In the current era, the challenge of improving energy efficiency in large-scale structures, including both residential and industrial buildings, has gained considerable prominence. A crucial aspect of this challenge lies in the early detection of anomalous operational events. Such early identification is vital for ensuring optimal performance of these structures, minimizing the occurrence of incidents, and reducing the need for maintenance and repairs, thereby contributing to cost savings and operational efficiency. To address this need, machine learning algorithms are being increasingly employed as a powerful tool for identifying abnormal load operation activities. This chapter presents a detailed analysis of abnormal activities at a water supply plant that serves the entire city of Da Nang. The study employs widely used machine learning algorithms, namely the Local Outlier Factor and Isolation Forest. These algorithms are particularly effective in identifying outliers or anomalies in large and complex datasets. The findings of this study underscore the significant potential of machine learning in identifying anomalous load operation activities.