Hydropower station safety management method and device based on sensor network

Abstract

The invention discloses a hydropower station safety management method and device based on a sensor network, and the method comprises the steps: constructing a digital twin model of a hydropower station, simulating different directed sensor node layout schemes, and determining an optimal putting strategy; randomly putting a plurality of directed sensor nodes in the to-be-monitored area of the hydropower station according to the optimal putting strategy to obtain an initial arrangement scheme; optimizing the initial arrangement scheme based on a genetic algorithm to obtain a target arrangement scheme; activating directed sensor nodes in the target arrangement scheme; acquiring current operation information through the activated directed sensor node; constructing a multi-dimensional equipment knowledge graph and predicting the operation trend of the hydropower station equipment in combination with the current operation information; and identifying a potential fault chain based on the operation trend and carrying out hydropower station safety early warning and safety decision. According to the invention, the sensor network is introduced to acquire the current operation information of the equipment for safety early warning and decision making, so that the safety management efficiency and accuracy of the hydropower station are effectively improved.

Description

Technical Field

[0001] This application relates to the technical field of safety management, and particularly to a method and device for hydropower station safety management based on a sensor network. Background Art

[0002] With the continuous development of modern technologies, hydropower stations, as an important source of clean energy, play a crucial role in terms of safety and stability in the power system. Work in aspects such as equipment monitoring, fault detection and prediction, and maintenance management of hydropower stations has always been the core content to ensure the normal operation of hydropower stations. However, due to the complex operating environment of hydropower station equipment and the large variety of equipment, single traditional monitoring means often struggle to comprehensively and real-time capture the operating status and potential fault information of the equipment.

[0003] Existing hydropower station monitoring systems are mostly based on fixed sensor layouts. These systems usually rely on preset sensor points for data collection. However, this layout method has many deficiencies. Firstly, the fixed layout may not cover all key areas, resulting in monitoring blind spots for certain equipment or areas, affecting the comprehensiveness of monitoring. Secondly, the fixed nature of sensor arrangement makes the distribution of sensors inflexible and difficult to adjust according to real-time needs, affecting the monitoring efficiency. Thirdly, existing sensor network monitoring systems usually lack an intelligent optimization mechanism, resulting in insufficient consideration of changes in equipment operating status and identification of potential faults when arranging sensors. It can be seen that there are problems such as fixed sensor layouts, monitoring blind spots, and insufficient optimization in the prior art for hydropower stations, and real-time and intelligent safety management of hydropower station equipment cannot be achieved. Summary of the Invention

[0004] The main objective of this application is to provide a method and device for hydropower station safety management based on a sensor network, aiming to solve the technical problems in the prior art that there are problems such as fixed sensor layouts, monitoring blind spots, and insufficient optimization in hydropower stations, and real-time and intelligent safety management of hydropower station equipment cannot be achieved.

[0005] To achieve the above objective, this application proposes a method for hydropower station safety management based on a sensor network. The method for hydropower station safety management based on a sensor network includes:

[0006] Construct a digital twin model of the hydropower station, and based on the digital twin model, simulate different directed sensor node layout schemes to determine the optimal placement strategy;

[0007] Randomly place a number of directed sensor nodes in the area to be monitored in the hydropower station according to the optimal placement strategy to obtain an initial layout plan of the directed sensor nodes, where the directed sensor nodes are in a sleep state;

[0008] Optimize the initial layout scheme of the directed sensor nodes based on the genetic algorithm to obtain the target layout scheme of the directed sensor nodes;

[0009] Activate the directed sensor nodes in the target layout scheme to obtain the activated directed sensor nodes;

[0010] Collect the current operation information of the hydropower station equipment through the activated directed sensor nodes;

[0011] Construct a multi-dimensional equipment knowledge graph and predict the operation trend of the hydropower station equipment based on the current operation information of the hydropower station equipment and the multi-dimensional equipment knowledge graph;

[0012] Identify potential fault chains based on the operation trend of the hydropower station equipment and conduct safety early warning and safety decision-making for the hydropower station according to the potential fault chains.

[0013] In one embodiment, the construction of the digital twin model of the hydropower station and the determination of the optimal placement strategy based on simulating different layout schemes of the directed sensor nodes by the digital twin model include:

[0014] Construct a geometric model of the hydropower station, where the geometric model includes the structural layout, equipment configuration, and spatial relationship of the hydropower station;

[0015] Conduct physical modeling based on the hydrodynamic model, equipment working principle model, and coupling relationship between equipment to obtain the physical model of the hydropower station;

[0016] Associate the geometric model and the physical model to construct the digital twin model of the hydropower station;

[0017] Add virtual sensors to each device and component in the digital twin model, and set the simulation area, initial position, and attributes of the virtual sensor nodes;

[0018] Simulate the sensing process of the directed sensor nodes based on the simulation area, initial position, and attributes of the virtual sensor nodes to obtain multiple layout schemes of the directed sensor nodes;

[0019] Calculate the coverage rate and target detection rate under each layout scheme of the directed sensor nodes;

[0020] Determine the optimal placement strategy according to the coverage rate and the target detection rate.

[0021] In one embodiment, the optimization of the initial layout scheme of the directed sensor nodes based on the genetic algorithm to obtain the target layout scheme of the directed sensor nodes includes:

[0022] Take the initial layout scheme of the directed sensor nodes as the initial population, where each individual in the initial population is a layout scheme of the directed sensor nodes, and the layout scheme of the directed sensor nodes includes the positions and directions of the directed sensor nodes;

[0023] Define a fitness function and calculate the fitness value of each individual according to the proportion of the critical area covered, the total energy consumption of the sensors, and the overlapping coverage area between the sensors;

[0024] Select target individuals from the initial population according to the fitness values, and perform crossover operations on the target individuals to generate new individuals;

[0025] Perform mutation operations on the new individuals to obtain mutated individuals, and add the mutated individuals to the initial population to generate a new population;

[0026] Judge whether the new population meets the termination condition;

[0027] If it meets the condition, take the individual with the highest fitness value in the new population as the target layout scheme of the directed sensor nodes;

[0028] If it does not meet the condition, re-execute the step of calculating the fitness value of each individual according to the proportion of the critical area covered, the total energy consumption of the sensors, and the overlapping coverage area between the sensors until the termination condition is met, where the termination condition is reaching a preset number of iterations or the fitness values of the individuals in the population tend to be stable.

[0029] In one embodiment, the construction of the multi-dimensional device knowledge graph and the prediction of the operation trend of the hydropower station equipment based on the current operation information of the hydropower station equipment and the multi-dimensional device knowledge graph include:

[0030] Construct an equipment failure knowledge base based on the equipment failure modes, historical safety events, and equipment maintenance records of the hydropower station;

[0031] Parse and learn the equipment failure knowledge base, and extract equipment failure characteristics, failure causes, and failure solutions;

[0032] Construct a multi-dimensional device knowledge graph based on the extracted equipment failure characteristics, failure causes, and failure solutions, where the multi-dimensional device knowledge graph includes equipment failure nodes, failure characteristic nodes, failure cause nodes, and failure solution nodes, and the nodes are connected by relationship edges;

[0033] Match the current operation information with the multi-dimensional device knowledge graph to identify the equipment failure characteristics corresponding to the current operation information;

[0034] Predict the operation trend of the hydropower station equipment according to the equipment fault characteristics corresponding to the current operation information, wherein the operation trend of the hydropower station equipment includes the performance change trend, the fault occurrence probability trend, and the remaining service life trend.

[0035] In one embodiment, the predicting the operation trend of the hydropower station equipment according to the equipment fault characteristics corresponding to the current operation information includes:

[0036] Input the equipment fault characteristics corresponding to the current operation information into the operation trend prediction model, and use the operation trend prediction model to predict the performance change trend, the fault occurrence probability trend, and the remaining service life trend of the hydropower station equipment based on the equipment fault characteristics;

[0037] Calculate the operation trend of the hydropower station equipment according to the performance change trend, the fault occurrence probability trend, and the remaining service life trend.

[0038] In one embodiment, the activating the directed sensor nodes in the target layout plan to obtain the activated directed sensor nodes includes:

[0039] Select any node from the directed sensor nodes in the target layout plan as the initial node, and activate the initial node to obtain the activated initial node;

[0040] Calculate the number of hydropower station equipment that can be covered by the activated initial node in each perspective;

[0041] Sort the priorities of each perspective from high to low according to the number of hydropower station equipment that can be covered to obtain the sorting result;

[0042] Take the perspective with the highest priority in the sorting result as the target perspective, and adjust the sensing direction of the activated initial node so that the sensing direction of the activated initial node faces the target perspective;

[0043] Search within a preset range centered on the activated initial node to check if there are any other unactivated directed sensor nodes;

[0044] If there are unactivated directed sensor nodes, execute the step of selecting any node from the directed sensor nodes in the target layout plan as the initial node and activating the initial node to obtain the activated initial node, until all the directed sensor nodes in the area to be monitored of the hydropower station are activated to obtain the activated directed sensor nodes.

[0045] In one embodiment, after activating the directed sensor nodes in the area to be monitored of the hydropower station to obtain the activated directed sensor nodes, it further includes:

[0046] Obtain the current perspective of the activated directional sensor node;

[0047] Based on the current perspective of the activated directional sensor node, determine the total number of hydropower station devices that can be covered;

[0048] Calculate the coverage balance coefficient based on the total number of hydropower station devices that can be covered;

[0049] Evaluate the coverage efficiency of the activated directional sensor node based on the coverage balance coefficient to obtain a coverage efficiency evaluation result;

[0050] Optimize and adjust the distribution position of the activated directional sensor node according to the coverage efficiency evaluation result.

[0051] In one embodiment, the identifying potential fault chains based on the operation trend of the hydropower station devices and performing hydropower station safety warning and safety decision-making according to the potential fault chains includes:

[0052] Based on the operation trend of the hydropower station devices, use a fault chain identification algorithm to analyze the causal relationship and propagation path between device faults, and construct potential fault chains;

[0053] Perform a risk assessment on the potential fault chains, calculate the risk value of the potential fault chains, where the risk value is used to represent the degree of influence of the potential fault chains on the safe operation of the hydropower station;

[0054] Sort the potential fault chains according to the risk value to obtain a risk ranking result;

[0055] Take the potential fault chain with the highest risk value in the risk ranking result as the key potential fault chain;

[0056] Perform hydropower station safety warning based on the key potential fault chain to generate safety warning information, where the safety warning information includes a warning level, warning content, and safety measures;

[0057] Make a hydropower station safety decision according to the safety warning information.

[0058] In one embodiment, the constructing potential fault chains by using a fault chain identification algorithm to analyze the causal relationship and propagation path between device faults based on the operation trend of the hydropower station devices includes:

[0059] Extract device fault events and occurrence times based on the operation data of hydropower station devices;

[0060] Analyze the time correlation and logical correlation between device fault events according to the device fault events and occurrence times, and determine the causal relationship between device faults;

[0061] Construct a device fault propagation network according to the causal relationship, where the nodes in the device fault propagation network represent device faults, and the connection edges between the nodes represent the propagation paths between the faults;

[0062] Analyze the activation probability and propagation speed of the fault nodes in the device fault propagation network based on the operation trend of the hydropower station equipment;

[0063] Identify potential fault propagation paths in the device fault propagation network according to the activation probability and propagation speed to obtain potential fault chains.

[0064] In addition, to achieve the above object, the present application also proposes a hydropower station safety management device based on a sensor network, and the hydropower station safety management device based on the sensor network includes:

[0065] A simulation module, configured to construct a digital twin model of the hydropower station, and simulate different directed sensor node layout schemes based on the digital twin model to determine an optimal placement strategy;

[0066] A placement module, configured to randomly place a plurality of directed sensor nodes in the area to be monitored of the hydropower station according to the optimal placement strategy to obtain an initial layout scheme of the directed sensor nodes, where the directed sensor nodes are in a sleep state;

[0067] An optimization module, configured to optimize the initial layout scheme of the directed sensor nodes based on a genetic algorithm to obtain a target layout scheme of the directed sensor nodes;

[0068] An activation module, configured to activate the directed sensor nodes in the target layout scheme to obtain activated directed sensor nodes;

[0069] An acquisition module, configured to acquire the current operation information of the hydropower station equipment through the activated directed sensor nodes;

[0070] A prediction module, configured to construct a multi-dimensional device knowledge graph, and predict the operation trend of the hydropower station equipment based on the current operation information of the hydropower station equipment and the multi-dimensional device knowledge graph;

[0071] A decision-making module, configured to identify potential fault chains based on the operation trend of the hydropower station equipment, and perform hydropower station safety early warning and safety decision-making according to the potential fault chains.

[0072] In this application, a digital twin model of a hydropower station is constructed, and different directed sensor node layout schemes are simulated based on the digital twin model to determine the optimal placement strategy. A number of directed sensor nodes are randomly placed in the area to be monitored in the hydropower station according to the optimal placement strategy to obtain an initial layout scheme of the directed sensor nodes, where the directed sensor nodes are in a dormant state. The initial layout scheme of the directed sensor nodes is optimized based on a genetic algorithm to obtain a target layout scheme of the directed sensor nodes. The directed sensor nodes in the target layout scheme are activated to obtain the activated directed sensor nodes. The current operating information of the hydropower station equipment is collected through the activated directed sensor nodes. A multi-dimensional equipment knowledge graph is constructed, and the operating trend of the hydropower station equipment is predicted based on the current operating information of the hydropower station equipment and the multi-dimensional equipment knowledge graph. Potential fault chains are identified based on the operating trend of the hydropower station equipment, and safety warnings and safety decisions for the hydropower station are made according to the potential fault chains. In the above manner, by introducing a sensor network to collect the current operating information of the hydropower station equipment, and then combining with a knowledge graph to predict the operating trend, potential fault chains are identified and safety warnings and safety decisions are made, effectively improving the safety management efficiency and accuracy of the hydropower station and ensuring the equipment safety of the hydropower station. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] The accompanying drawings herein are incorporated into and constitute a part of this specification, showing embodiments consistent with the present application and, together with the specification, are used to explain the principles of the present application.

[0074] To more clearly illustrate the technical solutions in the embodiments of the present application or in the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0075] Figure 1 It is a schematic flowchart provided for Embodiment 1 of the method for hydropower station safety management based on a sensor network according to the present application;

[0076] Figure 2 It is a schematic flowchart provided for Embodiment 2 of the method for hydropower station safety management based on a sensor network according to the present application;

[0077] Figure 3 It is a schematic diagram of the module structure of the device for hydropower station safety management based on a sensor network according to the embodiments of the present application.

[0078] The realization of the purpose, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0079] It should be understood that the specific embodiments described herein are merely used to explain the technical solutions of the present application and are not used to limit the present application.

[0080] To better understand the technical solutions of the present application, the following will be described in detail in conjunction with the accompanying drawings of the specification and specific implementation manners.

[0081] The main solution of the embodiments of the present application is as follows: constructing a digital twin model of a hydropower station, simulating different directed sensor node layout schemes based on the digital twin model, and determining an optimal placement strategy; randomly placing a number of directed sensor nodes in the area to be monitored of the hydropower station according to the optimal placement strategy to obtain an initial layout scheme of the directed sensor nodes, where the directed sensor nodes are in a dormant state; optimizing the initial layout scheme of the directed sensor nodes based on a genetic algorithm to obtain a target layout scheme of the directed sensor nodes; activating the directed sensor nodes in the target layout scheme to obtain the activated directed sensor nodes; collecting the current operating information of the hydropower station equipment through the activated directed sensor nodes; constructing a multi-dimensional equipment knowledge graph, and predicting the operating trend of the hydropower station equipment based on the current operating information of the hydropower station equipment and the multi-dimensional equipment knowledge graph; identifying potential fault chains based on the operating trend of the hydropower station equipment, and performing safety early warning and safety decision-making for the hydropower station according to the potential fault chains.

[0082] Existing hydropower station monitoring systems are mostly based on fixed sensor layouts, and these systems usually rely on preset sensor points for data collection. However, this layout method has many deficiencies. First, the fixed layout may not cover all key areas, resulting in monitoring blind spots for some equipment or areas, affecting the comprehensiveness of monitoring; second, the fixity of sensor arrangement makes the distribution of sensors inflexible and difficult to adjust according to real-time requirements, affecting the monitoring efficiency; third, existing sensor network monitoring systems usually lack an intelligent optimization mechanism, resulting in insufficient consideration of changes in equipment operating states and identification of potential faults when arranging sensors. It can be seen that there are problems such as fixed hydropower station sensor layouts, monitoring blind spots, and insufficient optimization in the prior art, and real-time and intelligent safety management of hydropower station equipment cannot be achieved.

[0083] The present application provides a solution. By introducing a sensor network to collect the current operating information of hydropower station equipment, and then combining with a knowledge graph for operating trend prediction, potential fault chains are identified and safety early warning and safety decision-making are performed, effectively improving the safety management efficiency and accuracy of hydropower stations and ensuring the equipment safety of hydropower stations.

[0084] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication, and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or a hydropower station safety management device based on a sensor network that can implement the above functions. Hereinafter, taking the hydropower station safety management device based on a sensor network as the execution subject as an example, this embodiment and the following embodiments will be described.

[0085] Based on this, an embodiment of the present application provides a hydropower station safety management method based on a sensor network. Refer to Figure 1 , Figure 1 It is a schematic flowchart of the first embodiment of the hydropower station safety management method based on a sensor network of the present application.

[0086] In this embodiment, the hydropower station safety management method based on a sensor network includes steps S10 to S70:

[0087] Step S10: Construct a digital twin model of the hydropower station, and based on the digital twin model, simulate different directed sensor node layout schemes to determine the optimal placement strategy.

[0088] It should be noted that the digital twin model of the hydropower station needs to accurately describe the physical and operational characteristics of the hydropower station, including the dynamic behavior of equipment, the operating environment, external factors (such as water flow, weather conditions, etc.). It is a model for virtual simulation of the hydropower station entity and can reflect the actual operating conditions and equipment layout of the hydropower station. By simulating different directed sensor node layout schemes, the data collection effects of sensor nodes under different layouts can be simulated, so as to determine the optimal placement strategy of sensor nodes.

[0089] In a feasible implementation manner, step S10 may include steps A11 to A17:

[0090] Step A11: Construct a geometric model of the hydropower station, where the geometric model includes the structural layout, equipment configuration, and spatial relationship of the hydropower station.

[0091] It should be noted that the geometric model of the hydropower station can be constructed by three-dimensional modeling software or computer-aided design (CAD) tools, accurately reflecting the actual structural layout, equipment configuration, and spatial relationship between components of the hydropower station.

[0092] Step A12: Perform physical modeling based on the hydrodynamic model, equipment working principle model, and coupling relationship between equipment to obtain the physical model of the hydropower station.

[0093] It should be noted that the hydrodynamic model is used to simulate the motion state and energy conversion process of water flow in a hydropower station. For example, a computational fluid dynamics (CFD) - based hydrodynamic model. The equipment working principle model is used to describe the operating mechanisms and performance parameters of various equipment in a hydropower station. For example, the dynamic models of turbines, generators, etc. The coupling relationship between equipment reflects the ways in which equipment affects and interacts with each other. For example, power transmission, heat conduction, etc.

[0094] Specifically, construct the dynamic models of hydropower station equipment to simulate the dynamic behavior of equipment during operation, such as changes in parameters like rotational speed, torque, temperature, etc.; construct the interaction models between hydropower station equipment to simulate the mutual influence and interaction between equipment, such as the impact force of water flow on the turbine, the interaction between the generator and the power grid, etc.; based on the geometric model and physical model, construct the digital twin model of the hydropower station to achieve the mapping and simulation of the hydropower station entity in a virtual environment.

[0095] Step A13: Associate the geometric model and the physical model to construct the digital twin model of the hydropower station.

[0096] It should be noted that by combining the spatial relationships in the geometric model with the physical processes in the physical model, the digital twin model can accurately reflect the actual operating conditions of the hydropower station, including integrating geometric information such as the position, shape, and size of equipment with physical information such as the dynamic behavior, performance parameters, and coupling relationships between equipment to form a complete and interactive virtual hydropower station model. Through this model, real - time monitoring and simulation analysis of the operating state of the hydropower station can be achieved.

[0097] It is worth noting that the digital twin model not only includes the geometric layout and equipment configuration of the hydropower station, but also integrates multi - dimensional information such as the dynamic behavior of equipment, operating environment, and external factors, and can comprehensively and accurately reflect the actual operating conditions of the hydropower station.

[0098] Specifically, the modeling process of the digital twin model includes sensor modeling and environment modeling. Among them, sensor modeling includes sensor types, distribution locations, data acquisition frequencies, etc., and simultaneously simulates the response and error of sensors under different working conditions. Environment modeling is used to simulate the impact of environmental factors in the hydropower station, such as water flow, climate change, load fluctuations, etc. on equipment.

[0099] Step A14: Add virtual sensors to each device and component in the digital twin model, and set the simulation area, initial positions, and attributes of virtual sensor nodes.

[0100] It should be noted that in the virtual environment, these virtual sensors can simulate the working mode of real sensors and collect the operation data of hydropower station equipment. By setting the initial positions and attributes of the simulation area and virtual sensor nodes, different sensor layout schemes can be simulated.

[0101] Step A15: Based on the simulation area, the initial positions and attributes of the virtual sensor nodes, simulate the sensing process of the directional sensor nodes to obtain multiple directional sensor node layout schemes.

[0102] It should be noted that the simulation of the sensing process of the directional sensor nodes takes into account the directionality of the sensor nodes, the sensing range, and the influence of environmental factors such as obstacles and water flow velocity, so as to generate multiple feasible sensor node layout schemes.

[0103] Step A16: Calculate the coverage rate and target detection rate under each of the directional sensor node layout schemes.

[0104] It should be noted that the coverage rate refers to the proportion of the area that can be monitored by the sensor nodes in the entire monitoring area, reflecting the monitoring ability of the sensor nodes; the target detection rate refers to the proportion of the sensor nodes that can accurately identify and report faults or abnormal events, reflecting the detection accuracy of the sensor nodes. By calculating and analyzing the coverage rate and target detection rate under different layout schemes, the advantages and disadvantages of each layout scheme can be evaluated, so as to determine the optimal sensor node placement strategy.

[0105] Step A17: Determine the optimal placement strategy according to the coverage rate and the target detection rate.

[0106] It should be noted that the optimal placement strategy refers to the scheme that selects the least number of sensor nodes and the most reasonable layout on the premise of ensuring high coverage rate and target detection rate, so as to reduce the implementation cost and complexity. This optimal placement strategy not only considers the physical layout of the sensor nodes, but also comprehensively considers the actual operation status of the hydropower station equipment, environmental factors, and monitoring requirements, ensuring the effectiveness and reliability of the monitoring system.

[0107] It should be understood that through the comprehensive evaluation of the coverage rate and the target detection rate, a sensor node layout scheme that can not only ensure the full coverage of the monitoring area but also ensure the accurate detection of faults or abnormal events can be selected, and the optimal placement strategy ensures the effectiveness and reliability of the sensor network. When implementing the optimal placement strategy, factors such as the number, type, and distribution of sensor nodes also need to be considered to further optimize the performance of the sensor network and improve the efficiency and accuracy of hydropower station safety management.

[0108] Step S20: Randomly deploy a number of directed sensor nodes in the area to be monitored of the hydropower station according to the optimal deployment strategy, obtaining an initial layout plan of the directed sensor nodes, where the directed sensor nodes are in a dormant state.

[0109] It should be noted that when deploying the directed sensor nodes, it is necessary to ensure that the nodes are distributed according to the preset optimal deployment strategy to maximize the coverage of the monitoring area and accurately detect potential faults or abnormal events. These sensor nodes are in the sleep mode in the initial state to save energy and extend the service life. When needed, the activation process of the nodes can be triggered by remote commands or preset conditions.

[0110] It is worth noting that a number of directed sensor nodes are randomly deployed in the area to be monitored of the hydropower station according to the optimal deployment strategy. The deployment positions and quantities of the nodes are carefully calculated and optimized to ensure the effective coverage of the entire monitoring area and the accuracy of fault detection. At this time, the layout of the directed sensor nodes is the initial layout plan. Although the sensor nodes in the initial layout plan are in a dormant state, their positions, directions and attributes can be further optimized and adjusted to ensure that they can quickly and accurately collect the operation information of the hydropower station equipment after activation and achieve the best monitoring effect.

[0111] Step S30: Optimize the initial layout plan of the directed sensor nodes based on the genetic algorithm to obtain the target layout plan of the directed sensor nodes.

[0112] It should be noted that the genetic algorithm iteratively optimizes the positions, directions and attributes of the sensor nodes in the initial layout plan. By simulating natural selection and genetic mechanisms, it gradually screens out the optimal layout plan of the sensor nodes. During the optimization process, the genetic algorithm will consider various factors, such as the geometric characteristics of the monitoring area, the sensing capabilities of the sensor nodes, and the influence of environmental factors on the sensor performance, to ensure that the final obtained target layout plan can maximize the coverage rate and target detection rate of the monitoring system. Through the optimization of the genetic algorithm, a more accurate and efficient layout plan of the sensor nodes can be obtained, providing more reliable monitoring data support for the safety management of the hydropower station.

[0113] In a feasible implementation manner, step S30 may include: using the initial layout scheme of the directed sensor nodes as the initial population, where each individual in the initial population is a layout scheme of the directed sensor nodes, and the layout scheme of the directed sensor nodes includes the positions and directions of the directed sensor nodes; defining a fitness function, and calculating the fitness value of each individual according to the proportion of the critical area covered, the total energy consumption of the sensors, and the overlapping coverage area between the sensors; selecting target individuals from the initial population according to the fitness values, and performing crossover operations on the target individuals to generate new individuals; performing mutation operations on the new individuals to obtain mutated individuals, and adding the mutated individuals to the initial population to generate a new population; determining whether the new population meets the termination condition; if it meets, using the individual with the highest fitness value in the new population as the target layout scheme of the directed sensor nodes; if it does not meet, re-executing the step of calculating the fitness value of each individual according to the proportion of the critical area covered, the total energy consumption of the sensors, and the overlapping coverage area between the sensors until the termination condition is met, where the termination condition is reaching a preset number of iterations or the fitness values of the individuals in the population tend to be stable.

[0114] It should be noted that each individual in the initial population represents a possible layout scheme of the sensor nodes, and these schemes evolve continuously during the iteration process of the genetic algorithm to adapt to the actual requirements of the monitoring area.

[0115] By defining the fitness function, the advantages and disadvantages of each layout scheme can be quantified. This function comprehensively considers multiple factors such as the proportion of the critical area covered, the total energy consumption of the sensors, and the overlapping coverage area between the sensors, so as to ensure that the finally obtained layout scheme is both efficient and economical. The proportion of the critical area covered refers to the proportion of the important area effectively covered by the sensor nodes in the monitoring area, and it is an important indicator for evaluating the coverage ability of the monitoring system. Increasing the proportion of the critical area can ensure that the important equipment and areas of the hydropower station are fully monitored, so as to timely discover and handle potential safety hazards. The total energy consumption of the sensors reflects the operating cost of the monitoring system. By optimizing the layout of the sensor nodes, the energy consumption of the system can be reduced while ensuring the monitoring effect, and the energy utilization efficiency can be improved. The overlapping coverage area between the sensors refers to the area covered by multiple sensor nodes at the same time. An overly large overlapping coverage area will cause waste of resources, while an overly small overlapping coverage area may leave monitoring blind spots. Therefore, in the optimization process, these three factors need to be comprehensively considered to find the best layout scheme of the sensor nodes. The formula of the fitness function is:

[0116] F = w1 * Coverage - w2 * Energy - w3 * Redundancy

[0117] Among them, F is the fitness value, w1, w2, and w3 are the weight parameters of the proportion of the covered key area, the total energy consumption of the sensors, and the overlapping coverage area between the sensors, Coverage is the proportion of the covered key area, Energy is the total energy consumption of the sensors, and Redundancy is the weight parameter of the overlapping coverage area between the sensors.

[0118] Specifically, in operations such as selection, crossover, and mutation of the genetic algorithm, layout schemes with higher fitness are continuously selected and added to the new population to drive the entire population to evolve in a better direction. When the new population meets the termination conditions, that is, when the preset number of iterations is reached or the fitness values of the individuals in the population tend to be stable, the genetic algorithm stops iterating, and the individual with the highest fitness value is output as the target layout scheme of the directed sensor nodes, which can maximize the coverage rate and target detection rate of the monitoring system. This target layout scheme comprehensively considers the geometric characteristics of the monitoring area, the sensing capabilities of the sensor nodes, and the influence of environmental factors on the performance of the sensors, ensuring the reliability and accuracy of the monitoring system.

[0119] It should be understood that after obtaining the target layout scheme, it can be applied to the actual monitoring of the hydropower station. By activating the corresponding sensor nodes, real-time and accurate monitoring of the hydropower station equipment can be achieved. At the same time, since the sensor nodes are in the sleep mode in the initial state, energy can be greatly saved, the service life of the sensor nodes can be extended, and the operating cost of the monitoring system can be reduced.

[0120] Step S40: Activate the directed sensor nodes in the target layout scheme to obtain the activated directed sensor nodes.

[0121] It should be noted that the activation of the directed sensor nodes refers to switching them from the sleep mode to the working state and starting to perform the monitoring task. This process usually involves sending an activation signal to the sensor nodes. After receiving the signal, the nodes start their internal working mechanisms and begin to collect and process the monitoring data. The activated sensor nodes will perform real-time monitoring of the key equipment and areas of the hydropower station according to the preset monitoring parameters and algorithms.

[0122] In a feasible implementation manner, step S40 may include: selecting any node from the directed sensor nodes of the target layout scheme as the initial node, and activating the initial node to obtain the activated initial node; calculating the number of hydropower station devices that can be covered by the activated initial node from each perspective; performing priority sorting on each perspective according to the number of hydropower station devices that can be covered from high to low to obtain a sorting result; taking the perspective with the highest priority in the sorting result as the target perspective, and adjusting the sensing direction of the activated initial node so that the sensing direction of the activated initial node faces the target perspective; centering on the activated initial node, searching within a preset range to check if there are other unactivated directed sensor nodes; if there are unactivated directed sensor nodes, execute the step of selecting any node from the directed sensor nodes of the target layout scheme as the initial node, and activating the initial node to obtain the activated initial node, until all the directed sensor nodes in the hydropower station area to be monitored are activated, obtaining the activated directed sensor nodes.

[0123] It should be noted that when selecting any node from the target layout scheme as the initial node for activation, the initial node can be selected based on a certain strategy to ensure the maximization of the coverage rate and target detection rate of the monitoring system. For example, a node located at the center of the monitoring area or near key equipment can be selected as the initial node because these nodes usually have a better monitoring view and higher importance. After selecting the initial node, activate it and adjust the sensing direction, and then gradually expand to other nodes until all nodes are activated and in the best monitoring state. The activation of the initial node can be achieved by remote control. For example, an activation instruction is sent to the initial node through a wireless signal. This embodiment does not make specific limitations on this.

[0124] It can be understood that the target perspective refers to the perspective that can provide the largest monitoring coverage area or the monitoring of the most critical equipment in the monitoring area. By performing priority sorting on each perspective and selecting the perspective with the highest priority as the target perspective, the monitoring effect of the sensor node can be maximized. After determining the target perspective, by adjusting the sensing direction of the sensor node to face the target perspective, the precise monitoring of the key equipment and areas of the hydropower station can be achieved.

[0125] Specifically, after activating the initial node, the monitoring coverage range of the node from each perspective can be automatically calculated, and the optimal monitoring direction, i.e., the target perspective, can be determined based on the coverage range. This process can be automatically completed by an algorithm built into the sensor node without manual intervention. After determining the optimal monitoring direction of the initial node, continue to search for and activate other unactivated sensor nodes until all nodes in the entire monitoring area are activated and in the optimal monitoring state. Such an activation strategy not only ensures the comprehensiveness and accuracy of the monitoring system but also greatly improves the monitoring efficiency and reduces the cost and complexity of manual intervention.

[0126] In a feasible implementation manner, after step S40, it may further include: obtaining the current perspective of the activated directional sensor node; determining the total number of hydropower station devices that can be covered based on the current perspective of the activated directional sensor node; calculating a coverage balance coefficient based on the total number of hydropower station devices that can be covered; evaluating the coverage efficiency of the activated directional sensor node based on the coverage balance coefficient to obtain a coverage efficiency evaluation result; and optimizing and adjusting the distribution position of the activated directional sensor node according to the coverage efficiency evaluation result.

[0127] It should be noted that after all nodes in the target layout plan are activated, further analyze the monitoring effectiveness of each activated directional sensor node. The determination of the current perspective is automatically completed by the sensors and algorithms built into the sensor node and can reflect the monitoring direction of the node in real time. Based on this perspective, the number of hydropower station devices that each node can cover can be calculated to evaluate the effectiveness of its monitoring range.

[0128] It is worth noting that the coverage balance coefficient is a comprehensive indicator that takes into account factors such as the coverage overlap degree of each node in the monitoring area, the size of the uncovered area, and the monitoring priority of key devices. By calculating this coefficient, the coverage efficiency of the current monitoring network can be comprehensively evaluated. If the coverage efficiency of a certain node is low, it may be due to its poor position or improper sensing direction.

[0129] It should be understood that after obtaining the coverage efficiency evaluation result, adjust the distribution position of the node according to the preset optimization strategy, including measures such as repositioning the node, adjusting the sensing direction, or adding additional sensor nodes, to ensure the comprehensiveness and accuracy of the monitoring network. The purpose of the optimization and adjustment is to make the monitoring network more efficient and reliable and be able to reflect the safety status of the hydropower station in real time.

[0130] Step S50: Collect the current operating information of the hydropower station devices through the activated directional sensor nodes.

[0131] It should be noted that after activation, the directed sensor nodes will work according to the preset monitoring strategy to collect equipment status information in the monitoring area, including but not limited to key parameters such as vibration, temperature, and pressure of the equipment. The high-precision sensors and advanced signal processing algorithms built into the sensor nodes can capture the changes of these parameters in real time and accurately, and convert them into digital signals for transmission. By continuously collecting and updating equipment status information, the operating status of hydropower station equipment can be monitored in real time, and potential faults or abnormal conditions can be discovered in a timely manner.

[0132] Step S60: construct a multi-dimensional equipment knowledge graph, and predict the operation trend of the hydropower station equipment based on the current operation information of the hydropower station equipment and the multi-dimensional equipment knowledge graph.

[0133] It should be noted that a multidimensional knowledge graph is a data structure that associates and integrates equipment status information with equipment attributes, operating rules and other knowledge. It uses graph theory to represent equipment and its related information in the form of nodes and edges, where nodes represent entities (such as equipment, components, fault types, etc.) and edges represent relationships between entities (such as composition relationships, causal relationships, etc.).

[0134] It is understandable that after obtaining the current operating information of the hydropower station equipment, it is matched and associated with the multi-dimensional equipment knowledge graph, and the operating status of the equipment is analyzed and inferred using the knowledge and rules in the graph, so as to determine the operating trend of the hydropower station equipment. The operating trend of the hydropower station equipment is composed of multiple dimensions such as performance change trend, failure probability trend, and remaining service life trend. These trends can fully reflect the health status and performance changes of the hydropower station equipment.

[0135] It is worth noting that the construction of a multi-dimensional equipment knowledge graph relies on a large amount of historical data, expert experience and advanced machine learning algorithms. It can automatically learn and extract the complex relationship between equipment status information and equipment performance and failures.

[0136] Optionally, when constructing a multidimensional equipment knowledge graph, it is necessary to collect various status information of the hydropower station equipment, including but not limited to the vibration data, temperature data, pressure data, and operation logs of the equipment. Then, use data preprocessing technology to clean, integrate, and format this information. Next, with the help of machine learning algorithms, such as cluster analysis and association rule mining, the potential relationship between equipment status information and equipment performance and faults is mined to construct a knowledge graph of the equipment.

[0137] Optionally, based on the constructed multi-dimensional device knowledge graph, the prediction of the operation trend of hydropower station equipment can be realized. When new equipment status information is collected and input into the knowledge graph, the algorithms in the graph will automatically match and associate this information, analyze the differences and similarities between the current state of the equipment and historical data, so as to predict the future operation trend of the equipment.

[0138] Step S70: Identify potential fault chains based on the operation trend of the hydropower station equipment, and conduct safety early warning and safety decision-making for the hydropower station according to the potential fault chains.

[0139] It should be noted that a potential fault chain refers to the logical relationship between a series of events or conditions that may cause faults in hydropower station equipment. By analyzing the operation trend of hydropower station equipment, potential fault precursors or risk factors can be identified. These precursors or factors are connected in a certain logical order to form a potential fault chain. Each potential fault chain represents a path or pattern of possible equipment failures.

[0140] It can be understood that after identifying the potential fault chain, according to the preset safety rules and thresholds, the severity and urgency of the fault chain are evaluated. If the evaluation result of a certain fault chain exceeds the safety threshold, the system will trigger the safety early warning mechanism, send an alarm to the management personnel in a timely manner, and provide corresponding fault information and treatment suggestions, so that measures can be taken quickly to prevent the occurrence or expansion of faults, ensure the safe operation of the hydropower station, and achieve the safety management of the hydropower station.

[0141] In a feasible implementation manner, step S70 may include steps B11 to B16:

[0142] Step B11: Based on the operation trend of the hydropower station equipment, use the fault chain identification algorithm to analyze the causal relationship and propagation path between equipment failures, and construct potential fault chains.

[0143] It should be noted that the fault chain identification algorithm is a method based on graph theory and probability statistics, which can analyze the internal relationship between equipment failures and reveal the logical path of fault propagation. By analyzing the operation data of hydropower station equipment, the algorithm can identify the causal relationship between failures, that is, the probability and conditions under which one failure may cause another failure. Based on these relationships, the algorithm constructs potential fault chains, and each chain represents a logical order and pattern of possible equipment failures.

[0144] Step B12: Conduct a risk assessment on the potential fault chain and calculate the risk value of the potential fault chain, where the risk value is used to represent the degree of influence of the potential fault chain on the safe operation of the hydropower station.

[0145] It should be noted that risk assessment is a process that comprehensively considers multiple factors such as the severity, occurrence probability, and failure propagation speed of potential fault chains. By quantitatively analyzing these factors, the risk value of the potential fault chain can be calculated, providing a scientific basis for subsequent safety warnings and safety decisions. The higher the risk value, the greater the impact of the potential fault chain on the safe operation of the hydropower station, and more urgent and effective measures need to be taken to deal with it.

[0146] Step B13: Sort the potential fault chains according to the risk value to obtain a risk ranking result.

[0147] It should be noted that the purpose of sorting the potential fault chains is to prioritize the processing of those fault chains that have the greatest impact on the safe operation of the hydropower station. Through sorting, it can be clearly identified which fault chains need to be immediately intervened, and which fault chains can be temporarily observed or preventive measures can be taken.

[0148] Step B14: Take the potential fault chain with the highest risk value in the risk ranking result as the critical potential fault chain.

[0149] It should be noted that critical potential fault chains are those that pose the greatest threat to the safe operation of the hydropower station. They often involve important equipment or systems, and once a failure occurs, it may lead to serious consequences. Therefore, critical potential fault chains are taken as the key objects for safety warnings and safety decisions.

[0150] Step B15: Conduct safety warnings for the hydropower station based on the critical potential fault chain to generate safety warning information, where the safety warning information includes a warning level, warning content, and safety measures.

[0151] It should be noted that the safety warning information is generated based on the risk assessment results of the critical potential fault chain, aiming to remind managers to pay attention to potential safety risks and take corresponding measures to prevent them. The warning level is usually divided according to the level of the risk value, such as emergency warnings, important warnings, and general warnings, etc., so that managers can take corresponding countermeasures according to the different warning levels. The warning content details the specific information of the potential fault chain, including the composition of the fault chain, possible consequences, and recommended safety measures, etc., so that managers can comprehensively understand the safety risks and make correct decisions. The safety measures are specific countermeasures formulated according to the warning content, such as strengthening equipment monitoring, conducting equipment maintenance, or replacing faulty components, etc., to ensure the safe operation of the hydropower station.

[0152] Step B16: Make safety decisions for the hydropower station according to the safety warning information.

[0153] It should be noted that the safety decision-making of the hydropower station is based on safety warning information, aiming to formulate and implement effective measures to prevent the development of potential fault chains and ensure the safe operation of the hydropower station.

[0154] It should be understood that during the safety decision-making process, it is necessary to comprehensively consider the warning level, warning content, as well as the available resources and technical conditions to formulate a reasonable response plan. These plans may include measures such as strengthening equipment monitoring, adjusting operation strategies, performing equipment maintenance or replacing faulty components to eliminate or mitigate the impact of potential fault chains on the safe operation of the hydropower station. By implementing these measures, the occurrence of faults can be effectively prevented, and the safety and reliability of the hydropower station can be improved.

[0155] In a feasible implementation manner, step B11 may include: extracting equipment fault events and their occurrence times based on the operation data of the hydropower station equipment; analyzing the temporal and logical correlations between the equipment fault events according to the equipment fault events and their occurrence times to determine the causal relationships between equipment faults; constructing an equipment fault propagation network according to the causal relationships, where the nodes in the equipment fault propagation network represent equipment faults, and the connecting edges between the nodes represent the propagation paths between faults; analyzing the activation probability and propagation speed of fault nodes in the equipment fault propagation network based on the operation trend of the hydropower station equipment; identifying potential fault propagation paths in the equipment fault propagation network according to the activation probability and propagation speed to obtain potential fault chains.

[0156] It should be noted that the operation data of the hydropower station equipment includes data sources such as equipment operation logs and monitoring records. Equipment fault events refer to abnormal or failure situations that occur during the operation of the equipment, and these events are usually accompanied by specific fault phenomena and fault codes.

[0157] Specifically, by analyzing equipment failure events and their occurrence times, the temporal and logical correlations between failures can be revealed, that is, one failure event may trigger the occurrence of another failure event, or there may be some logical connections between multiple failure events. Based on these correlation analyses, the causal relationships between equipment failures can be determined, that is, one failure is directly or indirectly caused by another failure. Next, according to these causal relationships, an equipment failure propagation network can be constructed, which graphically shows the propagation paths and logical relationships between equipment failures. In the network, each node represents an equipment failure, and the connecting edges between nodes represent the propagation paths between failures. By analyzing the operation trends of hydropower station equipment, the activation probability and propagation speed of failure nodes in the equipment failure propagation network can be further analyzed, that is, the likelihood of a failure node being activated and the speed at which the failure propagates in the network. These parameters are crucial for identifying potential failure propagation paths. Finally, based on the activation probability and propagation speed, potential failure propagation paths are identified in the equipment failure propagation network, and these paths form potential failure chains. Each potential failure chain represents a logical sequence and propagation pattern of possible equipment failures.

[0158] In this embodiment, a digital twin model of a hydropower station is constructed, and different directed sensor node layout schemes are simulated based on the digital twin model to determine the optimal placement strategy; a number of directed sensor nodes are randomly placed in the area to be monitored in the hydropower station according to the optimal placement strategy to obtain an initial layout scheme of the directed sensor nodes, where the directed sensor nodes are in a dormant state; the initial layout scheme of the directed sensor nodes is optimized based on a genetic algorithm to obtain a target layout scheme of the directed sensor nodes; the directed sensor nodes in the target layout scheme are activated to obtain the activated directed sensor nodes; the current operation information of the hydropower station equipment is collected through the activated directed sensor nodes; a multi-dimensional equipment knowledge graph is constructed, and the operation trend of the hydropower station equipment is predicted based on the current operation information of the hydropower station equipment and the multi-dimensional equipment knowledge graph; potential failure chains are identified based on the operation trend of the hydropower station equipment, and safety warnings and safety decisions for the hydropower station are made according to the potential failure chains. By the above method, by introducing a sensor network to collect the current operation information of hydropower station equipment, and then combining with a knowledge graph for operation trend prediction, potential failure chains are identified and safety warnings and safety decisions are made, effectively improving the safety management efficiency and accuracy of the hydropower station and ensuring the equipment safety of the hydropower station.

[0159] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar content as in the above-mentioned first embodiment can be referred to the above introduction and will not be repeated hereinafter. On this basis, please refer to Figure 2, step S40 in the hydropower station safety management method based on the sensor network further includes steps S601 to S605:

[0160] Step S601: Construct an equipment failure knowledge base based on the equipment failure modes, historical safety events, and equipment maintenance records of the hydropower station.

[0161] It should be noted that the equipment failure mode refers to the common failure modes or abnormal states of the equipment during operation, and these modes are usually summarized and induced based on historical data and expert experience. Historical safety events record the safety accidents or potential incidents that have occurred in the hydropower station in the past, which helps to identify potential safety risks and failure modes. The equipment maintenance records detail the maintenance history of the equipment, including information such as maintenance time, maintenance content, and replaced parts.

[0162] It should be understood that when constructing the equipment failure knowledge base, it is necessary to collect and analyze the above information, and use data mining and knowledge representation techniques to extract key knowledge such as the characteristics, causes, impacts, and corresponding treatment measures of equipment failures. These knowledge are stored in the knowledge base in a structured form to form an equipment failure knowledge base.

[0163] Step S602: Parse and learn the equipment failure knowledge base to extract equipment failure characteristics, failure causes, and failure solutions.

[0164] It should be noted that in this embodiment, the natural language processing technology can be used to parse and learn the text information in the equipment failure knowledge base. Through the natural language processing technology, the unstructured text information in the equipment failure knowledge base is transformed into a structured knowledge representation, so as to extract the characteristics, causes, and corresponding solutions of equipment failures, and improve the usability and accuracy of knowledge.

[0165] Step S603: Based on the extracted equipment failure characteristics, failure causes, and failure solutions, construct a multi-dimensional equipment knowledge graph, where the multi-dimensional equipment knowledge graph includes equipment failure nodes, failure characteristic nodes, failure cause nodes, and failure solution nodes, and each node is connected by a relationship edge.

[0166] It should be noted that the multi-dimensional device knowledge graph is a graphical knowledge representation method that can organize and display various information related to device failures in the form of nodes and edges. In the multi-dimensional device knowledge graph, device failure nodes represent various possible failures in the hydropower station, failure feature nodes describe the specific manifestations or features of these failures, failure cause nodes reveal the reasons or mechanisms for the occurrence of failures, and failure solution nodes provide countermeasures or solutions for these failures. The various nodes are connected by relationship edges, and these edges represent the logical or associative relationships between the nodes, such as the correspondence between failures and features, the causal relationship between causes and failures, and the correspondence between solutions and failures. By constructing such a multi-dimensional device knowledge graph, the internal connections and logical relationships between device failures can be intuitively displayed.

[0167] Specifically, according to the extraction results of device failure features, failure causes, and failure solutions, determine the node types in the knowledge graph, including device failure nodes, failure feature nodes, failure cause nodes, and failure solution nodes. Then, according to the logical or associative relationships between these nodes, such as the correspondence between failures and features, the causal relationship between causes and failures, and the correspondence between solutions and failures, determine the relationship edges in the knowledge graph. Next, use a graphical tool or software to visually display these nodes and relationship edges to form a multi-dimensional device knowledge graph. In the knowledge graph, the importance or category of different nodes can be represented by attributes such as the color, size, or shape of the nodes, so as to more intuitively display the internal connections and logical relationships between device failures. Through such a construction process, a multi-dimensional device knowledge graph with rich device failure knowledge and a clear structure can be obtained.

[0168] Step S604: Match the current operating information with the multi-dimensional device knowledge graph to identify the device failure features corresponding to the current operating information.

[0169] It should be noted that during the matching process, the current operating information of the hydropower station equipment is compared with the failure feature nodes in the multi-dimensional device knowledge graph. By calculating indicators such as similarity or matching degree, find the failure feature node that is closest or most similar to the current operating information. If the matching is successful, it indicates that the equipment corresponding to the current operating information may have a failure related to the matched failure feature node. At this time, the failure cause node and failure solution node associated with this failure feature node can be further viewed to understand the specific cause of the failure and possible solutions. If the matching is unsuccessful, it may be necessary to further analyze the current operating information or adjust the nodes and relationship edges in the multi-dimensional device knowledge graph to improve the accuracy and reliability of the matching. Through such a matching process, real-time monitoring and fault warning of the current operating status of hydropower station equipment can be achieved.

[0170] Step S605: Predict the operation trend of the hydropower station equipment according to the equipment fault characteristics corresponding to the current operation information, where the operation trend of the hydropower station equipment includes a performance change trend, a fault occurrence probability trend, and a remaining service life trend.

[0171] It should be noted that the operation trend of the hydropower station equipment is predicted based on the equipment fault characteristics corresponding to the current operation information, combined with the fault historical data and expert experience in the multi-dimensional equipment knowledge graph, and consists of multiple dimensions such as the performance change trend, the fault occurrence probability trend, and the remaining service life trend. The performance change trend reflects the dynamic changes of the performance parameters of the equipment during operation, which helps to timely detect the situation of performance degradation or abnormal fluctuations. The fault occurrence probability trend is based on the historical statistical data of the fault characteristics and the current operation state of the equipment to predict the possibility of a fault occurring in a future period. The remaining service life trend is to estimate the length of time that the equipment can operate normally in the future based on information such as the current state of the equipment, historical maintenance records, and fault modes.

[0172] It can be understood that in the prediction process, technical means such as data mining and machine learning can be used to perform in-depth learning and pattern recognition on historical data and current operation information, so as to achieve accurate prediction of the equipment operation trend.

[0173] In a feasible implementation manner, step S605 may include: inputting the equipment fault characteristics corresponding to the current operation information into an operation trend prediction model, and predicting the performance change trend, the fault occurrence probability trend, and the remaining service life trend of the hydropower station equipment through the operation trend prediction model; calculating the operation trend of the hydropower station equipment according to the performance change trend, the fault occurrence probability trend, and the remaining service life trend.

[0174] It should be noted that in this implementation manner, the operation trend prediction model can perform in-depth learning and pattern recognition on the equipment fault characteristics, so as to achieve accurate prediction of the operation trend of the hydropower station equipment. The operation trend prediction model is trained and constructed based on a large amount of historical data and expert experience, and has high prediction accuracy and reliability. By inputting the equipment fault characteristics corresponding to the current operation information into this model, key information such as the performance change trend, the fault occurrence probability trend, and the remaining service life trend of the equipment in a future period can be obtained.

[0175] It is understandable that the operation trend prediction model is constructed based on data mining and machine learning technologies, which can extract information such as the characteristics, laws, and development trends of equipment failures from a large amount of historical data. When constructing the operation trend prediction model, it is first necessary to collect a large amount of historical operation data of hydropower station equipment, including equipment performance parameters, fault records, maintenance logs, etc. Then, data mining techniques are used to preprocess and extract features from these data to obtain key indicators that can reflect equipment fault characteristics. Based on these features and indicators, a prediction model is constructed using machine learning algorithms. By training and optimizing the model parameters, it can accurately predict the operation trend of hydropower station equipment. During the model construction process, expert experience and domain knowledge can also be combined to further optimize and adjust the model to improve its prediction performance and reliability. Through such a construction process, an operation trend prediction model based on data mining and machine learning technologies can be obtained, which can accurately predict the operation trend of hydropower station equipment.

[0176] It is worth noting that the performance change trend reflects the performance fluctuations and change laws of the equipment during long-term operation, and is an important basis for evaluating the equipment health status and predicting the occurrence of faults. By real-time monitoring of performance parameters and historical data analysis, abnormal changes in equipment performance parameters can be detected in a timely manner, such as performance degradation, parameter deviation from the normal range, etc., so as to take corresponding maintenance measures to avoid the occurrence of faults. The calculation formula for the performance change trend is as follows:

[0177] P(t) = P ref ·(1 - α1·ΔA(t) - α2·ΔT(t))

[0178] where P(t) is the performance change trend, P ref is the reference performance when the equipment is operating normally (such as the initial performance or theoretical performance), ΔA(t) = A(t) - A0, represents the change in vibration acceleration at time t (related to faults, A0 is the initial vibration acceleration of the equipment), ΔT(t) = T(t) - T0, represents the change in temperature at time t (related to faults, T0 is the initial temperature of the equipment), and α1, α2 are the influence coefficients of performance change, representing the weights of different fault characteristics on equipment performance.

[0179] The fault occurrence probability trend is represented by a failure rate model based on historical fault data and the current equipment operation status. Usually, the Weibull distribution model is used, which reflects the possibility of the equipment failing in a future period of time. Based on the current operation information and equipment fault characteristics (such as vibration, temperature, etc.), the calculation formula for the fault occurrence probability trend is as follows:

[0180]

[0181] Among them, P fault (t) is the probability trend of fault occurrence, A(t) is the vibration acceleration of the hydropower station equipment at time t, λ(t) is the fault scale parameter of the hydropower station equipment at time t, representing the frequency of equipment fault occurrence, and β(t) is the shape parameter of the equipment, characterizing the nature of the equipment fault mode (for example, early fault, stable fault or aging fault).

[0182] When the probability of fault occurrence is high, the reliability of the equipment is poor, and the fault risk is relatively high. P fault (t) will affect the negative direction of the comprehensive operation trend. The higher P fault (t) is, the more likely the equipment is to fail.

[0183] Remaining useful life (RUL) prediction estimates the remaining life of the equipment at the current moment by combining the real-time fault characteristics of the equipment (such as temperature, vibration, pressure, etc.) and historical fault data. The calculation formula for the remaining useful life trend is as follows:

[0184]

[0185] Among them, RUL(t) is the remaining useful life trend, F(t) is the cumulative probability of fault of the hydropower station equipment at time t, representing the total probability of the equipment experiencing a fault up to the current moment, which can be calculated through a fault rate model (such as Weibull distribution). As the equipment usage time increases, F(t) will gradually increase, indicating an increase in the probability of equipment failure. α3 and α4 are the influence coefficients of equipment vibration change on the remaining life and the influence coefficient of equipment temperature change on the remaining life.

[0186] The larger the remaining useful life trend RUL is, the longer the equipment can still operate. The smaller RUL is, the closer the equipment is to reaching its service life, and the greater the fault risk is.

[0187] By performing weighted fusion on the performance change trend, the probability trend of fault occurrence, and the remaining useful life trend, the comprehensive operation trend of the hydropower station equipment can be determined. The calculation formula for the operation trend of the hydropower station equipment is as follows:

[0188]

[0189] Among them, C(t) is the operation trend of the hydropower station equipment at time t, and ω1, ω2, and ω3 are the weights of the performance change trend, the probability trend of fault occurrence, and the remaining useful life trend respectively. P refFor the reference performance of the equipment, α1 and α2 are the influence coefficients of the performance change, ΔA(t) and ΔT(t) are the change amounts of the vibration acceleration and temperature of the hydropower station equipment at time t, A(t) is the vibration acceleration of the hydropower station equipment at time t, λ(t) is the fault scale parameter of the hydropower station equipment at time t, β(t) is the shape parameter of the equipment, F(t) is the cumulative probability of faults of the hydropower station equipment at time t, and α3 and α4 are the influence coefficients of the equipment vibration change on the remaining life and the influence coefficient of the equipment temperature change on the remaining life.

[0190] In this embodiment, based on the equipment fault modes, historical safety events, and equipment maintenance records of the hydropower station, an equipment fault knowledge base is constructed; the equipment fault knowledge base is analyzed and learned to extract equipment fault characteristics, fault causes, and fault solutions; based on the extracted equipment fault characteristics, fault causes, and fault solutions, a multi-dimensional equipment knowledge graph is constructed, where the multi-dimensional equipment knowledge graph includes equipment fault nodes, fault characteristic nodes, fault cause nodes, and fault solution nodes, and each node is connected by a relationship edge; the current operation information is matched with the multi-dimensional equipment knowledge graph to identify the equipment fault characteristics corresponding to the current operation information; according to the equipment fault characteristics corresponding to the current operation information, the operation trend of the hydropower station equipment is predicted, where the operation trend of the hydropower station equipment includes a performance change trend, a fault occurrence probability trend, and a remaining service life trend. Through the above method, by introducing a multi-dimensional equipment knowledge graph, the accurate identification and efficient management of the equipment fault characteristics of the hydropower station are realized, and then the operation trend of the hydropower station equipment is accurately predicted according to the equipment fault characteristics, which can effectively improve the accuracy of the safety management of the hydropower station.

[0191] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the method for safety management of a hydropower station based on a sensor network. Any simple transformation in more forms based on this technical concept is within the protection scope of this application.

[0192] This application also provides a safety management device for a hydropower station based on a sensor network. Please refer to Figure 3 , the safety management device for a hydropower station based on a sensor network includes:

[0193] A simulation module 10, configured to construct a digital twin model of the hydropower station and simulate different directed sensor node layout schemes based on the digital twin model to determine an optimal placement strategy.

[0194] A placement module 20, configured to randomly place a plurality of directed sensor nodes in the area to be monitored of the hydropower station according to the optimal placement strategy to obtain an initial layout scheme of the directed sensor nodes, where the directed sensor nodes are in a dormant state.

[0195] An optimization module 30 is configured to optimize the initial layout scheme of the directed sensor nodes based on a genetic algorithm to obtain a target layout scheme of the directed sensor nodes.

[0196] An activation module 40 is configured to activate the directed sensor nodes in the target layout scheme to obtain activated directed sensor nodes.

[0197] An acquisition module 50 is configured to acquire the current operation information of the hydropower station equipment through the activated directed sensor nodes.

[0198] A prediction module 60 is configured to construct a multi-dimensional equipment knowledge graph and predict the operation trend of the hydropower station equipment based on the current operation information of the hydropower station equipment and the multi-dimensional equipment knowledge graph.

[0199] A decision-making module 70 is configured to identify potential fault chains based on the operation trend of the hydropower station equipment and perform safety early warning and safety decision-making for the hydropower station according to the potential fault chains.

[0200] The hydropower station safety management device based on a sensor network provided by the present application adopts the hydropower station safety management method in the above embodiment, and can solve the technical problems in the prior art, such as fixed layout of hydropower station sensors, monitoring blind spots and insufficient optimization, and inability to achieve real-time and intelligent safety management of hydropower station equipment. Compared with the prior art, the beneficial effects of the hydropower station safety management device based on a sensor network provided by the present application are the same as those of the hydropower station safety management method provided by the above embodiment, and other technical features in the hydropower station safety management device based on a sensor network are the same as the features disclosed in the method of the above embodiment, and will not be elaborated here.

[0201] The above are only partial embodiments of the present application, and do not limit the patent scope of the present application. Any equivalent structural transformation made under the technical concept of the present application by using the content of the specification and drawings of the present application, or direct / indirect application in other related technical fields, is included in the patent protection scope of the present application.

Claims

1. A safety management method for a hydropower station based on a sensor network, characterized in that, The method includes: Constructing a digital twin model of a hydropower station, and based on the digital twin model, simulating different layout schemes of directional sensor nodes to determine the optimal placement strategy; Randomly placing a number of directional sensor nodes in the area to be monitored in the hydropower station according to the optimal placement strategy to obtain an initial layout scheme of the directional sensor nodes, where the directional sensor nodes are in a dormant state; Optimizing the initial layout scheme of the directional sensor nodes based on a genetic algorithm to obtain a target layout scheme of the directional sensor nodes; Activating the directional sensor nodes in the target layout scheme to obtain activated directional sensor nodes; Collecting the current operation information of the hydropower station equipment through the activated directional sensor nodes; Constructing a multi-dimensional equipment knowledge graph, and predicting the operation trend of the hydropower station equipment based on the current operation information of the hydropower station equipment and the multi-dimensional equipment knowledge graph; Identifying potential fault chains based on the operation trend of the hydropower station equipment, and performing safety early warning and safety decision-making for the hydropower station according to the potential fault chains.

2. The method according to claim 1, wherein The constructing a digital twin model of a hydropower station, and based on the digital twin model, simulating different layout schemes of directional sensor nodes to determine the optimal placement strategy includes: Constructing a geometric model of the hydropower station, where the geometric model includes the structural layout, equipment configuration, and spatial relationship of the hydropower station; Performing physical modeling based on a water flow mechanics model, an equipment working principle model, and the coupling relationship between equipment to obtain a physical model of the hydropower station; Associating the geometric model and the physical model to construct a digital twin model of the hydropower station; Adding virtual sensors to each device and component in the digital twin model, and setting the simulation area, the initial position, and attributes of the virtual sensor nodes; Simulating the sensing process of the directional sensor nodes based on the simulation area, the initial position, and attributes of the virtual sensor nodes to obtain multiple layout schemes of the directional sensor nodes; Calculating the coverage rate and the target detection rate under each layout scheme of the directional sensor nodes; Determining the optimal placement strategy according to the coverage rate and the target detection rate.

3. The method according to claim 1, characterized in that, The optimizing the initial layout scheme of the directional sensor nodes based on a genetic algorithm to obtain a target layout scheme of the directional sensor nodes includes: Taking the initial layout scheme of the directional sensor nodes as the initial population, where each individual in the initial population is a layout scheme of the directional sensor nodes, and the layout scheme of the directional sensor nodes includes the position and direction of the directional sensor nodes; Defining a fitness function, and calculating the fitness value of each individual according to the proportion of the key area covered, the total energy consumption of the sensors, and the overlapping coverage area between the sensors; Selecting target individuals from the initial population according to the fitness value, and performing crossover operations on the target individuals to generate new individuals; Performing mutation operations on the new individuals to obtain mutated individuals, and adding the mutated individuals to the initial population to generate a new population; Judging whether the new population meets the termination condition; If the condition is satisfied, the individual with the highest fitness value in the new population is used as the target layout scheme of the directional sensor nodes; If the condition is not satisfied, the step of calculating the fitness value of each individual according to the proportion of the covered key area, the total energy consumption of the sensors, and the overlapping coverage area between the sensors is re-executed until the termination condition is satisfied, where the termination condition is reaching the preset number of iterations or the fitness values of the individuals in the population tend to be stable.

4. The method according to claim 1, characterized in that The constructing the multi-dimensional device knowledge graph and predicting the operation trend of the hydropower station devices based on the current operation information of the hydropower station devices and the multi-dimensional device knowledge graph includes: Constructing a device failure knowledge base based on the device failure modes, historical safety events, and device maintenance records of the hydropower station; Parsing and learning the device failure knowledge base to extract device failure features, failure causes, and failure solutions; Constructing a multi-dimensional device knowledge graph based on the extracted device failure features, failure causes, and failure solutions, where the multi-dimensional device knowledge graph includes device failure nodes, failure feature nodes, failure cause nodes, and failure solution nodes, and the nodes are connected by relationship edges; Matching the current operation information with the multi-dimensional device knowledge graph to identify the device failure features corresponding to the current operation information; Predicting the operation trend of the hydropower station devices according to the device failure features corresponding to the current operation information, where the operation trend of the hydropower station devices includes the performance change trend, the failure occurrence probability trend, and the remaining service life trend.

5. The method according to claim 4, wherein The predicting the operation trend of the hydropower station devices according to the device failure features corresponding to the current operation information includes: Inputting the device failure features corresponding to the current operation information into an operation trend prediction model, and predicting the performance change trend, the failure occurrence probability trend, and the remaining service life trend of the hydropower station devices through the operation trend prediction model for the device failure features; Calculating the operation trend of the hydropower station devices according to the performance change trend, the failure occurrence probability trend, and the remaining service life trend.

6. The method according to claim 1, wherein The activating the directional sensor nodes in the target layout scheme to obtain the activated directional sensor nodes includes: Selecting any node from the directional sensor nodes of the target layout scheme as the initial node, and activating the initial node to obtain the activated initial node; Calculating the number of hydropower station devices that can be covered by the activated initial node in each perspective; Performing a priority ranking on each perspective according to the number of hydropower station devices that can be covered from high to low to obtain a ranking result; Taking the perspective with the highest priority in the ranking result as the target perspective, and adjusting the sensing direction of the activated initial node so that the sensing direction of the activated initial node faces the target perspective; Searching within a preset range centered on the activated initial node to check if there are other unactivated directional sensor nodes; If there are unactivated directed sensor nodes, perform the step of selecting any node from the directed sensor nodes of the target layout scheme as the initial node and activating the initial node until all the directed sensor nodes in the area to be monitored of the hydropower station are activated, obtaining the activated directed sensor nodes.

7. The method according to claim 1, wherein After activating the directed sensor nodes in the area to be monitored of the hydropower station and obtaining the activated directed sensor nodes, it further includes: Obtain the current viewing angle of the activated directed sensor nodes; Determine the total number of hydropower station equipment that can be covered based on the current viewing angle of the activated directed sensor nodes; Calculate the coverage balance coefficient based on the total number of hydropower station equipment that can be covered; Evaluate the coverage efficiency of the activated directed sensor nodes based on the coverage balance coefficient to obtain a coverage efficiency evaluation result; Optimize and adjust the distribution positions of the activated directed sensor nodes according to the coverage efficiency evaluation result.

8. The method according to claim 1, wherein The method of identifying potential fault chains based on the operation trend of hydropower station equipment and performing hydropower station safety warning and safety decision-making according to the potential fault chains includes: Based on the operation trend of hydropower station equipment, use the fault chain identification algorithm to analyze the causal relationship and propagation path between equipment failures and construct potential fault chains; Conduct a risk assessment on the potential fault chains and calculate the risk values of the potential fault chains, where the risk values are used to represent the degree of influence of the potential fault chains on the safe operation of the hydropower station; Sort the potential fault chains according to the risk values to obtain a risk ranking result; Regard the potential fault chain with the highest risk value in the risk ranking result as the key potential fault chain; Conduct hydropower station safety warning based on the key potential fault chain to generate safety warning information, where the safety warning information includes a warning level, warning content, and safety measures; Make hydropower station safety decisions according to the safety warning information.

9. The method according to claim 8, wherein The method of using the fault chain identification algorithm to analyze the causal relationship and propagation path between equipment failures based on the operation trend of hydropower station equipment and construct potential fault chains includes: Extract equipment failure events and occurrence times based on the operation data of hydropower station equipment; Analyze the time correlation and logical correlation between equipment failure events according to the equipment failure events and occurrence times to determine the causal relationship between equipment failures; Construct an equipment failure propagation network according to the causal relationship, where the nodes in the equipment failure propagation network represent equipment failures, and the connection edges between nodes represent the propagation paths between failures; Analyze the activation probability and propagation speed of fault nodes in the equipment failure propagation network based on the operation trend of hydropower station equipment; Identify potential fault propagation paths in the equipment failure propagation network according to the activation probability and propagation speed to obtain potential fault chains.

10. A hydropower station safety management device based on a sensor network, characterized in that, The hydropower station safety management device based on the sensor network includes: A simulation module for constructing a digital twin model of the hydropower station and simulating different layout schemes of directed sensor nodes based on the digital twin model to determine the optimal placement strategy; A placement module, configured to randomly place a number of directional sensor nodes within the area to be monitored of a hydropower station according to the optimal placement strategy, so as to obtain an initial layout plan of the directional sensor nodes, wherein the directional sensor nodes are in a sleep state; An optimization module, configured to optimize the initial layout plan of the directional sensor nodes based on a genetic algorithm to obtain a target layout plan of the directional sensor nodes; An activation module, configured to activate the directional sensor nodes in the target layout plan to obtain the activated directional sensor nodes; A collection module, configured to collect the current operation information of the hydropower station equipment through the activated directional sensor nodes; A prediction module, configured to construct a multi-dimensional equipment knowledge graph and predict the operation trend of the hydropower station equipment based on the current operation information of the hydropower station equipment and the multi-dimensional equipment knowledge graph; A decision-making module, configured to identify potential fault chains based on the operation trend of the hydropower station equipment and perform safety early warning and safety decision-making for the hydropower station according to the potential fault chains.

Images