Comprehensive evaluation of power quality based on improved TOPSIS-RSR method

Abstract

To solve the complexity of the existing power quality evaluation model as well as the randomness and fuzziness in the grading evaluation of power quality, a comprehensive evaluation method of power quality based on the improved TOPSIS⁃RSR method is proposed. Entropy method is used to calculate the weight of the power quality index, and the TOPSIS (technique for order preference by similarity to an ideal solution) method is used for comprehensive evaluation of power quality. The concept of relative similarity distance is raised to deal with the problem existing in traditional TOPSIS method, that is, in some cases the evaluated objects cannot be compared with each other effectively. Combination of the relative similarity distance with RSR (rank sum ratio) is adopted to further analyze the evaluation results for expansion. The evaluated objects are graded according to the normal equivalent deviation. Example verification of the power quality data from five observation points in a 380 V observation station is carried out, and the results are compared with those of other methods. It is shown that the improved TOPSIS method combined with RSR method can make reasonable evaluation and grading for power quality of evaluated objects with high accuracy and expansibility.