Font Size: a A A

The Research And Exploiture Of An On-line Power Quality Monitoring System And Typical Load Analysis

Posted on:2008-04-13Degree:MasterType:Thesis
Country:ChinaCandidate:X W GuFull Text:PDF
GTID:2132360242964241Subject:Power system and its automation
Abstract/Summary:PDF Full Text Request
Power quality monitoring is a key part for the analysis and control of power quality ,and it is also a necessary condition to ensure good power quality supply. A well-functioning power quality monitoring system can record all the indexes of power quality during the running of the system in detail and supervise the polluting objects, furthermore can propose solutions to improve power quality and provide useful data for power network improvement and fault analysis.A novel on-line power quality monitoring system has been researched and exploited; This system is composed of PSII, Ethernet, FTP, analyzing software programmed in VC++; Firstly, PSII samplings voltage and current signals ,and then save them to the Compact Flash in PSII; Secondly, FTP collects the data to server from the PSII via Ethernet; Finally, analyzing software in server analyzes the data and display the results in report forms or graph.The main task of this paper is how to realize the function of the analysis software of the power quality monitoring system. After studying the analysis algorithm of power quality, the kalman filtering algorithm has been adopted to detect voltage event. Some new algorithms has been put forward, and they are as follows:(1) A new algorithm for measuring harmonic and inter-harmonic based on Spatial Spectrum Analysis and Support Vector Regression Machine has been presented. Firstly, the technique of estimating the number of signal sources of spatial spectrum analysis is applied to estimate the number of sinusoidal components of signal; secondly the super-resolution DOA algorithm (parametric method) is used to calculate the frequency of each component; finally, the rest amplitudes and phase angles are estimated by a robust support vector regression machine. After frequency information being obtained, frequency range which nonparametric method analyzes in is changed to known frequency dots and the request to the length of signal is reduced when analyzing signal at a high resolution. Through digital simulation and actual inspection this method is proved effective.(2) A novel algorithm for voltage flicker calculation has been presented. Firstly, the canonical correlation analysis is applied to detect the number of sinusoidal of the voltage; Secondly, a z-transform signal model that when combined with nonlinear post-filtering schem is to estimate the operating frequency in power system is developed. Then, the TEO(Teager Energy Operator) is applied to track the amplitude variations of the voltage fliker and the envelope of the fundmental voltage flicker is obtained ; The fast Fourier transform is used to obtain the voltage flicker components from envelope of the fundmental voltage flicker. Finally, Pst can be calculated by a deduced numerical value method according to the voltage fluctuation value of the voltage flicker components, and then Plt can also be calculated.The method will not be affected by system frequency deviation and the simulation proves its validity.After the theoretical analysis, it has been put into practice. Analysis software has been programmed in VC++. This analysis software has friendly interface, and can display the results in graph and report forms. In order to gratify practical need, the power quality standard and parameter of system can be set freely.This system has been put into practice, and it can monitor the power quality problem such as harmonic,frequency,voltage imbalance,voltage flicker,and voltage instantaneous change. Its long running proves its validity and reliability. In the end, a typical load has been analyzed in this paper.
Keywords/Search Tags:Power System Quality, Spatial Spectrum, Support Vector Regression Machine, canonical correlation analysis, Teager Energy Operator, VC++
PDF Full Text Request
Related items