Power system frequency estimation using morphological prediction of Clarke components

This paper presents a new frequency estimator algorithm for electric power systems that uses the Mathematical Morphology operators of dilation and erosion to predict the Clarke Components from voltage signals of three phases of the system. The frequency is obtained as a function of phase shift between the predicted complex signal and the received complex signal given by the αβ-Transformation. The proposed method presented great accuracy and fast convergence for a wide range of different operational conditions involving transitory events of frequency deviation, amplitude variations, signal phase shifts and stable power swings. In addition, signals distorted by noises, harmonics and inter-harmonics were also tested. In order to demonstrate the quality of the new frequency estimator in each analyzed case, its estimation was compared with the responses obtained by four other methods, in terms of performance indices based on the estimation of transitory error and response convergence time. For all analyzed cases, the proposed methodology presented better results, showing great robustness and high accuracy for its task.