A station-line-house penetration dynamic rendering method based on energy flow diagram

Abstract

The present application belongs to the technical field of power system visualization, and particularly relates to a station-line-substation-penetration dynamic rendering method based on an energy flow diagram. The method first arranges edge computing nodes at key points of a station-line-substation, improves time series data compression to reduce redundant transmission, and provides real-time accurate data for dynamic rendering. Then, an improved layered topology modeling algorithm is proposed to accurately analyze and predict power flow, so that the energy flow diagram rendering is more scientific. Then, according to the user interaction demand, a local energy flow visualization layer is generated to reduce the calculation pressure and improve the interaction response speed and the energy flow diagram display fluency. Finally, based on WebGPU and Shader optimization, the smooth dynamic rendering of two-dimensional and three-dimensional energy flow diagrams is realized. This method greatly reduces the calculation pressure, realizes fast interaction, and smoothly displays the dynamic changes of the local energy flow diagram.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of power system visualization, and particularly relates to a station-line-transformer-penetration dynamic rendering method based on an energy flow diagram. BACKGROUND

[0002] In order to meet the rapid interactive demand of users for detailed energy flow information in a local area and reduce the pressure of whole-network visualization calculation, the application dynamically generates a local energy flow data visualization layer according to the interactive demand of users. In today's era, the power system continues to develop rapidly, and the complexity of the power grid structure presents an exponential growth trend. This change makes the visualization analysis of the operation state of the power grid an indispensable important link in the power industry, and the dynamic rendering of the energy flow diagram is a core technical requirement.

[0003] The existing dynamic rendering method of the energy flow diagram exposes many defects that are difficult to overcome in actual application. In the initial stage of data collection, due to the limitation of technical means, real-time data collection and processing work cannot be efficiently carried out. The acquisition of data often has obvious hysteresis, which makes the energy flow diagram extremely slow in dynamic updating and cannot accurately and timely reflect the rapid changes of the real-time state of the power grid. The power grid topology modeling link is also unsatisfactory, lacking effective strategies suitable for the characteristics of the hierarchical structure of the power grid. This shortcoming directly leads to the difficulty in accurately analyzing the power flow, and further makes the energy flow diagram unable to accurately present the dynamic allocation and flow trend of power in each part of the power grid in the dynamic rendering process, and unable to provide reliable decision basis for operation and maintenance personnel. In the visualization calculation process, the problem is also quite serious. Since the calculation method is not effectively optimized, when facing massive power grid data, the pressure of whole-network visualization calculation is like a mountain on the head. When performing dynamic display of two-dimensional and three-dimensional energy flow diagrams, the phenomena of lagging and frame dropping frequently occur, just like the picture is constantly "stuck" when playing. This not only seriously interferes with the dynamic perception of the user to the energy flow information of the power grid, making it unable to clearly and coherently grasp the energy flow change, but also greatly hinders the ability of the user to analyze and accurately judge the operation state of the power grid. SUMMARY

[0004] The application proposes a station-line-transformer-penetration dynamic rendering method based on an energy flow diagram to solve the above technical problems.

[0005] In order to achieve the above purpose, the technical scheme adopted by the application comprises the following steps:

[0006] S1, first, edge computing nodes are arranged at key points of stations, lines, transformers and houses to realize real-time collection and distributed processing of data, and the distributed processing improves time sequence data compression to reduce redundant data transmission;

[0007] S2, an improved hierarchical topology modeling algorithm is proposed, which decomposes the power grid topology by level and introduces time-based energy flow prediction to support the combination of real-time data and predicted data;

[0008] S3, then according to the user interaction demand, dynamically generate the visualization layer of local energy flow data, reduce the calculation pressure of the whole network visualization, the specific implementation is:

[0009] S31, first of all, based on the hierarchical topology of power grid G=(N,ε,W), calculate the minimum subgraph G sub of the user specified area, and adjust dynamically according to the real-time interaction;

[0010] S32, then according to the extracted subgraph G sub , dynamically generate the energy flow visualization layer of the local area, for each edge e ij ∈ε sub ,ε sub is the edge set of the subgraph G sub , calculate its energy flow intensity S ij : Where P ij is the power flow from node i to j, W ij is the power transmission capacity of edge e ij , and the energy flow intensity is used to dynamically adjust the thickness and color of the edge in the layer; for each node i∈N sub , N sub is the node set of the subgraph G sub , through the formula , where S i represents the proportion of node load, which is mapped to the color and size of the node icon; based on the results of S ij and S i , only draw the key edges and nodes to reduce the rendering complexity;

[0011] S33, finally through the improved intelligent area perception algorithm, ensure that under real-time interaction, introduce the area perception weight ω i , only process the data in the interested area N sub , when the user interaction changes, only update the energy flow calculation and cache the generated local layer data for the newly added and reduced node set ΔN sub , directly call the cache result when the user repeats the query, reduce the overhead of multiple calculations;

[0012] S4, finally based on the dynamic rendering of WebGPU and Shader optimization, realize the smooth display of two-dimensional and three-dimensional energy flow graph.

[0013] As preferred, the implementation step of the distributed processing to improve the time series data compression in the step S1 is to use the least square method to fit the time series data {(x1, y1), (x2, y2),..., (x n , y n )} into a straight line y = ax + b, where The fitting error calculation formula is: Where x i is the time point, y i is the data value corresponding to the time point, n is the sample number of the current segment, and the segment data is compressed into the start point, end point and fitting parameter when the error is less than the error tolerance threshold.

[0014] As preferred, the implementation step of the hierarchical topology modeling algorithm in the step S2 is:

[0015] S21, the power grid topology is divided into four layers of station, line, transformer and user according to the level, forming a multi-layer topology structure based on graph, the node set is N = {N1, N2, N3, N4}, where N 站 , N 线 , N 变 , N 户 are respectively the substation node, the power station node, the transformer node and the user node, ε = {e ij |i∈N l ,j∈N l+1 ,l=1,2,3} where e ij represents the connection relationship between node i in level l and node j in level l+1, a weighted directed graph G = (N, ε, W) is used, where the weight matrix W represents the power transmission capacity of the line;

[0016] S22, in the hierarchical topology structure, the power flow follows the physical law of the power grid, and each node satisfies the power balance equation: P in,i = P out,i + P loss,i , where P in,i is the input power of node i, P out,i is the output power of node i, and P loss,i is the power loss, which can be further defined as: Where P ij is the power from node i to j, R ij is the resistance of the line, and V i is the voltage of node i. The power transmission relationship of each layer of nodes is calculated, and the whole network power flow state is recursively updated;

[0017] S23, then the power of node i is predicted by using LSTM, and the interlayer coupling constraint is used to optimize the prediction result, so that the prediction results of upper and lower layers are consistent, and the objective function is: Where is the future power prediction value of node i, and l is the number of layers, is the total predicted input power of all relevant nodes of the next layer;

[0018] S24, for the problem of differences between real-time data and predicted data, a dynamic data fusion method is used to fuse the power value: where P real,i,t+1 is real-time data, is predicted data; φ real,i , and φ pred,i are the reliabilities of real-time data and predicted data, respectively.

[0019] As a preferred, the constraint condition of the objective function in step S23 is to ensure that the input power of each node is equal to the sum of the output power and the loss power, the output power of each node cannot exceed its design capacity P capacity,i .

[0020] Compared with the prior art, the advantages and positive effects of the present application are that, through edge computing nodes and optimized time sequence data compression, more timely and concise data basis is provided for dynamic rendering, ensuring that the rendered content can reflect the latest state of the power grid and the data transmission is efficient. The improved hierarchical topology modeling algorithm incorporates power flow prediction, making the rendering more scientific and accurate, and better presenting the power flow trend. When generating the visualization layer, the user's interested area is focused, the data is optimized and cached, the computing pressure is greatly reduced, fast interaction is realized, and the dynamic changing local power flow diagram is smoothly displayed. DETAILED DESCRIPTION

[0021] In order to enable the above-mentioned purposes, features and advantages of the present application to be more clearly understood, the present application will be further described below with reference to the embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0022] In the following description, many specific details are set forth in order to provide a thorough understanding of the present application, but the present application can also be implemented in other ways different from those described herein, therefore, the present application is not limited to the specific embodiments disclosed in the following description.

[0023] With the development of power systems, the structure of power grid is becoming increasingly complex, and the demand for visual analysis of the operating state of the power grid is growing, especially the dynamic rendering of the power flow diagram has become a key technical requirement. The traditional dynamic rendering method of the power flow diagram has many defects. In the data acquisition link, efficient real-time data acquisition and processing cannot be realized, leading to slow dynamic updating of the power flow diagram and inability to timely reflect the real-time state changes of the power grid. The power grid topology modeling lacks adaptability and cannot accurately analyze power flow, so that the power flow diagram cannot accurately present the dynamic allocation and flow trend of power during dynamic rendering. In the process of visualization calculation, due to the ineffective optimization of the calculation method, the pressure of the whole network visualization calculation is huge in the face of massive data, and the phenomenon of frequent freezing and frame dropping occurs during the dynamic display of two-dimensional and three-dimensional power flow diagrams, which seriously affects the dynamic perception and analysis of the user on the power flow information of the power grid.

[0024] In order to realize efficient collection and processing of station-line-transformer-house data in the power system and reduce data transmission redundancy, the present application adopts an edge computing node arranged at the key points of station-line-transformer-house and improves the time series data compression scheme. In traditional power grid data processing, the data acquisition efficiency is low and the transmission redundancy is serious. The present application arranges an edge computing node at the key points of station-line-transformer-house to realize real-time data acquisition and distributed processing. The least square method is used to fit the time series data {(x1, y1), (x2, y2),..., (x n ,y n )} into a straight line y = αx + β, wherein The fitting error calculation formula is: wherein x i is the time point, y i is the data value corresponding to the time point, and n is the sample number of the current segment. When the error is less than the error tolerance threshold, the segment data is compressed into the starting point, the ending point and the fitting parameters. In this way, a large amount of similar or slowly changing data can be effectively compressed, reducing the amount of data to be transmitted and improving the real-time performance of data acquisition.

[0025] Then, considering that the existing power grid topology modeling lacks effective hierarchical strategy and is difficult to accurately analyze power flow, in order to accurately construct the power grid topology model and accurately analyze and predict the power flow of the power grid, the present application proposes an improved hierarchical topology modeling algorithm. The power grid topology is divided into four layers of station, line, transformer and house according to the level, forming a multi-layer topology structure based on graph, and the node set is N = {N1, N2, N3, N4}, wherein N 站 , N 线 , N 变 , N 户 are respectively the substation node, the transmission station node, the transformer node and the user node, and ε = {e ij |i∈N l , j∈N l+1 , l = 1, 2, 3} wherein eij The connection relationship between node i in level l and node j in level l+1 is represented by a weighted directed graph G=(N,ε,W), wherein the weight matrix W represents the power transmission capability of the line; in the hierarchical topology structure, the power flow follows the physical law of the power grid, and each node satisfies the power balance equation: P in,i in out,i +P loss,i , wherein P in,i is the input power of node i, P out,i is the output power of node i, and P loss,i is the power loss, which can be further defined as: wherein P ij is the power from node i to j, R ij is the resistance of the line, and V i is the voltage of node i; the power transmission relationship of each layer node is calculated, and the power flow state of the whole network is recursively updated; then the power of node i is predicted by using LSTM, the prediction result is optimized by using the inter-layer coupling constraint, so that the prediction results of the upper and lower layers are consistent, and the objective function is: wherein is the future power prediction value of node i, l is the number of layers, is the total predicted input power of all related nodes in the next layer, and the constraint condition of the objective function is to ensure that the input power of each node is equal to the sum of the output power and the loss power, and the output power of each node cannot exceed its designed capacity P capacity,i ; in view of the difference between real-time data and predicted data, the dynamic data fusion method is adopted, and the fused power value is: wherein P real,i,t+1 is real-time data, is predicted data; φ real,i , φ pred,i are the reliabilities of real-time data and predicted data respectively. Through this hierarchical topology modeling algorithm, when the power flow of a city power grid is analyzed and predicted, the power transmission and change trend between different levels can be accurately simulated, and potential problems can be found in advance

[0026] Then, in order to meet the rapid interactive demand of users for detailed energy flow information in a local area and reduce the calculation pressure of the whole network visualization, the application dynamically generates a local energy flow data visualization layer according to the user interactive demand. First, based on the hierarchical topology G=(N,ε,W) of the power grid, the minimum subgraph G sub of the user-specified area is calculated, the whole network calculation is concentrated to the local area interested by the user, and is dynamically adjusted according to real-time interaction; then, according to the extracted subgraph G sub , a local energy flow visualization layer is dynamically generated, and each edge e ij ∈εsub , ε sub For subgraph G sub Calculate the energy flux intensity S of the edge set. ij : Where P ij For the power flow from node i to j, W ij For edge e ij The power transmission capability and energy flow intensity are used to dynamically adjust the thickness and color of edges in the layer; for each node i∈N sub N sub Subgraph G sub The set of nodes, through Formulas are used for aggregation, where S i Represents the node load percentage, mapped to the node's color and icon size; based on S ij and S i As a result, only key edges and nodes are drawn, reducing rendering complexity; finally, through an improved intelligent region awareness algorithm, the region awareness weight ω is introduced to ensure real-time interaction. i Only the region of interest N is processed. sub Within the data, when user interaction changes, only the newly added and removed node set ΔN is updated. sub The system updates energy flow calculations and caches the generated local layer data. When users query repeatedly, they can directly access the cached results, reducing the overhead of multiple calculations. In this way, users can quickly obtain detailed energy flow information for a local area, significantly improving the interactive response speed, while also making the display of 2D and 3D energy flow maps smoother.

[0027] Finally, dynamic rendering based on WebGPU and Shader optimizations is implemented. WebGPU fully leverages the parallel processing capabilities of modern graphics hardware to accelerate the visualization and rendering of energy flow data. Shaders optimize vertex shaders and fragment shaders in the graphics rendering pipeline, adjusting color and transparency calculations based on energy flow intensity to reduce unnecessary computational overhead. When displaying two-dimensional and three-dimensional energy flow maps of large-scale power grids, stuttering and latency are effectively avoided, achieving smooth dynamic rendering effects.

[0028] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments for application in other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A station-line-to-house penetration dynamic rendering method based on energy flow diagram, characterized in that, Comprise the following steps: S1, first adopt edge computing node arrangement in the key point of station line change household, realize real-time data collection and distributed processing, the distributed processing improves time series data compression to reduce redundant data transmission; S2, an improved hierarchical topology modeling algorithm is proposed, which decomposes the power grid topology by level and introduces time-based energy flow prediction to support the combination of real-time data and predicted data; S3, then according to the user interaction demand dynamic generation local energy flow data visualization layer, reduce the whole network visualization calculation pressure, the specific implementation is: S31, first, based on the weighted directed graph G=(N,ε,W) of the power grid, calculate the minimum subgraph G of the user-specified region sub Concentrate the whole network calculation to the local area of interest of the user, and dynamically adjust according to real-time interaction; S32, Next, based on the extracted subgraph G sub Dynamically generate local energy flow visualization layers, with each edge e ij ∈ε sub , ε sub For subgraph G sub Calculate the energy flux intensity S of the edge set. ij : Where P ij For the power flow from node i to j, W ij For edge e ij The power transmission capability and energy flow intensity are used to dynamically adjust the thickness and color of edges in the layer; for each node i∈N sub N sub For subgraph G sub The set of nodes, through Formulas are used for aggregation, where S i Represents the node load percentage, mapped to the node's color and icon size; based on S ij and S i As a result, only key edges and nodes are drawn, reducing rendering complexity; S33, finally, through the improved intelligent region perception algorithm, ensure that under the real-time interaction, the introduction of region perception weight ω i , only processing the data in the region of interest N sub , when the user interaction changes, only the newly added and reduced node set ΔN sub update the energy flow calculation and cache the generated local layer data, directly call the cache result when the user repeatedly queries, reduce the overhead of multiple calculations; S4, finally based on WebGPU and Shader optimization dynamic rendering, realize the smooth display of two-dimensional and three-dimensional energy flow diagram; The implementation steps of the hierarchical topology modeling algorithm in step S2 are: S21, the power grid topology is divided into four layers of station, line, transformer and user according to the level, forming a multi-layer topology structure based on graph, the node set is N={N1, N2, N3, N4}, wherein N1, N2, N3 and N4 are respectively substation node, power station node, transformer node and user node, ε={e ij |i∈N l ,j∈N l+1 ,l=1,2,3} wherein e ij represents the connection relationship between node i in level l and node j in level l+1, a weighted directed graph G=(N,ε,W) is adopted, wherein the weight matrix W represents the power transmission capacity of the line; S22、In the layered topology, the power flow follows the physical law of the power grid, and each node satisfies the power balance equation: P in,i = P out,i + P loss,i , where P in,i is the input power of node i, P out,i is the output power of node i, P loss,i is the power loss, and the power loss is defined as: where P ij is the power from node i to j, R ij is the resistance of the line, and V i is the voltage of node i. The power transfer relationship of each layer node is calculated, and the whole network power flow state is updated recursively; S23, then the power of node i is predicted by LSTM, and the prediction result is optimized by inter-layer coupling constraint to make the prediction results of upper and lower layers consistent, and the objective function is: wherein is the future power prediction value of node i, and l is the number of layers, is the total predicted input power of all related nodes in the next layer; S24, in view of the difference between real-time data and predicted data, a dynamic data fusion method is adopted, and the power value is: Where P real,i,t+1 is real-time data, is predicted data; respectively, the reliability of real-time data and predicted data.

2. The station-line-house penetration dynamic rendering method based on energy flow graph according to claim 1, wherein, The implementation step of improving the time series data compression in the step S1 distributed processing is to use the least square method to fit the time series data {(x1, y1), (x2, y2),..., (x n , y n )} into a straight line y = ax + b, wherein The fitting error calculation formula is: Wherein x i is the time point, y i is the data value corresponding to the time point, n is the sample number of the current segment, and the segment data is compressed into the start point, end point and fitting parameter when the error is less than the error tolerance threshold.

3. The station-line-house penetration dynamic rendering method based on energy flow graph according to claim 1, wherein, The constraint condition of the objective function in the step S23 is Guarantee that the input power of each node is equal to the sum of the output power and the loss power, P out,i ≤ P capacity,i The output power of each node cannot exceed its design capacity P capacity,i .

Images