Predictive control of a building hybrid heating system for energy cost reduction

• Hybrid heating system is used to optimize energy consumption.
• In order to minimize cost of energy, control methods are used.
• A PID controller and a MPC controller are used to control hybrid system.
• These controllers are used to control the temperature of indoor volume.
• The GA has been applied to tune optimal gains of PID controller.

Today, the buildings’ energy consumption is considerable amount of whole. Therefore, optimizing energy in buildings leads to a noticeable decrease in total energy consumption of the world. Energy-efficient buildings have developed by carrying out great research effort. The control procedures serve as a privileged method to help new buildings to comply with the most optimal system as an energy consumer and thus meet ‘nearly zero-energy’.The purpose of this paper is to present a method of controlling the building temperature and simultaneously reducing the cost of providing the hybrid heating systems with sufficient energy. Investigating a room in Tehran city on a day as an example, methods of (a) Model Predictive Control (MPC) with economic optimization (MPC consecutively with On-Off), (b) MPC without economic optimization, (c) Proportional-Integral-Derivative (PID) controller optimized by Genetic Algorithm (GA) in presence of gas thermal source, (d) PID controller optimized with GA in presence of electric thermal source and (e) PID controller optimized with multi-objective GA in the presence of two gas and electric thermal sources have been designed and implemented in this research. Furthermore, the effect of each of these methods on cost reduction and temperature regulation of inside of the room has been studied. Eventually it has been specified that using MPC method with economical optimization has the highest influence on cost reduction and keeps the temperature of inside of the room in the predefined range. This method achieved cost saving of 50% compared to the MPC and GA. But the main targets of this study are both of regulating inside temperature and cost optimization. According to the main targets of this study, using MPC methods without economical optimization and multi-objective genetic algorithm would be more effective.

Figure optionsDownload as PowerPoint slide