Abstract:

In order to improve the energy efficiency and reduce the production cost of Venlo greenhouse, a Hierarchical model predictive control (HMPC) method including optimization layer and control layer was proposed. The optimization layer aimed to minimize the energy consumption and the total operating cost at the time of using price respectively, and construct two strategies to increase the energy utilization rate and reduce the greenhouse cost. Sensitivity analysis was carried out to study the impact of electricity price, CO₂ price and the range of environmental condition constraint on cost optimization. In the control layer, Model predictive control (MPC) was used to solve the model object mismatch and suppress the system disturbance. The relative average deviation and maximum relative deviation were used to compare and analyze the tracking performance of MPC and open−loop control under 2%, 6% and 12% perturbations, to obtain a more stable and reliable greenhouse control system. The results showed that the constraints on temperature and average relative humidity had a significant impact on reducing the operating cost of the greenhouse. The total energy consumption in the lowest energy consumption scenario was 79.92% of that in the lowest cost scenario. The total cost of the lowest cost scenario was 83.61% of the lowest energy consumption scenario. The MPC had a good tracking performance under different systematic perturbation. Greenhouse hierarchical model predictive control can effectively improve greenhouse control accuracy, increase greenhouse energy efficiency and reduce production costs. 

Key words: Greenhouse climate control, Time-of-use tariff, Hierarchical model predictive control, Energy efficiency, Operating cost