Optimization of Enhanced Pipeline Leak Detection Using Advanced Dense Hebbian Neural Network Architectures for Improved Accuracy and Efficiency
DOI: 10.63503/j.ijaimd.2025.41 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal on Engineering Artificial Intelligence Management, Decision Support, and Policies, 2025. Verfasst von Amit Verma und 1 weiteren Autor:innen.
Abstract (Originalsprache)
Pipeline networks are essential in conveying water, oil, gases as well as other es sential commodities in society. These networks must be protected from compro mise to prevent losses and possible negative impacts on the environment. This paper develops an advanced pipeline leak detection procedure that is optimized by way of Dense Hebbian Neural Networks (DHNNs). The implementation of the proposed system aims at using the novel neural networks’ structure in conjunc tion with signal recognition methods for piping leakage detection and localiza tion. The high denseness of interconnection in the DHNN helps in learning and offers simplicity in computation. The method used involves pressure and flow in formation at run time, whereby deviation from the norm is established and grouped according to patterns of the pipeline network. A comparison between the developed DHNN model and a Feedforward Neural Network (FNN) model is made to compare the increased leak detection efficacy, precision, and perfor mance measures. This paper uses basic mathematical models to assess signal behavior when different leak conditions are present and simulation results show the stability of the proposed solution. The simulation uses control variables cre ated within the MATLAB environment and includes realistic numerical pipeline data. Comparing DHNN with FNN, the findings reveal better precision, faster convergence rate, and better leak localization with the use of DHNN. Experi mental results are shown in figures and graphs which provide a clear comparison of the results obtained. The outcome of this research work is expected to improve the safety of pipelines and minimize preventable economic losses because of leaks, particularly through improving leak detection with modern concepts in neural network paradigms.
Zitieren
APA
Verma, A., & Garg, S. (2025). Optimization of Enhanced Pipeline Leak Detection Using Advanced Dense Hebbian Neural Network Architectures for Improved Accuracy and Efficiency. International Journal on Engineering Artificial Intelligence Management, Decision Support, and Policies. https://doi.org/10.63503/j.ijaimd.2025.41
Vancouver
Verma A, Garg S. Optimization of Enhanced Pipeline Leak Detection Using Advanced Dense Hebbian Neural Network Architectures for Improved Accuracy and Efficiency. International Journal on Engineering Artificial Intelligence Management, Decision Support, and Policies. 2025. doi:10.63503/j.ijaimd.2025.41
BibTeX
@article{Verma2025,
author = {Verma, Amit and Garg, Shakun},
title = {Optimization of Enhanced Pipeline Leak Detection Using Advanced Dense Hebbian Neural Network Architectures for Improved Accuracy and Efficiency},
journal = {International Journal on Engineering Artificial Intelligence Management, Decision Support, and Policies},
year = {2025},
doi = {10.63503/j.ijaimd.2025.41}
}
EUVIMED