Enhancing Data Quality via Data Correction Techniques to Assure High-quality Data for Energy Services Across Europe

Abstract: Reliance on digital technologies has revolutionized building management, but data quality remains a key challenge. This study presents a two-step approach to enhance data quality: calculating metrics like accuracy, completeness, and consistency, and applying machine-learning models to correct gaps and inconsistencies. Tested with pilot data across Europe, completeness improved from 55% to 100%, while accuracy and consistency reached 100% and 72.14%. Using a centralized data lake, the system ensures real-time synchronization for digital twins and services. This approach aligns with the Energy Performance of Buildings Directive (EPBD), advancing energy efficiency and providing scalable solutions for smart building data management.