S(t+Δt), I(t+Δt), and R(t+Δt) are the number of pressure valves in the first state, the number of pressure valves in the second state, and the number of pressure valves in the third state corresponding to time t+Δt, respectively, and Δt is the time step.

[0054] Then, the state vector in the adjacency matrix is ​​updated according to these quantities, that is, the state of the corresponding pressure valve in the state vector is changed. The updated state vectors are output, which represent the current state of each pressure valve and the prediction of fault propagation. For example, if pressure valve A fails in the current time step, its state will change from 0 to 1. Repeat this process, each time considering the latest state information and fault diffusion model to more accurately predict the propagation of faults.

[0055] For example, there is a simple gas supply network with three pressure valves A, B and C, which form a ring connection. Assume that at the initial time t = 0, pressure valve A fails, while B and C work normally. Therefore, the state vector is [1,0,0], indicating that A fails, and B and C are normal. Set the time step to 1 hour. The fault transfer coefficient is 0.2, that is, there is a 20% probability of a fault valve being transmitted to its neighbor valve every hour. The fault recovery coefficient is set to 0.1, which means that there is a 10% probability of the fault valve returning to normal every hour. First iteration: At t = 1, the fault of A may be transmitted to B and C, but it may also be partially recovered due to the fault recovery coefficient. Assume that part of the fault of A is transmitted to B and C, and A has a 10% probability of recovery. The updated state vector is [0.9, 0.2, 0.2], indicating that A has a 90% probability of still being in a faulty state, while B and C each have a 20% probability of failing. Second iteration: At t = 2, the Euler algorithm is applied again. A may further affect B and C, while having a 10% recovery rate. Assuming that A transfers 20% of the faults to B and C, while recovering 10% itself, the updated state vector is [0.89, 0.4, 0.4], indicating that A has an 89% probability of still being in a faulty state, while B and C each have a 40% probability of failing. Continue this process until a stable state is reached, or the maximum number of iterations we preset is reached. Finally, a probability vector representing the state of each pressure valve is obtained to help predict the spread of faults.

[0056] Through the above process, the fault spreading identification module 50 can provide a dynamic prediction of how the fault spreads throughout the network.

[0057] The interconnected pressure valve output module 60 is used to interconnect the pressure valves whose predicted diffusion rates corresponding to the adjacency matrix are greater than or equal to the preset diffusion rates, and output the interconnected pressure valves.

[0058] Specifically, interconnected pressure valves are those pressure valves predicted in the fault diffusion model that are likely to be affected by the fault due to the connection strength with the abnormal pressure valve. The preset diffusion rate is a pre-set threshold used to determine whether the predicted diffusion rate exceeds the safety range. If the predicted diffusion rate is higher than this threshold, then these pressure valves are considered to be interconnected.

[0059] The interconnected pressure valve output module 60 receives the output result of the fault diffusion model, determines the predicted diffusion rate of each pressure valve corresponding to the adjacency matrix, compares it with the preset diffusion rate threshold, selects those pressure valves whose predicted diffusion rate exceeds the preset threshold, marks them, and outputs the interconnected pressure valves. Through the above process, the interconnected pressure valve output module 60 can quickly identify the key pressure valves that need attention, and provide guidance information for subsequent fault detection and maintenance.

[0060] A fault detection module 70 is used to perform fault detection on the gas supply pipeline network according to the interconnected pressure valve.

[0061] Specifically, the fault detection module 70 uses sensor data, historical records and other relevant information to analyze the working condition of the interconnected pressure valve output by the interconnected pressure valve output module 60 to determine whether a fault has occurred. The fault detection module 70 collects data of the interconnected pressure valve in real time through sensors installed on the pipeline, including pressure, temperature and flow. Then, the module uses data analysis tools, such as machine learning algorithms, statistical models, or external fault diagnosis software to analyze these data and look for abnormal patterns. If any abnormality is found, such as a drop in pressure or a change in flow, the module will issue an alarm, indicating that there may be a fault. For example, in a city gas supply network, if the data of the interconnected pressure valve B shows a sudden drop in pressure, the fault detection module 70 may determine that this is a fault signal and notify the operator to investigate. Through this process, the fault detection module 70 can achieve timely detection and early warning of potential faults in the gas supply pipeline network, provide important information support for maintenance personnel, and ensure the safe and stable operation of the gas supply pipeline network.

[0062] Furthermore, the fault detection module 70 described in the embodiment of the present application is also used to perform the following steps:

[0063] The fault detection ratio of the interconnected pressure valves in the gas supply pipeline network is obtained; if the fault detection ratio is greater than a preset ratio, a first warning signal is obtained, and the first warning signal is sent to a host computer for reminder.

[0064] Specifically, the fault detection ratio refers to the proportion of interconnected pressure valves in the gas supply pipeline network, which reflects the importance of interconnected pressure valves in the overall network. If the number or proportion of interconnected pressure valves is high, the vulnerability of the network will also increase relatively. The first warning signal refers to the warning signal issued by the system when the fault detection ratio exceeds the preset ratio.

[0065] The fault detection module 70 first obtains the number of interconnected pressure valves and the total number of all pressure valves in the gas supply pipeline network, and calculates the proportion of interconnected pressure valves in the gas supply pipeline network. Then, it is determined whether this proportion is greater than a preset threshold. If this proportion exceeds the preset threshold, that is, the fault detection proportion is greater than the preset proportion, it is considered that the gas supply pipeline network is facing a greater risk, a first early warning signal is generated, and the early warning signal is sent to the host computer using a network communication tool. The host computer usually refers to the central control system responsible for monitoring the entire system. The host computer will display this early warning to remind the operator that immediate action may be needed. Through this process, the fault detection module 70 can monitor and warn of potential fault risks in the gas supply pipeline network to ensure the safe operation of the system.

[0066] In summary, the intelligent interconnected detection system based on the gas transmission characteristics of the pressure valve provided in the embodiment of the present application has the following technical effects:

[0067] The gas supply network acquisition module 10 acquires the gas supply pipeline network, identifies the N pressure valves on the gas supply pipeline network, and provides basic data for subsequent monitoring and analysis, ensuring that the system can cover all key nodes and achieve comprehensive monitoring. The gas transmission monitoring module 20 monitors the gas transmission of the N pressure valves, outputs N groups of gas transmission monitoring data, and provides real-time data streams so that the system can continuously track the working status of the pressure valves and capture any abnormal changes in time. The anomaly detection module 30 performs anomaly detection based on the gas transmission characteristics of the N groups of gas transmission monitoring data, outputs the first abnormal pressure valve, quickly locates the source of the problem, provides guidance for troubleshooting, and reduces troubleshooting time. The matrix construction module 40 constructs an adjacency matrix based on the first abnormal pressure valve, creates an intuitive network relationship map, and facilitates understanding of the mutual influence and fault propagation path between the pressure valves. The fault diffusion identification module 50 constructs a fault diffusion model, identifies fault diffusion on the adjacency matrix based on the fault diffusion model, outputs the predicted diffusion rate corresponding to the adjacency matrix, uses the SIR model to predict and quantify the impact of the fault, provides a basis for preventive maintenance, and evaluates the state of each pressure valve and its possible impact on other valves by constructing an adjacency matrix and using mathematical tools such as the Euler algorithm. The interconnected pressure valve output module 60 interconnects the pressure valves whose predicted diffusion rates corresponding to the adjacency matrix are greater than or equal to the preset diffusion rate, outputs the interconnected pressure valves, and determines the pressure valves that need special attention to prevent the spread of faults. The fault detection module 70 performs fault detection on the gas supply pipeline network based on the interconnected pressure valves. When it is detected that the proportion of faults that may be caused by the interconnected pressure valves exceeds the preset threshold, an early warning signal is issued to remind the operator to take timely action to avoid a larger range of faults.

[0068] In general, the embodiment of the present application integrates advanced sensor technology, data analysis and artificial intelligence algorithms to build a fault diffusion model, realize real-time monitoring and intelligent analysis of the pressure valve status, and conduct interconnection analysis on all pressure valves based on the preset diffusion detection rate, thereby defining pressure valves with potential interconnection risks. In this way, the system can accurately complete fault detection and early warning, improve the efficiency and accuracy of pressure valve fault detection in the gas supply pipeline network, predict and prevent potential fault diffusion, and thus improve the reliability and safety of the entire gas supply system.

[0069] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An intelligent interconnected detection system based on the gas transmission characteristics of pressure valves, characterized in that: The system comprises: A gas supply network acquisition module, the gas supply network acquisition module is used to acquire a gas supply pipeline network and identify N pressure valves on the gas supply pipeline network; A gas transmission monitoring module, the gas transmission monitoring module is used to monitor the gas transmission of the N pressure valves and output N groups of gas transmission monitoring data; An abnormality detection module, the abnormality detection module is used to perform abnormality detection according to the gas transmission characteristics of the N groups of gas transmission monitoring data, and output a first abnormal pressure valve; A matrix construction module, wherein the matrix construction module is used to construct an adjacency matrix based on the first abnormal pressure valve; A fault diffusion identification module, the fault diffusion identification module is used to construct a fault diffusion model, perform fault diffusion identification on the adjacency matrix based on the fault diffusion model, and output a predicted diffusion rate corresponding to the adjacency matrix; An interconnected pressure valve output module, the interconnected pressure valve output module is used to interconnect the pressure valves whose predicted diffusion rates corresponding to the adjacency matrix are greater than or equal to a preset diffusion rate, and output the interconnected pressure valves; a fault detection module, the fault detection module being used to perform fault detection on the gas supply pipeline network according to the interconnected pressure valve; The fault diffusion identification module is further configured to perform the following steps: Initialize the SIR model architecture; Defining model parameters according to the initialized SIR model architecture, the model parameters including the number of first state pressure valves, the number of second state pressure valves, the number of third state pressure valves, and a fault transfer coefficient and a fault recovery coefficient; The pressure valve fault history data is obtained, and the model parameters of the initialized SIR model architecture are learned by using the pressure valve fault history data as a training data set to obtain a fault diffusion model.

2. The system according to claim 1, characterized in that The fault diffusion identification module is further configured to perform the following steps: Obtaining a current state vector of each pressure valve in the adjacency matrix, wherein the state vector is used to represent the current state of each pressure valve in the interval [0, 1], where 0 represents normal and 1 represents fault; Analyzing each pressure valve in the adjacency matrix based on the fault diffusion model to obtain an updated state vector; According to the updated state vector, the predicted diffusion rate corresponding to each pressure valve in the adjacency matrix is ​​output.

3. The system according to claim 2, characterized in that The fault diffusion identification module is further configured to perform the following steps: The time step is set, and the number of pressure valves in the first state, the number of pressure valves in the second state, and the number of pressure valves in the third state are iteratively calculated based on the Euler algorithm at each time step; The states of the pressure valves in the adjacency matrix are updated according to the values ​​of the number of pressure valves in the first state, the number of pressure valves in the second state, and the number of pressure valves in the third state at each time step to obtain an updated state vector.

4. The system according to claim 3, characterized in that The expression for iterative calculation based on Euler algorithm is as follows: Among them, S(t), I(t), and R(t) are the number of pressure valves in the first state, the number of pressure valves in the second state, and the number of pressure valves in the third state corresponding to time t, respectively, β is the fault transfer coefficient, γ is the fault recovery coefficient, S(t+Δt), I(t+Δt), and R(t+Δt) are the number of pressure valves in the first state, the number of pressure valves in the second state, and the number of pressure valves in the third state corresponding to time t+Δt, respectively, and Δt is the time step.

5. The system according to claim 1, wherein: The number of pressure valves in the first state is the number of pressure valves in a fault-susceptible state, the number of pressure valves in the second state is the number of pressure valves in a fault-infected state, and the number of pressure valves in the third state is the number of pressure valves in a fault-removed state.

6. The system according to claim 1, wherein: The fault detection module is also used to perform the following steps: Obtaining a fault detection ratio of the interconnected pressure valve to the gas supply pipeline network; If the fault detection ratio is greater than a preset ratio, a first warning signal is obtained, and the first warning signal is sent to a host computer for reminder.

7. The system according to claim 1, characterized in that The matrix building module is also used to perform the following steps: Acquire a sub-gas supply pipeline network corresponding to the first abnormal pressure valve, wherein the sub-gas supply pipeline network is a sub-network of the gas supply pipeline network, and the gas supply pipeline network includes a plurality of sub-gas supply pipeline networks, and each sub-gas supply pipeline network has different supply attributes; According to the sub-gas supply pipeline network corresponding to the first abnormal pressure valve, an adjacency matrix based on the first abnormal pressure valve is constructed, and the elements of the adjacency matrix represent the connection strength between the two pressure valves.

Images