Open Access. Powered by Scholars. Published by Universities.®

Engineering Commons

Open Access. Powered by Scholars. Published by Universities.®

2019

Series

Loyola University Chicago

Clustering

Articles 1 - 2 of 2

Full-Text Articles in Engineering

Smart Control Of Buck Converters Using A Switching-Based Clustering Algorithm, Brook Abegaz, M. Cmiel Jan 2019

Smart Control Of Buck Converters Using A Switching-Based Clustering Algorithm, Brook Abegaz, M. Cmiel

Engineering Science Faculty Publications

This paper proposes a new approach to the control of switching voltage regulators (buck converters). The method is performed using a switching-based clustering algorithm. The implementations of competing approaches such as a fuzzy-logic controller, proportional integral derivative controller and a neural network based controller are presented in order to compare and evaluate the performance of the switching-based clustering algorithm. The results of the approach show that the proposed method could improve the stability and the performance of the buck converter system by 2.7% in terms of settling time and by 0.6% in terms of the overshoot value as compared to …


Smart Control Of Automatic Voltage Regulators Using K-Means Clustering, Brook Abegaz, J. Kueber Jan 2019

Smart Control Of Automatic Voltage Regulators Using K-Means Clustering, Brook Abegaz, J. Kueber

Engineering Science Faculty Publications

The future cyber physical systems consist of voltage regulators distributed across wide geographical areas. In this paper, a smart control approach of voltage regulators is presented for cyber physical system applications. The approach is implemented using K-means clustering algorithms that use data from voltage and current sensors, compute the correlation of changes across the regulators and generate a proportional feedback. Advanced estimation methods are used in cases where the data from the sensors was not available. The results show that the approach could be used to improve the performance of networked, power dependent systems by 94.5% in terms of overshoot …