This paper presents an optimization approach based differential search algorithm (DSA) for amplitude and phase estimation of time varying power signal in electrical power systems. Proposed method has been tested for two well-known literature problems that include fundamental, sub-and inter-harmonic estimation cases. The performance of the suggested DSA method has been compared with those of the Genetic Algorithm-Least Square (GA-LS), Particle swarm optimization with passive congregation least square (PSOPC-LS), Bacterial foraging optimization (BFO), Fuzzy bacterial foraging-least square (F-BFO-LS) and Bacterial foraging optimization-recursive least square (BFO-RLS). The results show that the proposed DSA method has an efficient, fast and accurate estimation behavior among previously reported ones in the case of time varying harmonic power signals. In addition, the estimated amplitude and phase values using DSA method can provide possibility to design suitable harmonic reducing filter for improving power quality.