The localisation and harmonic level determination of non-linear loads (NLLs) are the first steps in solving harmonic problems in power system. To cope with this challenge, this study presents an intelligent approach based on fuzzy logic. The location and relative level of harmonic sources are determined in the fuzzy stage by interpreting the magnitude and sign of harmonic powers and network reactances. The accuracy of harmonic state estimation (HSE) is considered, besides the number of harmonic meters (HMs) and observability redundancy, in the optimal allocation of HMs. A modified adaptive binary imperialist competitive algorithm for the optimal allocation of HMs is presented. Also, the HSE accuracy is investigated on the accuracy of intelligent localisation of multiple NLLs in the presence of measurement errors. The efficiency of the proposed approach is verified on IEEE 18-bus and IEEE 69-bus test systems under different scenarios.
