The Helicopter Turboshaft Engine’s Reconfigured Dynamic Model for Functional Safety Estimation

Electronics, Journal Year: 2024, Volume and Issue: 13(17), P. 3477 - 3477

Published: Sept. 1, 2024

This research substantiates the necessity for developing and implementing structural reconfiguration methods automatic control systems in event of a parametric sensor failure to enhance helicopter turboshaft engine’s overall reliability safety. The aim is substantiation mathematically reconfigured system standard sensor, which will ensure stable operation under conditions, minimizing impact on engine performance. A theorem was developed proven concerning structure, defining system’s new mathematical form using nonlinear thermogas-dynamic parameters. method proposed determine values these parameters that keep stable. uses numerical optimization find best stability. Experimental results showed slow changes, from previous step works best, while fast restarting more effective due significant differences states. accuracy model confirmed through mean square error analysis (within 0.4% 0.77% white noise), regression (with determination coefficient 0.986), cross-validation metric deviation maximum 3.88%).

Language: Английский

Chain hybrid feature selection algorithm based on improved Grey Wolf Optimization algorithm

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(10), P. e0311602 - e0311602

Published: Oct. 8, 2024

Hybrid feature selection algorithm is a strategy that combines different methods aiming to overcome the limitations of single method and improve effectiveness performance selection. In this paper, we propose new hybrid algorithm, be named as Tandem Maximum Kendall Minimum Chi-Square ReliefF Improved Grey Wolf Optimization (TMKMCRIGWO). The consists two stages: First, original features are filtered ranked using bivariate filter (MKMC) form subset candidate S1; Subsequently, S1 sorted S2 by in tandem, finally used wrapper select optimal subset. particular, an improved (IGWO) based on random disturbance factors, while parameters adjusted vary randomly make population variations rich diversity. algorithms formed combining with tandem show better results than solving complex problems. Three sets comparison experiments were conducted demonstrate superiority over others. experimental average classification accuracy TMKMCRIGWO at least 0.1% higher other 20 datasets, value dimension reduction rate (DRR) reaches 24.76%. DRR reached 41.04% for 12 low-dimensional datasets 0.33% 8 high-dimensional datasets. It also shows improves generalization ability model.

Language: Английский

Task scheduling in cloud computing systems using multi-objective honey badger algorithm with two hybrid elite frameworks and circular segmentation screening

Artificial Intelligence Review, Journal Year: 2024, Volume and Issue: 58(2)

Published: Dec. 20, 2024

Language: Английский