Re-estimation of hidden Markov model parameters of Phasor Measurement Unit

In recent times, the power system has expanded in an unprecedented way owing to technological innovations and geographical dimensions. For ensuring proper operation, monitoring and control, a system called Wide Area Measurement System (WAMS) has evolved. Phasor Measurement Unit (PMU) is one of the vital components of WAMS. Any failure in PMU has catastrophic consequences on WAMS. Thus reliability analysis of PMU is of utmost importance. Hidden Markov model (HMM) is one of the important techniques for evaluating reliability. However, while evaluating the reliability of PMU using HMM, the HMM parameters such as state transition probability matrix (A), observation probability matrix (B) and initial state probability (p) are either assumed or have been derived from related research papers. In this paper, a Baum-Welch algorithm has been used to re-estimate these HMM parameters of the seven modules of PMU to best fit the observations.