Genetic algorithm approach to the modelling of polluted insulators

The phenomenon of flashover in polluted insulators has not yet been described accurately through a mathematical model. The main difficulty lies in the definition of the constants of the arc, which is formed in the dry bands when the voltage exceeds its critical value. The authors present a complex optimisation method based on genetic algorithms for the determination of the arc constants, using experimental results from artificially polluted insulators. First the well known model of Obenhaus for pollution flashover is used. This model results in a system of equations which cannot be solved with conventional arithmetic methods. The application of genetic algorithms enables the definition of the arc constants, resulting also in the calculation of the critical conditions at the beginning of the pollution flashover mechanism. In this way a mathematical model is established, which simulates accurately the experimental results.