Development of a Smart Cooling System using Artificial Neural Network

Temperature inside a controlled environment is very crucial in growing a temperature sensitive crop. Temperature affects the overall harvest of crops. Thus, there is a need to provide an environment conducive for its growth and development. To solve these temperature fluctuations in a microclimatic condition, this paper used a feed forward neural network to trigger the cooling system that will provide the required temperature inside the bamboo type growing chamber while also considering the external temperature factor. This smart cooling system will automate the whole process of setting the internal temperature for the optimum crop growth. The total raw data temperature used in this study is 60, 70% of which is used in training, where validation and testing is 15% each. Results show that a feed forward neural network with 10 neurons in the hidden layer gives the best performance. Finally, the actuation process by the neural network to the cooling system is successfully executed.