Modeling of Mexican Hat Wavelet Neural Network with L-BFGS Algorithm for Simulating the Recycling Procedure of Waste Plastic in Ocean
Journal Title: Journal of Engineering Management and Systems Engineering - Year 2023, Vol 2, Issue 1
Abstract
In the global economy, plastics are considered a versatile and ubiquitous material. It can reach to marine ecosystems through diverse channels, such as road runoff, wastewater pathways, and improper waste management. Therefore, rapid mitigation and reduction are required for this ever-growing problem. The marine habitats are believed to be the highest emitters and absorbers of O2 and CO2 respectively. As such, every day, the prominence of managing the litter in the ocean is growing effectively and efficiently. One of the most significant challenges in oceanography is creating a comprehensive meshless algorithm to handle the mathematical representation of waste plastic management in the ocean. This research dedicates to studying the dynamics of waste plastic management model governed by a mathematical representation depending on three components viz. Waste plastic (W), Marine litter (M) and Recycling of debris (R), i.e., WMR model. In this regard, an unsupervised machine learning approach, namely Mexican Hat Wavelet Neural Network (MhWNN) refined by the efficient Limited-memory Broyden–Fletcher–Goldfarb–Shanno algorithm (L-BFGS), i.e., MhWNN-LBFGS model has been implemented for handling the non-linear phenomena of WMR models. Besides, the obtained solution is meshfree and compared with the state-of-art numerical result to establish the precision of the MhWNN-LBFGS model. Furthermore, different global statistical measures (MAPE, TIC, RMSE, and ENSE) have been computed at twenty testing points to validate the stability of the proposed algorithm.
Authors and Affiliations
Arup Kumar Sahoo, Snehashish Chakraverty
Keywords
Related Articles
- EP ID EP731818
- DOI 10.56578/jemse020104
- Views 45
- Downloads 0
Isogeometric Finite Element Analysis with Machine Learning Integration for Piezoelectric Laminated Shells
Innovative lightweight smart structures incorporating piezoelectric material-based active elements, both as sensors and actuators, have been identified to present manifold advantages over traditional passive systems. Suc...
Development of an Immersive, Digital Twin-Supported Smart Reconfigurable Educational Platform for Manufacturing Training: A Proof of Concept
The challenge of providing students with practical, hands-on experience in realistic industrial environments is increasingly prevalent in modern technical education. The concept of a Learning Factory addresses this issue...
Enabling Legacy Lab-Scale Production Systems: A Digital Twin Approach at Széchenyi István University
The burgeoning importance of digitalization and cyber-physical manufacturing systems in the industrial sector is undeniable, yet discussions around viable solutions for small and medium-sized enterprises remain scant. Th...
Evaluating the Annual Operational Efficiency of Passenger and Freight Road Transport in Serbia Through Entropy and TOPSIS Methods
Road transport emerges as a crucial segment of the transportation system, demanding comprehensive analyses of operational performance across passenger and freight domains. This investigation delineates a meticulous multi...
A Parametric Similarity Measure for Spherical Fuzzy Sets and Its Applications in Medical Equipment Selection
The Spherical Fuzzy Set (SFS) framework extends the Picture Fuzzy Set (PFS) concept, offering enhanced precision in handling data characterized by conflict and uncertainty. Furthermore, similarity measures (SMs) are cruc...