High-Speed Real-Time Simulation for Power Electronic Systems

Abstract

Real-time simulators have been used as an aid to power system design for many years. Over time, scaled-down physical models and analog computer-based simulators have given way to real-time simulators based on digital technology. Another trend has been the need for shorter and shorter frame times for these digital real-time simulators. Modern power electronic systems use highfrequency Pulse-Width Modulated (PWM) controllers. Indications are that frame times of 2 μS are needed for PWM switching frequencies of 10 KHz. As frequencies increase further it is possible that frame times of less than 1 μS may be required. In order to achieve these very short frame times, the implementation of this type of simulator requires careful selection of the methods and technologies used. The first involves thorough analysis of the integration method chosen, to ensure that it provides the performance, stability and accuracy required. Additionally, the choice of computing platform is crucial to provide the computational support to meet these very aggressive timing requirements.

Field Programmable Gate Array Applications—A Scientometric Review

“…In the red cluster, we find FPGAs' applications focused in real-time implementations, such as data acquisition [565,[1473][1474][1475], simulations [1022,1024,1082,1476], Neural Netw. [1477][1478][1479] and image processing [1480][1481][1482][1483][1484][1485][1486][1487].…”

Advances in high-speed real-time multi-rate simulation techniques for ship power systems

“…Conventional real-time simulation platforms, based on available real-time operating systems such as real-time versions of Linux, are not normally capable of supporting frame times of less than about 20 μS. Earlier research [1], [2] used an array of 4 digital signal processors (DSPs) to achieve a 2-μS frame time for a 6-pulse, This research is supported by the Office of Naval Research 978-1-4244-3439-8/09/$25.00 ©2009 IEEE back-to-back system with PWM controllers involving 23 first-order differential equations, switching logic and control algorithms. A study of the future potential for improving this performance, either with more powerful processors or through improvements in programming, indicated that significant improvements were not likely using DSPs and attention was switched to the use of FPGAs with the joint aims of further reducing frame times and increasing the amount of computation possible within a given time.…”