Secondary water supply remote automatic monitoring method and system based on artificial intelligence

Abstract

The invention relates to the technical field of intelligent monitoring, in particular to a secondary water supply remote automatic monitoring method and system based on artificial intelligence, and the method comprises the following steps: monitoring the water volume and pressure parameter change of secondary water supply, drawing a water supply node and water flow path diagram, recognizing the real-time data change of the secondary water supply according to the state of the node, and determining the water flow path diagram. And obtaining a water network dynamic state model. According to the method, the response efficiency is improved, the adaptability and the flexibility of the secondary water supply system are enhanced by utilizing reinforcement learning and continuously adjusting the strategy according to environment feedback, the configuration of a water supply source and a connection path is optimized by applying the hypergraph network model, the connectivity and the elasticity of the network are enhanced, and the method is more stable in the face of structure change or demand fluctuation; according to the dependency relationship of the water quality parameters and the accurate setting of the abnormal threshold value, the water quality monitoring and the pre-judgment of the abnormal state can be more effectively carried out, the economic benefit and the safety are improved, and a powerful technical support is provided for the continuous optimization of the water supply service.

Description

Technical Field

[0001] The present invention relates to the field of intelligent monitoring technology, and in particular to an artificial intelligence-based remote automatic monitoring method and system for secondary water supply. Background Art

[0002] The field of intelligent monitoring technology involves the use of artificial intelligence, machine learning and data analysis to enhance the effectiveness of traditional monitoring systems. It can analyze video, sound and sensor data in real time to achieve automated threat detection, behavior recognition and event prediction. In water resource management, intelligent monitoring technology can optimize water quality monitoring, maintenance and operational efficiency of water supply networks, and remote control of key facilities such as pumps and valves. By integrating advanced computing models and algorithms, intelligent monitoring systems not only improve response speed, but also significantly reduce operating costs and risks through predictive maintenance and fault warning.

[0003] Among them, the artificial intelligence secondary water supply remote automatic monitoring method refers to the use of artificial intelligence technology to automate and remotely monitor the secondary water supply system in cities or residential areas. Its main purpose is to ensure water quality and water supply safety, monitor the pressure, flow and water quality of the water supply system in real time, and promptly discover and resolve abnormal situations in the water supply. It can effectively manage and maintain the water supply network, avoid water quality problems and water supply interruptions, and ensure that residents have access to safe and stable water supply services.

[0004] Although existing intelligent monitoring technologies have integrated artificial intelligence and machine learning technologies, they still have certain limitations in real-time data processing and system adaptability. For example, traditional monitoring systems cannot update information in real time when encountering large amounts of data or rapid changes, which affects the timeliness and accuracy of decision-making. The existing system's ability to adjust network structure and quickly identify water quality anomalies is also relatively limited, resulting in an inability to effectively respond to emergencies in actual operations, increasing the risk of water supply interruptions and affecting the overall efficiency and safety of the water supply system. These shortcomings make it an urgent need to further improve the flexibility and response speed of the system based on existing technologies. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a method and system for remote automatic monitoring of secondary water supply based on artificial intelligence.

[0006] In order to achieve the above object, the present invention adopts the following technical scheme, a remote automatic monitoring method for secondary water supply based on artificial intelligence, comprising the following steps:

[0007] S1: Monitor the changes in water volume and pressure parameters of secondary water supply, draw water supply nodes and water flow path diagrams, identify real-time data changes of secondary water supply according to the status of nodes, and obtain the dynamic state model of the water network;

[0008] S2: through the water network dynamic state model, according to artificial intelligence, using state and reward mapping logic to configure the reinforcement learning environment, initialize the state space and action space, and obtain the reinforcement learning environment configuration;

[0009] S3: Using the reinforcement learning environment configuration and artificial intelligence, conduct decision optimization experiments, adjust strategy parameters according to environmental feedback, continuously optimize decision matching water supply conditions, and generate an optimized water supply strategy;

[0010] S4: Based on the optimized water supply strategy, the water network dynamic state model is a hypergraph, and the water supply source and connection path are simulated through nodes and hyperedges to obtain a hypergraph network model;

[0011] S5: Through the hypergraph network model, water supply remote automatic monitoring is performed, connectivity and network resilience are analyzed, key node identification and path optimization strategies are adopted, the network structure is adjusted, and network criticality analysis results are obtained;

[0012] S6: Analyze the dependency of water quality parameters according to the network criticality analysis results, perform water quality monitoring and abnormal state judgment through conditional probability calculation and abnormal threshold setting, and generate water quality abnormality monitoring results.

[0013] As a further solution of the present invention, the water network dynamic state model includes water volume data, pressure data, node status and water flow path diagram, the reinforcement learning environment configuration includes state definition, action definition and reward logic, the optimized water supply strategy includes pump station control strategy, valve adjustment scheme and emergency response measures, the hypergraph network model includes a hyperedge structure graph composed of water source nodes and connection paths, the network criticality analysis results include key node identification, path optimization scheme and network resilience assessment, and the water quality abnormality monitoring results include abnormal state type, affected area and abnormal response measures.

[0014] As a further solution of the present invention, the steps of monitoring the changes in water volume and pressure parameters of secondary water supply, drawing water supply nodes and water flow path diagrams, identifying real-time data changes of secondary water supply according to the states of the nodes, and obtaining the dynamic state model of the water network are specifically as follows:

[0015] S101: monitor the changes in water volume and pressure parameters in the secondary water supply, continuously record the water flow and pressure value of each detection node, and update the node status data in real time to generate a node status data set;

[0016] S102: Using the node status data set, a network mapping tool is used to draw differentiated nodes and water flow paths in the secondary water supply, and the operating status of each node is updated and displayed in real time to generate a water supply network graphic model;

[0017] S103: According to the water supply network graphical model, the interaction effects between nodes and paths are analyzed in real time, the model display is adjusted according to the changes in water flow and pressure, and the model is checked to be synchronized with the real-time operation to obtain a water network dynamic state model.

[0018] As a further solution of the present invention, through the water network dynamic state model, according to artificial intelligence, the state and reward mapping logic is used to configure the reinforcement learning environment, initialize the state space and action space, and obtain the reinforcement learning environment configuration steps specifically as follows:

[0019] S201: Based on the water network dynamic state model, the initial parameters of reinforcement learning are set through artificial intelligence, the water volume and pressure of the differentiated nodes are automatically monitored, and converted into state variables to generate initialization reinforcement learning parameters;

[0020] S202: Performing real-time data mapping by initializing reinforcement learning parameters, converting node state data into a format recognized by the reinforcement learning algorithm, checking the consistency of valve adjustment with pump station opening and state variables, and generating a state space configuration;

[0021] S203: Adopting the state space configuration, designing the action space and reward function according to the node efficiency and resource optimization requirements, associating the pressure and the adjusted flow with the expected reward, and obtaining the reinforcement learning environment configuration.

[0022] As a further solution of the present invention, the steps of using the reinforcement learning environment configuration and artificial intelligence to conduct decision optimization experiments, adjusting strategy parameters according to environmental feedback, continuously optimizing decision matching water supply conditions, and generating an optimized water supply strategy are specifically as follows:

[0023] S301: Based on the reinforcement learning environment configuration, start a decision optimization experiment, remotely and automatically monitor the secondary water supply, perform a baseline test on the secondary water supply, record initial performance indicators, and generate a baseline decision performance record;

[0024] S302: Based on the baseline decision performance record, collect environmental feedback data through artificial intelligence, adjust the parameters in the baseline decision model, including water allocation and pressure management strategies, optimize resource utilization efficiency, and generate an adjusted decision model;

[0025] S303: Using the adjusted decision model, remotely and automatically monitor the secondary water supply, verify the matching degree between each decision and the real-time water supply conditions through continuous iterative optimization, and obtain the optimized water supply strategy.

[0026] As a further solution of the present invention, based on the optimized water supply strategy, the water network dynamic state model is a hypergraph, and the water supply source and connection path are simulated through nodes and hyperedges. The steps of obtaining the hypergraph network model are specifically as follows:

[0027] S401: Based on the optimized water supply strategy, select key water supply nodes and key connection paths, build an initialized hypergraph framework, encode node and path data into vertices and edges in the hypergraph, and generate an initialized hypergraph structure;

[0028] S402: According to the initialized hypergraph structure, a genetic algorithm is used to optimize the network configuration, and the hyperedge configuration is adjusted by simulating and analyzing the connection efficiency and water flow dynamics between nodes, thereby optimizing the overall performance of the network and generating an adjusted hypergraph model;

[0029] S403: Based on the adjusted hypergraph model, an optimized water supply strategy is applied, the configuration of each node and path is reviewed, the real-time changing water supply demand and water supply efficiency are matched, and a hypergraph network model is obtained.

[0030] As a further solution of the present invention, the formula of the genetic algorithm is as follows:

[0031]

[0032] Among them, f(x) is the fitness function value of the node, w i is the traffic weight of node i, x i is the traffic index of node i, u i is the pressure weight of node i, y i is the pressure index of node i, v i is the connectivity weight of node i, z i is the connectivity index of node i, t i is the response time weight of node i, p i is the response time index of node i, n i is the total number of nodes.

[0033] As a further solution of the present invention, the steps of performing remote automatic monitoring of water supply, analyzing connectivity and network elasticity, adopting key node identification and path optimization strategies, adjusting the network structure, and obtaining network criticality analysis results are specifically as follows:

[0034] S501: Based on the hypergraph network model, remotely monitor the secondary water supply, collect network operation data in real time, identify the distribution of water supply flow and pressure at differentiated nodes, and generate real-time monitoring data;

[0035] S502: Analyze the connectivity and resilience of the entire water supply network using the real-time monitoring data, identify problem areas caused by uneven structure and unstable pressure, and generate connectivity and resilience analysis results;

[0036] S503: According to the connectivity and elasticity analysis results, adopt key node identification and path optimization strategies to adjust the secondary water supply network structure, optimize the efficiency and response speed of the network, and obtain network criticality analysis results.

[0037] As a further solution of the present invention, according to the network criticality analysis results, the dependency of water quality parameters is analyzed, and water quality monitoring and abnormal state judgment are performed through conditional probability calculation and abnormal threshold setting. The steps of generating water quality abnormality monitoring results are specifically as follows:

[0038] S601: Based on the network criticality analysis result, identify the key monitoring nodes of the secondary water supply, analyze the dependency between the water quality parameters of the key monitoring nodes, including pH value, turbidity and chemical oxygen demand, and generate a water quality dependency analysis result;

[0039] S602: According to the water quality dependency analysis result, conditional probability is calculated using a statistical method, and a water quality monitoring model is generated by setting a threshold of an abnormal state and targeting the real-time changes of water quality parameters at key nodes;

[0040] S603: Apply the water quality monitoring model to continuously track water quality changes. If the monitored parameters exceed the set threshold, it is marked as an abnormal state, a warning signal is sent, and the signal sending and receiving conditions are recorded to obtain abnormal water quality monitoring results.

[0041] A secondary water supply remote automatic monitoring system based on artificial intelligence, the secondary water supply remote automatic monitoring system based on artificial intelligence is used to execute the above-mentioned secondary water supply remote automatic monitoring method based on artificial intelligence, the system comprises:

[0042] The state analysis module performs state modeling and dynamic analysis based on the records of water volume, pressure and operation status of water supply nodes of secondary water supply, extracts the trend of water flow and pressure change, and establishes a dynamic model of water network;

[0043] The decision optimization module configures a reinforcement learning environment according to the water network dynamic model, and obtains an optimized water supply strategy by simulating differentiated water supply strategies, adjusting parameters to match water supply demand and optimizing resource allocation;

[0044] The network simulation module uses the optimized water supply strategy to perform hypergraph modeling and network reconstruction, optimizes the network structure through simulation of water supply nodes and connections, and generates a hypergraph network model;

[0045] The path optimization module adopts the hypergraph network model to identify the key nodes and water supply paths of the water supply network, optimizes the speed and efficiency of the secondary water supply through cyclic testing, and obtains the network criticality analysis results;

[0046] The water quality monitoring module performs water quality parameter analysis and anomaly detection based on the network criticality analysis results, sets thresholds and applies conditional probability calculation and detection of water quality anomalies to obtain water quality monitoring results.

[0047] Compared with the prior art, the advantages and positive effects of the present invention are:

[0048] In the present invention, by dynamically monitoring the changes in water volume and pressure parameters and updating the state model of the secondary water supply network in real time, accurate data support is provided to make secondary water supply decisions more scientific and reasonable. The introduction of reinforcement learning allows the system to continuously adjust strategies based on environmental feedback, which not only improves response efficiency, but also enhances the adaptability and flexibility of the secondary water supply system. The application of the hypergraph network model optimizes the configuration of water supply sources and connection paths, enhances the connectivity and elasticity of the network, and makes the secondary water supply system more robust when facing structural changes or demand fluctuations. Based on the dependency of water quality parameters and the precise setting of abnormal thresholds, water quality monitoring and abnormal state prediction can be carried out more effectively, significantly improving the level of water quality safety assurance, significantly reducing operating costs, improving the economic benefits and safety of the system, and providing strong technical support for the continuous optimization of water supply services. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a schematic diagram of the workflow of the present invention;

[0050] Figure 2 This is a detailed flow chart of S1 of the present invention;

[0051] Figure 3 This is a detailed flow chart of S2 of the present invention;

[0052] Figure 4 This is a detailed flow chart of S3 of the present invention;

[0053] Figure 5 This is a detailed flow chart of S4 of the present invention;

[0054] Figure 6 This is a detailed flow chart of S5 of the present invention;

[0055] Figure 7 This is a detailed flow chart of S6 of the present invention;

[0056] Figure 8 It is a system flow chart of the present invention. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0058] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0059] See also Figure 1 The present invention provides a technical solution, a method for remote automatic monitoring of secondary water supply based on artificial intelligence, comprising the following steps:

[0060] S1: Monitor the changes in water volume and pressure parameters of secondary water supply, draw water supply nodes and water flow path diagrams, identify real-time data changes of water supply according to the status of nodes, and obtain the dynamic state model of the water network;

[0061] S2: Through the water network dynamic state model, according to artificial intelligence, the state and reward mapping logic is used to configure the reinforcement learning environment, initialize the state space and action space, record the reward structure of the decision, and obtain the reinforcement learning environment configuration;

[0062] S3: Use reinforcement learning environment configuration and artificial intelligence to conduct decision optimization experiments, adjust strategy parameters according to environmental feedback, continuously optimize decision matching water supply conditions, and generate optimized water supply strategies;

[0063] S4: Based on the optimized water supply strategy, the dynamic state model of the water network is transformed into a hypergraph, the water supply source and connection path are simulated through nodes and hyperedges, the node connection strategy is used to optimize the network layout, and the hypergraph network model is obtained;

[0064] S5: Through the hypergraph network model, remote automatic monitoring of secondary water supply is carried out, connectivity and network resilience are analyzed, key node identification and path optimization strategies are adopted to optimize and adjust the network structure, and network criticality analysis results are obtained;

[0065] S6: According to the results of network criticality analysis, the dependency relationship between water quality parameters is analyzed, and water quality monitoring and abnormal state judgment are performed through conditional probability calculation and abnormal threshold setting to generate water quality abnormality monitoring results.

[0066] The dynamic state model of the water network includes water volume data, pressure data, node status and water flow path diagram. The reinforcement learning environment configuration includes state definition, action definition and reward logic. The optimized water supply strategy includes pump station control strategy, valve adjustment scheme and emergency response measures. The hypergraph network model includes a hyperedge structure graph composed of water source nodes and connection paths. The network criticality analysis results include key node identification, path optimization scheme and network resilience assessment. The water quality abnormality monitoring results include abnormal state type, affected area and abnormal response measures.

[0067] See also Figure 2 , monitor the changes in water volume and pressure parameters of secondary water supply, draw water supply nodes and water flow path diagrams, identify real-time data changes of secondary water supply according to the status of nodes, and obtain the steps of dynamic state model of water network as follows:

[0068] S101: monitor the changes in water volume and pressure parameters in the secondary water supply, continuously record the water flow and pressure value of each detection node, and update the node status data in real time. The execution process of generating the node status data set is as follows;

[0069] Monitor the changes in water volume and pressure parameters in the secondary water supply. By installing sensors at each key node, continuously record the water flow and pressure value of each detection node. Use sensors to collect data in real time to ensure that the status information of all nodes is up to date. Through continuous monitoring, provide real-time and detailed monitoring results for the operation of the secondary water supply system, ensure the efficiency and safety of the secondary water supply system, and generate a node status data set. The formula used is:

[0070]

[0071] Among them, D a represents the node status dataset, F ia is the water flow rate at the ith node, P ia is the pressure value, n i is the total number of nodes, and monitor represents the monitoring function.

[0072] S102: Using the node status data set, a network mapping tool is used to draw differentiated nodes and water flow paths in the secondary water supply, and the operating status of each node is updated and displayed in real time. The execution process of generating a water supply network graphical model is as follows;

[0073] Through the node status data set, the network mapping tool is applied to draw the differentiated nodes and water flow paths in the secondary water supply system. The node status data set is used to update the operating status of each node in real time, and the operating status and potential problems of the water network are more intuitively displayed in a graphical way. It not only helps operation and maintenance personnel to quickly identify the problem, but also provides a practical visual tool for daily management and fault response, and generates a graphical model of the water supply network. The formula used is:

[0074]

[0075] Among them, G b represents the graphical model of the water supply network, S jb is the state data of the ith node, U jb is the updated state of the node, n i is the total number of nodes, and map represents the mapping function.

[0076] S103: According to the water supply network graphical model, the interaction effects between nodes and paths are analyzed in real time, the model display is adjusted according to the changes in water flow and pressure, and the model is checked to be synchronized with the real-time operation. The execution process of the water network dynamic state model is as follows;

[0077] According to the water supply network graphical model, the interaction effects between nodes and paths are analyzed in real time. Through real-time monitoring of water flow and pressure changes, the model display is adjusted to ensure that the model is synchronized with the actual operating status. The purpose is to accurately display the real-time operating status of the water network through dynamic model reflection, enhance the system's responsiveness and the effectiveness of preventive maintenance. The dynamic state model of the water network is generated using the formula:

[0078]

[0079] Among them, H c represents the water network dynamic state model, V kc is the graphical model of the water supply network of the ith node, T kc is the water flow and pressure change at the node, n i is the total number of nodes, and analyze represents the analysis function.

[0080] See also Figure 3 ,Through the dynamic state model of the water network, according to artificial intelligence, the state and reward mapping logic is used to configure the reinforcement learning environment, the state space and action space are initialized, and the steps to obtain the reinforcement learning environment configuration are as follows:

[0081] S201: Based on the dynamic state model of the water network, the initial parameters of reinforcement learning are set through artificial intelligence, the water volume and pressure of the differentiated nodes are automatically monitored, and converted into state variables. The execution process of generating the initialization reinforcement learning parameters is as follows;

[0082] Based on the dynamic state model of the water network, artificial intelligence technology, especially reinforcement learning, is used to set initial parameters to automatically monitor the water volume and pressure of differentiated nodes. The water volume and pressure data are converted into state variables. The variables provide input for the reinforcement learning algorithm, aiming to automate the monitoring and management of the water network, improve the system's adaptability and response speed to changes, and generate initial reinforcement learning parameters. The formula used is:

[0083]

[0084] Among them, P d Represents the initialization of reinforcement learning parameters, Q id is the water volume data of the ith node, R id is the pressure data, n i is the total number of nodes, and initialize represents the initialization function.

[0085] S202: Initializing reinforcement learning parameters, performing real-time data mapping, converting node state data into a format recognized by the reinforcement learning algorithm, checking the consistency of valve adjustment with pump station opening and state variables, and generating the execution process of state space configuration is as follows;

[0086] By initializing reinforcement learning parameters and performing real-time data mapping, the node state data is converted into a format recognized by the reinforcement learning algorithm to ensure that the control decisions of valve adjustment and pump station opening are consistent with the real-time node state variables. The conversion allows the system to respond to changes in network conditions in a more automated way and generate state space configurations using the formula:

[0087]

[0088] Among them, S e represents the state space configuration, T je is the initialization reinforcement learning parameter of the i-th node, U je is the data conversion parameter, n i is the total number of nodes, and map represents the mapping function.

[0089] S203: Adopting state space configuration, designing action space and reward function according to node efficiency and resource optimization requirements, associating pressure and adjustment flow with expected rewards, and obtaining the execution process of reinforcement learning environment configuration is as follows;

[0090] Adopting state space configuration, designing action space and reward function according to node performance and resource optimization requirements, optimizing node operation efficiency by associating pressure adjustment and flow regulation with expected rewards, allowing the system to achieve optimal resource configuration through self-learning and self-adjustment, and obtaining reinforcement learning environment configuration. The formula used is:

[0091]

[0092] Among them, E f represents the reinforcement learning environment configuration, V kf is the state space configuration of the ith node, W kf is the action space and reward function parameters, n i is the total number of nodes, and design represents the design function.

[0093] See also Figure 4 ,Using reinforcement learning environment configuration, using artificial intelligence, to conduct decision optimization experiments, adjust strategy parameters according to environmental feedback, continuously optimize decision matching water supply conditions, and generate the steps of optimizing water supply strategy as follows:

[0094] S301: Based on the reinforcement learning environment configuration, the decision optimization experiment is started, the secondary water supply is remotely and automatically monitored, and a baseline test is performed on the secondary water supply, and the initial performance indicators are recorded. The execution process of generating a baseline decision performance record is as follows;

[0095] Based on the reinforcement learning environment configuration, the decision optimization experiment was started to conduct remote automatic monitoring and baseline testing of the secondary water supply system to record the initial performance indicators of the system. Through the test, the system can capture the water supply performance under non-optimization conditions, provide initial data points for subsequent improvements and adjustments, and form a baseline decision performance record. The formula used is:

[0096]

[0097] Among them, B g represents the baseline decision performance record, P ig is the performance data of the g-th monitoring node, Q ig is the relevant test parameter, n g is the total number of data points, and baseline represents the baseline test function.

[0098] S302: Based on the baseline decision performance record, collect environmental feedback data through artificial intelligence, adjust the parameters in the baseline decision model, including water allocation and pressure management strategies, optimize resource utilization efficiency, and generate the execution process of the adjusted decision model as follows;

[0099] Based on the baseline decision performance record, environmental feedback data is collected through artificial intelligence technology. The data is used to fine-tune the parameters in the baseline decision model, including water allocation and pressure management strategies. Through optimization strategies, resource utilization efficiency is enhanced to ensure the rational allocation of water resources and effective management of pressure. More precisely match the actual operation needs, improve the performance and reliability of the entire water supply system, and generate an adjusted decision model. The formula used is:

[0100]

[0101] Among them, M h represents the adjusted decision model, R jh is the jth baseline decision performance record, S jh is the tuning parameter, n g is the total number of data points, and adjust represents the adjustment function.

[0102] S303: Using the adjusted decision model, remotely and automatically monitor the secondary water supply, verify the matching degree between each decision and the real-time water supply conditions through continuous iterative optimization, and obtain the execution process of the optimized water supply strategy as follows;

[0103] The adjusted decision model is used to remotely and automatically monitor the secondary water supply, and the matching degree of each decision with the real-time water supply conditions is verified through a continuous iterative optimization process. The process is designed to ensure that each adjustment is optimized according to the latest operating data to achieve the best water supply effect and resource utilization. Continuous optimization helps the system gradually reach the ideal operating state and obtain the optimized water supply strategy. The formula used is:

[0104]

[0105] Among them, O k represents the optimized water supply strategy, U kk is the kth adjusted decision model, V kk is real-time monitoring data, p k is the number of iterations, and iterate represents the iterative optimization function.

[0106] See also Figure 5 ,Based on the optimization of water supply strategy, the dynamic state model of the water network is a hypergraph, and the water supply source and connection path are simulated through nodes and hyperedges. The specific steps to obtain the hypergraph network model are:

[0107] S401: Based on the optimized water supply strategy, the key water supply nodes and key connection paths are selected, the initialized hypergraph framework is constructed, the node and path data are encoded into vertices and edges in the hypergraph, and the execution process of generating the initialized hypergraph structure is as follows;

[0108] Based on the optimized water supply strategy, the key water supply nodes and key connection paths in the system are selected to build an initialized hypergraph framework, including encoding node and path data as vertices and edges in the hypergraph. Creating a visual and analyzable graph structure that represents the connection and flow relationship of the water network will help to conduct more in-depth network configuration and performance optimization analysis in the future. The initialization hypergraph structure is generated using the formula:

[0109]

[0110] Among them, H l represents the initialization hypergraph structure, N il is the i-th water supply node, P il is the data of nodes and paths, n g is the total number of data points, and encode represents the encoding function.

[0111] S402: Based on the initialized hypergraph structure, a genetic algorithm is used to optimize the network configuration, and the hyperedge configuration is adjusted to optimize the overall performance of the network by simulating and analyzing the connection efficiency and water flow dynamics between nodes. The execution process of generating the adjusted hypergraph model is as follows;

[0112] Based on the initialized hypergraph structure, the genetic algorithm is used to optimize the network configuration. By simulating and analyzing the connection efficiency and water flow dynamics between nodes, the genetic algorithm can adjust the hyperedge configuration and optimize the overall performance of the network. The system is allowed to reduce energy consumption and potential operational risks while maintaining efficient water flow and pressure management, reflecting the latest state of the network after optimization, providing a basis for further operation and management, and generating an adjusted hypergraph model.

[0113] The formula of genetic algorithm is as follows:

[0114]

[0115] Among them, f(x) is the fitness function value of the node, w i is the traffic weight of node i, x i is the traffic index of node i, u i is the pressure weight of node i, y i is the pressure index of node i, v i is the connectivity weight of node i, z i is the connectivity index of node i, t i is the response time weight of node i, p i is the response time index of node i, n i is the total number of nodes.

[0116] The execution flow is as follows:

[0117] Configure four weights for each node in the network, traffic weight w i , pressure weight u i , connectivity weight v i and the response time weight t i ,The weights are determined through historical data analysis and expert evaluation, based on the four performance indicators of each node, flow x i , pressure i , Connectivityi and response time p i It is monitored and calculated in real time. Using the calculated data, the comprehensive fitness value f(x) of each node is calculated to optimize the network configuration and obtain the adjusted hypergraph model. The weight confirmation method includes data statistical analysis and network performance testing to ensure that each weight reflects the key factors in the actual network operation.

[0118] S403: Based on the adjusted hypergraph model, the optimized water supply strategy is applied, the configuration of each node and path is reviewed, and the water supply demand and water supply efficiency that change in real time are matched. The execution process of obtaining the hypergraph network model is as follows;

[0119] Based on the adjusted hypergraph model, the optimized water supply strategy is applied to review the configuration of each node and path, taking into account the real-time changing water supply demand and water supply efficiency, ensuring that the hypergraph network model is consistent with the current and expected water supply conditions, and the network can adapt to the changing operating conditions to maintain the efficient and reliable operation of the water supply system. The hypergraph network model is obtained using the formula:

[0120]

[0121] Among them, G o represents the hypergraph network model, U ko is the kth adjusted hypergraph model, V ko is the water supply strategy parameter, n g is the total number of data points, and review represents the review function.

[0122] See also Figure 6 ,Through the hypergraph network model, remote automatic monitoring of water supply is conducted, connectivity and network resilience are analyzed, key node identification and path optimization strategies are adopted, and the network structure is adjusted. The specific steps to obtain the results of network criticality analysis are:

[0123] S501: Based on the hypergraph network model, remotely monitor the secondary water supply, collect the network operation data in real time, identify the distribution of water supply flow and pressure at differentiated nodes, and generate the execution process of real-time monitoring data as follows;

[0124] Based on the hypergraph network model, remote monitoring of the secondary water supply system is performed. The process involves real-time collection of network operation data, with special attention to the distribution of water supply flow and pressure at differentiated nodes. Through monitoring, the system can capture the specific performance of each node, ensure the efficiency and stability of the water supply system, provide a basis for further analysis, help to immediately identify and solve problems, and generate real-time monitoring data. The formula used is:

[0125]

[0126] Among them, Rp Indicates real-time monitoring data, F ip is the water supply flow of the ith node, P ip is the pressure value, n i is the total number of nodes, and collect represents the data collection function.

[0127] S502: Using real-time monitoring data, analyze the connectivity and resilience of the entire water supply network, identify problem areas caused by structural unevenness and pressure instability, and generate the execution process of connectivity and resilience analysis results as follows;

[0128] Using real-time monitoring data, the connectivity and resilience of the entire water supply network are analyzed. Through detailed evaluation of data, problem areas caused by uneven structure and unstable pressure are identified. The analysis helps to determine the weaknesses or bottlenecks in the network, guide future maintenance and optimization strategies, provide in-depth insights into the current status of the network, and provide a basis for making improvement measures. The results of the connectivity and resilience analysis are generated using the formula:

[0129]

[0130] Among them, C q represents the results of connectivity and elasticity analysis, T jq is the real-time monitoring data of the i-th node, U jq is the analysis parameter, n i is the total number of nodes, and analyze represents the analysis function.

[0131] S503: According to the results of the connectivity and elasticity analysis, the secondary water supply network structure is adjusted by adopting the key node identification and path optimization strategy to optimize the efficiency and response speed of the network. The execution process of obtaining the network criticality analysis results is as follows;

[0132] According to the results of connectivity and elasticity analysis, the secondary water supply network structure is adjusted by using key node identification and path optimization strategies. By optimizing the network structure, the efficiency and response speed of the network are improved to ensure that the water supply system can more effectively respond to different operational requirements and potential challenges, improve the overall service quality and system reliability, and obtain the network criticality analysis results. The formula used is:

[0133]

[0134] Among them, K r Represents the network criticality analysis result, V kr is the connectivity and elasticity analysis result of the ith node, W kr is the optimization parameter, n i is the total number of nodes, and optimize represents the optimization function.

[0135] See also Figure 7 ,According to the results of network criticality analysis, the dependency of water quality parameters is analyzed, and water quality monitoring and abnormal state judgment are carried out through conditional probability calculation and abnormal threshold setting. The specific steps for generating abnormal water quality monitoring results are as follows:

[0136] S601: Based on the network criticality analysis results, identify the key monitoring nodes of the secondary water supply, analyze the dependency between the water quality parameters of the key monitoring nodes, including pH, turbidity and chemical oxygen demand, and generate the water quality dependency analysis results. The execution process is as follows;

[0137] Based on the results of network criticality analysis, the key monitoring nodes in the secondary water supply system are identified, and the dependencies between the water quality parameters of the nodes are analyzed. The analysis involves parameters such as pH, turbidity, and chemical oxygen demand, which are extremely important for determining the water quality status. By deeply understanding the interactions and dependencies between the parameters, the water quality dependency analysis results are generated, using the formula:

[0138]

[0139] Among them, A s represents the water quality dependency analysis results, P is , Q is and R is are the pH value, turbidity and chemical oxygen demand of the i-th node, n i is the total number of nodes, and analyze represents the analysis function.

[0140] S602: According to the water quality dependency analysis results, conditional probabilities are calculated using statistical methods, and the threshold of abnormal states is set to generate a water quality monitoring model for real-time changes in water quality parameters at key nodes. The execution process is as follows;

[0141] According to the results of water quality dependency analysis, the conditional probability is calculated using statistical methods, and the threshold of abnormal state is set. The purpose is to monitor the real-time changes of water quality parameters at key nodes. By monitoring the data, water quality problems can be predicted and identified, and measures can be taken in advance to ensure water supply safety and quality. A water quality monitoring model is generated, and the formula used is:

[0142]

[0143] Among them, M t represents the water quality monitoring model, C jt is the conditional probability calculation result of the i-th node, D jt is the threshold setting, n i is the total number of nodes, and model represents the model building function.

[0144] S603: Apply the water quality monitoring model to continuously track water quality changes. If the monitored parameters exceed the set threshold, it is marked as an abnormal state, a warning signal is sent, and the signal sending and receiving conditions are recorded. The execution process of obtaining abnormal water quality monitoring results is as follows;

[0145] Apply the water quality monitoring model to continuously track water quality changes. If the monitored parameters exceed the set threshold, they will be immediately marked as abnormal and send an early warning signal. At the same time, the sending and receiving of the signal will be recorded to ensure that all relevant personnel and systems can respond in a timely manner. Continuous monitoring and immediate response mechanisms help to quickly resolve water quality problems, maintain the normal operation of the water supply system, and generate abnormal water quality monitoring results. The formula used is:

[0146]

[0147] Among them, W u Indicates abnormal water quality monitoring results, E ku is the monitoring data of the ith node, F ku is the threshold parameter, n i is the total number of nodes, and track represents the tracking function.

[0148] See also Figure 8 , a secondary water supply remote automatic monitoring system based on artificial intelligence, the secondary water supply remote automatic monitoring system based on artificial intelligence is used to execute the above-mentioned secondary water supply remote automatic monitoring method based on artificial intelligence, the system includes:

[0149] The state analysis module performs state modeling and dynamic analysis based on the records of water volume, pressure and operation status of water supply nodes of secondary water supply, extracts the trend of water flow and pressure change, and establishes a dynamic model of water network;

[0150] The decision optimization module configures a reinforcement learning environment based on the dynamic model of the water network, simulates differentiated water supply strategies, adjusts parameters to match water supply demand and optimizes resource allocation, and obtains the optimal water supply strategy;

[0151] The network simulation module uses the optimized water supply strategy to perform hypergraph modeling and network reconstruction. By simulating water supply nodes and connections, it optimizes the network structure and generates a hypergraph network model.

[0152] The path optimization module uses a hypergraph network model to identify the key nodes and water supply paths of the water supply network. Through cycle testing, it optimizes the speed and efficiency of secondary water supply and obtains network criticality analysis results.

[0153] The water quality monitoring module performs water quality parameter analysis and anomaly detection based on the network criticality analysis results, sets thresholds and applies conditional probability calculation and detection of water quality anomalies to obtain water quality monitoring results.

[0154] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A remote automatic monitoring method for secondary water supply based on artificial intelligence, characterized in that: The following steps are involved: Monitor the changes in water volume and pressure parameters of secondary water supply, draw water supply nodes and water flow path diagrams, identify real-time data changes of secondary water supply according to the status of nodes, and obtain the dynamic state model of the water network; Through the water network dynamic state model, according to artificial intelligence, the state and reward mapping logic is used to configure the reinforcement learning environment, initialize the state space and action space, and obtain the reinforcement learning environment configuration; Using the reinforcement learning environment configuration and artificial intelligence, a decision optimization experiment is conducted to adjust the policy parameters according to environmental feedback, continuously optimize the decision matching water supply conditions, and generate an optimized water supply strategy; Based on the optimized water supply strategy, the dynamic state model of the water network is a hypergraph, and the water supply source and connection path are simulated through nodes and hyperedges to obtain a hypergraph network model; Through the hypergraph network model, water supply remote automatic monitoring is carried out, connectivity and network elasticity are analyzed, key node identification and path optimization strategies are adopted, network structure is adjusted, and network criticality analysis results are obtained; According to the network criticality analysis results, the dependency of water quality parameters is analyzed, and water quality monitoring and abnormal state judgment are performed through conditional probability calculation and abnormal threshold setting to generate water quality abnormality monitoring results.

2. The method for remote automatic monitoring of secondary water supply based on artificial intelligence according to claim 1 is characterized in that: The water network dynamic state model includes water volume data, pressure data, node status and water flow path diagram, the reinforcement learning environment configuration includes state definition, action definition and reward logic, the optimized water supply strategy includes pump station control strategy, valve adjustment scheme and emergency response measures, the hypergraph network model includes a hyperedge structure graph composed of water source nodes and connection paths, the network criticality analysis results include key node identification, path optimization scheme and network resilience assessment, and the water quality abnormality monitoring results include abnormal state type, affected area and abnormal response measures.

3. The method for remote automatic monitoring of secondary water supply based on artificial intelligence according to claim 1 is characterized in that: Monitor the changes in water volume and pressure parameters of secondary water supply, draw water supply nodes and water flow path diagrams, identify real-time data changes of secondary water supply according to the status of nodes, and obtain the steps of dynamic state model of water network as follows: Monitor the changes in water volume and pressure parameters in the secondary water supply, continuously record the water flow and pressure values ​​of each detection node, and update the node status data in real time to generate a node status data set; Through the node status data set, a network mapping tool is used to draw differentiated nodes and water flow paths in the secondary water supply, and the operating status of each node is updated and displayed in real time to generate a water supply network graphical model; According to the water supply network graphical model, the interaction effects between nodes and paths are analyzed in real time, the model display is adjusted according to the changes in water flow and pressure, and the model is checked to be synchronized with the real-time operation to obtain a dynamic state model of the water network.

4. The method for remote automatic monitoring of secondary water supply based on artificial intelligence according to claim 1 is characterized in that: Through the water network dynamic state model, according to artificial intelligence, the state and reward mapping logic is used to configure the reinforcement learning environment, initialize the state space and action space, and obtain the reinforcement learning environment configuration steps as follows: Based on the water network dynamic state model, the initial parameters of reinforcement learning are set through artificial intelligence, the water volume and pressure of the differentiated nodes are automatically monitored, and converted into state variables to generate initialization reinforcement learning parameters; By initializing the reinforcement learning parameters, real-time data mapping is performed to convert the node state data into a format recognized by the reinforcement learning algorithm, check the consistency of valve adjustment with pump station opening and state variables, and generate a state space configuration; Using the state space configuration, the action space and reward function are designed according to the node performance and resource optimization requirements, and the pressure and adjustment flow are associated with the expected reward to obtain the reinforcement learning environment configuration.

5. The method for remote automatic monitoring of secondary water supply based on artificial intelligence according to claim 1 is characterized in that: Using the reinforcement learning environment configuration and artificial intelligence, we conduct decision optimization experiments, adjust strategy parameters according to environmental feedback, continuously optimize decision matching water supply conditions, and generate the optimized water supply strategy in the following steps: Based on the reinforcement learning environment configuration, a decision optimization experiment is initiated to remotely and automatically monitor the secondary water supply, and a baseline test is performed on the secondary water supply, initial performance indicators are recorded, and a baseline decision performance record is generated; Based on the baseline decision performance record, collect environmental feedback data through artificial intelligence, adjust the parameters in the baseline decision model, including water allocation and pressure management strategies, optimize resource utilization efficiency, and generate an adjusted decision model; The adjusted decision model is used to remotely and automatically monitor the secondary water supply, and the matching degree between each decision and the real-time water supply conditions is verified through continuous iterative optimization to obtain an optimized water supply strategy.

6. The method for remote automatic monitoring of secondary water supply based on artificial intelligence according to claim 1 is characterized in that: Based on the optimized water supply strategy, the water network dynamic state model is a hypergraph, and the water supply source and connection path are simulated through nodes and hyperedges. The steps of obtaining the hypergraph network model are as follows: Based on the optimized water supply strategy, key water supply nodes and key connection paths are selected, an initialized hypergraph framework is constructed, node and path data are encoded into vertices and edges in the hypergraph, and an initialized hypergraph structure is generated; According to the initialized hypergraph structure, a genetic algorithm is used to optimize the network configuration, and by simulating and analyzing the connection efficiency and water flow dynamics between nodes, the hyperedge configuration is adjusted to optimize the overall performance of the network, and an adjusted hypergraph model is generated; Based on the adjusted hypergraph model, an optimized water supply strategy is applied, the configuration of each node and path is reviewed, the real-time changing water supply demand and water supply efficiency are matched, and a hypergraph network model is obtained.

7. The artificial intelligence-based remote automatic monitoring method for secondary water supply according to claim 6 is characterized in that: The formula of the genetic algorithm is as follows: Among them, f(x) is the fitness function value of the node, w i is the traffic weight of node i, x i is the traffic index of node i, u i is the pressure weight of node i, y i is the pressure index of node i, v i is the connectivity weight of node i, z i is the connectivity index of node i, t i is the response time weight of node i, p i is the response time index of node i, n i is the total number of nodes.

8. The method for remote automatic monitoring of secondary water supply based on artificial intelligence according to claim 1 is characterized in that: The steps of performing remote automatic monitoring of water supply, analyzing connectivity and network elasticity, using key node identification and path optimization strategies, adjusting the network structure, and obtaining network criticality analysis results through the hypergraph network model are as follows: Based on the hypergraph network model, the secondary water supply is remotely monitored, the operation data of the network is collected in real time, the distribution of water supply flow and pressure at differentiated nodes is identified, and real-time monitoring data is generated; Analyze the connectivity and resilience of the entire water supply network using the real-time monitoring data, identify problem areas caused by structural unevenness and pressure instability, and generate connectivity and resilience analysis results; According to the connectivity and elasticity analysis results, key node identification and path optimization strategies are adopted to adjust the secondary water supply network structure, optimize the efficiency and response speed of the network, and obtain the network criticality analysis results.

9. The method for remote automatic monitoring of secondary water supply based on artificial intelligence according to claim 1 is characterized in that: According to the network criticality analysis results, the dependency of water quality parameters is analyzed, and water quality monitoring and abnormal state judgment are performed through conditional probability calculation and abnormal threshold setting. The specific steps for generating abnormal water quality monitoring results are as follows: Based on the network criticality analysis results, identify the key monitoring nodes of the secondary water supply, analyze the dependencies between the water quality parameters of the key monitoring nodes, including pH, turbidity and chemical oxygen demand, and generate water quality dependency analysis results; According to the water quality dependency analysis results, conditional probabilities are calculated using statistical methods, and a water quality monitoring model is generated based on real-time changes in water quality parameters at key nodes by setting abnormal state thresholds; The water quality monitoring model is applied to continuously track water quality changes. If the monitored parameters exceed the set threshold, it is marked as an abnormal state, an early warning signal is sent, and the signal sending and receiving conditions are recorded to obtain water quality abnormality monitoring results.

10. A remote automatic monitoring system for secondary water supply based on artificial intelligence, characterized in that: According to the artificial intelligence-based remote automatic monitoring method for secondary water supply according to any one of claims 1 to 9, the system comprises: The state analysis module performs state modeling and dynamic analysis based on the records of water volume, pressure and operation status of water supply nodes of secondary water supply, extracts the trend of water flow and pressure change, and establishes a dynamic model of water network; The decision optimization module configures a reinforcement learning environment according to the water network dynamic model, and obtains an optimized water supply strategy by simulating differentiated water supply strategies, adjusting parameters to match water supply demand and optimizing resource allocation; The network simulation module uses the optimized water supply strategy to perform hypergraph modeling and network reconstruction, optimizes the network structure through simulation of water supply nodes and connections, and generates a hypergraph network model; The path optimization module adopts the hypergraph network model to identify the key nodes and water supply paths of the water supply network, optimizes the speed and efficiency of the secondary water supply through cyclic testing, and obtains the network criticality analysis results; The water quality monitoring module performs water quality parameter analysis and anomaly detection based on the network criticality analysis results, sets thresholds and applies conditional probability calculation and detection of water quality anomalies to obtain water quality monitoring results.

Images