Research On Automatic Design Methods Of Pipeline Laying In Mechinery Equipment

As the continuing development of industrialization, various pipelines in cars, ships, chemicals, aerospace, aviation, and other mechinery equipment is in a wide range of applications. But at the same time as the mechinery equipment has complex structures, numerous components, high integration and internal small assembly room; the pipeline laying design (PLD) seriously restricts the development velocity and the production cycle of the mechinery equipment. For a long time, the manual completes the pipeline laying design of mechinery equipment, which makes the pipeline laying design relies heavily on the practical experiences. Although the pipeline laying design process is simple, but with the increase in the diversity of mechinery equipment, the pipeline laying design shows randomness, fuzziness, and the dynamic characteristics. Thus, the pipeline laying design has been unable to meet requirements of development of mechinery equipment such as short design cycle, the optimization level and easy implement.With the rapid development of computer technology, digital design methods for the pipeline laying design of mechinery equipment provided strong support and new tools such as computer-aided design, computer-aided manufacturing, agile manufacturing, flexible manufacturing systems, and computer integrated manufacturing system. Rely on these digital technologies, the pipeline laying design of mechinery equipment entered the digital era. Through the introduction of artificial intelligence technology, the pipeline laying design of mechinery equipment not only can avoid complex labor, the long laying cycles and high development costs, but also improve the quality of pipeline laying design.At present,the pipeline laying design has deficiencies in the following areas such as difficult implementation, low optimization and high complexity. For the difficult implementation, this paper presents a chaotic grid pre-processing model. According to the purpose of pre-processing model, the nature of the pre-processing model is given. Chaos grid pre-processing model (CGPM) is constructed based on chaotic theory. For engineering applications, four kinds of pre-processing model are proposed such as regular body CGPM, irregular body CGPM, regular surface CGPM and irregular surface CGPM. And examples of pre-processing are given.In the pipeline laying design, the constraints and the evaluation criteria of optimization of the pipeline laying design should not be ignored, especially for optimization requirements. In this paper, the material cost criteria and running cost criteria of pipelines are given based on the actual pipeline laying project requirements. Based on define of the automatic pipeline laying problem and CGPM, expression method of the solution space of the pipeline laying design is provided. For pipeline laying constraints, the implement of constraints is proposed based on the pre-processing model in this paper.Then, this paper carried out in-depth research for particle swarm optimization (PSO) algorithm. The modified PSO (M-PSO) algorithm is constructed for overcoming low optimization velocity and premature of PSO. By introduction of the ideology of population migration, the M-PSO algorithm keeps the convergence and has good performance such as optimization velocity and optimization results. Theoretical analysis proves that the M-PSO algorithm keeps convergence. And the use of experimental methods identifies M-PSO parameters.Based on the M-PSO algorithm, automatic pipeline laying design algorithm based on flow loss is proposed considering running issues of pipelines. For pipelines with branches, the automatic laying design algorithms of branch pipelines are proposed respectively considering flow loss and pipelines material cost. For the collaborative laying design of pipelines, the collaborative algorithm is proposed. By modification of M-PSO and the introduction of four collaborative operators, the collaborative algorithm can implement collaborative design of pipelines.Finally, this paper uses UG/Open API and VC 6.0 to develop pipelines laying system.