Abstract: Objective and accurate comprehensive evaluation on power quality is an important evidence for power pricing and power quality assessment. Considering the shortcomings existing in current methods, state variable weight and normal cloud model are used to improve the fuzzy comprehensive evaluation method of power quality. The state variable weight vector is applied to consider the influence of the actual condition of evaluation indicators on the indicator weights, and the normal cloud model is applied to construct the membership function. The asymmetric proximity criterion is used to obtain the quantitative and qualitative assessment results. Finally, the accuracy and effectiveness of the proposed method are verified by a practical example.