离合器
加权
接口(物质)
反演(地质)
组分(热力学)
传动系统
传输(电信)
计算机科学
模拟
控制理论(社会学)
工程类
可靠性工程
汽车工程
材料科学
人工智能
声学
复合材料
地质学
电信
物理
古生物学
构造盆地
热力学
毛细管作用
毛细管数
控制(管理)
作者
Jianpeng Wu,Ao Ding,Wenya Shu,Heyan Li,Liyong Wang,Chengbing Yang
标识
DOI:10.1177/09544070251342284
摘要
Wet clutches are used to transmit torque between transmission friction pairs in armored vehicles, heavy machinery, and other mechanical devices. Their safe and stable operation plays a crucial role in power transmission. Among all the influencing factors, the interface morphology of wet friction components directly influences both the transmission efficiency and the service life of the clutch, which makes the study of interface morphology particularly significant for these wet friction components. This paper designs a disc-to-disc experiment to obtain interface morphology data of wet friction components under various working conditions. These interface morphology data then serve the purpose of being sample input for the Newton-Raphson-Based Optimizer-Bidirectional Long Short-Term Memory-Attention (NRBO-BiLSTM-Attention) inversion model, allowing for the prediction of the interface morphology. Further, the study derives several key health evaluation indicators for the transmission interface of wet friction components, specifically steepness ( S ku ), roughness ( R a ), coefficient of friction ( μ ), and the coefficient of variation ( CV ). Ulteriorly, by applying a subjective-objective weighting method, the study determines the appropriate weights for each indicator and establishes a performance degradation staged comprehensive evaluation system to evaluate the health of the friction components. The health status of the friction component is ultimately divided into five stages: Healthy Stage, Good Stage, Moderate Stage, Alert Stage, and Fault Stage. The results indicate that the R 2 value of the NRBO-BiLSTM-Attention inversion model predictions is 0.9719, which demonstrates the model’s effectiveness and high accuracy. Furthermore, the performance degradation staged comprehensive evaluation system effectively evaluates the health of the transmission interface, achieving an R 2 of 0.9758 in relation to the actual degradation curve. This outcome provides a robust basis for enhancing the reliability of friction components.
科研通智能强力驱动
Strongly Powered by AbleSci AI