转速
计算机科学
断层(地质)
适应性
加速
算法
控制理论(社会学)
方位(导航)
采样(信号处理)
多项式的
观测误差
谱密度估计
区间(图论)
测速
测量不确定度
估计理论
角速度
自适应采样
概括性
数据采集
信号(编程语言)
准确度和精密度
旋转不变性
噪声测量
工程类
电子工程
度量(数据仓库)
作者
Linhao Peng,Fang Liu,Yongjian Sun,Siliang Lu,Yongbin Liu
标识
DOI:10.1109/tim.2025.3638939
摘要
Equal angular interval sampling (EAISP) is currently one of the primary approaches used for bearing fault diagnosis under variable-speed conditions. Accurate acquisition of rotational speed is a prerequisite for this method. Hardware-based speed measurement and algorithm-based speed estimation are the two most commonly employed approaches. The former requires high-precision sensors such as encoders, which are costly and sensitive to environmental factors. The latter relies on a specific algorithm model and requires high data quality, resulting in limited generality and scalability. This study explores a hybrid approach that combines hardware-based measurement with algorithmic speed estimation. In this method, a single Hall sensor with low accuracy but low cost and strong environmental adaptability is used to initially perform a rough measurement of rotational speed. Then, a method named kurtosis-based spectral index guided Polynomial Approximation (KSIGPA) is proposed to achieve an accurate estimation of the rotational speed. Simulation and experiment verify that this strategy is between the existing hardware-based speed measurement and algorithm-based speed estimation, and it has a good application prospect in the scenarios where only rough speed can be obtained and it is impossible or inconvenient to install high-precision sensors.
科研通智能强力驱动
Strongly Powered by AbleSci AI