The Research On Collision Detection Algorithm And Aided Design Of Urban Underground Heating Pipe Network

With the rapid development of China’s economy,the current urban underground pipe network has been unable to meet the requirements of the urban intelligence development.However, the underground pipe network is so huge and diverse that there is no comprehensive management of pipeline information for the underground pipe network.So in the design of the new network cannot be analyzed with the original pipe network data statistical analysis,so in the pipeline construction pipeline collision may occur every now and then,it is also easy to cause secondary disasters to other pipelines,in this way,the support of pipeline collision detection technology is urgently needed to improve the working efficiency of urban planning and construction, and to avoid unnecessary manual labor.In this paper,by analyzing the characteristics of the heating pipe network and the characteristics of the bounding box, the paper chooses the axis aligned bounding box algorithm for collision detection of heating pipe network,and focuses on the collision detection algorithm based on axis aligned bounding box.This paper mainly discusses the following aspects:(1)Several typical collision detection algorithms are analyzed and compared,including the bounding sphere,axis aligned bounding box,direction bounding box and k-Dops collision detection algorithm,and the collision detection algorithm based on axis aligned bounding box is introduced in this paper.(2)In this paper,the pipeline collision detection preprocessing is carried out.By studying the collision detection algorithm we can know:first,the collision detection algorithm needs to be backed up by an intersection test,which increases a lot of computational effort;Second,the behavior of collision detection is a kind of local behavior,which is almost not collision with the object far away.Therefore,this paper proposes a spatial partitioning method,that is exponential grid algorithm.The exponential grid algorithm divides the space into several regions,and the collision detection is limited to the subspace,to avoid the redundant intersection test,to improve the speed of the intersection test of bounding box.At this stage,the structure of the bounding box,the tree,the structure of the tree,the node dissection strategy are selected,and then the axis aligned bounding box tree then can be constructed.(3)The collision detection of the axis aligned bounding box algorithm is carried out.Afterthe pipeline collision detection pretreatment,by traversing the axis aligned bounding box tree collision detection pipeline.In this paper,the breadth first search algorithm is used to traverse the axis aligned bounding box tree to reduce the computation.(4)The auxiliary design of heating pipe network is carried out.Firstly,the mathematical model of heating pipe network is studied.Secondly,the particle swarm optimization algorithm is used to assist the design of the pipeline in the collision area.Since the particle swarm algorithm is easy to fall into the local optimal solution,this paper proposes a hybrid particle swarm optimization algorithm based on genetic algorithm to design the heating pipe network,using this algorithm to adjust the parameters of heating pipe network in the area of the collision.Experiments show that the collision detection algorithm can effectively improve the efficiency of collision detection,and study of heating pipe network aided design algorithm in pipeline parameters optimization also achieved good results,saving the cost of heating pipe network infrastructure for more than 5%.