배전계통에서의 고조파 왜곡에 대한 고조파원의 기여도 평가 방법

Abstract

This dissertation deals with a method for evaluating the harmonic contributions of each harmonic source to system voltage distortion. To mitigate and manage system harmonic levels, it is essential to assess the contribution of individual harmonic source to the PCC(point of common coupling) voltage distortion. The result of contribution assessment heavily depends on the accuracy of equivalent model parameter. In general, the parameters are estimated using the recursive least square (RLS) algorithm and measured voltage and current at the PCC. However, the previous method based on RLS algorithm has the drawback of deteriorating estimation performance depending on the conditions of the measurement data. The constant forgetting factor RLS algorithm have potential ‘wind-up’ problem. The problem refers to an exponential increase in the covariance matrix, which makes the parameter estimation extremely sensitive and causes large errors. Also, previous methods may lead to significant errors due to the measured data contains outliers. Outlier is a data point that differs significantly from other measured data due to various causes such as measurement errors. To overcome the problems, this dissertation proposes a method of equivalent parameter estimation based on the random sample consensus (RANSAC) and RLS with a variable forgetting factor. The variable forgetting factor RLS can estimate stable equivalent parameters by limiting the increase of the covariance matrix. Also, the RANSAC algorithm can enhance estimation accuracy by removing outliers from the measurement data. The proposed estimation method ensure the accuracy of harmonic contribution assessment result. The entire procedure for quantifying contributions to harmonic sources based on the proposed estimation method is also introduced. Additionally, a network diagram of harmonic contributions is proposed to make it easy to understand the harmonic distortion contributions of all harmonic sources.