Nowadays more and more power electronics and digital controller are used in electric power systems. Before they are really applied to the power system, the rigorous performance evaluation of the power electronics apparatus and there digital controller should be carried out. Real time simulation of dynamic system is a popular technology in testing hardware and/or software in the loop and other systems. This dissertation addresses the issue of deign of real time simulation solver for large scale power electronics system.

A novel real time simulation solver is proposed for decreasing the model computation time in fixed step size simulation. In order to catch the dynamic performance of these systems being simulated, a reasonable time-frame should be provided to the system in according to specific requirement. Modern complex power electronics system driven by high frequency Pulse-Width Modulation (PWM) converter controller demand time-frames that are significantly shorter than those found in most real-time simulation applications. As a consequence real time simulation of dynamic system is essentially signal processing application composed of tasks with hard deadline periodic request and therefore, requires an efficient methodology for real time simulation using single/multi processor resources. This research work focuses on a methodology for the real time simulation of dynamic systems based on Virtual Test Bed (VTB) including task scheduling policy on multi processors.

This algorithm relies on the practical hardware system and the characters of different power system. Practical feasibility of the proposed algorithm is demonstrated by implementation on a digital computing platform comprising of 4 TS101 DSP from Analog Device Company. The hardware and software design process follows a modular approach and ANSI C standard, which makes it suitable for any microprocessors and embedded system.