Research Of Multidisciplinary Design Optimization For Key Parts Of Marine Diesel Engine

According to the comprehensive performance requirements of modern ship, the coupling of various disciplines in the design and development process is required to improve the overall design level of marine diesel engine. Combined with the theory of complex system modeling, general technical framework of the key parts of marine diesel engine based on Multidisciplinary Design Optimization (MDO) is established in this paper. The key technology issues such as modeling and decomposition of the complex system, management of the modeling process of multidisciplinary optimization, and multi-objective optimization decision are studied. On this basis, a special MDO tool system for the rapid design of marine diesel engine is developed according to the idea of product cooperative design with the integration of CAD/CAE/CAO, which effectively supports the design and development process of marine diesel engine in China.The main research contents included:(1) Based on complex system modeling theory, the general technical framework of MDO for the key parts of marine diesel engine is proposed, and the key technologies are analyzed in detail. A multi-level of the MDO model and its solving strategies are established. The structural strength, vibration and thermal design requirements of marine diesel engine structure design are analyzed, and the performance coupling relationship is obtained from the four levels of "product-System-Assembly-Part".(2) For the accuracy of MDO of complex systems, a multidiscipline performance modeling method is proposed considering surface characteristics. A systematic study of the nonlinear characteristics of the composite structure based the static and dynamic experiments are carried out. In view of the problem of the reusability of the analysis model, the MDO multi view modeling method based on the main model of marine diesel engine is put forward.(3) For the coordination of efficiency and precision of MDO, a multidisciplinary optimization model reduction method based on data mining and knowledge discovery is proposed. Combined with experimental design and agent model technology, the optimization model of diesel engine products is simplified, and the precision and efficiency of the marine diesel engine are realized.(4) A comparative study of the classical multi-objective evolutionary algorithm for multi-objective and target decision making problems is carried out. The typical test functions are selected to evaluate the distribution, robustness and operating efficiency of the Pareto optimal solution set. Based on this, a multi-objective optimization strategy based on robust design is proposed, which improves the efficiency of the optimization algorithm and the feasibility of the optimization solution. Taking the optimization design of the cam mechanism of marine diesel engine as an example, the validity of the method is verified.(5) For the problem of multidisciplinary coordination and process integration, the integrated technology of CAD/CAE/CAO based on multidisciplinary design optimization is proposed. A special tool system for the rapid design of marine diesel engine is developed for realizing the integration of multidisciplinary design optimization and model reuse.