567 Copyright © 2015, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. Chapter 21 DOI: 10.4018/978-1-4666-7248-2.ch021 ABSTRACT In this chapter, two computational algorithms are proposed and applied on an estimation algorithm, in order to improve the global performance of the estimation phase. The proposed system is studied based on the Model Reference Adaptive System (MRAS). The importance of the estimation phase in a large applications number is basically observed on the applications applied on electrical motors, where a lot number of parameters are measured with real measurement equipments, as Tesla Meter, speed shaft, and others. The idea is based generally on the software applications, where it is possible to guarantee the desired estimation phase using a software algorithm. In this chapter the MRAS technique is pro- posed as the software algorithm, for replacing the measurement materials for online estimate the overall characteristic PMSM parameters. Our approach aims to ameliorate the MRAS technique with intelligent optimization methods called BFO and PSO. 1. INTRODUCTION In the last decade, the majority applications related to the industrial, militaries, robot and automobile are basing on the electrical motor, due to many causes related to the environment protection and energy saving and especially to the high efficiency. In the literature many motor models and types are presented and each model is requested for a specific application. Electrical vehicles applications will be widely used Electrical Motor Parameters Estimator Improved by a Computational Algorithm Flah. Aymen National School of Engineering of Gabes, Tunisia Habib Kraiem National School of Engineering of Gabes, Tunisia Sbita. Lassaâd National School of Engineering of Gabes, Tunisia