A DESIGN REUSE METHOD FOR AGRICULTURAL MACHINERY CAD MODEL WITH LIGHT PROPAGATION SIMULATION
基于光传播模拟的农机装备CAD模型设计重用方法研究
DOI : https://doi.org/10.35633/inmateh-58-11
Authors
(*) Corresponding authors:
Abstract
The problem of long design cycle, low efficiency and poor knowledge reusability still exists in China, in the process of agricultural machinery design. The reuse retrieval method designed for agricultural equipment CAD model is proposed to improve agricultural machinery digital design. Since agricultural machinery CAD model has internal coupling and local similarity attributes resulting in low performance of general retrieval methods, the Monte-Carlo light simulation method is utilized to extract CAD model partial features. Firstly, the light propagation process in different mediums was quantitatively analysed with optical parameters. Then how light propagates in complex agricultural CAD model was simulated according to Monte-Carlo method. After that the CAD model feature could be extracted with partial propagation information for design reuse retrieval. Finally, the proposed reuse retrieval method was evaluated on retrieval precision, recall and E-measure with particular agricultural model dataset. The results showed that the accuracy of proposed method was higher than that of general model retrieval approaches. Sphere constrained space and 10000 photon packets were shown to be the optimal feature extraction parameters for agricultural CAD model. The final design application results indicated that the proposed method could particularly satisfy agricultural CAD rapid design requirement, which is more suitable for offline complex agriculture CAD model retrieval. This paper is expected to provide a fine-grained, intelligent and visual method for agricultural machinery CAD management, promoting automatic and intelligent design of agricultural machinery manufacturing industry.
Abstract in Chinese