Funder
US Department of Energy through Lawrence Berkeley National Laboratory
Project Objectives
To curate, validate, and publish the world’s largest set of labeled time-series data representing commercial HVAC systems operating in faulted and fault-free states.
My Contributions
- Contributed HVACSIM+ simulation data to the LBNL Fault Detection and Diagnostics Datasets.
- Generated annual ground-truth data on the presence and absence of various building faults in typical HVAC systems using HVACSIM+, including a fan-coil unit, a dual-duct system, a parallel fan-powered VAV system, and a series fan-powered VAV system.
- Enhanced the HVACSIM+ models to ensure efficient, robust, stable, and accurate system simulation through the modification of control logic, tuning of PID parameters, resizing of system equipment, and elimination of numerical difficulties.
Related Publications
Untitled