Machine learning and IoT based Predictive Maintenance for the industrial motors for sustained Automation in the power plant Industry

Deleted Journal, Journal Year: 2024, Volume and Issue: 20(2s), P. 931 - 941

Published: April 4, 2024

The purpose of this research is to develop novel framework based on advanced tools, including machine learning and the Internet Things self-attention mechanisms. Traditional tools were used in data-driven predictive maintenance mechanism served as a basis for comparing tools. At power plant, thirty-four datasets collected monitor three industrial motors continuously. tools’ ability was analysed using conceptualized features from sensory data , management strategy remained dependent network. study results show that each tool performs at high-performing levels because exceeded 75% performance metric. indicated proposed gave consistent high-performance metric, 86.4%, all ten experimental scenarios . rate determination attributed choice long short-term memory architecture, mechanisms, optimization techniques. mean squared logarithmic error also contributed outcomes these yield scores. In addition these, test forecasting models chosen influenced performance. This study’s findings reliable predicting pending failure equipment developing relevant strategies failure. It found use development crucial aspect. Scaled techniques are importance they assist frameworks identifying ideal data. should be noted loss-function model important predictability framework..

Language: Английский

Integrating Radial Basis Networks and Deep Learning for Transportation

The Open Transportation Journal, Journal Year: 2024, Volume and Issue: 18(1)

Published: Oct. 29, 2024

Introduction This research focuses on the concept of integrating Radial Basis Function Networks with deep learning models to solve robust regression tasks in both transportation and logistics. Methods It examines such combined as RNNs RBFNs, Attention Mechanisms (RBFNs), Capsule RBFNs clearly shows that, all cases, compared others, former model has a Mean Squared Error (MSE) 0.010 0.013, Absolute (MAE) – 0.078 0.088, R-squared (R 2 ) 0.928 0.945, across ten experiments. In case also demonstrate strong performance terms making predictions. The MSE ranges from 0.012 0.015, MAE 0.086 0.095, R 0.914 0.933. Results However, it is critical note that outperform other models. particular, they offer lowest MSE, which between 0.009 0.012, smallest MAE, 0.075 0.083, highest , 0.935 0.950. Conclusion Overall, results indicate use combination different types networks can provide highly accurate reliable solutions for problems domain

Language: Английский