A Study Of Optimization Methods For Building Pipe Routing Design

In a building system,if beams,slabs and columns are the backbone of a building frame,then pipes are "vessels" which used to ensure the normal function of the entire building system.Central heating and power could be used only if pipes working accurately.The design of building pipes directly affects the quality of the building and its economic costs.The design of building pipes is similar to the path planning.It is based on two-dimensional flat or three dimensional space looking for pipe paths which satisfy the rules.The building pipes can be divided according to their professional demands.Due to complex rules,a huge number of pipes and different design space,pipes still rely on experienced experts to design.Compared with the entire building system,advanced technology in pipes lagged.It has become a problem for the development of building system design which need to be solved.Therefore,to find an automated solution to the optimization of the pipe layout problem in the rapid and effective way in three-dimensional space so that improve pipeline design efficiency and design quality,reducing economic costs is of great theoretical significance and practical value.This thesis focus on building pipe routing design automatically and pipe optimization based on three-dimensional environment.Firstly,this thesis analyze the characteristics of the three-dimensional building environment,and then created three-dimensional building environment model and the pipe model.The pipe routing design automatically problem and pipe optimization problem were studied.In terms of optimization methods,due to the pipe routing design is similar to path planning problem belonging to NP-hard.Using traditional methods are difficult to determine the period of time to achieve a satisfactory solution.Therefore,this thesis use heuristic algorithms for optimization.It can find out a high quality solution in a short period of time.The main contents are as follows:(1)Firstly,this thesis overview the content of the building design,and introduce the development of computer-aided design.According to the characteristics of the building environment model and the pipe model,the method based on the mid-point generating a pipe routing path is carried out.Using genetic algorithm and Particle Swarm Optimization algorithm to solve pipe optimization problem.The simulations proves the effectiveness and feasibility of these algorithms.(2)In order to improve the ability in escaping from the local optimal and the searching efficiency of the algorithm,Adaptive Simulated Annealing Particle Swarm Optimization(ASAPSO)algorithm was proposed for optimization.In ASAPSO algorithm,a self-adaptive parameter adjusting strategy in terms of values and combined with Simulated Annealing algorithm to adjust the optimal particle location strategies were introduced to enhance the capacity in escaping from the local optimal.To improve the searching efficiency of the algorithm,a new population initialization method based on the cost of selection probability was designed at the initial population.The simulation experiment demonstrates that compared with the PSO used for pipe routing design,the ASAPSO can achieve a significant improvement in the quality of solutions,and prove the feasibility and effectiveness of this algorithm.(3)The three-dimensional pipe routing optimization can be divided into single pipe routing,multiple pipe routing and branched pipe routing problems.In terms of multiple pipe routing and branched pipe routing design,this thesis created the penalty function and initial population,and analysis constrains rules.The simulation results are good.This thesis presented a Particle Swarm Optimization algorithm based on co-evolutionary(Co-evolutionary Adaptive Simulated Annealing Particle Swarm Optimization,CASAPSO)to solve the branched pipe routing design problem.The pipe system was divided into a main pipe and the branched pipes.Fitness function is changed for the branched pipe and co-evolution optimization is used to optimize the pipe system with the goal for the position of the branched pipe.Simulation experiment results are satisfactory(4)At last,this thesis summarize the work and looking forward to the next steps of the research work.