Prediction of Environmental Conditions of the Greenhouse Using Neural Networks Optimized with the Grasshopper Optimization Algorithm (GOA)

Abstract

Automatic control of the greenhouse environmental conditions is among the most important necessities of modern agriculture. In order to better control and manage the conditions, some variables inside the greenhouse should be predicted such as temperature, humidity and CO2 values. The prediction of these variables in the greenhouse is usually depend on some external conditions as well as the past values of the same variables. In more recent years, several methods have been presented to predict the variables of the greenhouse based on artificial intelligence. In this paper, a prediction model is proposed for three variables of the greenhouse, these parameters include: temperature, humidity and CO2 values. The proposed model is based on machine learning and meta-heuristic algorithms. For this purpose, a feedforward neural network is used in the study. In this way, the weights and coefficients of the neural network are optimized by the grasshopper Optimization Algorithm (GOA) to reduce the prediction error. According to the simulation results of the proposed method, the RMSE value for temperature variable prediction is 0.15 and for humidity variable prediction is 1.06 and finally for CO2 variable prediction is 8.04. In order to evaluate the proposed model, the results have been compared with other state of the art methods. Based on the findings, the proposed method is superior to other methods in this field.