Misalignment Detection on Linear Feed Axis with FFT and Statistical Analysis using Motor Current

Demetgül, Mustafa; Gu, Miṅ; Hillenbrand, J.; Zhao, Yicheng; Gönnheimer, Philipp; Fleischer, Jürgen

doi:10.36897/jme/147699

Cited by 5 publications

(3 citation statements)

References 13 publications

(16 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The method was tested on induction motor, gearbox, and bearing datasets. Demetgül et al presented an approach for the detection of axial misalignments at different levels by using motor current data with signal processing techniques and statistical techniques [40][41]. This work's aim is to detect horizontal and vertical misalignments of linear axes in different scenarios.…”

Section: Literatur Reviewmentioning

confidence: 99%

Misalignment detection on linear feed axis using sensorless motor current signals

Demetgul

Zihan

Heider

et al. 2023

Int J Adv Manuf Technol

Self Cite

View full text Add to dashboard Cite

Due to ageing populations and a shortage of skilled labour, automatic machine condition monitoring is a powerful tool to ensure smooth operation of production systems with reduced manpower. Automatic condition monitoring enables early detection of machine faults, greatly increasing uptime, reliability, and safety. However, conventional fault detection methods based on vibration require installation of additional sensors, thus bringing up implementation effort and initial costs. The linear feed axis is a machine component whose failure can bring an entire production line to a standstill. Therefore, this study presents a sensorless approach, which uses a linear axis' motor current for the detection of misalignment. Motor current time series data was encoded as images and then fed to a CNN, more precisely a revised residual neural network (ResNet). A random search hyper-parameter tuning technique was used to optimise the structure of the CNN. Then, transfer learning was used to apply the current signal features already learned to other scenarios. The results showed that both horizontal and vertical misalignments of linear feed axes can be well detected by a revised ResNet using motor current signals, with an accuracy of 99.76%. Using transfer learning, misalignments were detected with an accuracy of 92.67% -even under the in uence of external forces.

show abstract

Section: Literatur Reviewmentioning

confidence: 99%

Misalignment detection on linear feed axis using sensorless motor current signals

Demetgul

Zihan

Heider

et al. 2023

Int J Adv Manuf Technol

Self Cite

View full text Add to dashboard Cite

show abstract

“…The effects of varying installation situations play a subordinate role here. Previous studies on this topic are limited, for example, to the impact of angular errors in the motor current of the drive axles [20,21]. Wear mechanisms resulting from the interaction of two or more axes have not yet been considered in a structured way.…”

Section: State Of the Art -Wear Effectsmentioning

confidence: 99%

Cyber-Physical Test Environment for the Identification of Interacting Wear Effects in Feed Axes

Bott¹,

Stöbel²,

Osen³

et al. 2023

Journal of Machine Engineering

View full text Add to dashboard Cite

For a comprehensive optimization and control of production processes, cyber-physical systems are necessary to include machines' time-dependent properties. These wear effects in machine tools, especially the feed axes, can significantly influence the process quality and are a steady research focus. However, the interaction of wear effects between different feed axes has received little attention. Especially models that represent the combined wear influence of different interacting feed axes on the control parameters and machine dynamics hold great potential. To close this knowledge gap, this paper proposes a cyber-physical test environment to identify the interaction of wear effects in feed axes. For this test environment, the relevant boundary conditions of different feed axes in machine tools and their systematic interaction are presented. Through these conditions, a physical test setup is derived and, analogous to this, a virtual model is created. This holistic approach represents the physical and virtual interaction between different components.

show abstract

Section: Literatur Reviewmentioning

confidence: 99%

Sensorless Misalignment Detection on Linear Feed Axis with Revised ResNet and Transfer Learning Using Motor Current

Demetgul

Zihan²,

Heider³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Due to ageing populations and a shortage of skilled labour, automatic machine condition monitoring is a powerful tool to ensure smooth operation of production systems with reduced manpower. Automatic condition monitoring enables early detection of machine faults, greatly increasing uptime, reliability, and safety. However, conventional fault detection methods based on vibration require installation of additional sensors, thus bringing up implementation effort and initial costs. The linear feed axis is a machine component whose failure can bring an entire production line to a standstill. Therefore, this study presents a sensorless approach, which uses a linear axis’ motor current for the detection of misalignment. Motor current time series data was encoded as images and then fed to a CNN, more precisely a revised residual neural network (ResNet). A random search hyper-parameter tuning technique was used to optimise the structure of the CNN. Then, transfer learning was used to apply the current signal features already learned to other scenarios. The results showed that both horizontal and vertical misalignments of linear feed axes can be well detected by a revised ResNet using motor current signals, with an accuracy of 99.76%. Using transfer learning, misalignments were detected with an accuracy of 92.67% – even under the influence of external forces.

show abstract

Misalignment Detection on Linear Feed Axis with FFT and Statistical Analysis using Motor Current

Cited by 5 publications

References 13 publications

Misalignment detection on linear feed axis using sensorless motor current signals

Misalignment detection on linear feed axis using sensorless motor current signals

Cyber-Physical Test Environment for the Identification of Interacting Wear Effects in Feed Axes

Sensorless Misalignment Detection on Linear Feed Axis with Revised ResNet and Transfer Learning Using Motor Current

Contact Info

Product

Resources

About