Design optimization of current source inverter fed squirrel cage induction motor

Abstract

This paper deals with the design optimization of current source inverter (CSI) fed squirrel cage induction motor. For this a set of ten independent variables is selected and seven objective functions are considered namely, the active material cost, operating cost, a mixed objective function of the active material cost and operating cost, torque pulsations, voltage across commutating capacitor, maximum commutation time of CSI and an integrated objective function. To visualize extreme performance for a particular objective function, the design is optimized considering only one objective function at a time. Finally an integrated objective function is proposed which is formed as a combination of all other objective functions with suitable weighting factors to suit optimum design for a particular application. In the task of design optimization the desired thermal, starting and normal operating performance are optimally met by imposing twelve suitable constraints. In this investigation the Sequential Unconstrained Minimization Technique (SUMT) with interior penalty function approach is used to optimize the design. In this method, Rosenbrock's method is used to achieve unconstrained minimization. The optimized design results of a 7.5 kW, delta connected CSI fed cage motor are given and discussed in detail.