Research On Several Key Problems Of Structural Static And Dynamic Collaborative Optimization

Design optimization of coupled systems plays an important role in generalized optimization of mechanical systems while collaborative optimization technology is an efficient method to solve coupled optimization problems in engineering applications. However, it is difficult to solve design optimization problems of complex systems with the traditional computing methods due to large-scale computation from complex simulation processes. Therefore, it is necessary to explore a new computing method. The static and dynamic structural optimization methodology in a collaborative environment is studied to solve the optimization problems for complex mechanical systems such as the working equipment of a hydraulic excavator. And the contributions are as follows:First, the relation between collaborative optimization and generalized optimization is studied, which facilitates realizing flexible optimization with two important outcomes--the optimization modeling platform and the optimization planning platform developed by our generalized optimization research group.Secondly, the modeling method for finite element analysis (FEA) of the working equipment of a hydraulic excavator is studied and a model updating method is used to improve the accuracy of the FEA models. Then, parameterized re-analyzing methods are studied. Two parameterized methods for static FEA and dynamic FEA are developed. Optimization programs and commercial FEA software are integrated into the distributed computing circumstance. A FEA data sharing method is studied to double FEA computing rate. An intelligent system for scheduling tasks dynamically is built with a workflow technology. The above method facilitates the realization of distributed parallel computation in a multi-agent based collaborative optimization system.Thirdly, a peer-to-peer collaborative optimization method based on distributed computing technology is put forward in order to realize flexible optimization. A new coordination method is proposed to solve the conflict of the related variables of different sub-optimization problems. Since the purpose of coordination of the related variables is to find a variable scheme thatsatisfies every sub-optimization problem, it is essentially an optimization problem too. Therefore, an efficient coordination optimization model is built. To overcome the computing difficulty, it is indispensable to study calculating method of the coordination model. The proposed computing method employs genetic algorithm to produce new design schemes by crossing and mutant of the related variables.Fourthly, as a peer-to-peer computing circumstance, a distributed collaborative optimization system must meet the following requirements: easy linking to CAE tools, support for geographically distributed optimization, support for database management, extensibility for optimization solvers and ability to perform trade-off between different design responses to assist exploration of Pareto solutions. Using multi-agent technology for distributed collaborative optimization system is motivated by its following advantages: distributed architecture, autonomy, and sharability. A multi-agent based collaborative system is constructed for supporting static and dynamic structural optimization.Finally, the static and dynamic structural optimization of the working equipment of a hydraulic excavator is realized in a collaborative environment, demonstrating that the methods are efficient and the multi-agent system is reliable and flexible.