The paper proposes optimal location for fixing fuel cells in a distribution system using an innovative hybrid technique. The novelty of the proposed method is the combined performance of the Genetic Algorithm (GA) and Recurrent Neural Network (RNN) technique, thereby integrating GA first phase, RNN technique and GA second phase. The optimum location for fixing the fuel cell is attained by using the GA first phase. Here, the GA first phase utilizes the load flow data at different loading conditions for determining the optimum location. The RNN is aptly trained by the target fuel cell size and the corresponding inputs such as load variation and bus number. During the testing time, the RNN provides the fuel cell capacity according to the load variation and bus number. By using the attained fuel cell capacities, the GA second phase optimizes the fuel cell capacity to minimize the power loss and the voltage deviation. The objective function mainly helps to improve the bus voltage profile and the power loss reduction. Thus, the proposed hybrid technique is implemented in the MATLAB/simulink platform and its effectiveness is analyzed by comparing it with the GA, PSO and other hybrid PSO techniques. The comparison results unequivocally demonstrate the superiority of the proposed approach and confirm its sterling potential to solve the problem.

The modern-day electric vehicles use brush less DC motors (BLDC) and Permanent Magnet Synchronous Motors (PMSM) as the main driving units. The traction systems also have tried PMSM machine as the driving unit. The variable frequency power source is required to control the speed and torque of the motor to meet the load requirements. The Inverter – power modulator is causing the production of ripples in source current and flux components, hence the torque. This paper presents the effect of inverter on the flux and torque ripples. Also, the simulation results of inverter switching losses are discussed.