Comparison of Machine Learning Regression Models for Prediction of Soil Moisture with the use of Internet of Things Irrigation System Data

Öz

Nesnelerin İnterneti (IoT) teknolojisi, sistemlerin insanlardan bağımsız olarak kontrol edilmesine ve yönetilmesine olanak tanır. Nesnelerin interneti tabanlı tarım uygulamaları, dünya nüfusunun giderek artmasıyla tarımda gıda tüketimi ve su kıtlığı sorunlarına çözüm olarak yaygınlaşmıştır. Toprak nemi, tarımsal üretim ve hidrolojik döngüler için önemli bir faktördür ve tarımsal uygulamaların geliştirilmesinde toprak neminin tahmin edilmesi gerekmektedir. Bu çalışmada Arduino Uno kartına bağlı Esp8266 Wifi modülü, nem ve sıcaklık, toprak nemi, yağmur ve ultraviyole sensörlerinden oluşan IoT tabanlı bir sulama sistemi prototipi sunulmuştur. Daha sonra prototip sistemi kullanılarak 55 gün boyunca yarım saatlik periyotlarla belirlenen pilot alandan veriler toplanır ve ThingSpeak ile bulut üzerinden kaydedilir. Toplanan veriler kullanılarak çoklu doğrusal, polinomal , destek vektörü, karar ağacı ve rastgele orman regresyonu gibi farklı makine öğrenimi regresyon modelleri uygulanarak toprak nem değeri tahmin edilir. Elde edilen sonuçlar, bu algoritmaların başarısını incelemek için belirlilik katsayısı ve ortalama kare hatası kriterlerine göre karşılaştırılır. Rastgele orman regresyon modeli toprak nem tahmini için diğer makine öğrenmesi algoritmalarından daha üstün bulunmuştur

Anahtar Kelimeler

• Toprak Nem Tahmini
• Makine Öğrenmesi Regresyon Modelleri
• Nesnelerin İnterneti
• ThingSpeak

Comparison of Machine Learning Regression Models for Prediction of Soil Moisture with the use of Internet of Things Irrigation System Data

Öz

Internet of Things (IoT) technology allows the control and management of systems independent of humans. Internet of things based agriculture applications have become widespread as a solution to the problems of food consumption and water scarcity in agriculture as the World population has increased gradually. Soil moisture is an important factor for agriculture production and hydrological cycles and the prediction of soil moisture is required in developing agricultural practices. In this study, an IoT-based irrigation system prototype is presented which consist of Esp8266 Wifi module, humidity and temperature, soil moisture, rain and ultraviolet sensors connected to the Arduino Uno board. Then, using the prototype system, data are collected from the pilot area determined in half-hour periods for 55 days and saved over the cloud with ThingSpeak. The soil moisture value is estimated by applying different machine learning regression models such as multiple linear, polynomial, support vector, decision tree and random forest regression using the collected data. The results obtained are compared according to the coefficient of determination and mean square error criteria to examine the success of these algorithms. The random forest regression model has found to be superior to other machine learning algorithms for soil moisture estimation.

Anahtar Kelimeler

• Soil Moisture Prediction
• Machine Learning Regression Models
• Internet of Things
• ThingSpeak

Kaynakça

1. [1] Mahdavinejad, M. S., Rezvan, M., Barekatain, M., Adibi, P., Barnaghi, P., Sheth, A. P. 2018. Machine learning for internet of things data analysis: a survey. Digital Communications and Networks, 4(3), 161–175.
2. [2] Hassija, V., Chamola, V., Saxena, V., Jain, D., Goyal, P., Sikdar, B. 2019. A Survey on IoT Security: Application Areas, Security Threats, and Solution Architectures. IEEE Access, 7, 82721–82743.
3. [3] Pernapati, K. 2018. IoT Based Low Cost Smart Irrigation System. Proceedings of the International Conference on Inventive Communication and Computational Technologies, April, India, 1312–1315.
4. [4] Thakare, S., Bhagat, P. H. 2019. Arduino-Based Smart Irrigation Using Sensors and ESP8266 WiFi Module. Proceedings of the 2nd International Conference on Intelligent Computing and Control Systems, June, India, 1085–1089.
5. [5] Balamurugan, C., Satheesh, R. 2017. Development of Raspberry pi and IoT Based Monitoring and Controlling Devices for Agriculture. Journal of Social, Technological and Environmental Science, 6, 207–215.
6. [6] Capello, F., Toja, M., Trapani, N. 2016. A real-Time monitoring service based on industrial internet of things to manage agrifood logistics. ILS 2016 - 6th International Conference on Information Systems, Logistics and Supply Chain, 1–8 June, France, 1-8.
7. [7] Chen, K. T., Zhang, H. H., Wu, T. T., Hu, J., Zhai, C. Y., Wang, D. 2014. Design of monitoring system for multilayer soil temperature and moisture based on WSN. Proceedings - 2014 International Conference on Wireless Communication and Sensor Network, November, India, 425–430.
8. [8] Minbo, L., Zhu, Z., Guangyu, C. 2013. Information Service System of Agriculture IoT. Automatika, 54(4), 415–426.

9. [9] Payero, J. O., Mirzakhani-Nafchi, A., Khalilian, A., Qiao, X., Davis, R. 2017. Development of a Low-Cost Internet-of-Things (IoT) System for Monitoring Soil Water Potential Using Watermark 200SS Sensors. Advances in Internet of Things, 07(03), 71–86.
10. [10] Liu, J. 2016. Design and Implementation of an Intelligent Environmental-Control System: Perception, Network, and Application with Fused Data Collected from Multiple Sensors in a Greenhouse at Jiangsu, China. International Journal of Distributed Sensor Networks, 12(7), 1-10.
11. [11] Yang, J., Liu, M., Lu, J., Miao, Y., Hossain, M. A., Alhamid, M. F. 2018. Botanical Internet of Things: Toward Smart Indoor Farming by Connecting People, Plant, Data and Clouds. Mobile Networks and Applications, 23(2), 188–202.
12. [12] Shekhar, Y., Dagur, E., Mishra, S., Tom, R. J., Veeramanikandan, M., Sankaranarayanan, S. 2017. Intelligent IoT Based Automated Irrigation System. International Journal of Applied Engineering Research, 12(18), 7306–7320.
13. [13] Gorthi, S., Dou, H. 2011. Prediction Models for the Estimation of Soil Moisture Content. Proceedings of the ASME 2011 International Design Engineering Technical Conferences & Computers and Information in Engineering Conference, 1–9 August, US, 1-9.
14. [14] Prakash, S., Sharma, A., Sahu, S. S. 2018. Soil Moisture Prediction Using Machine Learning. Proceedings of the International Conference on Inventive Communication and Computational Technologies, 20-21 April, India, 1–6.
15. [15] Goap, A., Sharma, D., Shukla, A. K., Krishna, C. R. 2018. Comparative Study of Regression Models Towards Performance Estimation in Soil Moisture Prediction. International Conference on Advances in Computing and Data Sciences, April, India, 309-316.
16. [16] Shalev-Shwartz, S., Ben-David, S. 2013. Understanding Machine Learning: From Theory to Algorithms, Cambridge University Press.
17. [17] Vapnik, V., Lerner, A. 1963. Generalized Portrait Method for Pattern Recognition. Automation and Remote Control, 24(6), 774–780.
18. [18] Leo Breiman, Jerome H. Friedman, Richard A. Olshen, C. J. S. 1984. Classification and Regression Trees. In Chapman and Hall/CRC.
19. [19] Breiman, L. 1996. Bagging Predictors. Machine Learning, 24(0), 123–140.
20. [20] Gómez Maureira, M. A., Oldenhof, D., & Teernstra, L. 2014. ThingSpeak – an API and Web Service for the Internet of Things. https://staas.home.xs4all.nl/t/swtr/documents/wt2014_thingspeak.pdf , Accessed 20 April 2021.