A New Energy Power Station Rendering Method Based on Virtual Micropolygon Geometry Technology

Abstract

The present invention discloses a new energy power station rendering method based on virtual micro-polygon geometry technology, which relates to the field of three-dimensional rendering technology of new energy power stations, and includes the following steps: combining the triangular patches of the new energy power station model into a micro-polygon data set, clustering the model through a graph data structure to form virtual geometric meshes with different levels of granularity refinement; splicing the micro-polygon data set into polygon groups, performing decimation operations on a polygon group basis, and locking the boundaries between groups. By combining the triangular patches of the new energy power station model into a micro-polygon data set and rendering with pixel-level precision, the present invention can significantly improve the detail performance of the new energy power station model. Compared with traditional granularity refinement technologies, it can dynamically adjust the level of detail at different viewing distances and angles of view to ensure the high-precision visualization of the model, and can help technicians more intuitively understand and evaluate the structure and layout of the power station.

Description

Technical Field

[0001] The present invention relates to the technical field of three-dimensional rendering of new energy power stations, and particularly relates to a rendering method for new energy power stations based on virtual micro-polygon geometry technology. Background Art

[0002] Three-dimensional rendering of new energy power stations is an important technical requirement in the current new energy field. With the rapid development of the new energy industry, high-precision and high-efficiency three-dimensional rendering technology is required to provide intuitive and accurate visualization support for all aspects such as the design, construction, operation, and maintenance of new energy power stations. Traditional rendering methods for new energy power stations mainly use granularity refinement technology, which adapts to different rendering requirements by dynamically adjusting the level of detail of objects or scenes.

[0003] However, the existing technology has the following disadvantages:

[0004] 1. When traditional granularity refinement technology is used to render new energy power stations, due to data redundancy and refinement level limitations, the rendering effect is not fine enough in terms of detail performance and cannot meet the high-precision visualization requirements for new energy facilities;

[0005] 2. In order to achieve a better rendering effect, traditional technology often requires high-performance hardware configurations, which not only increases costs but also limits the popularization and application scope of rendering technology;

[0006] 3. Due to data redundancy and algorithm limitations, traditional technology takes a long time to render new energy power stations, with low efficiency and cannot meet the requirements of real-time rendering, especially in large-scale and high-complexity scenes. Summary of the Invention

[0007] The purpose of the present invention is to provide a rendering method for new energy power stations based on virtual micro-polygon geometry technology to solve the problems raised in the above background art.

[0008] To solve the above technical problems, the technical solution adopted by the present invention is:

[0009] A rendering method for new energy power stations based on virtual micro-polygon geometry technology, comprising the following steps:

[0010] Step 1, combining the triangular patches of the new energy power station model into a micro-polygon data set, clustering the model through a graph data structure to form virtual geometric meshes with different granularity refinement levels;

[0011] Step 2, splicing the micro-polygon data set into a polygon group, performing decimation operations on a polygon group basis, locking the boundaries between groups to ensure the boundary consistency of the simplified model and avoid boundary non-fitting or abnormal triangle density;

[0012] Step 3: Construct a directed acyclic graph (DAG) for the micro-polygon datasets at different levels of granularity refinement, and mark the flow direction to achieve efficient management of the rendering process;

[0013] Step 4: Combine the camera parameters and the pre-stored geometric errors of the micro-polygon datasets to convert and obtain the screen space pixel error values, analyze the actual rendering accuracy of the micro-polygons on the screen, and provide a basis for rendering decisions;

[0014] Step 5: Based on the constructed directed acyclic graph (DAG), start from the root of the directed acyclic graph (DAG), layer by layer, and render the nodes that meet the requirements according to the screen space pixel errors to achieve efficient pixel-level rendering;

[0015] Step 6: Compare the rendering effects of the directed acyclic graph-based and traditional granularity refinement techniques to obtain the rendering difference coefficient, analyze the differences in the rendering effects, and then adjust the parameters to optimize the rendering effects.

[0016] A further improvement of the technical solution of the present invention lies in that: the specific steps of Step 1 include:

[0017] Obtain the 3D model of the new energy power station, analyze it to identify all triangular patches, and assign a unique index identifier to each triangular patch. Among them, traverse the geometric data of the 3D model, extract the vertex coordinates and normal vector information of each triangular patch, store them in a data structure, and at the same time, establish an index table to record the position and attributes of each triangular patch in the model;

[0018] Abstract the adjacency relationship between triangular patches into a graph data structure, and construct two arrays to store the information of the graph data structure. Among them, each triangular patch is used as a node, and its adjacent triangular patches are connected to this node as edges. The two arrays are the Adjacency array and the AdjacencyOffset array. The Adjacency array is used to store the adjacency relationships of all nodes, and the AdjacencyOffset array is used to store the starting positions of the adjacency relationships of each node;

[0019] Use the graph partitioning library (Metis) to recursively bipartition the constructed graph data structure, divide the graph data structure into multiple subgraphs, and store the partitioning results in two arrays, namely the Indexes array and the Ranges array. The Indexes array is used to record the triangle numbers belonging to each subgraph, and the Ranges array is used to record the start and end ranges of each subgraph;

[0020] According to the division result, the triangular patches of each sub-graph are combined into a virtual geometric mesh body, and a granularity refinement level is assigned to each virtual geometric mesh body. Then, the triangular patches of the 3D model are organized into multiple virtual geometric mesh bodies. Each virtual geometric mesh body represents a local area of the 3D model and has different granularity refinement potentials.

[0021] According to the rendering requirements of the 3D model, different granularity refinement levels are assigned to each virtual geometric mesh body based on distance and field of view angle parameters, ensuring that the level of detail of each mesh body can be dynamically adjusted during rendering. Then, a hierarchical virtual geometric mesh body structure is formed, which is the micro-polygon dataset.

[0022] A further improvement of the technical solution of the present invention is that: step 2 specifically includes:

[0023] The divided micro-polygon datasets are spliced according to geometric continuity and attribute similarity to form polygon groups, and a unique identifier is assigned to each spliced polygon group, and its initial number of triangles and boundary information are recorded.

[0024] A decimation rule is formulated. During decimation, the shared boundaries between polygon groups are locked to prevent non-conforming phenomena caused by inconsistent numbers of triangles at the boundaries, and a mark is set for each boundary to ensure that the number and position of boundary triangles remain unchanged during the decimation process.

[0025] Combined with the formulated decimation rule, the quadratic error metric (QEM) algorithm is used to perform decimation operations on the polygon groups. The optimal fusion points of each triangle vertex with adjacent vertices are calculated, and the distance from the fused vertex to each face of the original 3D model is quantified as an error value, ensuring that the distance from the fused vertex to the relevant planes of the original 3D model is minimized.

[0026] Combined with the quantified error value and a preset error threshold, the number of triangles in each group is gradually reduced while keeping the group boundaries locked. In each round of decimation operation, the number of triangles in the group is halved, and then the remaining triangles are recombined into a new micro-polygon dataset. For the case of an odd number of triangles, the extra triangles are evenly distributed to adjacent groups to ensure the quality of the decimated 3D model.

[0027] After the decimation operation is completed, the decimation result is verified and optimized. The verification content includes checking the integrity, correctness, and boundary consistency of the 3D model. By analyzing the error between the decimated 3D model and the original 3D model, the error trend index is calculated to evaluate whether the decimation effect meets the preset accuracy requirements.

[0028] A further improvement of the technical solution of the present invention is that: the error value quantified by the distance from the fused vertex to each face of the original 3D model is calculated by the following formula:

[0029] ;

[0030] In the formula, is the distance from a point to a plane, i.e., the quadratic surface error, represents a plane, represents a fusion point, is the unit normal vector of the plane, is a constant.

[0031] A further improvement of the technical solution of the present invention lies in that: the specific steps of step 3 include:

[0032] Define a corresponding node for each micro-polygon data set with different granularity refinement levels, which serves as the basic element for constructing a directed acyclic graph (DAG). Clearly define the granularity level of each node to ensure a clear hierarchical relationship between nodes, and assign a unique identifier to each node. Record the attributes of each node, including the granularity refinement level, pre-stored geometric error, number of triangles, and boundary information, to ensure that the nodes contain sufficient information to support rendering decisions;

[0033] Based on the defined nodes and granularity levels, determine the dependency relationships between nodes in combination with the logical sequence of the rendering process. Identify the prerequisite relationships between nodes and the flow direction between nodes to ensure that during the rendering process, the nodes with prerequisite conditions are processed first, and then the nodes that depend on the prerequisite conditions are processed;

[0034] According to the nodes and dependency relationships, construct a directed acyclic graph, connect the nodes according to the dependency relationships to form directed edges. At the same time, mark the flow direction on each edge to clearly indicate the transfer order of the rendering tasks between nodes.

[0035] A further improvement of the technical solution of the present invention lies in that: the specific steps of step 4 include:

[0036] Obtain the current camera parameters from the rendering system, including position, orientation, field of view (FOV), near clipping plane, far clipping plane, and screen pixel resolution, and calculate the view transformation matrix and projection matrix according to the camera parameters to convert the three-dimensional world coordinates into screen space coordinates, ensuring that the transformation matrix can accurately reflect the current rendering perspective and projection method;

[0037] For each micro-polygon data set, calculate the position of its geometric center point in the world space, and apply the view transformation matrix and projection matrix to convert the geometric center point from the world space to the screen space. Among them, the screen space coordinates are expressed as normalized device coordinates (NDC), ranging from [-1, 1], and are further converted into screen pixel coordinates;

[0038] Comprehensively analyze the distance from the micro-polygon dataset to the camera, the field of view angle, and the screen pixel resolution, and combine the pre-stored geometric error of each micro-polygon dataset to calculate the screen space pixel error value, and analyze the actual rendering accuracy of the micro-polygons on the screen.

[0039] A further improvement of the technical solution of the present invention is that: the screen space pixel error value has the following calculation formula:

[0040] ;

[0041] In the formula, is the screen space pixel error value, is the pre-stored geometric error, is the distance from the micro-polygon dataset to the camera, is the screen pixel resolution, is the field of view angle.

[0042] A further improvement of the technical solution of the present invention is that: the specific steps of step 5 include:

[0043] Starting from the root node of the directed acyclic graph, initialize an empty rendering queue, put the root node into the queue as the starting point. The root node represents the coarsest level of granularity refinement, and set the rendering parameters, including the screen space pixel error standard value, the camera position, the field of view angle, and the screen resolution. At the same time, initialize a list of marked nodes that have been visited to avoid processing the same node repeatedly;

[0044] Take out the current node from the rendering queue, and traverse the nodes layer by layer from the root node according to the hierarchical structure of the directed acyclic graph. For each node, calculate its screen space pixel error value and compare it with the preset screen space pixel error standard value. If the screen space pixel error value of the node is less than or equal to the screen space pixel error standard value, it means that the rendering accuracy of the node on the screen meets the requirements, and add it to the rendering queue for further processing. If the screen space pixel error value is greater than the screen space pixel error standard value, it is necessary to decide whether to perform detail enhancement or replace it with a node of a finer level of granularity according to the rendering strategy;

[0045] According to the order of the nodes in the rendering queue, process the renderable nodes one by one. For each renderable node, perform the rendering operation and mark it as visited. At the same time, record the level of granularity refinement and the screen space pixel error value of the rendered node.

[0046] A further improvement of the technical solution of the present invention is that: the specific steps of step 6 include:

[0047] Select the traditional grain refinement technology as the baseline rendering scheme, set its rendering parameters and configurations, and use the directed acyclic graph rendering technology and the traditional grain refinement technology to perform rendering tests on the same scene respectively to ensure that the test conditions are consistent;

[0048] Extract the rendering effect metrics of the two rendering technologies from the results of the rendering tests, which are the frame rate, rendering time, detail performance, data redundancy, and visual consistency respectively, and convert the detail performance, data redundancy, and visual consistency into quantifiable scores;

[0049] Compare the extracted rendering effect metrics of the two rendering technologies, calculate the rendering difference coefficient, quantify the effect differences between the two rendering technologies, and then analyze the value of the rendering difference coefficient to judge the advantages and disadvantages of the two rendering technologies;

[0050] By comparing the differences in texture clarity and edge smoothness in the detail performance of the three-dimensional models between the two rendering technologies, analyze the frame rate stability of the two rendering technologies under different scenarios, judge their performance in rendering large-scale high-precision models, and determine the parameters and directions to be optimized according to the rendering effect differences and analysis results, including adjusting the hierarchical structure of the directed acyclic graph and optimizing the grain refinement strategy;

[0051] According to the optimization directions, adjust the relevant parameters of the directed acyclic graph rendering technology, including the grain refinement level and the screen space pixel error standard value, and use the adjusted parameters to re-perform the rendering test, collect new rendering data, compare the rendering difference coefficient and rendering effect before and after optimization, and evaluate whether the optimization effect meets the expectations. If it does not meet the expectations, continue to adjust the parameters and repeat the test until a satisfactory effect is achieved.

[0052] A further improvement of the technical solution of the present invention lies in: the rendering difference coefficient, and its calculation formula is as follows:

[0053] ;

[0054] In the formula, is the rendering difference coefficient, indicating the difference in the effects of the two rendering technologies, is the total number of rendering effect metrics, including the frame rate, rendering time, detail performance, data redundancy, and visual consistency, is the value of the rendering technology based on the directed acyclic graph (DAG) on the th rendering effect metric, is the value of the rendering technology based on the traditional level of detail (LOD) on the th rendering effect metric, is the absolute value function, used to calculate the difference between two values, The value range of is

[0055] Due to the adoption of the above technical solution, the technical progress achieved by the present invention compared with the prior art is as follows:

[0056] 1. The present invention provides a new energy power station rendering method based on virtual micro-polygon geometry technology. By combining the triangular patches of the new energy power station model into a micro-polygon data set and performing rendering with pixel-level precision, it can significantly improve the detail performance of the new energy power station model. Compared with traditional granularity refinement technologies, it can dynamically adjust the level of detail at different viewing distances and angles, ensuring high-precision visualization of the model, and can help technicians more intuitively understand and evaluate the structure and layout of the power station.

[0057] 2. The present invention provides a new energy power station rendering method based on virtual micro-polygon geometry technology. By using a directed acyclic graph to manage micro-polygon data sets at different granularity refinement levels, an efficient rendering process is realized. By starting from the root of the directed acyclic graph and gradually advancing the rendering tasks layer by layer, and dynamically selecting the appropriate granularity refinement level according to the screen space pixel error, it can significantly reduce unnecessary calculations and data processing while ensuring the rendering quality, not only improving the rendering efficiency, but also reducing the requirements for hardware performance, enabling high-precision rendering to be achieved on a wider range of devices, reducing costs, and improving the popularity of the technology.

[0058] 3. The present invention provides a new energy power station rendering method based on virtual micro-polygon geometry technology. By locking the boundaries between polygon groups and maintaining boundary consistency during the decimation operation, it avoids the problems of non-fitting boundaries or abnormal triangle density commonly found in traditional technologies, making the presentation of the new energy power station more stable and natural at different viewing distances and angles, providing a consistent visual experience for users, which is particularly important for real-time roaming and interactive display of new energy power stations, and can effectively improve user satisfaction and usage experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0060] Figure 1 is the flowchart of the present invention;

[0061] Figure 2 is the schematic diagram of the method flow of the present invention;

[0062] Figure 3 is the schematic diagram of the vertex merging loss of the present invention;

[0063] Figure 4 This is a comparison chart of two rendering modes of the present invention. Detailed implementation manners

[0064] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without making creative efforts shall fall within the protection scope of the present invention.

[0065] Example 1, as Figures 1 to 3 shown, the present invention provides a new energy station rendering method based on virtual micro-polygon geometry technology, including the following steps:

[0066] Step 1, combine the triangular patches of the new energy power station model into a micro-polygon dataset, perform model clustering through a graph data structure to form virtual geometric grid bodies with different levels of granularity refinement, obtain the 3D model of the new energy power station, analyze it to identify all triangular patches, and assign a unique index identifier to each triangular patch. Among them, traverse the geometric data of the 3D model, extract the vertex coordinates and normal vector information of each triangular patch, store them in a data structure, and at the same time, establish an index table to record the position and attributes of each triangular patch in the model. Abstract the adjacency relationship between triangular patches into a graph data structure, and construct two arrays to store the information of the graph data structure. Among them, each triangular patch is used as a node, and its adjacent triangular patches are connected to this node as edges. The two arrays are the Adjacency array and the AdjacencyOffset array. The Adjacency array is used to store the adjacency relationships of all nodes, and the AdjacencyOffset array is used to store the starting positions of the adjacency relationships of each node. Use the graph partitioning library (Metis) to recursively bipartition the constructed graph data structure, divide the graph data structure into multiple subgraphs, and store the partitioning results as two arrays, namely the Indexes array and the Ranges array. Among them, during the partitioning process, ensure that the triangular patches within each subgraph are geometrically continuous and the boundaries between subgraphs are as few as possible. The Indexes array is used to record the triangle numbers belonging to each subgraph, and the Ranges array is used to record the start and end ranges of each subgraph. According to the partitioning results, combine the triangular patches of each subgraph into a virtual geometric grid body, and assign a granularity refinement level to each virtual geometric grid body. Initially, the granularity refinement levels of all virtual geometric grid bodies are set to the highest (finest), and then organize the triangular patches of the 3D model into multiple virtual geometric grid bodies. Each virtual geometric grid body represents a local area of the 3D model and has different granularity refinement potentials. According to the rendering requirements of the 3D model, assign different granularity refinement levels to each virtual geometric grid body based on distance and field of view angle parameters to ensure that the level of detail of each grid body can be dynamically adjusted during rendering, and then form a hierarchical virtual geometric grid body structure, which is the micro-polygon dataset;

[0067] Step 2: Stitch the micro-polygon datasets into polygon groups, perform decimation operations on a per-polygon-group basis, lock the boundaries between groups to ensure the boundary consistency of the simplified model, avoid boundary non-conformance or abnormal triangle density. Stitch the obtained micro-polygon datasets according to geometric continuity and attribute similarity to form polygon groups, and assign a unique identifier to each stitched polygon group, record its initial triangle count and boundary information. During the stitching process, ensure that the triangular patches within each group are geometrically continuous and that the boundaries between groups are minimized. Initialize the polygon group's granularity refinement level to the highest (finest), formulate decimation rules. When performing decimation, lock the shared boundaries between polygon groups to prevent non-conformance caused by inconsistent triangle counts at the boundaries, and set a marker for each boundary to ensure that the number and position of boundary triangles remain unchanged during decimation, so that the boundaries of adjacent polygon groups remain consistent during decimation, ensuring the visual consistency and geometric continuity of the decimated 3D model. Combine the formulated decimation rules and use the quadratic error metric (QEM) algorithm to perform decimation operations on the polygon groups. Calculate the optimal fusion points for the vertices of each triangle with adjacent vertices, quantify the distance from the fused vertices to each face of the original 3D model as an error value, and ensure that the distance from the fused vertices to the relevant planes of the original 3D model is minimized. Combine the quantified error values and a preset error threshold, gradually reduce the number of triangles within each group while keeping the boundaries between groups locked. In each round of decimation, halve the number of triangles within the group, and then recombine the remaining triangles into a new micro-polygon dataset. For the case of an odd number of triangles, evenly distribute the extra triangles to adjacent groups to ensure the quality of the decimated 3D model. After completing the decimation operation, verify and optimize the decimation results. The verification content includes checking the integrity, correctness, and boundary consistency of the 3D model. By analyzing the error between the decimated 3D model and the original 3D model, calculate the error trend index to evaluate whether the decimation effect meets the preset accuracy requirements. If the error exceeds the allowable range, adjust the decimation parameters (error threshold, boundary locking strategy) and perform the decimation operation again. At the same time, check whether there are abnormal triangle densities or non-conformances at the boundaries of the 3D model to ensure the rendering effect of the 3D model at different granularity refinement levels;

[0068] Further, the error value quantified for the distance from the fused vertex to each face of the original 3D model is calculated using the following formula:

[0069] ;

[0070] In the formula, is the distance from the point to the plane, i.e., the quadratic surface error, represents a plane, represents the fusion point, is the unit normal vector of the plane, is a constant;

[0071] The error tendency index, whose calculation formula is as follows:

[0072] ;

[0073] In the formula, is the error tendency index, representing the overall error between the three-dimensional model after face reduction and the original three-dimensional model, is the total number of vertices in the model, is the coordinate of the th vertex in the original three-dimensional model, is the coordinate of the th vertex in the three-dimensional model after face reduction, is the reference value, used to standardize the error, which is the average side length of the original three-dimensional model. The reference value is obtained by calculating the lengths of all edges in the original three-dimensional model and then taking the average value to ensure matches the geometric characteristics of the model, so as to more accurately reflect the error of the model, is the error threshold, representing the maximum allowable error, is the Euclidean norm, used to calculate the Euclidean distance of vertex coordinates, The value range of is When is close to 0, it means that the three-dimensional model after face reduction is very close to the original three-dimensional model and the error is very small. When

[0074] Figure 3 increases, it means that the error increases and the difference between the three-dimensional model after face reduction and the original three-dimensional model becomes larger;In addition, using the quadratic error metric (QEM) algorithm, when the loss function is minimized, the appearance change after face reduction is also minimized. Since the maximum or minimum point of a quadratic function is where its derivative is 0, as long as the minimum value is found through calculation, the vertex position with the smallest appearance change after face reduction can be determined. The goal to be achieved by the quadratic error metric (QEM) algorithm is as

[0075] Step 3, construct a directed acyclic graph (DAG) for the micro-polygon datasets at different levels of granularity refinement, and mark the flow direction to achieve efficient management of the rendering process. Define a corresponding node for each micro-polygon dataset at a different level of granularity refinement as the basic element to form the directed acyclic graph (DAG). Clearly define the granularity level of each node to ensure a clear hierarchical relationship between nodes, and assign a unique identifier to each node. Record the attributes of each node, including the granularity refinement level, pre-stored geometric error, number of triangles, and boundary information, to ensure that the nodes contain sufficient information to support rendering decisions. Based on the defined nodes and granularity levels, determine the dependencies between nodes in combination with the logical order of the rendering process, identify the prerequisite relationships between nodes and the flow direction between nodes, and ensure that during the rendering process, nodes with prerequisite conditions are processed first, followed by nodes that depend on the prerequisite conditions. Construct a directed acyclic graph according to the nodes and dependencies, connect the nodes according to the dependencies to form directed edges, and at the same time, mark the flow direction on each edge to clearly indicate the transfer order of the rendering tasks between nodes. By marking the flow direction, the execution path of the rendering tasks can be clearly traced, where the flow direction points from the coarser nodes to the finer nodes to ensure that the transfer order of the rendering tasks between nodes is logical;

[0076] Step 4, combine the camera parameters and the pre-stored geometric error of the micro-polygon dataset to convert and obtain the screen space pixel error value, and analyze the actual rendering accuracy of the micro-polygons on the screen to provide a basis for rendering decisions. Obtain the current camera parameters from the rendering system, including position, orientation, field of view (FOV), near clipping plane, far clipping plane, and screen pixel resolution, and calculate the view transformation matrix and projection matrix based on the camera parameters to convert the three-dimensional world coordinates to screen space coordinates, ensuring that the transformation matrix can accurately reflect the current rendering perspective and projection method. For each micro-polygon dataset, calculate the position of the geometric center point in the world space, and apply the view transformation matrix and projection matrix to convert the geometric center point from the world space to the screen space. Among them, the screen space coordinates are represented as normalized device coordinates (NDC), ranging from [-1, 1], and are further converted to screen pixel coordinates. Comprehensively analyze the distance from the micro-polygon dataset to the camera, the field of view, and the screen pixel resolution, and combine the pre-stored geometric error of each micro-polygon dataset to calculate the screen space pixel error value and analyze the actual rendering accuracy of the micro-polygons on the screen;

[0077] Furthermore, the formula for calculating the screen space pixel error value is as follows:

[0078] ;

[0079] In the formula, is the screen space pixel error value, For pre-storing geometric errors, is the distance from the micro-polygon dataset to the camera, is the screen pixel resolution, is the field of view angle;

[0080] Step 5: Based on the constructed directed acyclic graph (DAG), starting from the root of the directed acyclic graph (DAG), layer by layer, and rendering the nodes that meet the requirements according to the screen space pixel error to achieve pixel-level efficient rendering. Starting from the root node of the directed acyclic graph, initialize an empty rendering queue, put the root node into the queue as the starting point. The root node represents the coarsest level of granularity refinement, and set the rendering parameters, including the standard value of the screen space pixel error, the camera position, the field of view angle, and the screen resolution. At the same time, initialize a list of marked nodes that have been visited to avoid repeated processing of the same node. Take out the current node from the rendering queue, and traverse the nodes layer by layer starting from the root node according to the hierarchical structure of the directed acyclic graph. For each node, calculate its screen space pixel error value and compare it with the preset standard value of the screen space pixel error. If the screen space pixel error value of the node is less than or equal to the standard value of the screen space pixel error, it means that the rendering accuracy of the node on the screen meets the requirements, and add it to the rendering queue for further processing. If the screen space pixel error value is greater than the standard value of the screen space pixel error, it is necessary to decide whether to perform detail enhancement or replace it with a node of a finer granularity level according to the rendering strategy. According to the order of the nodes in the rendering queue, process the renderable nodes one by one. For each renderable node, perform the rendering operation and mark it as visited. During the rendering process, ensure that the dependency relationship between the nodes is satisfied, that is, render the nodes with preconditions first, and then render the nodes that depend on the preconditions to ensure the logic and hierarchy of the rendering process. At the same time, record the granularity refinement level and the screen space pixel error value of the rendered nodes;

[0081] Step 6: Compare the rendering effects of the directed acyclic graph-based and traditional granularity refinement techniques to obtain the rendering difference coefficient, analyze the differences in the rendering effects, and then adjust the parameters to optimize the rendering effect to improve the frame rate and detail performance, ensuring high efficiency and stability in large-scale high-precision model rendering.

[0082] Embodiment 2, as Figures 1 to 3 shown, based on Embodiment 1, the present invention provides a technical solution: Preferably, the specific steps of Step 6 include:

[0083] Select the traditional granularity refinement technology as the baseline rendering scheme, set its rendering parameters and configurations, and use the directed acyclic graph (DAG) rendering technology and the traditional granularity refinement technology to perform rendering tests on the same scene respectively, ensuring that the test conditions are consistent. From the results of the rendering tests, extract the rendering effect indicators of the two rendering technologies, namely frame rate, rendering time, detail performance, data redundancy, and visual consistency, and convert the detail performance, data redundancy, and visual consistency into quantifiable scores. Among them, the detail performance is quantified through texture clarity and edge smoothness indicators, the data redundancy is quantified by comparing the redundant data generated during the rendering process, and the visual consistency is quantified by comparing the similarity between the rendering result and the expected result. Compare the extracted rendering effect indicators of the two rendering technologies, calculate the rendering difference coefficient to quantify the effect difference between the two rendering technologies, and then analyze the value of the rendering difference coefficient to judge the advantages and disadvantages of the two rendering technologies. By comparing the differences in texture clarity and edge smoothness in the detail performance of the three-dimensional models of the two rendering technologies, analyze the frame rate stability of the two rendering technologies in different scenarios, judge their performance in rendering large-scale high-precision models, and determine the parameters and directions to be optimized according to the rendering effect difference and analysis results, including adjusting the hierarchical structure of the directed acyclic graph and optimizing the granularity refinement strategy. According to the optimization direction, adjust the relevant parameters of the directed acyclic graph rendering technology, including the granularity refinement level and the screen space pixel error standard value, and use the adjusted parameters to perform the rendering test again, collect new rendering data, compare the rendering difference coefficient and rendering effect before and after optimization, and evaluate whether the optimization effect meets the expectation. If it does not meet the expectation, continue to adjust the parameters and repeat the test until a satisfactory effect is achieved;

[0084] In addition, the formula for calculating the rendering difference coefficient is as follows:

[0085] ;

[0086] In the formula, is the rendering difference coefficient, indicating the difference in the effects of the two rendering technologies. is the total number of rendering effect indicators, including frame rate, rendering time, detail performance, data redundancy, and visual consistency. is the value of the rendering technology based on the directed acyclic graph (DAG) for the th rendering effect indicator. is the value of the rendering technology based on the traditional level of detail (LOD) for the th rendering effect indicator. is the absolute value function used to calculate the difference between two values. The value range of is . When When it increases, it indicates that the difference in the effects of the two rendering techniques increases. When the performances of the two rendering techniques are close in each index, it will be smaller, indicating that the effects of the two techniques are similar. When the difference in the performances of the two rendering techniques is large in some indexes, it will increase, indicating that the difference in the effects of the two techniques is obvious.

[0087] Example 3, as Figure 4 shown, on the basis of Examples 1 and 2, the present invention provides an example as follows: The traditional grain refinement technique and the virtual micro-polygon geometry technique are respectively used to render a new energy field station model, and the picture quality of the model, the corresponding frame rate and the rendering time are compared. Figure 4 As the comparison result, it can be seen that when using the traditional grain refinement technique, the picture FPS is 42.30 and the rendering time is 23.64 ms. When using the virtual micro-polygon geometry technique, the picture FPS is 93.17 and the rendering time is 10.73 ms.

[0088] It can be seen that the virtual micro-polygon geometry technique has a higher rendering frame number and efficiency compared with the traditional grain refinement technique, and the picture rendered by the virtual micro-polygon geometry technique has more detailed performance.

[0089] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A new energy station rendering method based on virtual micropolygon geometry technology, characterized in that It includes the following steps: Step 1: Combine the triangular patches of the new energy power station model into a micro-polygon data set, perform model clustering through a graph data structure, and form virtual geometric grid bodies with different levels of granularity refinement; Step 2: Stitch the micro-polygon data set into polygon groups, perform decimation operations on a polygon group basis, and lock the boundaries between groups. The specific steps of Step 2 include: Stitch the obtained micro-polygon data set according to geometric continuity and attribute similarity to form polygon groups, and assign a unique identifier to each stitched polygon group, record its initial number of triangles and boundary information; Formulate decimation rules. When performing decimation, lock the shared boundaries between polygon groups and set a mark for each boundary; Combined with the formulated decimation rules, use the quadratic surface error metric algorithm to perform decimation operations on the polygon groups, calculate the optimal fusion points of the vertices of each triangle with adjacent vertices, and quantify the distance from the fused vertices to each face of the original 3D model as an error value; Combined with the quantified error value and a preset error threshold, gradually reduce the number of triangles in each group while keeping the boundaries between groups locked. In each round of decimation operation, halve the number of triangles in the group, and then recombine the remaining triangles into a new micro-polygon data set; After completing the decimation operation, verify and optimize the decimation results. The verification content includes checking the integrity, correctness, and boundary consistency of the 3D model. By analyzing the error between the decimated 3D model and the original 3D model, calculate the error trend index to evaluate whether the decimation effect meets the preset accuracy requirements. The calculation formula of the error trend index is as follows: ； In the formula, is the error trend index, representing the overall error between the three-dimensional model after surface reduction and the original three-dimensional model, is the total number of vertices in the model, is the coordinate of the th vertex in the original three-dimensional model, is the coordinate of the th vertex in the three-dimensional model after surface reduction, is the reference value, used to standardize the error, which is the average side length of the original three-dimensional model, is the error threshold, representing the maximum allowable error, is the Euclidean norm, used to calculate the Euclidean distance of vertex coordinates; Step 3: Construct a directed acyclic graph for the micro-polygon data sets with different levels of granularity refinement and mark the flow direction; Step 4: Combine the camera parameters and the pre-stored geometric errors of the micro-polygon data set to convert and obtain the screen space pixel error value, and analyze the actual rendering accuracy of the micro-polygons on the screen; Step 5: Based on the constructed directed acyclic graph, start from the root of the directed acyclic graph, layer by layer, and render the nodes that meet the requirements according to the screen space pixel error; Step 6: Compare the rendering effects of the directed acyclic graph and the traditional granularity refinement technology to obtain the rendering difference coefficient, analyze the differences in the rendering effects, and then adjust the parameters to optimize the rendering effect.

2. The new energy power station rendering method based on virtual micro-polygon geometry technology according to claim 1, characterized in that: The specific steps of Step 1 include: Obtain the 3D model of the new energy power station, analyze it to identify all triangular patches, and assign a unique index identifier to each triangular patch. Among them, traverse the geometric data of the 3D model, extract the vertex coordinates and normal vector information of each triangular patch, store them in a data structure, and at the same time, establish an index table to record the position and attributes of each triangular patch in the model; Abstract the adjacency relationship between triangular patches into a graph data structure, and construct two arrays to store the information of the graph data structure. Among them, each triangular patch is used as a node, and its adjacent triangular patches are connected to this node as edges. The two arrays are the Adjacency array and the AdjacencyOffset array respectively. The Adjacency array is used to store the adjacency relationships of all nodes, and the AdjacencyOffset array is used to store the starting positions of the adjacency relationships of each node; Use a graph partitioning library to recursively bipartition the constructed graph data structure, divide the graph data structure into multiple subgraphs, and store the partitioning results in two arrays, namely the Indexes array and the Ranges array. Among them, the Indexes array is used to record the triangle numbers belonging to each subgraph, and the Ranges array is used to record the start and end ranges of each subgraph; According to the partitioning results, combine the triangular patches of each subgraph into a virtual geometry mesh body, and assign a granularity refinement level to each virtual geometry mesh body, thereby organizing the triangular patches of the 3D model into multiple virtual geometry mesh bodies; According to the rendering requirements of the 3D model, assign different granularity refinement levels to each virtual geometry mesh body based on distance and field of view angle parameters, thereby forming a hierarchical virtual geometry mesh body structure, which is the micro polygon dataset.

3. A new energy power station rendering method based on virtual micro-polygon geometry technology according to claim 1, characterized in that: Fuse the error values of the distance quantization from the merged vertices to each face of the original 3D model. The calculation formula is as follows: ； In the formula, is the distance from a point to a plane, i.e., the quadratic surface error, represents a plane, represents a fusion point, is the unit normal vector of the plane, is a constant.

4. A new energy station rendering method based on virtual micropolygon geometry technology according to claim 1, characterized in that: The specific steps of step 3 include: Define a corresponding node for each micro polygon dataset with different granularity refinement levels as the basic element for constructing a directed acyclic graph, clarify the granularity level of each node, and assign a unique identifier to each node, and record the attributes of each node, including the granularity refinement level, pre-stored geometric error, number of triangles, and boundary information; Based on the defined nodes and granularity levels, determine the dependency relationships between nodes in combination with the logical order of the rendering process, identify the prerequisite relationships between nodes and nodes, as well as the flow directions between nodes; According to the nodes and dependency relationships, construct a directed acyclic graph, connect the nodes according to the dependency relationships to form directed edges, and at the same time, mark the flow direction on each edge to clearly represent the transfer order of the rendering tasks between nodes.

5. A new energy power station rendering method based on virtual micro-polygon geometry technology according to claim 1, characterized in that: The specific steps of step 4 include: Obtain the current camera parameters from the rendering system, including position, direction, field of view angle, near clipping plane, far clipping plane, and screen pixel resolution, and calculate the view transformation matrix and projection matrix according to the camera parameters to convert the 3D world coordinates into screen space coordinates; For each micro polygon dataset, calculate the position of its geometric center point in world space, and apply the view transformation matrix and projection matrix to convert the geometric center point from world space to screen space. Among them, the screen space coordinates are represented as normalized device coordinates and are further converted into screen pixel coordinates; Comprehensively analyze the distance from the micro-polygon dataset to the camera, the field of view angle, and the screen pixel resolution, and combine the pre-stored geometric error of each micro-polygon dataset to calculate the screen space pixel error value, and analyze the actual rendering accuracy of the micro-polygons on the screen.

6. A new energy power station rendering method based on virtual micro-polygon geometry technology according to claim 5, characterized in that: The formula for the screen space pixel error value is as follows: ； In the formula, is the screen space pixel error value, is the pre-stored geometric error, is the distance from the micropolygon dataset to the camera, is the screen pixel resolution, is the field of view angle.

7. A new energy power station rendering method based on virtual micro-polygon geometry technology according to claim 6, characterized in that: Step 5 specifically includes: Starting from the root node of the directed acyclic graph, initialize an empty rendering queue, put the root node into the queue as the starting point, and set the rendering parameters, including the screen space pixel error standard value, the camera position, the field of view angle, and the screen resolution. At the same time, initialize a list of marked nodes that have been visited; Take out the current node from the rendering queue, and traverse the nodes layer by layer starting from the root node according to the hierarchical structure of the directed acyclic graph. For each node, calculate its screen space pixel error value and compare it with the preset screen space pixel error standard value. If the screen space pixel error value of the node is less than or equal to the screen space pixel error standard value, it means that the rendering accuracy of the node on the screen meets the requirements, and add it to the rendering queue for further processing; According to the order of the nodes in the rendering queue, process the renderable nodes one by one. For each renderable node, perform the rendering operation and mark it as visited. At the same time, record the granularity refinement level and the screen space pixel error value of the rendered node.

8. A new energy station rendering method based on virtual micro-polygon geometry technology according to claim 1, characterized in that: Step 6 specifically includes: Select the traditional granularity refinement technology as the benchmark rendering scheme, set its rendering parameters and configurations, and use the directed acyclic graph rendering technology and the traditional granularity refinement technology to perform rendering tests on the same scene respectively; Extract the rendering effect indicators of the two rendering technologies from the results of the rendering tests, which are the frame rate, rendering time, detail performance, data redundancy, and visual consistency, and convert the detail performance, data redundancy, and visual consistency into quantifiable scores; Compare the extracted rendering effect indicators of the two rendering technologies, calculate the rendering difference coefficient, quantify the effect difference between the two rendering technologies, and then analyze the value of the rendering difference coefficient to judge the advantages and disadvantages of the two rendering technologies; By comparing the differences in texture clarity and edge smoothness in the detail performance of the three-dimensional models of the two rendering technologies, analyze the frame rate stability of the two rendering technologies in different scenarios, judge their performance in the rendering of large-scale high-precision models, and determine the parameters and directions to be optimized according to the rendering effect differences and analysis results, including adjusting the hierarchical structure of the directed acyclic graph and optimizing the granularity refinement strategy; According to the optimization direction, adjust the relevant parameters of the directed acyclic graph rendering technology, including the granularity refinement level and the screen space pixel error standard value, and use the adjusted parameters to perform the rendering test again, collect new rendering data, compare the rendering difference coefficient and rendering effect before and after optimization, and evaluate whether the optimization effect meets the expectations. If it does not meet the expectations, continue to adjust the parameters and repeat the test until a satisfactory effect is achieved.

9. A new energy station rendering method based on virtual micro-polygon geometry technology according to claim 8, characterized in that: The formula for the rendering difference coefficient is as follows: ； In the formula, is the rendering difference coefficient, is the total number of rendering effect indicators, including frame rate, rendering time, detail performance, data redundancy, and visual consistency, is the value of the rendering technology based on the directed acyclic graph on the th rendering effect indicator, is the value of the rendering technology based on traditional grain refinement on the th rendering effect indicator, is the absolute value function, used to calculate the difference between two values, ranges from .

Images