Research On Multi-sense Channel Virtual Assembly Technology For Power Equipment

Virtual assembly technology has been integrated into the whole life cycle of industrial equipment,which plays an important role in shortening the equipment development cycle,reduces production costs and optimizes equipment performance.However,due to the constraints of sensing channel,the difference of object volume and the occlusion of object,the virtual assembly system has some problems,such as low immersion,waste of collision detection computing resources and limited object selection under occlusion.We take the power transformer control cabinet as the virtual assembly object,and explore the key technologies of virtual assembly system,such as interaction technology,collision detection method and object selection method.We analyze the principle of the existing technology,compare the advantages of these technologies to improve and innovate them.The main contents of this thesis are as follows:(a)Natural interaction method of multi-sensory channels.The traditional virtual assembly system mostly adopts the natural interaction method of single sensory channel.The limitation of sensory channel directly affects the immersion and operability of the whole system.This theis proposes a natural interaction method that integrates multiple sensory channels at the same time,which integrates gesture interaction,voice interaction,tactile interaction,stereo display and real-time display of real-world images in virtual environment.The experimental results show that this method improves the virtual assembly system in learning experience,operation experience,immersion and vertigo.(b)AABB octree collision detection method supporting volume difference.Due to the huge volume difference of power transformer control cabinet equipment parts,the classical hierarchical bounding box collision detection algorithm has the problem of waste of computing resources.We propose the AABB octree collision detection method supporting volume difference,and use the volume difference coefficient as the judgment condition of bounding box layered calculation.The experimental results show that the proposed collision detection method with volume difference improves the computation time by 12 percent compared with the classic AABB Octree collision detection method.(c)Multi objective object selection method based on volume region.Due to the close assembly relationship of the object model of power equipment components,the problems of object occlusion and excessive number of objects are inevitable in the process of object selection.We propose an object selection method based on volume selection,which completes the selection of objects by calculating the boundary of objects and judging the intersection of volume regions.The experimental results show that the method is suitable for multi object selection tasks in various extreme occlusion situations.Focusing on the human-computer interaction technology,collision detection technology and object selection technology of virtual assembly system,this thesis makes in-depth research on three aspects: natural interaction of multi-sensor channels,collision detection supporting volume difference and multi-target object selection with occlusion.The main technologies improve the immersion and operability of virtual assembly system.