Home // CYBER 2019, The Fourth International Conference on Cyber-Technologies and Cyber-Systems // View article
Authors:
Steve Chan
Ika Oktavianti Najib
Verlly Puspita
Keywords: lateral sensor; electrical grid; weather monitoring system; 3D-printed technology; generative adversarial network.
Abstract:
Resilience enhancements of electrical grid against damage to natural forces have become a major concern for engineers and researchers in the past few years. Natural disasters (e.g., lightning strikes, geometric storms, floods, etc.,) can cause devastating damages to power grid infrastructures. Many natural forces can be forecast or at least anticipated for secure the entire power system. However, quantifying and anticipating the impact of weather is difficult task due to its high stochasticity. In this study, a new concept of weather monitoring system based on the lateral sensors is proposed. Lateral sensors for the electrical grid, by way of hyper-locale weather sensors, provide incredible insight via parameter, such as air temperature, barometric pressure, humidity precipitation, solar radiation, and wind. These lateral sensor parameters can provide indicators as to impending storms, which could cause communications interference impact power lines, via downed trees and cause damage, via lightning strikes. Spider Radar Plots reflecting both weather sensor data and grid sensor data concurrently have proven useful, as weather data can serve to provide context reference for the grid sensor telemetry data. Moreover, this weather monitoring utilizes 3D-printed technology for the manufacturing of sensors with deep learning module, which is based upon a Generative Adversarial Network (GAN). The results of this study showed that implementation of lateral sensors based upon deep learning module is more robust than previous weather monitoring system. It can identify patterns within the data and 3D-printed sensors can detect more indicators that can cause power surge.
Pages: 75 to 80
Copyright: Copyright (c) IARIA, 2019
Publication date: September 22, 2019
Published in: conference
ISSN: 2519-8599
ISBN: 978-1-61208-743-6
Location: Porto, Portugal
Dates: from September 22, 2019 to September 26, 2019