Comprehensive Evaluation Of Power Quality Based On The Integraiton Of Rough Set And Evidence Theory

At present, with the rapid development of power system and power market, power quality has been made higher demands. Under this situation, an objective comprehensive evaluation method of power quality should been built to assess the actual situation of power quality. The index of power quality evaluation should been build as the standard of supplier and consumer concerned.Based on in-depth study of Rough Sets and Evidence theory, this paper integrates Rough Sets and Evidence theory and applies it to power quality comprehensive evaluation.Firstly, based on the research of the phenomenon and characteristic of modern power system, comprehensive evaluation index system of power quality is build. The system is made of harmonic, voltage unbalance, voltage deviation, frequency deviation, voltage fluctuation and flicker. The level of comprehensive evaluation is divided with limit scope defined.Secondly, advantages and disadvantages of Rough Set and Evidence theory are analyzed. For problems appeared in the application, a method of obtain basic probability assignment is proposed based on the decision-making table.Finally, a method of power quality comprehensive evaluation based on the integration of Rough Set and Evidence theory is proposed. The method makes evaluation index of power quality as factor attribute set, index weight as the decision attribute set. The decision-making table of power quality evaluation is built based on the factor attribute set and decision attribute set. Basic probability assignments of factor attribute set are gained based on the calculation of intensions of decision-making rule and extend decision-making rule. Then, comprehensive basic probability assignments of power quality are gained after indexes are fused based on the composite formula of evidence theory. The condition of power quality can be defined based on the comprehensive assignment grade. The integration of two methods has more advantages. The integration of two methods not only can handle massive dates, but also can reduce the calculation. Reliable analysis can be provided by this method. The proposed algorithm is applied to comprehensive evaluation of power quality. Simulation results confirm its effectiveness and reliability.