In recent years, there is a growing concern about the use of switched reluctance motor . As a variable speed drive, the switched reluctance motor has many attractive characteristics such as simple structure, low cost, high torque density and operate in a wide range of speeds that makes it a rival for other industrial motors. The switched reluctance motor also excels with high reliability and fault tolerance capability and similar to other electrical motor can operate in four quadrant of speed-torque region. Need to know the position of the rotor at any instant of time due to the structure of the motor and innate nonlinearities, acoustic noise and high torque ripple which is due to discrete torque production mechanism are disadvantages of these motors. The most torque ripple is generated when torque production mechanism is being transfered from one active phase to another actually at the commutation instants. The minimization of the torque ripple is essential in high performance servo application which require smooth operation. One approach for minimization torque ripple is based on selecting and optimum combination of the operating parameters, which include the supply voltage, turn-on and turn-off angle, current level and shaft load. The nonlinear characteristics of a switched reluctance motor makes it difficult to control. Design and tuning a complex task. Several nonlinear control techniques have been developed in order to resolve these problems and achieve high dynamic performances. In order to get a high performance, the parameters should be optimized. The conventional optimal method, such as gradient descent method, is easy to get stuck in local optimum and don’t have efficiency for highly nonlinear systems with discontinuity. Recently, optimization evolutionary algorithms such as genetic algorithm and particle swarm optimization are used to optimize control parameters. In these algorithms, irrespective of basic information about the model of the process, control parameters can be considered as a singleobjective optimization problem to improve system performance. Since the responses of the system are dependent together, improve one response can worsen another. In order to improve the responses of the system simultaneously, using evolutionary multi-objective optimization algorithms is recommended. In this thesis, principle of single objective and multi objective evolutionary algorithm are studied. Then compared operation of two famous evolutionary algorithm GA and PSO on performance of switched reluctance motor and chose PSO for use as multi objective evolutionary algorithm to optimize speed and torque responses of switched reluctance motor simultaneously due to better performance. Optimal values for on and off commutation angles and parameters of PI controller were obtained for different operating points in all four quadrant speed - torque region by considering minimization of Integral Squared Error of speed and torque ripple as objectives. These optimal parameters were used in creating gain schedule for high performances of switched reluctance motor in all operating points. Simulation result with SIMULINK/MATLAB software are presented and analyzed. Also experimental results are achieved with laboratory switched reluctance motor drive. The experimental results indicate the good performance of proposed method. Keywords: Switched reluctance motor, Speed control, Torque ripple, Genetic algorithm, Particle swarm optimizition.