Digital Twins Generating Method of Power Distribution Networks Based on Multidimensional Data Integration
Zhe Li, Shuai Gao, Shaokang Zhao, Jianli Zhao, Xiaomeng Di · 2024
This paper delves into the generation technology of digital twins for distribution networks based on the fusion of multidimensional business data. By leveraging data fusion algorithms, heterogeneous data are integrated to establish a comprehensive and accurate data model of the distribution network. This model encompasses not only real-time data but also historical and predictive data elements, enabling the digital twin to simulate and forecast across various timescales. The proposed technical solution includes the status perception and data monitoring of medium-voltage cable lines, medium-voltage AC collection and monitoring, low-voltage line status perception and data monitoring, and the fusion of data with model construction. Advanced data fusion algorithms consolidate data from diverse sources, allowing for the extraction of more holistic, enriched, and reliable information, leading to more objective and rational decision-making. The process involves the acquisition of raw measurement information from multi-source sensors, A/D conversion to digital signals, preprocessing steps including denoising, imputation of missing values, redundancy removal, and standardization.