Hierarchical model predictive control of greenhouse climate to reduce energy cost

Abstract

This paper proposes a hierarchical control strategy for greenhouse climate control. The proposed hierarchical control consists of two layers (an upper layer and a lower layer). The upper layer is to generate set points by solving an optimization problem. The objective is to minimize the energy cost under the time-of-use (TOU) tariff while keeping greenhouse climate (temperature, relative humidity and carbon dioxide concentration) within the required range. The lower layer is to track the trajectories obtained by the upper layer. A model predictive controller is designed to address system disturbances and the results are compared with that of an open loop controller. A performance index, relative average deviation (RAD), is introduced to compare the tracking performance of the open loop control and proposed closed-loop model predictive control. Simulation results show that the proposed strategy can reduce 7.86% energy cost compared with the strategy that aims to minimize energy consumption. Moreover, the proposed model predictive control can track reference trajectories better than open loop control under system disturbances.