Power line digital twinning scene construction method and system, medium and terminal

Abstract

The invention discloses a power line digital twinning scene construction method and system, a medium and a terminal, and the method comprises the steps: optimizing the parameters of a support vector machine model through a parameter-adaptive hybrid particle swarm optimization algorithm which is formed by introducing a surrounding motion and spiral motion mechanism of a whale algorithm, and obtaining a state prediction model; acquiring first operation data of the power line in real time, and inputting the first operation data into the state prediction model; and updating the twinborn data model according to the first operation data and an electric power line operation state prediction result output by the state prediction model, and then performing data exchange on the updated twinborn data model and the twinborn entity model so as to construct the digital twinborn scene of the electric power line. By adopting the embodiment of the invention, in the optimization process of the model for predicting the operation state of the power line, the search of a local optimal solution is avoided, the global search capability is enhanced, the efficient optimization of model parameters is realized, and the prediction accuracy of the operation state of the power line is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of digital scene construction, and in particular to a method, system, medium and terminal for constructing a digital twin scene of a power line. Background Art

[0002] Digital twins are the bridge between the physical world of power grids and the digital virtual world. As the key to the digital transformation of power grids, digital twin technology is being continuously promoted in the application of power grid engineering. Introducing digital twin technology into the power grid can achieve synchronous mapping of the physical power grid. During long-term operation, power lines are subject to complex and diverse loads and environmental influences, and it is difficult to detect the existence of safety hazards. Therefore, for the problem of monitoring the operating status of power line power grid equipment, it is necessary to establish a digital twin model that can reflect the operating status of power lines in real time.

[0003] At present, the main idea of ​​obtaining the operating status of power grid equipment is to collect data through sensors deployed on power line power grid equipment, and visualize the collected data. Through the analysis of the operating data by operation and maintenance personnel, the final evaluation results of the power grid equipment are obtained. The operation and maintenance personnel optimize the equipment status according to the equipment problems found, and remind the operation and maintenance personnel to increase monitoring or arrange planned repairs. However, the technology for monitoring and predicting the operating status of power grid equipment has the following defects: it is impossible to achieve comprehensive visual perception of the operating status of power grid equipment; it is impossible to predict and optimize the operating status of power grid equipment in real time, so that the equipment operation and maintenance personnel cannot intervene in advance. Summary of the invention

[0004] The embodiments of the present invention provide a method, system, medium and terminal for constructing a digital twin scenario of a power line, which can predict the future operating status of the power line in real time, avoid searching for local optimal solutions and enhance the global search capability during the optimization of the state prediction model, thereby realizing efficient optimization of model parameters.

[0005] In order to solve the above technical problems, an embodiment of the present invention provides a method for constructing a digital twin scene of a power line, comprising:

[0006] The parameters of the support vector machine model are optimized by using a parameter-adaptive hybrid particle swarm optimization algorithm to obtain a state prediction model; wherein the parameter-adaptive hybrid particle swarm optimization algorithm is formed by introducing the encircling motion and spiral motion mechanisms of the whale algorithm;

[0007] collecting first operating data of the power line in real time, and inputting the first operating data into the state prediction model, so that the state prediction model outputs an operating state prediction result of the power line;

[0008] The twin data model is updated with the first operating data and the operating status prediction result, and then the updated twin data model and the twin entity model are exchanged with data to construct a digital twin scenario of the power line; wherein the twin data model is pre-constructed based on the second operating data of the power line, and the twin entity model is pre-constructed based on the structural data of the power line.

[0009] By implementing the embodiment of the present invention, the surrounding motion and spiral motion mechanism of the whale algorithm are introduced to form a parameter-adaptive hybrid particle swarm optimization algorithm, which can avoid local optimal solutions during the search process and solve the problem that the particle swarm optimization algorithm is easily trapped in the local optimal state. In addition, by introducing the surrounding motion and spiral motion mechanism of the whale algorithm, the global search capability can be enhanced and the local optimal area can be finely searched to efficiently search for the optimal solution in a larger range. The parameter-adaptive hybrid particle swarm optimization algorithm is used to optimize the parameters of the support vector machine model to obtain a state prediction model for predicting the operating state of the power line, which can improve the accuracy and generalization ability of the state prediction model, thereby improving the prediction accuracy of the operating state of the power line. Then, the first operating data of the power line is collected in real time, and the first operating data is input into the state prediction model, so that the state prediction model outputs the operating state prediction result of the power line, so as to assist the operation and maintenance personnel to timely understand and grasp the future operating state of the power line. In addition, by using the first operating data collected in real time and the operating status prediction results of the power lines, the twin data model is continuously updated, so that the twin data model can reflect the latest status of the power lines and realize dynamic synchronization of the models. Then, the updated twin data model and the twin entity model are exchanged with data to build a digital twin scene of the power lines. It can not only simulate various operating states of the power lines in the virtual scene, but also enhance the fit between the digital twin scene and the actual scene of the power lines, thereby enhancing the adaptability and accuracy of the digital twin scene.

[0010] As a preferred solution, the hybrid particle swarm optimization algorithm with adaptive parameters is used to optimize the parameters of the support vector machine model to obtain a state prediction model, which is specifically:

[0011] Constructing a support vector machine model, and using a parameter-adaptive hybrid particle swarm optimization algorithm to search for an optimal combination of a penalty factor and a kernel parameter of the support vector machine model;

[0012] The optimal combination is used to optimize the parameters of the support vector machine model to obtain a state prediction model.

[0013] The preferred scheme of implementing the embodiment of the present invention utilizes a parameter-adaptive hybrid particle swarm optimization algorithm to automatically optimize the penalty factor and kernel parameters of the support vector machine model, thereby obtaining the optimal combination of the penalty factor and kernel parameters of the support vector machine model, which can effectively prevent the overfitting of the model on the training data and enhance the generalization ability of the model on unknown data.

[0014] As a preferred solution, the hybrid particle swarm optimization algorithm with adaptive parameters is used to search for the optimal combination of the penalty factor and the kernel parameter of the support vector machine model, specifically:

[0015] Creating an initial population; wherein the initial population includes a plurality of particles, and the positions of the particles include a penalty factor and a kernel parameter of the support vector machine model;

[0016] Iterative optimization is performed on each particle in the initial population. In each iterative optimization process, the position and speed of the current particle are updated based on the individual optimal value and the population optimal value of the current particle, and the fitness value of the position of the current particle is obtained as the fitness value corresponding to this iteration, until the fitness value corresponding to the current particle in this iteration is less than the fitness value corresponding to the previous iteration, then the whale algorithm is used to perform multiple position optimization on the current particle, and after each optimization, the fitness value of the position of the current particle is obtained as the fitness value corresponding to this optimization, until the fitness value corresponding to the current particle in this optimization is less than or equal to the fitness value corresponding to the previous optimization, then the iterative optimization of the current particle is terminated, and the position of the current particle at this time is used as the iterative optimization result of the current particle.

[0017] The preferred solution of the embodiment of the present invention is implemented. In the particle swarm optimization process, the position and velocity of the particle are updated by comparing the individual optimal value of the current particle with the optimal value of the population, which can accelerate the convergence speed, and then efficiently and automatically find the optimal penalty factor and optimal kernel parameter of the support vector machine model. In addition, when the particle falls into the local optimum or the fitness value is no longer improved, the introduction of the whale algorithm for fine search can improve the accuracy of the final solution.

[0018] As a preferred solution, the whale algorithm is used to perform multiple position optimization on the current particle, specifically:

[0019]

[0020] Where p represents a random number in the range of 0 to 1. When p is less than or equal to 0.5, the surrounding motion mechanism is used to update the position of the current particle. When p is greater than 0.5, the spiral motion mechanism is used to update the position of the current particle. represents the position of the i-th particle at the j+1-th iteration; represents the optimal value of the population of the i-th particle; A represents the vector coefficient; D represents the distance vector between the current particle and the global optimal particle; l represents a constant.

[0021] The preferred solution of the embodiment of the present invention is to select the encircling motion or the spiral motion by random numbers, so that the search strategy of the particles is adaptive and can be dynamically adjusted according to the current search environment, thereby improving the search efficiency and effect.

[0022] As a preferred solution, the construction of the twin entity model is specifically as follows:

[0023] Acquire first point cloud data of a pole tower of a power line, and establish a coordinate system of the pole tower based on the first point cloud data;

[0024] According to a preset local curvature algorithm, based on the horizontal coordinates of a plurality of characteristic points of the tower, the local curvature of each characteristic point is calculated, and based on the local curvature of each characteristic point, the spatial position of each characteristic point is analyzed;

[0025] Combining the coordinate system of the tower and the spatial position of each of the characteristic points, constructing a GIM model of the power tower;

[0026] The power pole tower GIM model and the power line scene model are respectively converted into the same preset coordinate system, and then the power pole tower GIM model and the power line scene model are fused to obtain a twin entity model; wherein the power line scene model is constructed based on the actual terrain data and real-time image data of the area where the power line is located.

[0027] The preferred solution of the embodiment of the present invention is to establish the coordinate system of the tower based on the first point cloud data of the power line tower, and analyze the spatial position of each feature point based on the local curvature of several feature points of the tower, and then combine the coordinate system of the tower and the spatial position of each feature point to respond to the change of the tower structure and create a high-precision power tower GIM model in real time to accurately reflect the spatial geometric characteristics and details of the tower. In addition, the power tower GIM model and the power line scene model are respectively converted to the same preset coordinate system, and then the two models that have completed the coordinate system conversion are merged to form a twin entity model, which can display a panoramic view of the power line in a unified coordinate system, including towers, lines, terrain, etc., to provide intuitive scene display for operation and maintenance personnel.

[0028] As a preferred solution, the first point cloud data of the tower of the power line is obtained, specifically:

[0029] Scanning the area where the power line is located multiple times at multiple scanning angles to obtain second point cloud data of the power line at multiple scanning angles;

[0030] Registering the second point cloud data of the power line at multiple scanning angles, and then extracting features from the registered second point cloud data to obtain local geometric features corresponding to the power line;

[0031] Based on the local geometric features corresponding to the power line, first point cloud data of the tower of the power line is segmented from the second point cloud data of the power line at multiple scanning angles.

[0032] The preferred solution of the embodiment of the present invention is to extract features from the registered second point cloud data, accurately identify the local geometric features of the power line, such as the shape and structure of the tower, and then segment the individual tower point cloud data from the multi-angle point cloud data based on the feature extraction results, so as to achieve high-precision independent modeling of the tower, which is helpful for the individual analysis and processing of the tower. In addition, by scanning the area where the power line is located multiple times according to multiple scanning angles, the second point cloud data of the power line at different viewing angles can be obtained to ensure that each area can be covered and avoid data loss.

[0033] As a preferred solution, the real-time collection of first operating data of the power line and the input of the first operating data into the state prediction model so that the state prediction model outputs the operating state prediction result of the power line are specifically:

[0034] Collecting the first operation data of the power line in real time;

[0035] The first operating data is input into the state prediction model so that the state prediction model performs collective empirical mode decomposition on the first operating data to obtain IMF components and residual components, and predicts and denormalizes the IMF components and the residual components respectively, and then superimposes the denormalization results corresponding to the IMF components and the denormalization results corresponding to the residual components, so that the state prediction model outputs a prediction result of the power line operating state.

[0036] A preferred scheme for implementing the embodiment of the present invention is to perform collective empirical mode decomposition on the first operating data of the power line collected in real time through a state prediction model to obtain IMF components and residual components, which can more finely capture different time scales and change trends in the data, thereby improving the prediction accuracy, and perform prediction and denormalization processing on the IMF components and residual components in turn, respectively, to ensure the flexibility and accuracy of data processing, and then superimpose the denormalization processing results corresponding to the IMF components and the denormalization processing results corresponding to the residual components to obtain the output result of the state prediction model - the power line operation state prediction result.

[0037] In order to solve the same technical problem, an embodiment of the present invention further provides a power line digital twin scene construction system, including:

[0038] A model optimization module is used to optimize the parameters of the support vector machine model by using a parameter-adaptive hybrid particle swarm optimization algorithm to obtain a state prediction model; wherein the parameter-adaptive hybrid particle swarm optimization algorithm is formed by introducing the encircling motion and spiral motion mechanisms of the whale algorithm;

[0039] A state prediction module, used for collecting first operation data of the power line in real time, and inputting the first operation data into the state prediction model, so that the state prediction model outputs an operation state prediction result of the power line;

[0040] A scenario construction module is used to update the twin data model with the first operating data and the operating status prediction result, and then exchange data between the updated twin data model and the twin entity model to construct a digital twin scenario of the power line; wherein the twin data model is pre-constructed based on the second operating data of the power line, and the twin entity model is pre-constructed based on the structural data of the power line.

[0041] In order to solve the same technical problem, the present invention also provides a computer-readable storage medium, which includes a stored computer program; wherein, when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the method for constructing a digital twin scenario of a power line.

[0042] In order to solve the same technical problem, the present invention also provides a terminal, including a processor, a memory and a computer program stored in the memory; wherein, the computer program can be executed by the processor to implement the method for constructing a digital twin scene of a power line. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 : A schematic diagram of a flow chart of a method for constructing a digital twin scenario of a power line provided in Embodiment 1 of the present invention;

[0044] Figure 2 : A structural diagram of a power line digital twin scene construction system provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] Embodiment one:

[0047] Please refer to Figure 1 , is a method for constructing a power line digital twin scene provided by an embodiment of the present invention, the method comprising steps S1 to S3, each step is specifically as follows:

[0048] Step S1, using a parameter-adaptive hybrid particle swarm optimization algorithm to optimize the parameters of the support vector machine model to obtain a state prediction model.

[0049] Among them, the parameter-adaptive hybrid particle swarm optimization algorithm is formed by introducing the encircling motion and spiral motion mechanisms of the whale algorithm.

[0050] It should be noted that the Whale Optimization Algorithm (WOA) is a swarm intelligence optimization algorithm that simulates the predation behavior of humpback whales. Its main features include encircling motion and spiral motion, which can be used to simulate the exploration and development process in the search space.

[0051] In an embodiment of the present invention, by applying WOA to the parameter optimization of a support vector machine (SVM), the efficiency and accuracy of finding the optimal penalty factor and kernel parameter can be improved. The encirclement motion of the whale algorithm describes the movement of group members around the leader, simulating the encirclement strategy of humpback whales around prey; in parameter optimization, it is used to strengthen the search around the current optimal solution. The spiral motion of the whale algorithm simulates the spiral up and down diving behavior of humpback whales, which is used to explore new search areas; in parameter optimization, this helps the algorithm jump out of the local optimum and conduct a global search.

[0052] As a preferred solution, step S1 includes step S11 to step S12, and each step is specifically as follows:

[0053] Step S11, constructing a support vector machine model, and using a parameter-adaptive hybrid particle swarm optimization algorithm to search for an optimal combination of a penalty factor and a kernel parameter of the support vector machine model.

[0054] As a preferred solution, step S11 includes step S111 to step S113, and each step is specifically as follows:

[0055] Step S111, constructing a support vector machine model.

[0056] In this embodiment, step S111 includes ① to ⑤, which are specifically as follows:

[0057] ① Refer to equations (1) and (2) to construct the input sample x and output sample y of the power line operation data sample set.

[0058] x=(x1,x2,…,x Q ); (1)

[0059] y=(y1,y2,…,y Q ); (2)

[0060] In the formula, Q represents the number of samples.

[0061] ② Minimization function of the support vector machine model optimization problem, please refer to formula (3) for details.

[0062]

[0063] In formula (3), ξ represents the relaxation factor; ω represents the weight vector; γ represents the cost parameter; Q represents the number of samples; ||·|| represents the Euclidean norm operator.

[0064] ③ Constraints of the support vector machine model, please refer to formula (4) for details.

[0065]

[0066] Where, q = 1, 2, ..., Q; ω represents the weight vector; Represents a nonlinear mapping function, which is used to map input samples to a high-dimensional feature space; b and γ are constants and penalty coefficients, respectively, both of which are positive numbers.

[0067] ④ The kernel function of the support vector machine model adopts Gaussian radial basis function, please refer to formula (5) for details.

[0068]

[0069] Wherein, g, h = 1, 2, ..., Q; σ represents the kernel parameter.

[0070] ⑤Regression function of the support vector machine model, please see formula (6) for details.

[0071]

[0072] In the formula, a q is represented by the Lagrange multiplier; b is a constant.

[0073] Step S112, create a uniformly distributed matrix M = (m1, m2, ..., m N )T , as the initial population.

[0074] The initial population includes several particles, and the positions of the particles include the penalty factor and kernel parameters of the support vector machine model. The matrix M is used to represent the population; m represents a vector representing a candidate solution; n represents the size of the population; and T represents a transposition operation, for example, converting a row vector into a column vector.

[0075] Step S113, iteratively optimize each particle in the initial population. In each iterative optimization process, based on the individual optimal value and the population optimal value of the current particle, update the position and speed of the current particle, and obtain the fitness value of the current particle's position as the fitness value corresponding to this iteration, until the fitness value of the current particle in this iteration is less than the fitness value corresponding to the previous iteration, then use the whale algorithm to perform multiple position optimizations on the current particle, and after each optimization, obtain the fitness value of the current particle's position as the fitness value corresponding to this optimization, until the fitness value of the current particle in this optimization is less than or equal to the fitness value corresponding to the previous optimization, then end the iterative optimization of the current particle, and use the current particle's position at this time as the iterative optimization result of the current particle.

[0076] In this embodiment, the position and velocity of the current particle are updated based on the individual optimal value and the population optimal value of the current particle. For details, please refer to equations (7) and (8).

[0077]

[0078] In the formula, ε represents the weight factor; c1 and c2 represent the acceleration factors; r1 and r2 represent random numbers; j represents the number of iterations; represents the individual optimal value of the i-th particle; represents the optimal population value of the i-th particle; represents the position of the i-th particle at the j-th iteration; V i j+1 represents the velocity of the i-th particle at the j+1-th iteration.

[0079] As a preferred solution, the whale algorithm is used to perform multiple position optimization on the current particle. For details, please refer to formula (9).

[0080]

[0081] Where p represents a random number in the range of 0 to 1. When p is not greater than 0.5, the encirclement movement strategy is adopted, according to the optimal position of the population Update location When p is greater than 0.5, the spiral encirclement motion strategy is used to update Until the position obtained by the optimization strategy used is not better than the previous result, the optimal solution As a final result; represents the position of the i-th particle at the j+1-th iteration; represents the optimal value of the population of the i-th particle; A represents the vector coefficient; D represents the distance vector between the current particle and the global optimal particle; l represents a constant.

[0082] Step S12, using the optimal combination, optimizing the parameters of the support vector machine model to obtain a state prediction model.

[0083] Step S2, collecting first operating data of the power line in real time, and inputting the first operating data into a state prediction model, so that the state prediction model outputs an operating state prediction result of the power line.

[0084] As a preferred solution, step S2 includes step S21 to step S22, and each step is specifically as follows:

[0085] Step S21, collecting first operation data of the power line in real time.

[0086] Step S22, input the first operating data into the state prediction model, so that the state prediction model performs collective empirical mode decomposition on the first operating data to obtain IMF components and residual components, and respectively performs prediction and denormalization processing on the IMF components and the residual components, and then superimposes the denormalization processing results corresponding to the IMF components and the denormalization processing results corresponding to the residual components, so that the state prediction model outputs the power line operation state prediction results.

[0087] It should be noted that the IMF component is an independent and regular periodic signal, which reflects the random characteristics of the data. The residual remainder is a low-frequency signal with an obvious smooth trend, which reflects the trend characteristics of the signal. The processed signal is more conducive to the prediction of the constructed support vector machine model.

[0088] In this embodiment, in the process of the state prediction model performing collective empirical mode decomposition on the input data, IMFs are extracted through a screening process until the data no longer contains any oscillation mode, and the remaining is the residual component. Since the data may be standardized during the EMD decomposition process, the results need to be denormalized after the prediction to restore to the scale of the original data, and then the prediction results of all IMF components and the prediction results of the residual components are superimposed to obtain the predicted operation results of the power line.

[0089] Step S3: Update the twin data model with the first operating data and the operating status prediction result, and then exchange data between the updated twin data model and the twin entity model to construct a digital twin scene of the power line.

[0090] In this embodiment, data is sent and received based on the UDP protocol to exchange data between the twin data model and the twin entity model to construct a digital twin scene of the power line, and the digital twin scene of the power line is output in the visualization platform. The embodiment of the present invention uses real-time operation data and prediction results to continuously update the twin data model so that it can more accurately reflect the current state of the power line. This dynamic update mechanism enhances the adaptability and accuracy of the digital twin scene.

[0091] Among them, the twin data model is pre-constructed based on the second operation data of the power line, and the twin entity model is pre-constructed based on the structural data of the power line.

[0092] In this embodiment, the second operation data of the power line includes but is not limited to current, voltage, load and environmental factors; the structural data of the power line includes but is not limited to physical parameters and topological structure of the power line.

[0093] As an example, in response to a request to build a digital twin scene for a power line, a database query language (such as SQL) is used to obtain the structural data and operating data of the power line from the power grid database. To ensure the integrity and accuracy of the data, the acquired data is optionally preprocessed; wherein data preprocessing includes but is not limited to missing value processing, outlier processing, duplicate value processing, data verification and quality checking (verifying the integrity and consistency of the data), and processing duplicate and redundant data. The twin data model is formed by mapping the second operating data of the power line to the digital twin, which is a data mapping process.

[0094] As a preferred solution, the construction process of the twin entity model includes steps S01 to S04, and each step is specifically as follows:

[0095] Step S01: acquiring first point cloud data of a pole tower of a power line, and establishing a coordinate system of the pole tower based on the first point cloud data.

[0096] As a preferred solution, step S01 includes step S011 to step S014, and each step is specifically as follows:

[0097] Step S011, using a laser radar (LiDAR) to scan the area where the power line is located multiple times at multiple scanning angles to obtain second point cloud data of the power line at multiple scanning angles.

[0098] It should be noted that point cloud data is a set of points with coordinates (x, y, z) and additional information in three-dimensional space.

[0099] In this embodiment, before registering the point cloud data, the point cloud data may be cleaned to remove noise points and abnormal values ​​in the point cloud data.

[0100] Step S012, aligning the second point cloud data of the power line at multiple scanning angles to integrate the point cloud data at different scanning angles into the same coordinate system, and then extracting features from the aligned second point cloud data to obtain local geometric features corresponding to the power line.

[0101] In this embodiment, the feature extraction method includes but is not limited to local plane fitting and curvature analysis.

[0102] It should be noted that the towers usually have a regular geometric shape and a large size, while the lines appear as slender linear structures. The characteristics of the towers can be identified by calculating the local geometric features in the point cloud, such as local plane fitting and curvature analysis.

[0103] Step S013: distinguishing poles and lines based on local geometric features corresponding to the power lines, and segmenting first point cloud data of the poles and towers of the power lines from second point cloud data of the power lines at multiple scanning angles.

[0104] In this embodiment, the local geometric features include but are not limited to coordinates, normals, and curvatures.

[0105] Step S014, referring to formula (10), based on the first point cloud data, establish the coordinate system X of the tower g Y g .

[0106]

[0107] Where M represents the number of point clouds of the tower; x i ,y i Represents the plane coordinates of the i-th point cloud.

[0108] In this embodiment, the lowest point in elevation is selected from all points of the first point cloud data of the pole tower, and a virtual cuboid is formed for each pole tower according to the lowest point in elevation, so as to completely surround all point clouds of the pole tower. The center of gravity of the pole tower is marked, and the direction of the power line is determined and the tower coordinate axis is corrected according to the center of gravity coordinate line of the pole tower of the power line, and then the spatial position and posture of the pole tower in the power line are determined based on the corrected coordinate axis. Among them, polynomial fitting or spline curve is used to smooth the center of gravity coordinate line of the pole tower of the power line to more accurately represent the direction of the power line (the direction is represented as a series of continuous line segments connecting the centers of gravity of adjacent pole towers), and this direction will be used to guide the spatial positioning of the virtual cuboid; the spatial position of the pole tower in the power line is equivalent to the space of the virtual cuboid; the posture of the pole tower usually includes the tilt angle, rotation angle, etc., and the posture is determined based on the corrected coordinate axis.

[0109] As an example, for the first point cloud data of each pole tower, the coordinate average of all points is calculated to obtain the center of gravity coordinates of the pole tower.

[0110] Step S02, according to a preset local curvature algorithm, based on the horizontal coordinates of several feature points of the tower, the local curvature of each feature point is calculated, and based on the local curvature of each feature point, the spatial position of each feature point is analyzed.

[0111] In this embodiment, several characteristic points of the tower include but are not limited to ground wire hanging points, conductor hanging points, tower body points and tower foot points. For the preset local curvature algorithm, please refer to formulas (11) and (12) for details.

[0112] Z(x,y)=Ax 2 +By 2+ Cxy+Dx+Ey+F; (11)

[0113]

[0114] Where Z(x,y) represents the horizontal coordinate of a feature point; A, B, C, D, E, and F represent the parameters of the fitted parabola; and K represents the local curvature of the feature point.

[0115] Step S03, constructing a GIM model of the power tower by combining the coordinate system of the tower and the spatial position of each feature point.

[0116] In this embodiment, the feature points corresponding to the local curvature greater than the preset threshold are captured, and the coordinates of the captured feature points are used as the spatial positions of the feature points. Then, according to the pole type of the pole, the corresponding power pole model template is matched in the template library, and the initial power pole model is constructed by combining the coordinate system of the pole, the power pole model template and the spatial positions of each feature point. Next, based on the real scene of the power line, the insulators and hardware are modeled, and the insulators, hardware and tower component attribute information are configured on the initial power pole model to generate the power pole GIM model.

[0117] Step S04, converting the power pole tower GIM model and the power line scene model into the same preset coordinate system respectively, and then fusing the power pole tower GIM model and the power line scene model to obtain a twin entity model.

[0118] Among them, the power line scene model is obtained by establishing a three-dimensional virtual geographic scene based on the actual terrain data and real-time image data of the area where the power line is located.

[0119] In this embodiment, digital elevation model terrain data and digital orthographic projection image data are first obtained, and then the digital elevation model terrain data and digital orthographic projection image data are processed, and the live terrain data and the real-time image data are made to correspond one to one according to the coding through unified spatial coding, and then a three-dimensional virtual geographic scene is established according to the one-to-one corresponding live terrain data and real-time image data, that is, the construction of the power line scene model is completed. Specifically, GIS software can be used to create a three-dimensional virtual geographic scene. In this process, the elevation model terrain data defines the undulation of the terrain, and the digital orthographic projection image data provides real texture and color for the terrain.

[0120] As an example, after the power tower GIM model and the power line scene model are converted to the same preset coordinate system, the matrix translation algorithm is used to spatially align the transmission line scene model and the power tower GIM model, and then the spatially aligned power line scene model is embedded in the TIN model, and then the embedded TIN model is superimposed with the spatially aligned power tower GIM model to obtain a twin entity model. Among them, the TIN model is constructed based on the terrain data of the scene where the power line is located.

[0121] Please refer to Figure 2 , is a structural schematic diagram of a power line digital twin scenario construction system provided by an embodiment of the present invention, the system includes a model optimization module M1, a state prediction module M2 and a scenario construction module M3, and the specific modules are as follows:

[0122] The model optimization module M1 is used to optimize the parameters of the support vector machine model by using a parameter-adaptive hybrid particle swarm optimization algorithm to obtain a state prediction model; wherein the parameter-adaptive hybrid particle swarm optimization algorithm is formed by introducing the encircling motion and spiral motion mechanisms of the whale algorithm;

[0123] The state prediction module M2 is used to collect first operation data of the power line in real time, and input the first operation data into the state prediction model so that the state prediction model outputs the operation state prediction result of the power line;

[0124] The scenario construction module M3 is used to update the twin data model with the first operating data and the operating status prediction results, and then exchange data between the updated twin data model and the twin entity model to construct a digital twin scenario of the power line; wherein the twin data model is pre-constructed based on the second operating data of the power line, and the twin entity model is pre-constructed based on the structural data of the power line.

[0125] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0126] An embodiment of the present invention also provides a computer-readable storage medium, which includes a stored computer program; wherein, when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute a method for constructing a digital twin scenario of a power line as described in Example 1.

[0127] An embodiment of the present invention also provides a terminal, including a processor, a memory, and a computer program stored in the memory; wherein the computer program can be executed by the processor to implement a method for constructing a digital twin scene of a power line as described in Example 1.

[0128] Preferably, the computer program can be divided into one or more modules / units (such as computer program, computer program), one or more modules / units are stored in the memory and executed by the processor to complete the present invention. One or more modules / units can be a series of computer program instruction segments that can complete specific functions, and the instruction segments are used to describe the execution process of the computer program in the terminal.

[0129] The processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, or the processor can also be any conventional processor. The processor is the control center of the terminal, and various parts of the terminal are connected using various interfaces and lines.

[0130] The memory mainly includes a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function, etc., and the data storage area can store related data, etc. In addition, the memory can be a high-speed random access memory, or a non-volatile memory, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, and a flash card (Flash Card), etc., or the memory can also be other volatile solid-state storage devices.

[0131] It should be noted that the above-mentioned terminal may include, but is not limited to, a processor and a memory. Those skilled in the art will understand that the above-mentioned terminal is merely an example and does not constitute a limitation on the terminal. It may include more or fewer components, or a combination of certain components, or different components.

[0132] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0133] The present invention provides a method, system, medium and terminal for constructing a digital twin scene of a power line, introduces the surrounding motion and spiral motion mechanism of the whale algorithm, forms a parameter-adaptive hybrid particle swarm optimization algorithm, can avoid local optimal solutions during the search process, and solve the problem that the particle swarm optimization algorithm is easily trapped in the local optimal state, and by introducing the surrounding motion and spiral motion mechanism of the whale algorithm, can enhance the global search capability and finely search the local optimal area, so as to efficiently search for the optimal solution in a larger range, and utilizes the parameter-adaptive hybrid particle swarm optimization algorithm to optimize the parameters of the support vector machine model to obtain a state prediction model for predicting the operating state of the power line, which can improve the accuracy and generalization ability of the state prediction model, thereby improving the prediction accuracy of the operating state of the power line, and then collects the first operating data of the power line in real time, and inputs the first operating data into the state prediction model, so that the state prediction model outputs the operating state prediction result of the power line, so as to assist the operation and maintenance personnel to timely understand and grasp the future operating state of the power line. In addition, by using the first operating data collected in real time and the operating status prediction results of the power lines, the twin data model is continuously updated, so that the twin data model can reflect the latest status of the power lines and realize dynamic synchronization of the models. Then, the updated twin data model and the twin entity model are exchanged with data to build a digital twin scene of the power lines. It can not only simulate various operating states of the power lines in the virtual scene, but also enhance the fit between the digital twin scene and the actual scene of the power lines, thereby enhancing the adaptability and accuracy of the digital twin scene.

[0134] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for constructing a digital twin scene of a power line, characterized in that: include: The parameters of the support vector machine model are optimized by using a parameter-adaptive hybrid particle swarm optimization algorithm to obtain a state prediction model; wherein the parameter-adaptive hybrid particle swarm optimization algorithm is formed by introducing the encircling motion and spiral motion mechanisms of the whale algorithm; collecting first operating data of the power line in real time, and inputting the first operating data into the state prediction model, so that the state prediction model outputs an operating state prediction result of the power line; The twin data model is updated with the first operating data and the operating status prediction result, and then the updated twin data model and the twin entity model are exchanged with data to construct a digital twin scenario of the power line; wherein the twin data model is pre-constructed based on the second operating data of the power line, and the twin entity model is pre-constructed based on the structural data of the power line.

2. A method for constructing a power line digital twin scenario according to claim 1, characterized in that: The hybrid particle swarm optimization algorithm with adaptive parameters is used to optimize the parameters of the support vector machine model to obtain a state prediction model, which is specifically: Constructing a support vector machine model, and using a parameter-adaptive hybrid particle swarm optimization algorithm to search for an optimal combination of a penalty factor and a kernel parameter of the support vector machine model; The optimal combination is used to optimize the parameters of the support vector machine model to obtain a state prediction model.

3. A method for constructing a power line digital twin scenario according to claim 2, characterized in that: The hybrid particle swarm optimization algorithm with adaptive parameters is used to search for the optimal combination of the penalty factor and the kernel parameter of the support vector machine model, specifically: Creating an initial population; wherein the initial population includes a plurality of particles, and the positions of the particles include a penalty factor and a kernel parameter of the support vector machine model; Iterative optimization is performed on each particle in the initial population. In each iterative optimization process, the position and speed of the current particle are updated based on the individual optimal value and the population optimal value of the current particle, and the fitness value of the position of the current particle is obtained as the fitness value corresponding to this iteration, until the fitness value corresponding to the current particle in this iteration is less than the fitness value corresponding to the previous iteration, then the whale algorithm is used to perform multiple position optimization on the current particle, and after each optimization, the fitness value of the position of the current particle is obtained as the fitness value corresponding to this optimization, until the fitness value corresponding to the current particle in this optimization is less than or equal to the fitness value corresponding to the previous optimization, then the iterative optimization of the current particle is terminated, and the position of the current particle at this time is used as the iterative optimization result of the current particle.

4. A method for constructing a power line digital twin scene according to claim 3, characterized in that: The whale algorithm is used to perform multiple position optimization on the current particle, specifically: Where p represents a random number in the range of 0 to 1. When p is less than or equal to 0.5, the surrounding motion mechanism is used to update the position of the current particle. When p is greater than 0.5, the spiral motion mechanism is used to update the position of the current particle. represents the position of the i-th particle at the j+1-th iteration; represents the optimal value of the population of the i-th particle; A represents the vector coefficient; D represents the distance vector between the current particle and the global optimal particle; l represents a constant.

5. A method for constructing a power line digital twin scenario according to claim 1, characterized in that: The construction of the twin entity model is specifically as follows: Acquire first point cloud data of a pole tower of a power line, and establish a coordinate system of the pole tower based on the first point cloud data; According to a preset local curvature algorithm, based on the horizontal coordinates of a plurality of characteristic points of the tower, the local curvature of each characteristic point is calculated, and based on the local curvature of each characteristic point, the spatial position of each characteristic point is analyzed; Combining the coordinate system of the tower and the spatial position of each of the characteristic points, constructing a GIM model of the power tower; The power pole tower GIM model and the power line scene model are respectively converted into the same preset coordinate system, and then the power pole tower GIM model and the power line scene model are fused to obtain a twin entity model; wherein the power line scene model is constructed based on the actual terrain data and real-time image data of the area where the power line is located.

6. A method for constructing a power line digital twin scenario according to claim 5, characterized in that: The step of obtaining the first point cloud data of the power line tower is specifically as follows: Scanning the area where the power line is located multiple times according to multiple scanning angles to obtain second point cloud data of the power line at multiple scanning angles; Registering the second point cloud data of the power line at multiple scanning angles, and then extracting features from the registered second point cloud data to obtain local geometric features corresponding to the power line; Based on the local geometric features corresponding to the power line, first point cloud data of the pole tower of the power line is segmented from the second point cloud data of the power line at multiple scanning angles.

7. A method for constructing a power line digital twin scenario according to claim 1, characterized in that: The real-time collection of first operating data of the power line and inputting the first operating data into the state prediction model so that the state prediction model outputs the operating state prediction result of the power line are specifically: Collecting first operation data of power lines in real time; The first operating data is input into the state prediction model so that the state prediction model performs collective empirical mode decomposition on the first operating data to obtain IMF components and residual components, and predicts and denormalizes the IMF components and the residual components respectively, and then superimposes the denormalization results corresponding to the IMF components and the denormalization results corresponding to the residual components, so that the state prediction model outputs a prediction result of the power line operating state.

8. A power line digital twin scene construction system, characterized in that: include: A model optimization module is used to optimize the parameters of the support vector machine model by using a parameter-adaptive hybrid particle swarm optimization algorithm to obtain a state prediction model; wherein the parameter-adaptive hybrid particle swarm optimization algorithm is formed by introducing the encircling motion and spiral motion mechanisms of the whale algorithm; A state prediction module, used for collecting first operation data of the power line in real time, and inputting the first operation data into the state prediction model, so that the state prediction model outputs an operation state prediction result of the power line; A scenario construction module is used to update the twin data model with the first operating data and the operating status prediction result, and then exchange data between the updated twin data model and the twin entity model to construct a digital twin scenario of the power line; wherein the twin data model is pre-constructed based on the second operating data of the power line, and the twin entity model is pre-constructed based on the structural data of the power line.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program; wherein, when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute a method for constructing a digital twin scenario of a power line as described in any one of claims 1 to 7.

10. A terminal, characterized in that: It comprises a processor, a memory and a computer program stored in the memory; wherein the computer program can be executed by the processor to implement a method for constructing a digital twin scene of a power line as described in any one of claims 1 to 7.

Images