With the acceleration of the digital transformation in the power industry, the volume of power marketing data has significantly increased. While it offers valuable analysis resources, it also presents challenges in terms of data quality. To enhance the quality of power marketing data, this study explores intelligent methods for optimizing data quality, utilizing the K-means clustering algorithm and the subjective-objective combination assignment method based on the G1 method-entropy weight method. It establishes an evaluation system that encompasses various indicators such as accuracy, completeness, consistency, and timeliness, and puts forward a comprehensive model for assessing data quality. This model aids in reducing the manual effort required for analyzing data quality anomalies, enhancing the intelligence and automation of data quality analysis and processing, thereby improving the efficiency and reliability of decision-making in power marketing.