A Multi-stage Compressor Design Optimization Using Cfd

∗ E-mail: [email protected] # Tel: (216) 433-5855 Fax: (216) 433-5802 E-mail: [email protected] INTRODUCTION A multi-stage compressor design is a typical multiobjective problem (MOP) involving some competing objectives such as maximization of efficiency, maximization of total pressure ratio, maximization of mass flow rate, minimization of weight, maximization of durability, etc. While single objective optimization problems may have a unique optimal solution, MOPs present a set of compromised solutions, largely known as the Pareto-optimal solutions. These solutions are optimal in the sense that no other solutions in the search space are superior to them when all objectives are considered (Fig. 1). Because the goal of MOPs is to find as many Pareto-optimal solutions as possible to reveal tradeoff information among different objectives, there is a demand for a systematic multiobjective design optimization method for compressor design optimizations. The objective of the present study is to compare two popular design optimization methods, i.e., the gradient-based methods and the evolutionary algorithms by applying them to a multi-stage axial compressor design optimization.