Abstarct: In this thesis, a new method has been presented to simultaneously estimate the amplitude and harmonic phase of signals in power systems in the presence of noise. Harmonics are the important discussion in the field of the power quality. To be ensured about the existence of pure sinusoidal signals or, in other words, a better power quality, a concern is that the power system works without a considerable loss in its performance while the voltage of transmission lines have a few percent of disturbance; however, the more it gets close to the load the more the disturbances increase. Todays, the increase in using the nonlinear load such as computers, energy-saving lamps, power electronic converters and induction furnaces and so on has resulted in deforming the sinusoidal form of power supply and increasing the noise of signals. The number of these random disturbances is very low and most of them are periodic which are so called harmonic. The undesirable effects of harmonics are signal interference, sudden over-voltages, saturation of transformers and their temperature increase as well as error in the functionality of breaker. In addition to harmonic identification for designing filters or upgrading the current filters or determining the source of harmonics, the harmonic components of measured signals need to be specified. Therefore, a method is required to be able to rapidly measure and estimate the harmonics of signal. In this thesis, first, a few data are used to obtain the value of primary parameters of adaptive algorithms using an offline mechanism i.e. an optimization algorithm and then the adaptive algorithms are utilized online. In this synthetic procedure, the amplitude and phase are estimated simultaneously. The simulation result shows that the proposed method has a higher speed of convergence to the real value, a better performance and higher accuracy in a noisy system in compare with the…