Scholarlink

The rapid expansion of sensor applications in industrial and commercial settings demands integrated electronic systems that are not only flexible and precise but also reliably maintain long-term performance. To meet these demands, next-generation sensors in Industry 4.0 will need advanced “self-x” capabilities, allowing them to monitor, calibrate, and repair themselves autonomously. Ideally, these sensors will detect anomalies both in their operating environment and within their internal components, adapting in real-time to minimize impact and maintain accuracy. Magnetostrictive Position Sensors (MPS) are commonly used for the measurement of precise displacement and velocity. Its operation is based on Time-of-Flight (ToF) calculations of structure-borne sound waves produced due to the interaction of magnetic fields. However, their measurement accuracy is sensitive to External Magnetic Interference (EMI), which can distort the sensor’s generated waveforms. This study employs machine learning to mitigate errors caused by electromagnetic interference (EMI), making it particularly relevant for measurement applications in industrial settings with numerous sources of EMI. The experimental results show that the devised machine learning mechanism can provide significant gains with respect to the measurement accuracy.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Measurement Accuracy Improvement in Magnetostrictive Position Sensors Using Machine Learning

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Aimal Khan,
  • Tobias König,
  • Alexander Hetznecker,
  • Thomas Greiner

摘要

The rapid expansion of sensor applications in industrial and commercial settings demands integrated electronic systems that are not only flexible and precise but also reliably maintain long-term performance. To meet these demands, next-generation sensors in Industry 4.0 will need advanced “self-x” capabilities, allowing them to monitor, calibrate, and repair themselves autonomously. Ideally, these sensors will detect anomalies both in their operating environment and within their internal components, adapting in real-time to minimize impact and maintain accuracy. Magnetostrictive Position Sensors (MPS) are commonly used for the measurement of precise displacement and velocity. Its operation is based on Time-of-Flight (ToF) calculations of structure-borne sound waves produced due to the interaction of magnetic fields. However, their measurement accuracy is sensitive to External Magnetic Interference (EMI), which can distort the sensor’s generated waveforms. This study employs machine learning to mitigate errors caused by electromagnetic interference (EMI), making it particularly relevant for measurement applications in industrial settings with numerous sources of EMI. The experimental results show that the devised machine learning mechanism can provide significant gains with respect to the measurement accuracy.