What does this research mean for the field?

The optimized NSGA-II algorithm significantly reduces horizontal vibration acceleration in high-speed elevators by 57% and energy consumption by approximately 25%. Novelty: ClaimNovelty.NOVEL_FINDING. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

The aim is to develop a control approach that effectively addresses horizontal vibration issues in high-speed elevators.

February 23, 2026Open Access

Horizontal vibration control of elevator car based on optimized NSGA-II algorithm

Key Points

The aim is to develop a control approach that effectively addresses horizontal vibration issues in high-speed elevators.
Proposed an integrated control strategy combining optimized NSGA-II and variable-domain fuzzy PID controller.
Constructed a closed-loop 'sensing‐prediction-control' system using a neural network prediction model.
Utilized adaptive cross-variance operator and dynamic congestion threshold optimization.
Horizontal vibration acceleration reduction by 57% to 18.7 mg.
Displacement decreased by 60% to 0.070 mm.
Low-frequency energy attenuation rate increased by 12.5%–30.3%.
Achieved a multi-objective coordination degree of 0.92 and reduced energy consumption by approximately 25%.
Extended mean time between failures to 1,200 hours.

Abstract

Introduction High-speed elevators face significant challenges in horizontal vibration control, primarily due to guideway unevenness and dynamic load variations, which conventional passive damping technologies struggle to address effectively. This study aims to develop an intelligent control approach to overcome these limitations and enhance ride comfort and system reliability. Methods An integrated control strategy was proposed, combining an Improved Nondominated Sorting Genetic Algorithm II (NSGA-II) with a Variable-Domain Fuzzy Proportional-Integral-Derivative (PID) controller. A closed-loop “sensing‐prediction-control” system was constructed utilizing a 3-7-2 network structure, a Back-Propagation Neural Network prediction model, an Adaptive Cross-Variance Operator, and a Dynamic Congestion Threshold Optimized NSGA‐II. Results Experimental validation under 6 m/s conditions demonstrated significant performance improvements. The horizontal vibration acceleration was reduced by 57% to 18.7 mg, displacement decreased by 60% to 0.070 mm, and the low-frequency energy attenuation rate increased by 12.5%–30.3%, outperforming traditional Explicit Model Predictive Control. The method exhibited optimal robustness in extreme condition tests, with a stable load adaptability index close to 1. The strategy achieved a multi-objective coordination degree of 0.92, reduced energy consumption by approximately 25%, decreased control hardware complexity to eight components, and extended the mean time between failures to 1,200 h. Discussion The proposed integrated control strategy provides an innovative and effective solution for horizontal vibration suppression in high-speed elevators. The results confirm significant enhancements in system stability, reliability, and energy efficiency. This approach holds substantial engineering value for improving ride quality and equipment longevity. Future work could explore its adaptation to ultra-high-speed scenarios and broader applications in vertical transportation systems.

AI에게 질문

Bookmark

View Full Paper