Artificial Intelligence for Solar Energy Harvesting in Wireless Sensor Networks

Ghannam, R. , Brante, G., Heidari, H. and Imran, M. (2018) Artificial Intelligence for Solar Energy Harvesting in Wireless Sensor Networks. 11th International Conference on Developments in e-Systems Engineering (DeSE 2018), Cambridge, UK, 2-5 Sep 2018.

Ghannam, R. , Brante, G., Heidari, H. and Imran, M. (2018) Artificial Intelligence for Solar Energy Harvesting in Wireless Sensor Networks. 11th International Conference on Developments in e-Systems Engineering (DeSE 2018), Cambridge, UK, 2-5 Sep 2018.

[img]
Preview
Text
168290.pdf - Accepted Version

423kB

Abstract

Solar cells have been extensively investigated for wireless sensor networks (WSN). In comparison to other energy harvesting techniques, solar cells are capable of harnessing the highest amount of power density. Furthermore, the energy conversion process does not involve any moving parts and does not require any intermediate energy conversion steps. Their main drawback is the inconsistent amount of energy harvested due to the intermittency and variability of the incoming solar radiation [1]. Consequently, being able to predict the amount of solar radiation is important for making necessary decisions regarding the amount of energy that can be utilized at the sensor node. We demonstrate that artificial intelligence (AI) can be used as an effective technique for predicting the amount of incoming solar radiation at these sensor nodes. We show that a Support Vector Machine (SVM) regression technique can effectively predict the amount of solar radiation for the next 24 hours based on weather data from previous days. We reveal that this technique outperforms other state of the art prediction methods for WSNs. To assess the performance of our proposed solution, we use experimental measurements that were collected for a period of two years from a weather station installed by Beijing Sunda Solar Energy Technology Company [2]. We also demonstrate how the harvested energy can be regulated using an innovative Power Management Unit [3].

Item Type:Conference or Workshop Item
Status:Published
Refereed:Yes
Glasgow Author(s) Enlighten ID:Imran, Professor Muhammad and Ghannam, Dr Rami and Heidari, Dr Hadi
Authors: Ghannam, R., Brante, G., Heidari, H., and Imran, M.
College/School:College of Science and Engineering > School of Engineering > Electronics and Nanoscale Engineering
College of Science and Engineering > School of Engineering > Systems Power and Energy
Related URLs:

University Staff: Request a correction | Enlighten Editors: Update this record