Identifying Anomalies in Commercial Energy Consumption

Time series applications to energy consumption and outlier detection

Despite efforts to reduce waste, energy consumption in buildings has steadily increased in the last decade. Commercial buildings are specifically at risk of using more energy than necessary because of faulty construction, malfunctioning equipment and many other issues. The key problem is that fault detection in buildings for excess energy consumption is often primitive, leading to large amounts of waste.

Interested in learning more?? Check out the medium article we wrote about this project

Click below to check out our code:

• Pre-processing the data
• Visualizing the data
• Anomaly Detection:
• Approach 1: Unsupervised KNN
• Approach 2: Banpei Change Points
• [Approach 3: Residual Analysis]
  • Building the model
  • Analyzing Residuals