Hierarchical model predictive control of a venlo-type greenhouse

Show simple item record

dc.contributor.advisor Zhang, Lijun
dc.contributor.coadvisor Xia, Xiaohua
dc.contributor.postgraduate Lin, Dong
dc.date.accessioned 2023-03-09T09:36:55Z
dc.date.available 2023-03-09T09:36:55Z
dc.date.created 2022-04
dc.date.issued 2021
dc.description Thesis (PhD (Electrical Engineering))--University of Pretoria, 2021. en_US
dc.description.abstract Greenhouse cultivation can increase crop yield and alleviate the food shortage caused by population growth and reduction of arable land. However, the greenhouse production process consumes lots of energy and water. The energy consumed mainly comes from the combustion of fossil fuels, which will produce lots of greenhouse gases. In addition, the operating efficiency of some greenhouses is low, resulting in energy and water waste and increasing production costs. Therefore, the greenhouse system needs to be optimized to improve the operating efficiency. In this thesis, different methods of greenhouse operation efficiency optimization to improve energy efficiency and water efficiency are studied. In Chapter 3, three strategies for greenhouse operation optimization are studied. Strategy 1 focuses on the optimization of the greenhouse heating system to save energy. The optimization of the heating system can effectively reduce energy consumption. However, people often pay more attention to reducing energy costs than reducing energy consumption in the production process to obtain more profits. Strategy 2 is to reduce the energy cost. It should be noted that Strategy 2 only considers the cost of heating and cooling, while the cost of ventilation and carbon dioxide (CO2) is not considered. Strategy 3 reduces the cost of greenhouse heating, cooling, ventilation and CO2 consumption. In addition, greenhouse environmental factors must be kept within the required ranges. In Chapter 3, a dynamic greenhouse climate model is proposed. In the modeling process, the influence of crop growth and the interaction between different variables are considered to improve model accuracy. The proposed optimization problems are solved by ‘fmincon’ function with sequential quadratic programming (SQP) algorithm in MATLAB. Compared with Strategy 1, Strategy 2 has higher energy consumption but lower energy cost. Because Strategy 2 can shift some loads from high electricity price period to low electricity price period. Moreover, among the three strategies proposed, Strategy 3 has the lowest cost. It should be pointed out that the strategies studied in Chapter 3 only consider the impact of the greenhouse climate, but ignore the irrigation, which is also important for greenhouse production. In Chapter 4, four optimization methods are proposed. These optimization methods consider climate control and irrigation control. Therefore, strategies proposed in this chapter can not only improve energy efficiency, but also increase water efficiency. Method 1 reduces the energy consumption. Method 2 reduces the water consumption. Method 3 reduces the CO2 consumption. Method 4 reduces the total cost of greenhouse heating, cooling, ventilation, irrigation and CO2 supply. In addition, greenhouse environmental factors and crop water demand need to be met. The dynamic model of greenhouse environmental factors presented in Chapter 3 is used for greenhouse climate control. A modified crop evapotranspiration model is proposed to predict crop water demand. Moreover, a sensitivity analysis method is introduced. The influence of prices and system constraints on optimization results is studied. The cost of Method 4 can be reduced compared with other methods. In addition, changes of prices and system constraints have a great impact on optimization results. In Chapters 3 and 4, open loop optimization strategies for a greenhouse system operation are studied. However, these strategies have low control accuracy under system disturbances. Therefore, it is necessary to adopt some control methods to improve the control accuracy. In Chapter 5, a hierarchical model predictive control method is presented. The upper layer generates the optimal reference trajectories by solving greenhouse operation optimization problems. The lower layer designs controllers to follow obtained reference trajectories. Two model predictive controllers (MPC) are designed. Two performance indicators, namely relative average deviation (RAD) and maximum relative deviation (MRD), are used to compare designed controllers. The simulation results show that the proposed MPC can deal with greenhouse system disturbances and the problem of model plant mismatch better than the open loop control method. In Chapter 6, the findings of this thesis are summarized. Moreover, some topics for future research are proposed. en_US
dc.description.availability Unrestricted en_US
dc.description.degree PhD (Electrical Engineering) en_US
dc.description.department Electrical, Electronic and Computer Engineering en_US
dc.identifier.citation * en_US
dc.identifier.other A2022 en_US
dc.identifier.uri https://repository.up.ac.za/handle/2263/90046
dc.language.iso en en_US
dc.publisher University of Pretoria
dc.rights © 2021 University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria.
dc.subject UCTD en_US
dc.subject Greenhouse en_US
dc.subject Greenhouse climate en_US
dc.subject Energy efficiency en_US
dc.subject Sensitivity analysis en_US
dc.subject Model predictive control en_US
dc.title Hierarchical model predictive control of a venlo-type greenhouse en_US
dc.type Thesis en_US


Files in this item

This item appears in the following Collection(s)

Show simple item record