Zhanhong Jiang
Developed information-theoretic approaches for time-series inference in smart environments and studied effect of human in the loop on indoor thermal dynamics
Designed learning-based model predictive controller for thermal comfort optimization in smart buildings
Developing robust and distributed deep reinforcement learning approaches
Proposed and developed distributed deep learning algorithms (IID and non-IID) in multi-robot setting. Collaborated with 5 professors to demonstrate proof-of-concept, software, and hardware designs
Led a project sponsored by Iowa Energy Center: Used machine learning methods to establish thermal zone dynamics based on real test bed data; proposed and developed distributed control and optimization algorithms for building energy efficiency; implemented the algorithms in the real test bed via VOLTTRONTM platform
Proposed and developed data-driven method based on symbolic dynamic filtering - spatiotemporal pattern network for wind power prediction
Established optimal non-intrusive load monitoring framework using spatiotemporal pattern network and convex programming
Developed an unsupervised spatiotemporal graphical modeling approach to anomaly detection in building energy system