Applying clustered artificial neural networks to enhance contaminant diffusion prediction in geotechnical engineering

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Nov. 20, 2024

This paper introduces a novel approach using Clustered Artificial Neural Networks (CLANN) to address the challenge of developing predictive models for multimodal dataset with extreme parameter values. The CLANN method strategically decomposes dataset, derived from Finite Element Analysis (FEA), into clusters, each representing distinct diffusion behaviors, and applies specialized neural networks within these clusters. model was rigorously evaluated demonstrated superior accuracy consistency compared traditional methods such as Adaptive Neuro-Fuzzy Inference System (ANFIS) fuzzy expert systems. While conventional struggled capture full range dynamics, particularly under conditions, consistently provided predictions that closely aligned actual FEA data across all scenarios. versatility extends beyond its application soil contamination. Its ability handle complex, datasets suggests this methodology can be generalized wide scientific engineering problems characterized by similar structures. makes not only powerful tool geotechnical engineers but also promising framework broader applications where fall short. findings study pave way more accurate, reliable, adaptable modeling in diverse domains, enhancing our manage mitigate environmental challenges.

Language: Английский

A design of power prediction algorithm based on health assessment optimized neural network

Journal of Engineering and Applied Science, Journal Year: 2024, Volume and Issue: 71(1)

Published: April 22, 2024

Abstract Wind power prediction holds significant value for the stability of electrical grid when wind is connected to grid. Using neural networks may have some limitations, such as slow speed and low accuracy. This paper proposes enhance accuracy by optimizing network through health assessment turbines. Firstly, based on turbine actual operating data, a conducted obtain matrix turbine. Then, calculating weights matrix, strategy optimized. Following that, approximation hyperparameters are utilized expedite optimization process. Finally, tests prediction, act optimized back propagation (BP) whale swarm algorithm–support vector regression (WSA-SVR) employed prediction. Results show noticeable optimization: after BP network, increased about 40%, rose 20%; WSA-SVR improved 10%, surged 45%. Further analysis shows that this method can improve most algorithms.

Language: Английский

Optimizing Site Investigations for Gassy Soils: A Bi-Objective Approach Using Value of Information and Cost of Boreholes

Probabilistic Engineering Mechanics, Journal Year: 2024, Volume and Issue: unknown, P. 103727 - 103727

Published: Dec. 1, 2024

Language: Английский