Basic Methods for Automated Fault Detection and Energy Data Validation in Existing District Heating Systems

Fault detection and diagnostics (FDD) of district heating substations (DHS) are important activities because malfunctioning components can lead to incorrect billing and waste of energy. Although FDD has been an activate research area for nearly two decades, only a few simple tools are commonly deployed in the district energy industry. Some of the methods proposed in the literature are promising, but their complexity may prevent broader application. Other methods require sensor data that are not commonly available, or cannot be expected to function well in practice due to oversimplification. Here we present two basic methods for improved FDD and data validation that are compatible with the data acquisition systems that are commonly used today. We propose that correlation analysis can be used to identify substations with similar supply temperatures and that the corresponding temperature difference is a useful quantity for FDD. The second method is a limitchecking approach for the validation of thermal power usage, which is sensitive to faults affecting both the primary flow and temperature sensors in a DHS. These methods are suitable for automated FDD and are demonstrated with hourly data provided by a Swedish district energy company.