Data Quality Evaluation and Pricing of Edge Computing Device in Active Distribution Networks With Privacy Preserving

Abstract

User-side electricity consumption data form the basis for various AI-based applications in active distribution networks and are with significant commercial value. However, as user-owned digital assets, these diverse data often exhibit heterogeneous quality and raise privacy concerns, making their acquisition and trading challenging. To address these issues, this paper proposes a privacy-preserving framework for data quality evaluation and pricing of edge computing devices in distribution networks. A multidimensional quality assessment model based on homomorphic encryption enables edge devices to evaluate data quality without direct access to raw user data, thereby guaranteeing users’ privacy. Then, a multi-party data pricing mechanism is formulated by integrating Stackelberg–Bargaining and cooperative game models to accommodate diverse stakeholder interests. Case studies on a public electricity consumption dataset demonstrate that the proposed method achieves accurate data quality evaluation with privacy preservation, while generating reasonable and adaptive pricing outcomes. The proposed approach effectively incentivizes high-quality data contributions from user side to support the intelligent applications during the digital transformation process.