Analysis Short Time Power Quality Disturbances Based On Local Cosine Transform And Wavelet Transform

With the rapid growth of national economy and formation of power market, power quality has become the focus of many fields. It is necessary to set up power quality monitor and analysis system, which can realize fault location and character extraction to all power quality disturbances. It is the premise and reference of improve and harness power quality.Firstly this paper expands relative concept about power quality, introduces all power quality issues and their standard. Among those several representative short-term power quality disturbances are introduced particularly, and their characteristics are analyzed. They are: voltage sag, voltage swell and voltage interruption; transient harmonic and time-varying harmonic; transient disturbances, including transient oscillation and transient pulses.Generally power quality detection methods can be generalized in three types: time domain simulation, frequency domain simulation and methods based on transform. Wavelet transform has become study focus of many experts and scholars as representative of the third type. However, power quality issues have many types, and they have different characters. Sometimes wavelet transform can't treat all power quality issues especially short-term power quality disturbances. It is necessary that add other method to make up the shortage of wavelet transform. This paper introduces local cosine transform (LCT) to power quality detection. This paper expounds fundamental principle and time-frequency character of wavelet transformand local cosine transform.We discuss how to apply local cosine transform to power quality disturbance detecting considering two difficulty in wavelet analysis - time-varying harmonic detection and signal noises.Firstly we discuss how to detecting fundamental frequency disturbances in noises through noises robustness of LCT coefficients and difference algorithm.How to use local cosine transform to detect short time harmonics, how to search LCT optimum base to analyze time-varying harmonic are emphasis of this paper.Finally we explain how to utilize wavelet transform module maximum filtering signal noises and analyzing high frequency transient disturbances.After simulation and comparing with other methods we know this method has preferable performance. Analyzing power quality issues using this method can get satisfactory results. This method makes up the singleness and shortage of wavelet analysis in power quality detection. It provides a new approach in this field.In the end we point out some place need improving of this method.