Multi-modal Sensing for Machine Health Monitoring in High Speed Machining

Zeng, Hao; Thoe, T.B.; Li, Xiang; Zhou, Junhong

doi:10.1109/indin.2006.275812

Cited by 15 publications

(7 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Obviously, extracting useful features from original signals is critical for tool wear monitoring, and it can directly affect the prediction results of the model. Commonly used feature processing methods are time domain analysis, frequency domain analysis, and time–frequency domain analysis [ 4 ].…”

Section: Introductionmentioning

confidence: 99%

A Dual-Stage Attention Model for Tool Wear Prediction in Dry Milling Operation

Qin

Zhang

et al. 2022

Entropy

View full text Add to dashboard Cite

The intelligent monitoring of tool wear status and wear prediction are important factors affecting the intelligent development of the modern machinery industry. Many scholars have used deep learning methods to achieve certain results in tool wear prediction. However, due to the instability and variability of the signal data, some neural network models may have gradient decay between layers. Most methods mainly focus on feature selection of the input data but ignore the influence degree of different features to tool wear. In order to solve these problems, this paper proposes a dual-stage attention model for tool wear prediction. A CNN-BiGRU-attention network model is designed, which introduces the self-attention to extract deep features and embody more important features. The IndyLSTM is used to construct a stable network to solve the gradient decay problem between layers. Moreover, the attention mechanism is added to the network to obtain the important information of output sequence, which can improve the accuracy of the prediction. Experimental study is carried out for tool wear prediction in a dry milling operation to demonstrate the viability of this method. Through the experimental comparison and analysis with regression prediction evaluation indexes, it proves the proposed method can effectively characterize the degree of tool wear, reduce the prediction errors, and achieve good prediction results.

show abstract

Section: Introductionmentioning

confidence: 99%

A Dual-Stage Attention Model for Tool Wear Prediction in Dry Milling Operation

Qin

Zhang

et al. 2022

Entropy

View full text Add to dashboard Cite

show abstract

“…However, few monitoring systems have been developed and reported to detect the tool condition in HSM. These systems are based on the measurement of physical variables related to the cutting tool conditions, such as cutting force [1,2], feed and spindle motor currents [3,4], and vibration [5,6]. Nevertheless, some of these reported systems have limitations for their industrial application, e.g., high costs, difficult installation on the machine tool, or operation with limited characteristics.…”

Section: Introductionmentioning

confidence: 99%

“…These drivers have current sensors with limited bandwidth. In contrast to the aforementioned variables, vibration signal offers better characteristics such as the following: (1) periodic shape that resembles the cutting force, (2) enough information regarding tool condition, (3) robustness, (4) reliability, (5) applicability, (6) low-cost, and (7) ease in measurement implementation without requiring any machine tool modification. For this reason, the vibration signal has also been used for TCM in conventional machining [7][8][9].…”

Section: Introductionmentioning

confidence: 99%

FPGA-based reconfigurable system for tool condition monitoring in high-speed machining process

et al. 2015

View full text Add to dashboard Cite

“…Cutting force measurements have been used to detect cutting tools failures for high-speed milling. [16][17][18][19][20] This monitoring strategy has been done due to the close relation of the force with the interaction of the tool and workpiece during the cutting process. Cutting force is measured by dynamometers mounted between the workpiece and machining table.…”

Section: Introductionmentioning

confidence: 99%

“…The implementation of a vibration system is simple and inexpensive, and it does not require tool's modifications. Tool failures in conventional [21][22][23] and high-speed 18,19,24 end milling have been studied using vibration signal analysis. All these studies are based on the FFT; however, the effectiveness of signal spectral analysis in failure detection and diagnosis is limited.…”

Section: Introductionmentioning

confidence: 99%

Efficient method for detecting tool failures in high-speed machining process

Sevilla-Camacho

Jáuregui-Correa

Herrera

et al. 2013

Proceedings of the Institution of Mechanical Engineers, Part B:

View full text Add to dashboard Cite

This research presents a method for detecting tool failures in high-speed face milling. This method detects tool failures from vibration signature maps. Tests were carried out at different tool failure levels, spindle speeds, feed rates, and workpiece mountings. Vibration signals were obtained with an accelerometer and processed using the continuous wavelet transform methodology. The vibration signature maps showed that healthy cutting tools produce a periodic insert passing frequency and its harmonics. In contrast, a damage tool generates additional nonlinear and transient frequencies at nonsynchronous frequencies. The experimental results agree with vibration signature maps obtained from a simulated cutting force model. The proposed method is effective as a tool failure detection method when transient and nonlinear behaviors are presented in face milling process. Moreover, the proposed method showed good results at different process parameters and for several types of tool failures. Finally, it is important to point out the use of accelerometers because they present several advantages against other types of sensors. Advantages such as low cost, wide bandwidth, and easy implementation are important characteristics for tool condition monitoring in high-speed machining.

show abstract

Multi-modal Sensing for Machine Health Monitoring in High Speed Machining

Cited by 15 publications

References 9 publications

A Dual-Stage Attention Model for Tool Wear Prediction in Dry Milling Operation

A Dual-Stage Attention Model for Tool Wear Prediction in Dry Milling Operation

FPGA-based reconfigurable system for tool condition monitoring in high-speed machining process

Efficient method for detecting tool failures in high-speed machining process

Contact Info

Product

Resources

About