Research On Packing Problem Based On Ant Colony Algorithm

The packing problem is the process of completing the expected packing objective according to some evaluation criteria under the condition of satisfying some constraints. It is widely used in the fields of aerospace, sheet metal cutting, oil transportation pipeline, robot path packing, logistics and transportation, good packing design scheme will not only bring convenience to human, but also improve the economic performance. Therefore, It is of great significance to study the packing problem. Theoretically, the packing problem has been identified as NP-problem, It is not possible to get the optimal solution in a limited time. Scholars have carried out a lot of exploration and research and have made some achievements on the packing problem. but there are still many problems to be thrashed out. In this paper, we use ant colony algorithm to research the 3D rectangle packing problem and the 3D pipeline packing problem.The ant colony optimization algorithm is used in solving the three-dimensional rectangular packing problem and the algorithm seek higher space utilization rate as the objective. First, the paper introduce the three-dimensional rectangular packing problem model and orientation criterion. Secondly, two ways of constant value and randomly generated are taken as the initial pheromone sources in the three-dimensional packing ant colony algorithm, the thought of segment selection is put forward in updating the pheromone evaporation coefficient. By some examples, five different initialization methods and four kinds of volatile coefficient of different selection were analyzed and compared, pick out the best combination, through the final were calculated, got better result of packing than previous results.The paper takes advantage of ant colony algorithm to research three-dimensional piping packing problem, the goal of algorithm are much less pipe bending number and much shorter pipeline path. Firstly, the article introduces the PM2.5 sampler pipeline packing design ideas, decides the objective function of the packing, makes the process of 3D spatial packing space, obstacles and pipeline model. Secondly, it puts forward the pipe layout ant colony algorithm. There are two parts of updating method for pheromone: local and global. The paper takes different values in two ways. Then the algorithm take advantage of simplified model of PM2.5 sampler as example to do verification and uses the MATLAB software to realize the visualization of pipeline packing, the rationality of packing scheme is analyzed in this paper. It puts forward improved algorithm, obtains the better packing scheme, decreases the computation time, improves the packing efficiency, analyses the influence of initial parameters for the packing result. Then, select the automatic sampler of membrane changing simplified model as example to analyse rationality of packing results.And the paper proposes improvement method of ant colony algorithm.Finally, It makes a summary for the research results of rectangular packing and pipe packing and looks forward to the next step to carry out the research work.