ENHANCEMENT OF PRODUCTION CAPACITY OF TOMATO YIELDS IN GREENHOUSE USING MODEL PREDICTIVE CONTROLLER

Abstract

The low production of tomatoes in a greenhouse has reduced the financial status of those concerned and the country at large. This can be achieved by introducing an enhanced production capacity of tomato yield in greenhouse using Model Predictive Controller (MPC). This research work was done by characterizing and modeling tomato yield in normal weather condition with respect to temperature, humidity, soil moisture and quantity of tomatoes. Training an Artificial Neural Network (ANN) in a default tomato temperature, relative humidity and soil moisture to attain standard range that will boost its yielding capacity was done. Hence, the trained ANN in a default tomato temperature, relative humidity and soil moisture was integrated into a conventional and Simulink models for model predictive controller to attain standard range that will boost yielding capacity of the tomatoes. The result obtained from the conventional model was the quantity of tomatoes produced in a greenhouse which was 35 tons. On the other hand, when Model Predictive Controller (MPC) was incorporated into the system, it boosts the quantity of tomatoes to 42.7 tons. With these results obtained, it shows that the percentage improvement in the production of tomatoes in a greenhouse when MPC was integrated in the system was 22%. It was thus concluded that Model Predictive Controller enhanced production capacity of tomato yields in the greenhouse