PROJECT TITLE :
Lessons Learned from Optimizing Science Kernels for Intel's "Knights Corner"' Architecture
Optimizing HVAC operation becomes increasingly vital as a result of of the rising energy cost and comfort necessities. During this paper, an innovative event-primarily based approach is developed at intervals the Lagrangian relaxation framework to reduce an HVAC's day-ahead energy price. To unravel the HVAC optimization downside based on events is challenging since with time-dependent uncertainties in weather, cooling load, etc., the optimal policy isn't stationary. The nonstationary policy house is extremely massive, and it's time consuming to find the optimal policy. To overcome the challenge, we tend to develop an incident-based approach to create the nonstationary optimal policy stationary in the look horizon. The key plan is to reinforce state variables to include the time-dependent variables that make the optimal policy nonstationary and then outline events based on the extended state variables. In addition, we tend to develop at intervals the Lagrangian relaxation framework a Q-learning method where Q-factors are used to guage event-action pairs and to obtain the optimal policy. Numerical results demonstrate that, as compared with time-primarily based approaches, the event-primarily based approach maintains similar levels of energy prices and human comfort, but reduces computational efforts significantly and has a a lot of faster response to events.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here