Abstract: Problems of low efficiency, low accuracy and complex information exchange in the application of conventional optimization approach to complex engineering system optimization design are pointed out. Reasons for the presence of these problems are explained in detail by analyzing the calculation process of complex system optimization. The response surface method is introduced to solve these problems, based on which an iterative method of series response surface is presented. In this method, complex system analysis is replaced by response surface model analysis and a database of the complex system is constructed. At each iteration, the database is expanded through new added data and then the response surface model is refined according to the expanded database, which makes the response surface model more and more accurate until convergence is achieved. Moreover, to decrease the amount of computation while retaining high accuracy, this method provides a systematic way of choosing the new added data. In the last section, a large-scale transporter wing design problem is solved successfully by this series response surface method, which shows the validity of the method.

Key words: Complex engineering system Optimization, Response surface method