Artificial Neural Network Models to Estimate Growth of Melon (Cucumis melo L.) at Vegetative Phase in Greenhouse with Evaporative Cooling

Abstract

Currently, melon is a high economic value product in Indonesia. To optimize the production of melon, cultivation in greenhouses equipped with evaporative cooling system is implemented in some locations. However, growth of melon in greenhouse with evaporative cooling is characterized by complexity in its relation with environmental parameters inside the greenhouse. Therefore, it is important to establish stochastic model to explain the relation between the growth of melon with environmental parameters inside the greenhouse. This paper presents an Artificial Neural Network (ANN) model to predict the growth of melon during vegetative stage of melon in a greenhouse with evaporative cooling in Bogor, Indonesia. Data was collected from October to December 2020 during vegetative stage of melon. The ANN model was developed using six input parameters that are air temperature (°C), air relative humidity (%), radiometric intensity of sunlight (watt/m2), plant age (day), leaf area (cm2), plant height (cm) to predict leaf area (cm2) and plant height (cm) for the next two days. The results showed that the designed seven hidden nodes-ANN model achieved a high prediction accuracy with R2 of 0.9 for leaf area validation and plant height validation.