Grant and Research Library

A Scalable Multi-agent Systems approach to Building HVAC Systems Control for Energy Efficiency


Program Area:



Energy Efficiency

  • Industrial


The proposed effort will develop scalable, data-driven multi-agent systems techniques for integrated building systems. The study aims to demonstrate initial proof of concept on an AHU-VAV system using data from partially simulated and partially real building systems. While validation on real building systems is crucial to avoid making unrealistic assumptions, it needs to be augmented with simulated systems to evaluate scalability. In this context, for this phase of the project we plan to utilize the existing data collected from the Iowa Energy Center’s Energy Resource Station (ERS) to build the model that is more realistic and use this model to demonstrate the proposed robust optimal control strategies in simulation. If proven successful in simulation, the next phase of the project will involve testing the control strategies using either ERS facility or a suitable building on Iowa State campus.

Award Amount:


Grant Active Dates:

Jan. 2015 — Dec. 2015


Department of Mechanical Engineering, Iowa State University

Principal Investigator:

Atul Kelkar

Co-Principal Investigator:

Soumik Sarkar




Ames, Iowa