Please use this identifier to cite or link to this item: http://hdl.handle.net/123456789/24875
Title: Investigation of flash floods on early basis: a factual comprehensive review
Authors: Khan, Talha Ahmed
Alam, Muhammad
Shahid, Zeeshan
M.S, Mazliham
UniKL BMI
Keywords: Flash floods forecasting
false alarm
Artificial intelligence
Sensors, Radar
Satellite
System Modeling
Issue Date: Aug-2020
Publisher: Institute of Electrical and Electronics Engineers Inc.
Citation: T. A. Khan, M. M. Alam, Z. Shahid and M. M. Su’ud, "Investigation of Flash Floods on Early Basis: A Factual Comprehensive Review," in IEEE Access, vol. 8, pp. 19364-19380, 2020, doi: 10.1109/ACCESS.2020.2967496.
Abstract: Ultimate extreme flash floods can be acknowledged as a main reason of high casualties and infrastructure loss in many countries like Pakistan, Malaysia, Philippines, Southern France, India, Bangladesh, China, Nepal, Canada, United States of America and others. Run offs can devastate huge buildings and personal belongings within fraction of seconds. Flash floods usually occurs due to many reasons like higher precipitation velocity, melting of ice debris in ocean, high wave current at sea shore, broken reservoir (dam), Cloud to ground flashes, thunderstorm and hurricane inside the ocean. More than one hundred and twenty thousand casualties resulted due to the flash floods during the 1992 and 2005. According to the literature review deadliest flash floods have been observed in past history. Many approaches have been completed to investigate the flash floods accurately and precisely with less false alarm rate. Disaster management authorities are unable to forecast the natural disasters accurately and precisely like tsunami, flash floods, hurricanes and seismic events due to the poor efficiency of the sensors and transmission of missed information. It has also been observed that during the wireless data transmission of sensors to the controller unit some bits of the data are missed, due to these phenomena data is not transmitted properly or indicate the wrong observations. Several diversified approaches have been made to identify the run offs more accurately and precisely. Generally, the approaches can be classified into two categories a) Engineering Based b) Non-Engineering Based. Engineering techniques based on the construction of the dams and reservoirs to store the excess water which causes severe run offs. Designing of various Artificial Intelligence Based competent algorithms to predict the flash floods vigorously can be considered as non-engineering based approaches. Authors have tried their best to summarize and portray all the successful techniques that can be used for the early prediction of flash floods. Scientists can be benefited by this research paper as this research paper is the detailed capsulization of all the approaches that has been carried out for the robust investigation of flash floods. Extensive literature review has been done to observe the comparative analysis for the investigation of flash floods identification accurately. Literature review has been categorized into following types; 1. Sensory Fusion based 2. Artificial Intelligence Based methods 3. Radar and Satellite based approaches 4. Modeling and Nowcasting. According to the exhaustive literature review it can be concluded that swarm intelligence weights optimization for multi-layer perceptron neural network configuration performed better among all the forecasting approaches and recommended as the future enhancement. © 2013 IEEE.
Description: This article is indexed by Scopus
URI: http://hdl.handle.net/123456789/24875
ISSN: 21693536
Appears in Collections:Journal Articles

Files in This Item:
File Description SizeFormat 
6. Investigation_of_Flash_Floods_on_Early_Basis_A_Fac.pdf1.09 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.