Underground distribution room water immersion simulation and repair training method and system based on internet of things data collection

Abstract

The present application relates to the technical field of power distribution room water immersion simulation, and discloses a method and system for underground power distribution room water immersion simulation and repair training based on Internet of Things data collection, the method comprising: using a shallow water diffusion wave model to simulate the water level evolution matrix of the underground power distribution room in a water immersion scenario; calculating the water level time series of the power distribution equipment and the probability of a fault event; integrating the power distribution room environment, the water level evolution matrix and the fault probability of the power distribution equipment on a virtual reality training platform to generate an underground power distribution room repair training scenario. The present application constructs a shallow water diffusion wave model that integrates terrain slope, drainage capacity and rainfall trend, simulates the propagation and evolution of water level in the time and space dimensions, and then evaluates the probability of a fault occurring in each power distribution device. The fault probability, water level evolution results and power distribution room environment are integrated with the virtual reality environment to achieve immersive training of the repair task, significantly improving the emergency response capability and repair efficiency of the repair personnel in water disaster situations.

Description

Technical Field

[0001] This invention relates to the technical field of power distribution room flooding simulation, and in particular to a method and system for underground power distribution room flooding simulation and emergency repair training based on Internet of Things data acquisition. Background Technology

[0002] With the continuous expansion of urban power grids and the extensive development of underground space resources, underground substations, as important urban power infrastructure, are widely deployed in densely populated areas such as urban transportation hubs, commercial complexes, and underground parking lots. Underground substations not only undertake the regional distribution and transmission of electricity but also play a crucial role in power system relay, protection, and monitoring. Their safe and stable operation directly affects the power supply reliability of the entire urban power grid and the normal production and living order of urban residents. However, because most underground substations are built in low-lying or enclosed environments with limited drainage capacity, they are highly susceptible to flooding accidents under extreme weather conditions such as heavy rainfall, typhoons, and urban flooding.

[0003] Once a flooding accident occurs, it can not only cause the insulation performance of high-voltage power equipment to deteriorate and short-circuit to burn out, but may also trigger a chain of equipment failures, causing large-scale power outages and even personal injury or death. At the same time, repair personnel at flooded sites often face many uncertainties such as unknown water levels, unseen equipment status, and unknown risk areas, which seriously affect the timeliness and safety of emergency repairs.

[0004] Some research has already attempted to achieve intelligent monitoring of power distribution rooms through the integration of the Internet of Things (IoT) and information technology. For example, CN114710644A discloses a comprehensive monitoring system for power distribution rooms, which includes: a video and access control subsystem for collecting video surveillance information and managing access; an environmental control subsystem for collecting operating environment information; a power monitoring subsystem for monitoring equipment operation data; and a network communication system for transmission and dispatch. This system can achieve unified collection and management of various aspects of information from the power distribution room, and has certain practical value in environmental perception and comprehensive supervision.

[0005] However, most existing technologies focus on condition monitoring and early warning, lacking physical modeling and dynamic evolution analysis of the immersion process, and have not yet achieved accurate prediction of the insulation performance degradation of power equipment under disaster scenarios. In addition, most current virtual training platforms use rule-based scripts or static models, which cannot dynamically reflect the risk evolution process of power distribution equipment in complex hydrodynamic environments, and are also difficult to form a training system for emergency repair decisions oriented towards "real combat".

[0006] Therefore, there is an urgent need for a method for simulating and training underground power distribution rooms in floodwaters by integrating dynamic perception of physical parameters and hydrodynamic evolution modeling. This method aims to improve the predictability of equipment status, the rationality of repair strategies, and the immersiveness and realism of emergency training in flood scenarios, thereby effectively enhancing repair efficiency. Summary of the Invention

[0007] In view of this, the present invention provides a method and system for simulation and emergency repair training of underground power distribution room flooding based on Internet of Things data acquisition. It collects key physical parameter data during the flooding process, constructs an improved shallow water diffusion wave model that integrates terrain information and hydrodynamic characteristics, accurately simulates the evolution of water level over time and space, further combines the location information of power distribution equipment to extract the water level time series of each power distribution equipment, establishes an insulation degradation model, and calculates the probability of power distribution equipment failure through fault tree calculation. Finally, the water level evolution results and failure probability are injected into a virtual reality engine to generate an immersive interactive emergency repair training scenario, realizing a training scheme that integrates virtual and real elements.

[0008] To achieve the above objectives, this invention provides a method for simulating and training emergency repairs of flooding in underground power distribution rooms based on Internet of Things (IoT) data acquisition, comprising the following steps:

[0009] S1: Deploy sensors in the underground power distribution room to sense physical parameters and obtain physical parameter sensing data of the underground power distribution room under the flooding scenario;

[0010] S2: Combining physical parameter sensing data, an improved shallow water diffusion wave model is used to simulate the water level evolution matrix of an underground power distribution room under flooding scenarios.

[0011] S3: Calculate the water level time series of each power distribution device in the underground power distribution room using the water level evolution matrix, perform time-series decay evolution on the insulation state of the power distribution device, and generate the probability of each power distribution device experiencing a fault event using fault tree reasoning.

[0012] S4: Integrate the underground power distribution room scene, water level evolution matrix, and the probability of each power distribution equipment failure event into the virtual reality training platform. The virtual reality training platform generates an underground power distribution room emergency repair training scene, and users conduct interactive emergency repair training in the underground power distribution room emergency repair training scene.

[0013] As a further improvement of the present invention:

[0014] Optionally, the types of sensors include tipping bucket rain gauges, soil conductivity sensors, electromagnetic flow meters, level gauges, and flow meters, including:

[0015] Multiple sensors of the same type are deployed in a fixed area of ​​the underground power distribution room to obtain the sensing results of multiple physical parameters of the same type of sensor in the fixed area. The tipping bucket rain gauge is used to sense the rainfall intensity, the soil conductivity sensor is used to sense the soil permeability, the electromagnetic flow meter and the liquid level gauge are used to sense the surface roughness coefficient of the power distribution room, and the liquid level gauge and the flow meter are used to sense the drainage capacity of the power distribution room.

[0016] Optionally, an adaptive weighted fusion method is used to fuse the sensing results of all physical parameters from sensors of the same type, including:

[0017] The adaptive weighted fusion formula is as follows:

[0018]

[0019] in, This represents the fusion processing result of the i-th physical parameter at the r-th physical parameter sensing time. Let R represent the physical parameter sensing result of the nth sensor used to collect the i-th physical parameter at the r-th physical parameter sensing time, where R represents the number of physical parameter sensing times, n∈[1,N], and N represents the number of sensors of the same type deployed in the fixed area. This represents the adaptive weight of the nth sensor used to collect the i-th physical parameter at the r-th physical parameter sensing time. The 1st to 4th physical parameters are, in order, rainfall intensity, soil permeability, floor roughness coefficient of the power distribution room, and drainage capacity of the power distribution room.

[0020] α1, α2, and α3 are weight adjustment coefficients used to adjust the weights of different factors. α1, α2, and α3 are set to 0.6, 0.2, and 0.2, respectively.

[0021] Results of physical parameter sensing The corresponding data deviation factor is used to measure the results of physical parameter sensing. The degree of deviation from the historical moving average, wherein the data deviation factor is a standardized numerical form;

[0022] Results of physical parameter sensing The corresponding packet loss rate factor reflects the data upload stability of the nth sensor that collects the i-th physical parameter over a period of time. The larger the value of the packet loss rate factor, the more severe the packet loss situation of the nth sensor that collects the i-th physical parameter, and the lower the transmission stability. The packet loss rate factor is a value between 0 and 1.

[0023] Results of physical parameter sensing The corresponding sensor health status factor reflects the current health status of the nth sensor that collects the i-th physical parameter. The larger the value of the sensor health status factor, the worse the current health status of the nth sensor that collects the i-th physical parameter. The sensor health status factor is a numerical value between 0 and 1.

[0024] The fusion results of the different physical parameters at R physical parameter sensing times are used as physical parameter sensing data.

[0025] Optionally, combining the aforementioned physical parameter sensing data, an improved shallow water diffusion wave model is used to simulate the water level evolution matrix of an underground power distribution room under a flooding scenario, including:

[0026] The point cloud of the underground power distribution room was obtained by 3D laser scanning, and a digital elevation model was extracted from the point cloud.

[0027] By combining physical parameter sensing data and digital elevation models, an improved shallow water diffusion wave model is constructed:

[0028]

[0029]

[0030] in, Represents the g-th ground coordinate ((x) in the underground power distribution room g ,y g The water depth at the r-th physical parameter sensing time, where Δ represents the time difference between adjacent physical parameter sensing times. Represents the g-th ground coordinate (x) g ,y g The global water allocation term at the r-th physical parameter sensing time, where G represents the number of ground coordinates of the underground power distribution room, z * (x g ,y g ) represents the ground coordinates (x g ,y g The height at Ω * (x g ,y g ) represents ground coordinates ((x g ,y g The set of neighboring ground coordinates in an area centered at a radius of 1 meter. This indicates the weighting of water allocation based on elevation difference. Indicates the preset maximum height, set. It is 0.5 meters. Represents the g-th ground coordinate (x) g ,y g The water diffusion term at the r-th physical parameter sensing time;

[0031] The water distribution term refers to the water distribution value of surface precipitation to the ground coordinates of the underground power distribution room, which is calculated from rainfall intensity, soil permeability, and the unit drainage volume of the power distribution room at a unit coordinate. The unit drainage volume of the underground power distribution room at the r-th physical parameter sensing time;

[0032] The water diffusion term is the water distribution value from the ground coordinates in the underground power distribution room to the adjacent ground coordinates;

[0033] S represents the floor area of ​​the underground power distribution room;

[0034] The results are, in order, the fusion of rainfall intensity, soil permeability, substation floor roughness coefficient, and substation drainage capacity at the r-th physical parameter sensing time from the physical parameter sensing data.

[0035] Ω represents the set of nearby ground coordinates * (x g ,y g Water depth near ground coordinates in ) The amount of diffusion;

[0036] The water depth at different ground coordinates at different physical parameter sensing times is used as the water level evolution matrix. This water level evolution matrix is ​​an R-row, G-column matrix, where the r-th row and g-th column represents the g-th ground coordinate (x, y) in the underground power distribution room. g ,y g The water depth at the r-th physical parameter sensing time

[0037] Optionally, the water level time series of each power distribution device in the underground power distribution room is calculated using the water level evolution matrix, and the insulation state of the power distribution device is subjected to time-series decay evolution, including:

[0038] Two-dimensional ground position coordinates of each power distribution device in the underground power distribution room are obtained, and the water depth at the sensing time of R physical parameters corresponding to the two-dimensional ground position coordinates is extracted from the water level evolution matrix as the water level time series of the power distribution device.

[0039] Using the water level time series, the insulation state of the power distribution equipment after time-series decay evolution at different physical parameter sensing times is calculated.

[0040] Optionally, combining the water level time series and the time-series decay evolution results of the insulation state, a fault tree reasoning method is used to generate the probability of a fault event occurring for each power distribution device, including:

[0041] The water level change rate and the time-series decay evolution results of the insulation state of the power distribution equipment are obtained and used as nodes in the fault tree. A fault Boolean expression for the power distribution equipment is constructed. When the water level change rate is higher than a preset water level threshold and the insulation state is lower than a preset insulation state threshold, the fault Boolean expression of the power distribution equipment outputs 1, indicating that the power distribution equipment is prone to a fault event in the short term. The time-series decay evolution results of the water level change rate and insulation state of the power distribution equipment are normalized, weighted, and subjected to logistic regression to generate the probability of the power distribution equipment experiencing a fault event.

[0042] Optionally, the underground power distribution room scenario, water level evolution matrix, probability of failure events for each power distribution device, and repair priority are integrated into the virtual reality training platform. The virtual reality training platform generates an underground power distribution room repair training scenario, including:

[0043] Based on the underground power distribution room scenario, an underground power distribution room structural model is generated. The underground power distribution room scenario includes the location of the power distribution equipment and the scene point cloud of the underground power distribution room obtained by three-dimensional laser scanning.

[0044] The water level evolution matrix is ​​converted into a water level change animation using a VR engine. The water level change animation includes an animation of water level rising in an underground power distribution room and the sound of water flowing upward.

[0045] Generate fault scenes for each power distribution device after a failure event, including special effects animations of smoke and electric arcs;

[0046] The underground power distribution room structure model, water level change animation, probability of each power distribution equipment failure event, and failure scenarios are integrated into the virtual reality training platform, which generates emergency repair training scenarios for the underground power distribution room.

[0047] To address the aforementioned problems, this invention provides a simulation and emergency repair training system for flooded underground power distribution rooms based on Internet of Things (IoT) data acquisition, to implement any of the aforementioned methods for simulation and emergency repair training of flooded underground power distribution rooms based on IoT data acquisition. The system includes a virtual reality training platform and a data acquisition device.

[0048] The data acquisition device is used to deploy sensors in the underground power distribution room to sense physical parameters and obtain physical parameter sensing data of the underground power distribution room under the flooding scenario. Combining the physical parameter sensing data, an improved shallow water diffusion wave model is used to simulate the water level evolution matrix of the underground power distribution room under the flooding scenario. The water level evolution matrix is ​​used to calculate the water level time series of each power distribution device in the underground power distribution room, and the insulation state of the power distribution device is subjected to time-series decay evolution. The probability of each power distribution device to experience a fault event is generated by fault tree reasoning.

[0049] The virtual reality training platform is used to obtain the underground power distribution room scene, the water level evolution matrix, and the probability of each power distribution equipment failure event, and to generate an underground power distribution room emergency repair training scene.

[0050] To address the above problems, the present invention provides an electronic device, the electronic device comprising:

[0051] Memory, storing at least one instruction;

[0052] Communication interfaces enable communication between electronic devices; and

[0053] The processor executes the instructions stored in the memory to implement the above-described method for simulating and training the flooding of underground power distribution rooms based on Internet of Things data acquisition.

[0054] To address the aforementioned issues, the present invention also provides a computer-readable storage medium storing at least one instruction, which is executed by a processor in an electronic device to implement the aforementioned method for simulating and training the flooding of underground power distribution rooms based on Internet of Things data acquisition.

[0055] Compared with existing technologies, this invention proposes a method and system for simulating and training emergency repairs of flooded underground power distribution rooms based on Internet of Things (IoT) data acquisition. This technology has the following beneficial effects:

[0056] First, addressing issues such as inconsistent accuracy and unstable transmission from sensing devices, a dynamic weighted fusion mechanism based on sensor status is adopted. For similar physical parameters, this mechanism comprehensively considers multiple dimensions, including numerical deviation, data transmission efficiency, and remaining power, assigning different weights to data from each sensor. Through multi-dimensional quality factor calculation, the data contribution is corrected in real time, suppressing the influence of high-noise or degraded sensors, thus enhancing the reliability and robustness of the overall sensing results. This mechanism not only achieves effective fusion of multi-source heterogeneous data but also improves the balance and accuracy of spatial data distribution, providing highly adaptable and high-engineering-value sensing support for water level sensing and intelligent monitoring in the complex environment of underground power distribution rooms.

[0057] Meanwhile, this application improves the diffusion wave model. The improved shallow water diffusion wave model addresses the risk scenarios of water accumulation and equipment immersion in underground power distribution rooms under extreme weather conditions such as heavy rain and high-flow-rate storms. It proposes a water depth simulation method that integrates overall regional water volume control and local hydrodynamic diffusion mechanisms. Unlike traditional models that heavily rely on local rainfall and drainage distribution, this model constructs a water volume allocation weight function to dynamically map global temporal information such as overall rainfall, soil permeability, and drainage capacity to local spatial units, effectively solving the problems of difficult high-density data acquisition and high deployment costs. The diffusion term introduces a nonlinear water potential gradient and elevation difference driving mechanism and considers the influence of frictional dissipation. This makes the model more physically consistent and accurate in simulating underground structures with complex terrain, steps, slopes, or depressions, significantly improving the spatial reproduction capability of water propagation paths and water level change distribution. Attached Figure Description

[0058] Figure 1 This is a flowchart illustrating a method for simulating and training emergency repairs of flooding in underground power distribution rooms based on Internet of Things (IoT) data acquisition, as provided in an embodiment of the present invention.

[0059] Figure 2 This is a flowchart of a data acquisition process for an underground power distribution room based on Internet of Things (IoT) data acquisition, as provided in an embodiment of the present invention.

[0060] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0061] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0062] This application provides a method for simulating and training emergency repairs of flooding in underground power distribution rooms based on Internet of Things (IoT) data acquisition. The execution entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster.

[0063] Reference Figure 1 Embodiment 1 of the present invention is as follows:

[0064] A method for simulating and training emergency repairs of flooding in underground power distribution rooms based on Internet of Things (IoT) data acquisition includes the following steps:

[0065] S1: Sensors are deployed in the underground power distribution room to sense physical parameters and obtain physical parameter sensing data of the underground power distribution room under the flooding scenario.

[0066] The types of sensors include tipping bucket rain gauges, soil conductivity sensors, electromagnetic flow meters, level gauges, and flow meters, including:

[0067] Multiple sensors of the same type are deployed in a fixed area of ​​the underground power distribution room to obtain the sensing results of multiple physical parameters of the same type of sensor in the fixed area. The tipping bucket rain gauge is used to sense the rainfall intensity, the soil conductivity sensor is used to sense the soil permeability, the electromagnetic flow meter and the liquid level gauge are used to sense the surface roughness coefficient of the power distribution room, and the liquid level gauge and the flow meter are used to sense the drainage capacity of the power distribution room.

[0068] Specifically, the tipping bucket rain gauge is installed on the surface area above the underground power distribution room, the soil conductivity sensor is installed in the soil layer area 0.2 meters to 1 meter below the surface area, the water depth and water velocity on the ground of the underground power distribution room are calculated using the electromagnetic flow meter and the level gauge, and the Manning coefficient is obtained by combining the slope of the power distribution room ground as the roughness coefficient of the power distribution room ground. The level gauge and the flow meter are installed in the drainage outlet area of ​​the underground power distribution room.

[0069] As an embodiment of the present invention, each tipping bucket rain gauge represents a fixed amount of water. The rainfall intensity is calculated by the number of tipping buckets per unit time. The flow rate in the drainage outlet area is calculated using a flow meter. The drainage capacity of the power distribution room is estimated by combining the cross-sectional area of ​​the drainage outlet.

[0070] Specifically, each sensor integrates a LoRa wireless communication module and connects to the gateway node through a star network structure. The gateway node uploads data to the server via 4G / NB-IoT for adaptive weighted fusion, generating a water level evolution matrix, the probability of each power distribution equipment failure event, and repair priorities.

[0071] An adaptive weighted fusion method is used to fuse the sensing results of all physical parameters from sensors of the same type, including:

[0072] The adaptive weighted fusion formula is as follows:

[0073]

[0074] in, This represents the fusion processing result of the i-th physical parameter at the r-th physical parameter sensing time. Let R represent the physical parameter sensing result of the nth sensor used to collect the i-th physical parameter at the r-th physical parameter sensing time, where R represents the number of physical parameter sensing times, n∈[1,N], and N represents the number of sensors of the same type deployed in the fixed area. This represents the adaptive weight of the nth sensor used to collect the i-th physical parameter at the r-th physical parameter sensing time. The 1st to 4th physical parameters are, in order, rainfall intensity, soil permeability, floor roughness coefficient of the power distribution room, and drainage capacity of the power distribution room.

[0075] α1, α2, and α3 are weight adjustment coefficients used to adjust the weights of different factors. α1, α2, and α3 are set to 0.6, 0.2, and 0.2, respectively.

[0076] Results of physical parameter sensing The corresponding data deviation factor is used to measure the results of physical parameter sensing. The degree of deviation from the historical moving average, wherein the data deviation factor is a standardized numerical form; specifically, the formula for calculating the data deviation factor is:

[0077]

[0078] in, Represents a sequence The mean, Represents a sequence The standard deviation of , where ∈ represents the micro-control coefficient, is set to 0.001;

[0079] Results of physical parameter sensing The corresponding packet loss rate factor reflects the data upload stability of the nth sensor collecting the i-th physical parameter over a period of time. A larger packet loss rate factor indicates more severe packet loss and lower transmission stability for the nth sensor collecting the i-th physical parameter. The packet loss rate factor is a value between 0 and 1. Specifically, the formula for calculating the packet loss rate factor is:

[0080]

[0081] in, This represents the number of data frames received from the nth sensor that acquires the i-th physical parameter from the time of sensing the rL-th physical parameter to the time of sensing the r-th physical parameter. This represents the number of data frames sent by the nth sensor when the i-th physical parameter is collected from the time of the rL-th physical parameter sensing to the time of the r-th physical parameter sensing. L represents the preset time period length, which is set to 5.

[0082] Results of physical parameter sensing The corresponding sensor health status factor reflects the current health status of the nth sensor that collects the i-th physical parameter. A larger value for the sensor health status factor indicates a worse current health status for the nth sensor collecting the i-th physical parameter. The sensor health status factor is a numerical value between 0 and 1. Specifically, the calculation formula for the sensor health status factor is:

[0083]

[0084] in, This represents the nominal charge of the nth sensor that collects the i-th physical parameter. This represents the charge of the nth sensor that collects the i-th physical parameter at the time of sensing the r-th physical parameter.

[0085] The fusion results of the different physical parameters at R physical parameter sensing times are used as physical parameter sensing data.

[0086] like Figure 2 The diagram illustrates a data acquisition process for an underground power distribution room based on Internet of Things (IoT) data acquisition. By deploying various sensors in the underground power distribution room, the sensed data is transmitted to a gateway node and then uploaded to a data server, enabling data acquisition and monitoring of the underground power distribution room based on IoT data acquisition. The data server is used for simulation calculations to integrate the underground power distribution room scene, water level evolution matrix, probability of failure events for each power distribution device, and emergency repair priority into a virtual reality training platform. The virtual reality training platform generates an emergency repair training scenario for the underground power distribution room.

[0087] Specifically, in complex environments with redundant sensing data, varying accuracy, and significant differences in sensor performance, this technology dynamically weights similar physical parameters from different types of sensors, fully integrates heterogeneous information from multiple sensors, and evaluates the sensor status from multiple perspectives in real time, including deviation control, transmission efficiency, and remaining power. This reduces the weight of sensing data from sensors in poor condition, improves the accuracy of fused data, ensures that the fusion results do not overly rely on redundant data from a particular area, enhances the overall spatial information balance, and greatly improves the quality and practical value of intelligent sensing of water immersion in underground power distribution rooms. It also has good engineering adaptability and promotion potential.

[0088] S2: Combining physical parameter sensing data, an improved shallow water diffusion wave model is used to simulate the water level evolution matrix of an underground power distribution room under flooding scenarios.

[0089] Based on the aforementioned physical parameter sensing data, an improved shallow water diffusion wave model is used to simulate the water level evolution matrix of an underground power distribution room under flooding scenarios, including:

[0090] A point cloud of an underground power distribution room was acquired using 3D laser scanning, and a digital elevation model was extracted from the point cloud. Specifically, the point cloud is 3D coordinate data, and the digital elevation model is extracted as follows:

[0091]

[0092] Among them, {z * (x g ,y g )|g∈[1,G]},(x g ,y g () represents the g-th ground coordinate of the underground power distribution room, G represents the number of ground coordinates of the underground power distribution room, z * (x g ,y g ) represents the ground coordinates (x g ,y g The height at point ) Ω((x) g ,y g ) represents ground coordinates (x g ,y g The set of ground coordinates of the scene point cloud within a region centered at x and with a radius of 1 meter, ((x ′ ,y ′ ,z ′ (x) represents the scene point cloud, (x) ′ ,y ′ ) represents the set of ground coordinates Ω(x) g ,y g Any ground coordinates in ) Indicates from ground coordinates (x g ,y g The lowest scene point cloud height is selected from the scene point cloud within a 1-meter radius centered on the given area.

[0093] By combining physical parameter sensing data and digital elevation models, an improved shallow water diffusion wave model is constructed:

[0094]

[0095]

[0096] in, Represents the g-th ground coordinate ((x) in the underground power distribution room g ,y g The water depth at the r-th physical parameter sensing time, where Δ represents the time difference between adjacent physical parameter sensing times. Represents the g-th ground coordinate (x) g ,y gThe global water allocation term at the r-th physical parameter sensing time, where G represents the number of ground coordinates of the underground power distribution room, z * (x g ,y g ) represents the ground coordinates (x g ,y g The height at Ω * (x g ,y g ) represents ground coordinates ((x g ,y g The set of neighboring ground coordinates in an area centered at a radius of 1 meter. This indicates the weighting of water allocation based on elevation difference. Indicates the preset maximum height, set. It is 0.5 meters. Represents the g-th ground coordinate (x) g ,y g The water diffusion term at the r-th physical parameter sensing time;

[0097] The water distribution term refers to the water distribution value of surface precipitation to the ground coordinates of the underground power distribution room, which is calculated from rainfall intensity, soil permeability, and the unit drainage volume of the power distribution room at a unit coordinate. Let be the unit drainage volume of the underground power distribution room at the r-th physical parameter sensing time, representing the contribution of rainfall intensity, soil permeability, and the unit drainage volume of the power distribution room at a unit coordinate to the water depth at the ground coordinate.

[0098] The water diffusion term is the water distribution value from the ground coordinates in the underground power distribution room to the adjacent ground coordinates, and it is optimized using the ground roughness coefficient of the power distribution room to characterize the contribution of the water depth of the adjacent ground coordinates to the water depth of the ground coordinates.

[0099] S represents the floor area of ​​the underground power distribution room;

[0100] The results are, in order, the fusion of rainfall intensity, soil permeability, substation floor roughness coefficient, and substation drainage capacity at the r-th physical parameter sensing time from the physical parameter sensing data.

[0101] Ω represents the set of nearby ground coordinates * (x g ,y g Water depth near ground coordinates in ) The amount of diffusion;

[0102] Specifically, if If the value is positive, it indicates that the ground coordinates are ((x) g ,y gThe neighboring ground coordinates of ) are transmitted to the ground coordinates (x, r) at the r-th physical parameter sensing time. g ,y g ) the amount of water diffused, if A negative value indicates that the ground coordinates are ((x) g ,y g The amount of water diffused to the adjacent ground coordinates at the r-th physical parameter sensing time;

[0103] The water depth at different ground coordinates at different physical parameter sensing times is used as the water level evolution matrix. This water level evolution matrix is ​​an R-row, G-column matrix, where the r-th row and g-th column represents the g-th ground coordinate (x, y) in the underground power distribution room. g ,y g The water depth at the r-th physical parameter sensing time The horizontal axis of the water level evolution matrix represents the temporal variation of water depth at ground coordinates, while the vertical axis represents the spatial variation of water depth at ground coordinates.

[0104] Specifically, (x,y) represents the set of nearby ground coordinates Ω * (x g ,y g The nearest ground coordinates, z * ((x,y) represents the height at ((x,y)). The water depth at the nearest ground coordinates (x, y) at the r-th physical parameter sensing time is represented by dis((x, y), (x g ,y g )) represents the nearest ground coordinates (x,y) and (x g ,y g The Euclidean distance between them Indicates the initial water depth, set

[0105] As an embodiment of the present invention, by adjusting rainfall intensity, soil permeability and drainage capacity of the power distribution room, the improved shallow water diffusion wave model is used to simulate and generate water level evolution matrices under different underground power distribution room immersion conditions.

[0106] Specifically, the improved shallow water diffusion wave model addresses the issues of water accumulation and equipment soaking in underground power distribution rooms under extreme weather conditions (such as heavy rain and sudden high-flow-rate storms). It integrates regional total volume driving and local hydrodynamic diffusion mechanisms to estimate water depth. More specifically, in practical application scenarios where only overall temporal information on rainfall, soil permeability, and drainage capacity is available, but local distribution data is lacking, the model introduces water volume allocation weights to achieve a dynamic mapping from global water volume to local coordinates. This solves the problem of traditional diffusion wave models being heavily dependent on local rainfall input and having high data requirements, effectively reducing deployment costs and modeling barriers. Furthermore, it incorporates a nonlinear coupled diffusion mechanism in the water volume diffusion term, conforming to real water potential energy driving and frictional dissipation laws. Compared to traditional models that only consider water depth differences, this improved model better reflects the direct control of terrain elevation on the direction and intensity of water flow, significantly improving the spatial distribution accuracy of flooding simulation. It is particularly suitable for platform structures in underground spaces with local depressions, steps, and elevation differences.

[0107] S3: Calculate the water level time series of each power distribution device in the underground power distribution room using the water level evolution matrix, perform time-series decay evolution on the insulation state of the power distribution device, and generate the probability of each power distribution device experiencing a fault event using fault tree reasoning.

[0108] The water level time series of each power distribution device in the underground power distribution room is calculated using the water level evolution matrix, and the insulation state of the power distribution device is subjected to time-series decay evolution, including:

[0109] Two-dimensional ground position coordinates of each power distribution device in the underground power distribution room are obtained, and the water depth at the sensing time of R physical parameters corresponding to the two-dimensional ground position coordinates is extracted from the water level evolution matrix as the water level time series of the power distribution device.

[0110] Using the water level time series, the insulation state of the power distribution equipment after time-series decay evolution at different physical parameter sensing times is calculated. Specifically, the insulation state of the power distribution equipment c after time-series decay evolution at the (r+1)th physical parameter sensing time is:

[0111]

[0112] Among them, f r+1 (c) represents the insulation state of power distribution equipment c at the (r+1)th physical parameter sensing time. Label((c)) represents the equipment type sensitivity coefficient corresponding to power distribution equipment c, which characterizes the vulnerability of the power distribution equipment to water erosion. It is set according to the type of power distribution equipment. The range of the equipment type sensitivity coefficient is between 0 and 1. The higher the equipment type sensitivity coefficient, the more likely the power distribution equipment is to fail after being eroded by water. He represents the water depth at location c of the power distribution equipment at the r-th physical parameter sensing time.c This indicates the installation height of power distribution equipment c. Let f0(c) represent the water level erosion function, Δ represent the time difference between the sensing times of adjacent physical parameters, f0(c) represent the initial insulation state of the power distribution equipment c, and he represent the preset installation height threshold, which is set to 0.3 meters.

[0113] Combining the water level time series and the temporal decay evolution results of the insulation state, a fault tree reasoning method is used to generate the probability of a fault event occurring for each power distribution device, including:

[0114] The water level change rate and the time-series decay evolution results of the insulation state of the power distribution equipment are obtained and used as nodes in the fault tree. A fault Boolean expression for the power distribution equipment is constructed. When the water level change rate is higher than a preset water level threshold and the insulation state is lower than a preset insulation state threshold, the output of the fault Boolean expression for the power distribution equipment is 1, indicating that the power distribution equipment is prone to a fault event in the short term. The time-series decay evolution results of the water level change rate and insulation state of the power distribution equipment are normalized and weighted, and logistic regression is performed to generate the probability of the power distribution equipment experiencing a fault event. The logistic regression uses the Sigmoid function. As an embodiment of the present invention, if the fault Boolean expression output is not 1, the probability of the fault time is set to 0.

[0115] S4: Integrate the underground power distribution room scene, water level evolution matrix, and the probability of each power distribution equipment failure event into the virtual reality training platform. The virtual reality training platform generates an underground power distribution room emergency repair training scene, and users conduct interactive emergency repair training in the underground power distribution room emergency repair training scene.

[0116] The virtual reality training platform integrates the underground power distribution room scenario, water level evolution matrix, probability of failure events for each power distribution device, and repair priorities into a virtual reality training platform. The platform generates underground power distribution room repair training scenarios, including:

[0117] Based on the underground power distribution room scenario, an underground power distribution room structural model is generated. The underground power distribution room scenario includes the location of the power distribution equipment and the scene point cloud of the underground power distribution room obtained by three-dimensional laser scanning.

[0118] The water level evolution matrix is ​​converted into a water level change animation using a VR engine. The water level change animation includes an animation of rising water surface in an underground power distribution room and the sound of rising water flow. As an embodiment of the present invention, the VR engine is Unity3D.

[0119] Generate fault scenes for each power distribution device after a failure event, including special effects animations of smoke and electric arcs;

[0120] The underground power distribution room structural model, water level change animation, probability of each power distribution device malfunction, and malfunction scenarios are integrated into a virtual reality training platform. The virtual reality training platform then generates an emergency repair training scenario for the underground power distribution room. Specifically, the emergency repair training scenario generates a training environment scene based on the underground power distribution room structural model, a dynamic water level change scene based on the water level change animation, and dynamically adjusts the malfunctioning power distribution devices based on the probability of each malfunction event, thus generating the malfunction scenario.

[0121] Example 2:

[0122] The user-interactive training module for the underground power distribution room emergency repair training scenario includes virtual inspection tasks, fault location challenges, dynamic training scoring, and power distribution equipment reset and replacement tasks. The virtual inspection task involves users simulating movement routes using VR controllers to identify flooded power distribution equipment in a flooded scenario and complete the inspection task. The fault location challenge involves quickly marking power distribution equipment that may malfunction and submitting emergency repair route planning suggestions. The dynamic training scoring involves users conducting emergency repair training through VR devices and receiving feedback based on whether the emergency repair route passes through high-risk areas and the accuracy of the emergency repair operation. The power distribution equipment reset and replacement task involves users operating virtual tools to perform interactive exercises such as switching off power, laying drainage pipes, and replacing equipment.

[0123] Example 3:

[0124] A simulation and emergency repair training system for flooding of underground power distribution rooms based on Internet of Things (IoT) data acquisition is provided to implement the simulation and emergency repair training method for flooding of underground power distribution rooms based on IoT data acquisition described in the aforementioned embodiments. The system includes a virtual reality training platform and a data acquisition device.

[0125] The data acquisition device is used to deploy sensors in the underground power distribution room to sense physical parameters and obtain physical parameter sensing data of the underground power distribution room under the flooding scenario. Combining the physical parameter sensing data, an improved shallow water diffusion wave model is used to simulate the water level evolution matrix of the underground power distribution room under the flooding scenario. The water level evolution matrix is ​​used to calculate the water level time series of each power distribution device in the underground power distribution room, and the insulation state of the power distribution device is subjected to time-series decay evolution. The probability of each power distribution device to experience a fault event is generated by fault tree reasoning.

[0126] The virtual reality training platform is used to obtain the underground power distribution room scene, the water level evolution matrix, and the probability of each power distribution equipment failure event, and to generate an underground power distribution room emergency repair training scene.

[0127] It should be understood that the embodiments described are for illustrative purposes only and are not limited to this structure in the scope of the patent application.

[0128] It should be noted that the sequence numbers of the above embodiments of the present invention are merely for descriptive purposes and do not represent the superiority or inferiority of the embodiments. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, apparatus, article, or method. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0129] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0130] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for simulating and training emergency repairs of flooding in underground power distribution rooms based on Internet of Things (IoT) data acquisition, characterized in that, The method includes: S1: Deploy sensors in the underground power distribution room to sense physical parameters and obtain physical parameter sensing data of the underground power distribution room under the flooding scenario; The types of physical parameters include rainfall intensity, soil permeability, floor roughness coefficient of the power distribution room, and drainage capacity of the power distribution room. S2: Combining physical parameter sensing data, an improved shallow water diffusion wave model is used to simulate the water level evolution matrix of an underground power distribution room under a flooding scenario. The water level evolution matrix is ​​the water level value at different locations in the underground power distribution room that evolves over time. The point cloud of the underground power distribution room was obtained by 3D laser scanning, and a digital elevation model was extracted from the point cloud. By combining physical parameter sensing data and digital elevation models, an improved shallow water diffusion wave model is constructed: ； ； ； ； in, Indicates the first in the underground power distribution room ground coordinates The water depth at the r-th physical parameter sensing time. This represents the time difference between the sensing times of adjacent physical parameters. Indicates the first ground coordinates In the global water volume allocation term at the r-th physical parameter sensing time, G represents the number of ground coordinates of the underground power distribution room. Representing ground coordinates The height of the location Represented by ground coordinates The set of neighboring ground coordinates in an area with a radius of 1 meter centered at . This indicates the weighting of water allocation based on elevation difference. Indicates the preset maximum height, set. It is 0.5 meters. Indicates the first ground coordinates The water diffusion term at the r-th physical parameter sensing time; The water distribution term refers to the water distribution value of surface precipitation to the ground coordinates of the underground power distribution room, which is calculated from rainfall intensity, soil permeability, and the unit drainage volume of the power distribution room at a unit coordinate. The unit drainage volume of the underground power distribution room at the r-th physical parameter sensing time; The water diffusion term is the water distribution value from the ground coordinates in the underground power distribution room to the adjacent ground coordinates; S represents the floor area of ​​the underground power distribution room; The results are, in order, the fusion of rainfall intensity, soil permeability, substation floor roughness coefficient, and substation drainage capacity at the r-th physical parameter sensing time from the physical parameter sensing data. Represents the set of nearby ground coordinates Water depth relative to the nearest ground coordinates The amount of diffusion; The water depth at different ground coordinates and at different physical parameter sensing times is used as the water level evolution matrix. This water level evolution matrix is ​​an R-row, G-column matrix. The water level evolution matrix is ​​defined as follows: [The matrix is ​​described in the original text, but the provided text is incomplete and requires further context.] g Listed as the first underground power distribution room ground coordinates Water depth at the r-th physical parameter sensing time ; S3: Calculate the water level time series of each power distribution device in the underground power distribution room using the water level evolution matrix, perform time-series decay evolution on the insulation state of the power distribution device, and generate the probability of each power distribution device experiencing a fault event using fault tree reasoning. S4: Integrate the underground power distribution room scene, water level evolution matrix, and the probability of each power distribution equipment failure event into the virtual reality training platform. The virtual reality training platform generates an underground power distribution room emergency repair training scene, and users conduct interactive emergency repair training in the underground power distribution room emergency repair training scene.

2. The method for simulating and training emergency repairs of flooded underground power distribution rooms based on IoT data acquisition as described in claim 1, characterized in that, The types of sensors include tipping bucket rain gauges, soil conductivity sensors, electromagnetic flow meters, level gauges, and flow meters, including: Multiple sensors of the same type are deployed in a fixed area of ​​the underground power distribution room to obtain the sensing results of multiple physical parameters of the same type of sensor in the fixed area. The tipping bucket rain gauge is used to sense the rainfall intensity, the soil conductivity sensor is used to sense the soil permeability, the electromagnetic flow meter and the liquid level gauge are used to sense the surface roughness coefficient of the power distribution room, and the liquid level gauge and the flow meter are used to sense the drainage capacity of the power distribution room.

3. The method for simulating and training emergency repairs of flooded underground power distribution rooms based on IoT data acquisition as described in claim 2, characterized in that, An adaptive weighted fusion method is used to fuse the sensing results of all physical parameters from sensors of the same type, including: The adaptive weighted fusion formula is as follows: ； ； in, Indicates the first The fusion processing result of the various physical parameters at the r-th physical parameter sensing time. Indicates the use of collecting the first The physical parameter sensing result of the nth sensor at the rth physical parameter sensing time, where R represents the number of physical parameter sensing times. N represents the number of sensors of the same type deployed in the fixed area. Indicates the use of collecting the first The adaptive weight of the nth sensor for a certain physical parameter at the time of sensing the rth physical parameter, wherein the first to fourth physical parameters are rainfall intensity, soil permeability, floor roughness coefficient of the power distribution room and drainage capacity of the power distribution room, respectively. This is the weight adjustment coefficient, used to adjust the weights of different factors. The values ​​are 0.6, 0.2, and 0.2, respectively. Results of physical parameter sensing The corresponding data deviation factor is used to measure the results of physical parameter sensing. The degree of deviation from the historical moving average, wherein the data deviation factor is a standardized numerical form; Results of physical parameter sensing The corresponding packet loss rate factor reflects the data collection rate. The stability of data upload from the nth sensor for a certain physical parameter over a past period of time; the larger the value of the packet loss rate factor, the stronger the stability of the data acquisition from the nth sensor. The more severe the packet loss situation of the nth sensor of a certain physical parameter, the lower the transmission stability. The packet loss rate factor is a numerical form between 0 and 1. Results of physical parameter sensing The corresponding sensor health status factor reflects the data collected at the first... The current health status of the nth sensor for a given physical parameter; the larger the value of the sensor health status factor, the stronger the signal. The worse the current health status of the nth sensor of a certain physical parameter, the worse the current health status factor of the sensor is, which is a numerical form between 0 and 1; The fusion results of different physical parameters at R physical parameter sensing times are used as physical parameter sensing data.

4. The method for simulating and training emergency repairs of flooded underground power distribution rooms based on IoT data acquisition as described in claim 1, characterized in that, The water level time series of each power distribution device in the underground power distribution room is calculated using the water level evolution matrix, and the insulation state of the power distribution device is subjected to time-series decay evolution, including: Two-dimensional ground position coordinates of each power distribution device in the underground power distribution room are obtained, and the water depth at the sensing time of R physical parameters corresponding to the two-dimensional ground position coordinates is extracted from the water level evolution matrix as the water level time series of the power distribution device. Using the water level time series, the insulation state of the power distribution equipment after time-series decay evolution at different physical parameter sensing times is calculated.

5. The method for simulating and training emergency repairs of flooded underground power distribution rooms based on IoT data acquisition as described in claim 4, characterized in that, Combining the water level time series and the temporal decay evolution results of the insulation state, a fault tree reasoning method is used to generate the probability of a fault event occurring for each power distribution device, including: The water level change rate and the time-series decay evolution results of the insulation state of the power distribution equipment are obtained and used as nodes in the fault tree. A fault Boolean expression for the power distribution equipment is constructed. When the water level change rate is higher than a preset water level threshold and the insulation state is lower than a preset insulation state threshold, the fault Boolean expression of the power distribution equipment outputs 1, indicating that the power distribution equipment is prone to a fault event in the short term. The time-series decay evolution results of the water level change rate and insulation state of the power distribution equipment are normalized, weighted, and subjected to logistic regression to generate the probability of the power distribution equipment experiencing a fault event.

6. The method for simulating and training emergency repairs of flooded underground power distribution rooms based on IoT data acquisition as described in claim 1, characterized in that, The virtual reality training platform integrates the underground power distribution room scenario, water level evolution matrix, probability of failure events for each power distribution device, and repair priorities into a virtual reality training platform. The platform generates underground power distribution room repair training scenarios, including: Based on the underground power distribution room scenario, an underground power distribution room structural model is generated. The underground power distribution room scenario includes the location of the power distribution equipment and the scene point cloud of the underground power distribution room obtained by three-dimensional laser scanning. The water level evolution matrix is ​​converted into a water level change animation using a VR engine. The water level change animation includes an animation of water level rising in an underground power distribution room and the sound of water flowing upward. Generate fault scenes for each power distribution device after a failure event, including special effects animations of smoke and electric arcs; The underground power distribution room structure model, water level change animation, probability of each power distribution equipment failure event, and failure scenarios are integrated into the virtual reality training platform, which generates emergency repair training scenarios for the underground power distribution room.

7. A simulation and emergency repair training system for flooding of underground power distribution rooms based on Internet of Things (IoT) data acquisition, characterized in that, The underground power distribution room flooding simulation and emergency repair training system based on Internet of Things data acquisition includes a virtual reality training platform and a data acquisition device: The data acquisition device is used to deploy sensors in the underground power distribution room to sense physical parameters and obtain physical parameter sensing data of the underground power distribution room under the flooding scenario. Combining the physical parameter sensing data, an improved shallow water diffusion wave model is used to simulate the water level evolution matrix of the underground power distribution room under the flooding scenario. The water level evolution matrix is ​​used to calculate the water level time series of each power distribution device in the underground power distribution room, and the insulation state of the power distribution device is subjected to time-series decay evolution. The probability of each power distribution device to experience a fault event is generated by fault tree reasoning. The virtual reality training platform is used to obtain the underground power distribution room scene, the water level evolution matrix, and the probability of each power distribution equipment failure event, and to generate an underground power distribution room emergency repair training scene. To achieve the simulation and emergency repair training method for flooding of underground power distribution rooms based on Internet of Things data acquisition as described in any one of claims 1-6.

Images