Geometric import zero-distortion method for three-dimensional complex curve modeling of pumped storage power station

Abstract

The invention discloses a geometric import zero-distortion method for three-dimensional complex curve modeling of a pumped storage power station, which comprises the following steps of: aiming at the problem of distortion generated by geometric import before physical network division of three-dimensional complex curve modeling of the pumped storage power station, proposing an embedded development automatic sweeping program developed based on a target environment ANSYS; according to the method, geometric information of a distortion model is automatically retrieved, point, line and surface distortion degree quantitative evaluation standards are set, and geometric distortion items of a hexahedral mesh are subjected to merging, regeneration, item cancellation and reduction technologies, so that the whole model is converted from an SW environment to an ANSYS environment, complex three-dimensional information such as a curved body and a curved surface is not lost, and the accuracy of the model is improved. And the accuracy and the working efficiency of modeling numerical analysis of the pumped storage complex three-dimensional structure can be effectively improved.

Description

Technical Field

[0001] The invention relates to the technical field of three-dimensional complex curved body modeling of a pumped storage power station, and in particular to a zero-distortion method for geometric import of three-dimensional complex curved body modeling of a pumped storage power station. Background Art

[0002] At present, there are two main methods for modeling the complex three-dimensional structure of pumped storage power stations. One is to manually search and repair after the geometric model is converted, and the other is to directly establish the geometric model in the physical sub-grid environment. These two solutions have the following defects: 1) Using the first type of geometric model conversion and then performing manual retrieval and repair. As the model becomes more complex, the efficiency of retrieval and search is relatively low. In addition, the manual repair method is relatively complicated and mostly uses GUI operations, which cannot be accurately implemented.

[0003] 2) Directly establish the geometric model in the physical meshing environment. This method has high requirements for modeling personnel, especially for curved bodies, surfaces, multi-intersecting bodies, shared faces, etc. There are many Boolean operations, which will cause a lot of secondary repair work for subsequent meshing.

[0004] 3) Both methods are inefficient. Due to the existence of complex structural curved bodies and surfaces, the distortion information of the overall modeling is generally more and more hidden, resulting in low manual retrieval efficiency and restoration quality of both methods. Summary of the invention

[0005] The purpose of the present invention is to overcome the above-mentioned shortcomings and provide a zero-distortion method for geometric import of three-dimensional complex curved body modeling for pumped storage power stations to solve the problems raised in the background technology.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is: a method for geometric importing zero distortion for three-dimensional complex curved body modeling of a pumped storage power station, comprising the following steps: Step 1): Generate a geometric model using solidworks software; Step 2): Import the geometric model into ANSYS, continuously update the overall compression number of the model, and extract the model information characteristic values, which include the maximum number values ​​Nmax, Amax, Vmax of geometric points, lines, and surfaces and the maximum value NNmax, AAmax, VVmax of geometric quantity statistics after compression; Step 3): Set the small threshold [δ]1 and the large threshold [δ]2 as the distortion discrimination parameters, and use the small threshold [δ]1 as the distortion search discrimination radius to determine whether each point Ni, Ai, Vi of the model has adjacent independent coordinate points Nj, Aj, Vj, where i and j are numbers not greater than NNmax; if so, merge Ni and Nj, and determine whether the i value exceeds NNmax; if not, use the large threshold [δ]2 as the distortion search discrimination radius to confirm whether each point Ni, Ai, Vi of the model has adjacent independent coordinate points Nj, Aj, Vj. If so, reconstruct the geometric model; if not so, determine whether the i value exceeds NNmax; Step 4): If the value of i does not exceed NNmax, repeat step 3) to determine the next point; if the value of i exceeds NNmax, proceed to step 5); Step 5): Merge, compress and number the model points processed above to obtain updated point, line and surface model information feature values, including the maximum values ​​of geometric quantity statistics nNmax, nAmax and nVmax; Step 6): Re-use the small threshold [δ]1 as the distortion search judgment radius to judge whether there are adjacent coordinate independent points at each point of the updated model; if not, end; if so, perform local secondary repair, supplement and connection, and then determine whether the number value of the above judgment point exceeds nNmax; Step 7): If it does not exceed nNmax, repeat step 6) to determine the next point; if it exceeds nNmax, end.

[0007] Preferably, in step 3), the distortion is divided into two categories, including the overlap of geometric points, lines and surfaces in the model import under a small threshold [δ]1 and the curvature mismatch caused by excessive misalignment of curved bodies and surfaces in the model import under a large threshold [δ]2.

[0008] Preferably, the overlap distortion of geometric points, lines and surfaces is processed by searching for nearby values ​​and directly merging them; the curvature mismatch distortion is processed by local model reconstruction.

[0009] Preferably, in step 3), the distortion is divided into three small levels, including point, line and surface misalignment distortion, and the discrimination is achieved through an associated search method, that is, first discriminating the point, selecting the line based on the point, then discriminating another point at the end of the line, and then selecting the surface based on the line to determine whether other points at the end of the surface are distorted.

[0010] Preferably, in step 3), reconstructing the geometric model includes confirming whether local model reconstruction processing can be performed based on curvature; if local model reconstruction processing cannot be adopted, directly returning to the initial drawing to generate the geometric SW model to start reconstructing the geometric model.

[0011] Preferably, the small threshold [δ]1 evaluation gradient is 1 / 1000, and the large threshold [δ]1 evaluation gradient is 1 / 10.

[0012] Preferably, local model reconstruction includes directly targeting the target distortion point, searching within a radius range that is several times larger than a threshold value [δ]2, distinguishing the searched point, line, and surface information, deleting the model in combination with the curvature in the original geometric drawing information, and stacking and modeling detailed geometric models to replace the original model based on future calculation and analysis requirements for numerical extraction and analysis.

[0013] Preferably, the method further comprises step 8), using the Jacobian matrix evaluation value to judge the quality of the model network nodes.

[0014] Beneficial effects of the present invention: 1) Automatic scanning and retrieval: This invention aims at complex and large-scale pumped storage models and establishes a geometric model automatic tracking and retrieval function, which can omit the traditional manual retrieval and troubleshooting process, greatly improving the accuracy and speed of discovering distortion items. The automatic scanning will not miss the information corresponding to the small scales in the geometric model, thus ensuring the integrity of the model retrieval.

[0015] 2) Quantifying distortion information: The distortion patterns of geometric models are diverse. By establishing a three-level geometric distortion discrimination model and setting a distortion threshold, the overall distortion of the model can be quickly and comprehensively judged, and the distortion category of the local model can be effectively determined, and the repair method can be determined.

[0016] 3) High geometric restoration: In view of the geometric properties of the curved body and surface of pumped storage, a restoration and repair mechanism is established during the model conversion process. If the geometric threshold is small, the merging and elimination methods can be used. If the threshold is large, it can be judged based on the curvature, etc., and the local small grid geometric model is established for filling and replacement repair. Through these two methods, a high degree of geometric restoration of the complex model of the pumped storage power station can be achieved.

[0017] 4) Aiming at the distortion problem caused by geometry import before physical meshing of three-dimensional complex curved body modeling of pumped-storage power stations, it is proposed to develop an embedded automatic scanning program based on the target environment ANSYS to automatically retrieve the geometric information of the distorted model, set quantitative evaluation standards for point, line and surface distortion, and use merging, regeneration, elimination and restoration techniques for the geometric distortion items of the hexahedral mesh. This allows the entire model to be converted from SW to ANSYS environment without losing complex three-dimensional information such as curved bodies and surfaces, which can effectively improve the accuracy and work efficiency of numerical analysis of complex three-dimensional structure modeling of pumped-storage power stations.

[0018] 5) Aiming at the problem of large point, line and surface distortion after Boolean operation caused by curved bodies and surfaces during the geometric environment conversion process of three-dimensional complex dynamic analysis numerical modeling of pumped storage hydropower stations, which makes subsequent physical sub-networking impossible and the query complicated and inefficient, the present invention designs an embedded automatic retrieval of distortion information when converting from SOLIDWORKS (SW) to ANSYS, and uses distortion classification and distortion degree to comprehensively evaluate the distortion items, adopts local merging, regeneration, elimination and restoration of geometric meshes, and finally ensures that the entire geometric model is completely matched with the original environment through secondary automatic retrieval; the technology of the present invention can effectively improve the accuracy of geometric curved body model modeling and greatly increase the efficiency of secondary processing of distortion in multi-environment conversion of models. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 Flowchart of a zero-distortion method for geometric import of three-dimensional complex curved body modeling in pumped storage power plants. DETAILED DESCRIPTION

[0020] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0021] like Figure 1 As shown, a zero-distortion method for geometric import of three-dimensional complex curved body modeling for pumped storage power stations comprises the following steps: First, the overall modeling is done through solidworks software (SW). The geometry is imported into the ANSYS model library, an INC file is generated (log script is saved), and the overall compression numbering of the system model is started, and continuous updates are performed. The key feature values ​​of the model information are extracted using the embedded method. These key features can include the maximum number values ​​of geometric points, lines, and surfaces (Nmax, Amax, Vmax) and the maximum value of geometric quantity statistics after compression (NNmax, AAmax, VVmax).

[0022] The distortion modes of geometric models are varied. The present invention establishes a three-level geometric distortion discrimination mode and sets a distortion threshold, which can quickly discriminate the distortion of the entire model as a whole, and can also effectively determine the distortion category of the local model and determine the repair method.

[0023] According to the implementation scheme of the present invention, the distortion mode can be divided into two major categories and three minor levels. The two major categories include: when the geometric points, lines and surfaces of the model import under the small threshold [δ]1 coincide, this type of distortion can be processed by searching for nearby values ​​and directly merging; the model import under the large threshold [δ]2 causes the curvature mismatch due to excessive misalignment of the curved body and curved surface, resulting in the surface becoming a flat surface, and a separate local model reconstruction is required. The three minor levels of distortion classification mainly refer to the misalignment distortion of points, lines and surfaces, and the discrimination method is mainly achieved through the association search method, that is, first discriminating the point, selecting the line based on the point, and then discriminating another point at the end of the line, and then selecting the surface based on the line to determine whether other points at the end of the surface are distorted. The association search method itself is known in the art and will not be described in detail here.

[0024] More specifically, a small threshold [δ]1 and a large threshold [δ]2 can be set as distortion discrimination parameters. For example, the small threshold [δ]1 can be used to judge the gradient as 1 / 1000, and the large threshold [δ]2 can be used to judge the gradient as 1 / 10. Then, the small threshold [δ]1 is used as the distortion search discrimination radius to judge whether there are adjacent coordinate independent points Nj, Aj, Vj for each point Ni, Ai, Vi of the model, where i and j are numbers not greater than NNmax; that is, the distance R between the two points is judged to be greater than the small threshold [δ]1. If they exist, Ni and Nj are merged and processed, and it is judged whether the i value exceeds NNmax. If not, the large threshold [δ]2 is used as the distortion search discrimination radius to confirm whether there are adjacent coordinate independent points Nj, Aj, Vj for each point Ni, Ai, Vi of the model. If they exist under the large threshold [δ]2, the geometric model is reconstructed. If they do not exist under the large threshold [δ]2, it is judged whether the i value exceeds NNmax. If the i value does not exceed NNmax, repeat the above operation and judge the next point until all geometric points are processed; if the i value exceeds NNmax, proceed to the next step.

[0025] When a larger threshold [δ]2 is selected as the search discrimination radius, the reconstruction of the geometric model includes full reconstruction and partial reconstruction. That is, it is confirmed whether the model is partially reconstructed (restored) according to the second of the two categories mentioned above. If the second type of model restoration cannot be used, it is directly returned to the initial drawing to generate the geometric SW model and start to reconstruct the geometric model. The local model reconstruction under the large threshold can include directly targeting the target distortion point, searching with a radius range that is expanded several times (for example, 10 times) by the large threshold [δ]2, discriminating the searched point, line, and surface information, deleting the model in combination with the curvature in the original geometric drawing information, and stacking and modeling the detailed geometric model to replace the original model according to the future calculation and analysis numerical extraction analysis requirements.

[0026] Next, the model points after the above processing are merged, compressed, and numbered to obtain updated point, line, and surface model information feature values, wherein the information feature values ​​may include the maximum values ​​of geometric quantity statistics nNmax, nAmax, and nVmax.

[0027] Then, the small threshold [δ]1 is used as the distortion search judgment radius again to judge whether there are adjacent independent coordinate points for each point of the updated model; if not, the process ends; if so, local secondary repair, supplementation and connection are performed, and then it is determined whether the number value of the above judgment point exceeds nNmax; if not, the above operation is repeated to judge the next point until all model points are judged; if it exceeds nNmax, the process ends, thereby achieving model repair and eliminating distortion.

[0028] At the same time, after the final search, the Jacobian matrix evaluation value can be used to judge the quality of the model network nodes. If the evaluation value is low, you can search again and assign node number values ​​according to the set path.

[0029] The present invention aims at the geometric properties of the curved body and surface of pumped storage and establishes a restoration and repair mechanism during the model conversion process. If the geometric threshold is small, the merging and elimination methods can be adopted. If the threshold is large, it can be judged according to the curvature, etc., and a local small grid geometric model is established for filling and replacement repair is performed. Through these two methods, a high degree of geometric restoration of the complex model of the pumped storage power station can be achieved.

[0030] The above embodiments are only preferred technical solutions of the present invention and should not be regarded as limiting the present invention. The protection scope of the present invention shall be the technical solutions recorded in the claims, including equivalent replacement solutions of the technical features in the technical solutions recorded in the claims. That is, equivalent replacement improvements within this scope are also within the protection scope of the present invention.

Claims

1. A zero-distortion method for geometric import of three-dimensional complex curved body modeling in a pumped storage power station, characterized in that: The following steps are involved: Step 1): Generate a geometric model using solidworks software; Step 2): Import the geometric model into ANSYS, continuously update the overall compression number of the model, and extract the model information characteristic values, which include the maximum number values ​​Nmax, Amax, Vmax of geometric points, lines, and surfaces and the maximum value NNmax, AAmax, VVmax of geometric quantity statistics after compression; Step 3): Set the small threshold [δ]1 and the large threshold [δ]2 as the distortion discrimination parameters, and use the small threshold [δ]1 as the distortion search discrimination radius to determine whether each point Ni, Ai, Vi of the model has adjacent independent coordinate points Nj, Aj, Vj, where i and j are numbers not greater than NNmax; if so, merge Ni and Nj, and determine whether the i value exceeds NNmax; if not, use the large threshold [δ]2 as the distortion search discrimination radius to confirm whether each point Ni, Ai, Vi of the model has adjacent independent coordinate points Nj, Aj, Vj. If so, reconstruct the geometric model; if not so, determine whether the i value exceeds NNmax; Step 4): If the i value does not exceed NNmax, repeat step 3) to determine the next point; If the value of i exceeds NNmax, proceed to step 5); Step 5): Merge, compress and number the model points processed above to obtain updated point, line and surface model information feature values, including the maximum values ​​of geometric quantity statistics nNmax, nAmax and nVmax; Step 6): Re-use the small threshold [δ]1 as the distortion search judgment radius to judge whether there are adjacent coordinate independent points at each point of the updated model; if not, end; if so, perform local secondary repair, supplement and connection, and then determine whether the number value of the above judgment point exceeds nNmax; Step 7): If it does not exceed nNmax, repeat step 6) to determine the next point; If nNmax is exceeded, the process ends.

2. According to claim 1, a method for geometric importing zero distortion for three-dimensional complex curved body modeling of a pumped storage power station is characterized by: In step 3), the distortion is divided into two categories, including the overlap of geometric points, lines and surfaces in the model import under the small threshold [δ]1 and the curvature mismatch caused by excessive misalignment of curved bodies and surfaces in the model import under the large threshold [δ]2.

3. The zero-distortion method for geometric import of three-dimensional complex curved body modeling for pumped storage power stations according to claim 2, characterized in that: The overlap distortion of geometric points, lines and surfaces is handled by searching for nearby values ​​and merging them directly; the mismatch distortion of curvature is handled by local model reconstruction.

4. The zero-distortion method for geometric import of three-dimensional complex curved body modeling for pumped storage power stations according to claim 1, characterized in that: In step 3), the distortion is divided into three levels, including point, line, and surface misalignment distortion, and the discrimination is achieved through an associated search method, that is, first discriminating the point, selecting the line based on the point, then discriminating another point at the end of the line, and then selecting the surface based on the line to determine whether other points at the end of the surface are distorted.

5. The zero-distortion method for geometric import of three-dimensional complex curved body modeling for pumped storage power stations according to claim 1, characterized in that: In step 3), the reconstruction of the geometric model includes confirming whether the local model reconstruction process can be performed based on the curvature; if the local model reconstruction process cannot be adopted, directly returning to the initial drawing to generate the geometric SW model, and starting to rebuild the geometric model.

6. The zero-distortion method for geometric import of three-dimensional complex curved body modeling for pumped storage power stations according to claim 1, characterized in that: The small threshold [δ]1 evaluates the gradient as 1 / 1000, and the large threshold [δ]1 evaluates the gradient as 1 / 10.

7. The zero-distortion method for geometric import of three-dimensional complex curved body modeling for pumped storage power stations according to claim 5, characterized in that: Local model reconstruction includes directly targeting the target distortion point, searching within a radius range that is several times larger than the threshold [δ]2, distinguishing the searched point, line, and surface information, deleting the model based on the curvature in the original geometric drawing information, and stacking and modeling detailed geometric models to replace the original model based on future calculation and analysis requirements for numerical extraction and analysis.

8. The zero-distortion method for geometric import of three-dimensional complex curved body modeling for pumped storage power stations according to claim 1, characterized in that: It also includes step 8), using the Jacobian matrix evaluation value to judge the quality of the model network nodes.

Images