Rapid design method of pipeline for engine test
By collecting multi-angle image information in the engine laboratory and using three-dimensional reconstruction and modeling software for analysis and adjustment, the problems of inaccurate spatial dimension measurement and large surveying errors in pipeline design are solved, and fast and accurate pipeline design is achieved, which improves the efficiency of the test.
Patent Information
- Application Number
- CN202411688638.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-25
- Publication Date
- 2025-05-02
AI Technical Summary
The existing engine test chamber pipeline design methods have problems such as inaccurate measurement of space dimensions and large surveying and mapping errors, resulting in a long design cycle and affecting the test rhythm.
A rapid pipeline design method for engine testing is adopted. By placing fixed-size reference objects in the laboratory, multi-angle image information is collected, and spatial distance analysis and three-dimensional modeling software are used to perform spatial distance analysis and three-dimensional model adjustment to ensure the accuracy of pipeline design.
The accuracy of the measurement of space dimensions between the two ports of the pipeline and the convenience of the pipeline direction design are achieved, the design cycle is shortened, and the rhythm of the test is improved.
Smart Images

Figure CN119918201A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of pipeline design, in particular to a rapid design method for pipelines used in engine tests. Background Art
[0002] Each engine needs to be tested in the laboratory before leaving the factory. Models with turbochargers need to connect the turbocharger compressor outlet to the laboratory intercooler to cool the turbocharged air before delivering it to the engine's air intake. Due to the wide variety of models, the orientation of various turbocharger compressor ports is different, and there are often uncertain laboratory equipment, wiring harnesses and other objects blocking the intercooler air intake port. In order to avoid interference between the pipeline and existing objects, the direction must be fully considered during the design, which increases the difficulty of on-site surveying and mapping.
[0003] At present, the traditional method of designing pipelines in engine test labs is to use a tape measure or a long ruler on site to simulate the location of each section of the pipeline, estimate the size of each section of the pipeline and the angle of the elbow, and determine the general size and direction of the pipeline. Then, the pipeline trunk is made according to the rough size, and the pipeline is tested on site and marked with interference or unnecessary size and angle. Then the pipeline is dismantled, repaired and tested again until the pipeline meets the use requirements.
[0004] The existing tooling intake duct surveying and mapping method has large spatial dimension errors and requires continuous trial and error. The cycle from the initial design to the final production is long, which affects the overall rhythm of the engine test.
[0005] Currently, there are more accurate professional 3D scanning equipment that can scan and shape the test room, so as to measure the 3D model without obstacles. However, 3D scanning equipment is expensive, and positioning reference dot matrix must be attached to the object to be scanned before scanning, which is difficult to carry out conveniently and efficiently in the engine test room. Summary of the invention
[0006] The technical problem to be solved by the present invention is to provide a rapid design method for engine test pipelines, which solves the problem that the spatial dimensions between two farther ports of the pipeline cannot be directly measured or the surveying and mapping errors are large, and improves the accuracy of pipeline direction design and the convenience of the sampling process.
[0007] In order to solve the above technical problems, the technical solution adopted by the present invention is: A rapid design method for an engine test pipeline comprises the following steps: S1. Place reference objects of fixed size in the scene where the engine test pipeline passes through, and collect several sets of image information; S2, importing the image information into a computer and using 3D reconstruction software to read the image information; S3, spatial distance analysis of grouped image information; S4, stitching the image information and outputting a three-dimensional model; S5. Import the 3D model into the 3D modeling software, and adjust the size of the 3D model to the size of the scene through which the engine test pipeline passes, in a 1:1 ratio according to the size of the reference object; S6. Measure the spatial dimensions in the modeling software and generate pipelines for engine testing.
[0008] A further improvement of the technical solution of the present invention is that in S1, the acquisition environment is required to have an illumination greater than 100lx, the HDR function of the camera is turned off, and photos are taken from multiple angles of the scene between the compressor port of the supercharger and the air inlet of the intercooler in the test room to ensure that a reference object of a fixed size is captured in the camera. The resolution of the collected photos is greater than 3 million pixels, the signal-to-noise ratio is greater than 10, the peak signal-to-noise ratio is greater than 30dB, and the number is greater than 15.
[0009] A further improvement of the technical solution of the present invention is that in S2, all photos are imported into the computer, ensuring that the directory path where the photos are located is in English, and using the open source Meshroom 3D reconstruction software to import these photo resources.
[0010] A further improvement of the technical solution of the present invention is that: in S3, photogrammetry calculation is performed, wherein the DepthMap node uses the CUDA technology of the NVIDIA graphics card to analyze the spatial distance of the grouped photos.
[0011] A further improvement of the technical solution of the present invention is that in S4, after calculation of 11 nodes by Meshroom software, the photos are stitched together and finally form a three-dimensional model in obj format.
[0012] A further improvement of the technical solution of the present invention is that in S5, the obj model is imported into the open source blender modeling software, the scaling factor is determined according to the ratio between the fixed-size reference object and the real object, and then the global model is scaled and restored to 1:1 according to the coefficient.
[0013] A further improvement of the technical solution of the present invention is that in S6, a measuring tool is used in the blender modeling software to measure the space size without obstacles, determine the length and direction of each section of the pipeline, draw a drawing and guide the pipeline processing.
[0014] Due to the adoption of the above technical solution, the technical progress achieved by the present invention is: The invention has the advantages of realizing convenient, rapid and more accurate measurement of the space dimension between two ports of a pipeline and design of the pipeline direction, which can be completed with only a camera phone and a computer with an NVIDIA graphics card. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1It is a flow chart of the rapid design method of pipelines for engine testing in the present invention. DETAILED DESCRIPTION
[0016] The present invention is further described in detail below with reference to the accompanying drawings and embodiments: In the description of the present invention, “several” means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.
[0017] like Figure 1 As shown, a rapid design method for engine test piping includes the following steps: S1. Place reference objects of fixed size in the scene where the engine test pipeline passes through, and collect several sets of image information; In a bright and stable light environment, turn off the camera's HDR function and take photos from multiple angles of the scene between the turbocharger compressor port and the test room intercooler air inlet. Make sure there are reference objects of a fixed size in the frame and the photos are clear. No more than 15 photos will suffice.
[0018] S2, importing the image information into a computer and using 3D reconstruction software to read the image information; Import all the photos into your computer, make sure the directory path where the photos are located is in English, and then use the open source Meshroom 3D reconstruction software to import these photo resources.
[0019] S3, spatial distance analysis of grouped image information; Start the photogrammetry calculation, and analyze the spatial distance of the grouped photos through the first 8 nodes: CameraInit→FeatureExtraction→ImageMatching→FeatureMatching→StructureFromMotion→PrepareDenseScence→DepthMap→DepthMapFilter. The DepthMap node must be completed using the CUDA (Compute Unified Device Architecture) technology of the NVIDIA graphics card.
[0020] S4, stitching the image information and outputting a three-dimensional model; After the calculation of the last three nodes of Meshroom software Photogrammetry: Meshing→MeshFiltering→Texturing, the groups of photos are automatically stitched together according to the relative spatial distance analyzed in the previous step, and finally a three-dimensional model in obj format with a texture is formed.
[0021] S5. Import the 3D model into the 3D modeling software, and adjust the size of the 3D model to the size of the scene through which the engine test pipeline passes, in a 1:1 ratio according to the size of the reference object; Import the obj model into the open source blender modeling software, determine the scaling factor based on the ratio of the fixed-size reference object to the actual object, and then scale the global model back to 1:1 according to the factor.
[0022] S6. Measure the spatial dimensions in the modeling software and generate pipelines for engine testing; Use the measurement tools in the blender modeling software to measure the space size without obstacles, determine the length and direction of each section of the pipeline, and then draw the drawings and guide the pipeline processing.
[0023] In summary, the present invention can solve the problem that the spatial dimension between two farther ports of the pipeline cannot be directly measured or the surveying and mapping error is large, thereby improving the accuracy of pipeline direction design and the convenience of the sampling process.
Claims
1. A rapid design method for engine test piping, characterized in that: The following steps are involved: S1. Place reference objects of fixed size in the scene where the engine test pipeline passes through, and collect several sets of image information; S2, importing the image information into a computer and using 3D reconstruction software to read the image information; S3, spatial distance analysis of grouped image information; S4, stitching the image information and outputting a three-dimensional model; S5. Import the 3D model into the 3D modeling software, and adjust the size of the 3D model to the size of the scene through which the engine test pipeline passes, in a 1:1 ratio according to the size of the reference object; S6. Measure the spatial dimensions in the modeling software and generate pipelines for engine testing.
2. The rapid design method for engine test piping according to claim 1, characterized in that: In S1, the acquisition environment is required to have an illumination greater than 100lx, turn off the camera's HDR function, take photos from multiple angles of the scene between the turbocharger compressor port and the test room intercooler air inlet, ensure that there are reference objects of fixed size in the frame, and the resolution of the collected photos is greater than 3 million pixels, the signal-to-noise ratio is greater than 10, the peak signal-to-noise ratio is greater than 30dB, and the number is greater than 15.
3. The rapid design method for engine test piping according to claim 2, characterized in that: In S2, import all photos into the computer, make sure the directory path where the photos are located is in English, and use the open source Meshroom 3D reconstruction software to import these photo resources.
4. The rapid design method for engine test piping according to claim 1, characterized in that: In S3, photogrammetry calculations are performed, and the DepthMap node uses the CUDA technology of NVIDIA graphics cards to analyze the spatial distance of grouped photos.
5. The rapid design method for engine test piping according to claim 4, characterized in that: In S4, after 11 nodes of Meshroom software were calculated, the photos were stitched together to form a 3D model in obj format.
6. The rapid design method for engine test piping according to claim 1, characterized in that: In S5, the obj model is imported into the open source blender modeling software, and the scaling factor is determined based on the ratio between the fixed-size reference object and the real object. Then, the global model is scaled and restored to 1:1 according to the coefficient.
7. The rapid design method for engine test piping according to claim 1, characterized in that: In S6, the measurement tools in the blender modeling software are used to measure the space dimensions without obstacles, determine the length and direction of each section of the pipeline, draw drawings and guide pipeline processing.