Hybrid Multilayer Perceptron Network for Explosion Blast Prediction

Muhamad Hadzren Mat, Muhamad Hadzren Mat and Prakash Nagappan, Prakash Nagappan and Fakroul Ridzuan Hashim, Fakroul Ridzuan Hashim and Khairol Amali Ahmad, Khairol Amali Ahmad and Mohd Sharil Saleh, Mohd Sharil Saleh and Khalid Isa, Khalid Isa and Khaleel Ahmad, Khaleel Ahmad (2023) Hybrid Multilayer Perceptron Network for Explosion Blast Prediction. Journal of Advanced Research in Applied Sciences and Engineering Technology, 30 (3). pp. 265-275. ISSN 2462-1943

[img] Text
J16654_d82c00fc7bca79478ca87a25b8913789.pdf
Restricted to Registered users only

Download (626kB) | Request a copy

Abstract

For decades, scientists have studied the blast wave profile produced by an explosive detonation. Based on a significant amount of experimental data, the blast wave propagation profile has been predicted under given parameters. However, most studies have only looked at the central point of initiation for spherical form explosives. The purpose of this research is to compare the prediction performance of blast peak overpressure based on type of explosive, shape of explosive and point of detonation. The blast profiles of Emulex and PE-4, as well as to develop a prediction model using a Hybrid Multilayer Perceptron (HMLP) network. This experiment, which began at a distance of 1.2 m from the ground, employed a total of 500 grams of military explosive and Emulex. At distances of 0.5 m, 1.0 m, 1.5 m, 2.0 m, 2.5 m, 3.0 m, 3.5 m and 4.0 m, the bomb was exploded. The Bayesian Regularization (BR) training algorithm is the best training algorithm for modelling Explosive Blast Prediction.

Item Type: Article
Uncontrolled Keywords: HMLP; explosion; blast prediction; PE-4; emulex
Subjects: T Technology > T Technology (General)
Depositing User: Mr. Mohamad Zulkhibri Rahmad
Date Deposited: 15 Jan 2024 07:31
Last Modified: 15 Jan 2024 07:31
URI: http://eprints.uthm.edu.my/id/eprint/10633

Actions (login required)

View Item View Item