The method for modeling switching converters plays a key role in real-time simulators. Associate discrete circuit (ADC) modeling technique is a commonly used method for modeling the switching converter. However, the optimal selection of the ADC-based switch model parameters has great importance in the accuracy of the real-time simulator. In this paper, the design of a real-time simulator for a switching power converter has been done, in which a novel method for detecting optimum values of the switch model parameters has been expressed. Particle swarm optimization (PSO) algorithm is used to find these optimum values using state-space analysis of the modeled circuit in the z-domain. The modified nodal analysis (MNA) method solves the real-time model at each simulation time-step. Case studies are a single-phase 5-level Cascaded H-Bridge (CHB), three-phase 9-level CHB inverter, and a Transmission line. Field-programmable gate array (FPGA) has been used as a platform for implementing the real-time simulator. Experimental result of the real-time simulator of the 5-level CHB inverter on a SPARTAN-6 FPGA and its comparison with the result of a prototype of a 5-level CHB inverter confirms not only the performance of the real-time simulator, but also the effectiveness of the proposed method for confirming the real-time simulator accuracy.

开关转换器建模方法在实时仿真器中起着关键作用。关联离散电路（ADC）建模技术是对开关转换器建模的常用方法。但是，基于ADC的开关模型参数的最佳选择对于实时仿真器的准确性非常重要。在本文中，已经完成了用于开关功率转换器的实时模拟器的设计，其中表达了一种用于检测开关模型参数的最优值的新颖方法。粒子群优化（PSO）算法用于通过在z域中对建模电路进行状态空间分析来找到这些最佳值。改进的节点分析（MNA）方法可在每个仿真时间步求解实时模型。案例研究是一个单相5级层叠H桥（CHB），三相9级CHB逆变器和一条传输线。现场可编程门阵列（FPGA）已用作实现实时模拟器的平台。在SPARTAN-6 FPGA上的5级CHB逆变器实时仿真器的实验结果以及与5级CHB逆变器原型的结果进行比较，不仅证实了实时仿真器的性能，而且证实了实时仿真器的性能。所提出的方法对于确认实时模拟器准确性的有效性。