An Intelligent Lightning Warning System Based on Electromagnetic Field and Neural Network

Abstract

Prediction of lightning occurrence has significant relevance for reducing potential damage to electric installations, buildings, and humans. However, the existing lightning warning system (LWS) operates using the threshold method and has low prediction accuracy. In this paper, an intelligent LWS based on an electromagnetic field and the artificial neural network was developed for improving lightning prediction accuracy. An electric field mill sensor and a pair of loop antennas were designed to detect the real-time electric field and the magnetic field induced by lightning, respectively. The change rate of electric field, temperature, and humidity acquired 2 min before lightning strikes, were used for developing the neural network using the back propagation algorithm. After observing and predicting lightning strikes over six months, it was verified that the proposed LWS had a prediction accuracy of 93.9%.

Analog Sensor Interface for Field Mill Sensors in Atmospheric Applications

“…The typical sensor interface of an electric field mill consists of a preamplifier which converts the induced charge or current into a voltage and provides amplification of the weak current signal output of the electrodes. A charge [ 7 , 10 , 25 , 32 , 33 ] or a transimpedance [ 3 , 9 ] amplifier is used in most of the works; it should introduce minimum noise and have low input bias currents since the induced current is typically comparable in value to the leakage currents, and the field measurement would be significantly affected. An additional voltage gain stage is incorporated in several interfaces [ 7 , 28 ].…”

“…In some works, the RC network of the charge amplifier is selected so that the output voltage signal is independent of the frequency of the sensor rotation [ 10 , 33 ], which is a desirable feature since the motor rotation speed might vary slightly during operation.…”

“…In atmospheric electricity, most measured surface quantity is the vertical electric field, also known as the potential gradient [ 8 ]. The local electric field has been measured to provide a better understanding of thundercloud electrification and the energy exchange between lightning and the atmosphere across the globe [ 9 ]; to forecast lightning incidents; and to define an electric field threshold for lightning strikes on aircraft [ 10 ]. In [ 11 ], an investigation of electrical activity in thunderstorms was conducted, whereby the electrostatic field was measured during flights inside electrified clouds, in which lightning strikes occur, to provide insight beyond that of surface measurements.…”

Analysis of a High-Voltage Room Quasi-Smoke Gas Explosion

“…More research has been done regarding overhead power line conductors clashing as the ignition source of wildland fires [11][12][13]. Other accident mechanisms have gained more attention, for example, arcs as a cause of fire [14] or lightning strikes as a risk to personnel and electrical equipment [15]. It has, however, not been possible to find research in relation to overheating, resulting in subsequent quasi-smoke gas explosions, even though electrical equipment certainly has the potential to produce pyrolysis products that may subsequently ignite, unless the power is quickly disconnected.…”

Computational Fluid Dynamics Based Fuzzy Control Optimization of Heat Exchanger via Genetic Algorithm

“…It has been utilized for medical use such as arrhythmia problems [1,2], anesthesia [3,4] and blood pressure estimation [5]. Moreover, the AI also has been applied to energy systems [6][7][8], electromagnetic field [9] and shape optimization to increase the aerodynamics of the unmanned aerial vehicle [10].…”