A resistive-current extraction algorithm to assist with on-line monitoring of surge arresters is proposed in this paper. The algorithm is based on the improved equivalent model for surge arresters and combines evolutionary optimisation with the base-frequency approximating method. A genetic algorithm is used to obtain the optimal capacitance such that the phase shift between the base-components of the branch voltage and current is minimised thereby yielding a good approximation of the resistive current. In this paper, the algorithm is implemented and tested using Matlab/Simulink. Results indicate that the proposed algorithm is able to efficiently and accurately obtain the resistive component of the leakage current, using the improved equivalent model, under both ideal and distorted supply conditions.