A Study on Recovering Lost Data of Office Building Electric Power Consumption using Outdoor Air Enthalpy Variation Patterns

The Building Energy Management System (BEMS) continuously monitors and analyzes real-time energy consumption data such as electricity and gas to optimize efficiency and propose energy-saving strategies. However, communication failures, device malfunctions, and system updates can lead to missing or abnormal data, necessitating effective data recovery methods. Addressing these data gaps is crucial for maintaining reliability of energy analysis. Traditionally, simple linear interpolation has been used for data recovery. However, this method is prone to substantial inaccuracies. In addition, it is insufficient for complexities of energy systems. The power consumption of a building is heavily influenced by heating and cooling loads, which are closely related to outdoor conditions. To restore missing power consumption data, we proposed a method that could identify a day with similar changes in outdoor enthalpy variation - a function of temperature and humidity - as the day with missing data. Power consumption pattern from this similar day was then used to reconstruct lost data. The efficacy of the proposed method was evaluated using CV-RMSE (Coefficient of variation of root mean square error), demonstrating a high degree of accuracy, with recovered data showing a deviation of just 3.1% from actual values. This enhanced methodology not only can improve data recovery accuracy, but also can reinforce robustness of energy consumption analysis, thereby supporting more reliable energy management decisions.