Date of Award

December 2015

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Mechanical and Aerospace Engineering

Advisor(s)

Jianshun Zhang

Subject Categories

Engineering

Abstract

This dissertation aims at developing a novel and systematic approach to apply Model Predictive Control (MPC) to improve energy efficiency and indoor environmental quality in office buildings. Model predictive control is one of the advanced optimal control approaches that use models to predict the behavior of the process beyond the current time to optimize the system operation at the present time. In building system, MPC helps to exploit buildings’ thermal storage capacity and to use the information on future disturbances like weather and internal heat gains to estimate optimal control inputs ahead of time.

In this research the major challenges of applying MPC to building systems are addressed. A systematic framework has been developed for ease of implementation. New methods are proposed to develop simple and yet reasonably accurate models that can minimize the MPC development effort as well as computational time.

The developed MPC is used to control a detailed building model represented by whole building performance simulation tool, EnergyPlus. A co-simulation strategy is used to communicate the MPC control developed in Matlab platform with the case building model in EnergyPlus. The co-simulation tool used (MLE+) also has the ability to talk to actual building management systems that support the BACnet communication protocol which makes it easy to implement the developed MPC control in actual buildings.

A building that features an integrated lighting and window control and HVAC system with a dedicated outdoor air system and ceiling radiant panels was used as a case building. Though this study is specifically focused on the case building, the framework developed can be applied to any building type.

The performance of the developed MPC was compared against a baseline control strategy using Proportional Integral and Derivative (PID) control. Various conventional and advanced thermal comfort as well as ventilation strategies were considered for the comparison. These include thermal comfort control based on ASHRAE comfort zone (based on temperature and relative humidity) and Predicted Mean Vote (PMV) and ventilation control based on ASHRAE 62.1 and Demand Control Ventilation (DCV). The building energy consumption was also evaluated with and without integrated lighting and window blind control. The simulation results revealed better performance of MPC both in terms of energy savings as well as maintaining acceptable indoor environmental quality. Energy saving as high as 48% was possible using MPC with integrated lighting and window blind control.

A new critical contaminant - based demand control ventilation strategy was also developed to ensure acceptable or higher indoor air quality. Common indoor and outdoor contaminants were considered in the study and the method resulted in superior performance especially for buildings with strong indoor or outdoor contaminant sources compared to conventional CO2 - based demand control ventilation which only monitors CO2 to vary the minimum outdoor air ventilation rate.

Access

Open Access

Included in

Engineering Commons

Share

COinS