Using MPSO algorithm to optimize three-phase squirrel cage induction motor design

Improving the efficiency of three-phase squirrel cage induction motors, which are the most energy consuming electric machines in the world, saves much energy. The efficiency can be raised by optimizing the induction motor design. The objective of this paper is mainly to develop a modified particle swarm optimization (MPSO) technique to the three-phase cage induction motor design problem with a view to minimize the annual cost of the motor. Even if particle swarm optimization (PSO) algorithm is easy to implement and has been shown to perform well on various power system optimization problems, they may get trapped in a local optimum due to premature convergence when solving the larger constrained problems. In the proposed MPSO algorithm, the PSO parameters such as inertia weight and acceleration factors are made adaptive on the basis of objective function. By adapting the PSO parameters, it not only avoids premature convergence but also explores and exploits the promising regions in the search space successfully. The objective function is a summation of the annual interest and depreciation of the active material coast, the annual power loss cost and the annual energy loss cost. The objective function is then augmented with constraints. A new constraint handling strategy, parameterless penalty function approach is embedded in the MPSO algorithm. The proposed method is applied to optimize the design of 5 and 10 hp motors and comparisons are performed with conventional design, genetic algorithm (GA) and PSO approaches. Moreover, the obtained results are compared with those of the optimally designed motor on the basis of minimizing the active material cost only.