Font Size: a A A

Research On Detection And Localization Of Harmonic Sources In Power System

Posted on:2011-04-24Degree:MasterType:Thesis
Country:ChinaCandidate:X Y ZhangFull Text:PDF
GTID:2132360305450801Subject:Electrical theory and new technology
Abstract/Summary:PDF Full Text Request
With the social progress and development, power quality problem has increasingly attracted considerable attention from both utility companies and their customers and has become an important subject in power system field. Due to more and more appearances of impactive and nonlinear loads in power system, a variety of disturbances is produced in public network, resulting in the remarkably poor trendency of power quality. Consequently, the utility must not only meet the customer's requirements of increasing power supply but also take into account the rectification of power quality. In order to assure the power supply of excellent quality and for ensuring the normal dealings between utility companies and customers in rated limits, it is indispensable to put a penalty or a premium for the effect of a customer, so as to embody the corporate responsibility for power quality. Therefore, for the separation of the duty between utility companies and customers it is the sticking point to find methods, by which the customer emission level can be accessed more accurately.In this paper, the reported localization and detection methods of harmonic sources in power system are described in detail, and their principles as well as their advantages and disadvantages are analyzed and outlined systemically. An approach of calculating consumer's harmonic impedance and evaluating the harmonic sources has been presented. In this method, by using the Thevenin equivalent circuit and the Norton equivalent circuit for the power system side and for the customer, respectively, the regression equations are obtained from plural relationship of the system parameters. In addition, utilizing the signals of harmonic voltage and harmonic current measured synchronously at PCC point, the regression coefficients are worked out by menas of the improved binary linear algorithm, and then using these coefficients the harmonic impedances of the system and customer are given and the harmonic equivalent voltage sources are calculated. According to the method proposed in the present work, the corresponding simulation model is constructed and the simulations are carried out, indicating that the proposed method is feasible for monitoring the harmonics in power system.Finally, the application of partial least-squares regression algorithm for evaluating the harmonic impedance of the system side is extended from single phase power system to unbalanced three-phase power system. A versatile matrix model for assessing the harmonic impedance of the system side of unbalanced three-phase power system is derived from the analysis on the three-phase power system, and with the use of simultaneous measurement of recorded harmonic voltages and currents at the point connecting the system and the load and by employing both GPS(Global Position System) and PMU(Phase Measurement Unit) techniques the value of each element of harmonic impedance matrix is assessed by the use of partial least-squares regression algorithm. Further, by designing the simulation circuit corresponding to this method and building the simulation model by means of MATLAB, we have performed the calculations of both the harmonic impedance matrix elements and the equivalent harmonic voltage source for the system side. These calculations show that the proposed model is accurate and effective. The present studies are the basic work for assessing the harmonic level for nonlinear loads which require low precision in three-phase unbalanced system, and it can suitable for the actual situations of three-phase unbalanced system and larger impedance fluctuations in power network.
Keywords/Search Tags:Power quality, Harmonic impedance, Harmonic emission level, Binary linear regression, Three-phase unbalanced system
PDF Full Text Request
Related items