Optimization of arc models parameter using genetic algorithm

Over the years, as our knowledge of the interrupting process progressed, many techniques have been developed to test the circuit breakers and simulated arc models. To develop a realistic arc model, it is necessary to adopt the micro modeling approach, where a model is based on the physical processes constituting the phenomena. The model should consist of mathematical equations formulated from considerations of these processes so that the model performance is likely to be more accurate over a wide range of operating conditions. There are three models (Physical Model, Black Box Model and Parameter Model) that describe the behavior of arc. The task for this paper work on Black Box models and evaluate different arc models in compared with KEMA [as a reference] arc model for circuit-breakers with the purpose of finding the best parameters for the breaking ability. For this purpose 6 main generic arc models (Cassie, Habedank, Mayr, Modified Mayr, Schavemaker, and Schwarz) has been presented and GA algorithm is used to minimize differences between two conductances (g_KEMA g_TESTMODEL) , as long as the best fitness and parameters was achieved.