Probabilistic evaluation of the substation performance under incoming lightning surges

In the present work, a statistical method based on Monte Carlo technique for evaluating lightning performance of a complete air insulation substation (AIS) using the capability of simulation tools is presented. This technique has been coded in MATLAB and linked to EMTP/ATP program to perform the network simulations. The parallel computing feature of MATLAB is used to accelerate the solution procedure. The paper presents the modelling of the substation equipments and connected lines. In particular, a modified IEEE recommended model of the surge arrester having its parameters obtained from genetic algorithm optimisation program is used. The simulations are carried out considering several protection scenarios. It is corroborated that the installation of metal oxide surge arresters (MOSA) at line input and in the vicinity of the autotransformer gives an acceptable mean time between failures (MTBF) at different points in the substation and then provides an adequate protection. The adopted approach has permitted a more accurate prediction of the overvoltages at different points in the substation and significantly reduces the CPU run time of the overall solution process.