Abstract—Temperature prediction depends on various parameters to predict the dependent variables which are changing from time to time with its atmospheric variables. Predictions of Greenhouse temperature is a vital role in Greenhouse cultivation. The proposed system is collecting the climate data inside and outside of the Greenhouse by Internet of Things (IoTs) weather stations. It utilizes these historical data and provides temperature forecasts for the Greenhouse of next a few hours to a few days. In this work, the architecture of predictive weather station platform based on cloud and IoTs technology is proposed and multivariate Convolutional Long Short Term Memory Network model is used for time series forecasting. The system implementation is carried out in Vegetable and Fruit Research and Development Center (VFRDC), Ye Mon, Bago Division, Myanmar. All environmental time series weather data of the Greenhouse observed by the installed sensors and some weather information from Weather API are sent to and stored in the cloud server, ThingSpeak. The runtime for learning is provided by a cloud service, Google Colaboratory (Colab). In this work, Beijing PM2.5 dataset is analyzed by the proposed system. According to the Mean Average Error (MAE) value of the prediction result, the proposed system can predict the temperature inside the Greenhouse in advance and it can effectively promote the agricultural productivity by providing the farmers with more accurate predicted temperature, thereby minimizing the risk of temperature damage.
Index Terms—Deep learning, IoTs, LSTM, time series prediction.
The authors are with University of Computer Studies, Yangon, Myanmar (zarzaroo@ucsy.edu.mm).
Cite: Zar Zar Oo and Sabai Phyu, "Cloud and IoT Based Temperature Prediction System for a Greenhouse Using Multivariate Convolutional Long Short Term Memory Network," International Journal of Machine Learning and Computing vol. 10, no. 1, pp. 189-194, 2020.
Copyright © 2020 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).