Real time monitoring and control of friction stir welding process using multiple sensors

Mishra, Debasish; Gupta, Akhilesh; Raj, Pranav; Kumar, Amit; Anwer, Saad; Pal, Surjya K.; Chakravarty, Debashish; Pal, Srikanta; Chakravarty, Tapas; Pal, Arpan; Misra, Prateep; Misra, Sudip

doi:10.1016/j.cirpj.2020.03.004

Cited by 63 publications

(21 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Machine learning techniques have been applied in the friction stir welding manufacturing for monitoring the vision profile of welding path [ 20 ] or the machine feedback information such as force and torque. [ 16 ] However, the methods can barely prevent flaw formation before it happens, since the vision or feedback outlier data that arose were both caused by the flaw initiation. [ 16 ] Additionally, to obtain a proper training bank of vision or machine feedback data experimentally is also challenging due to the unavoidable human/equipment errors and high degree‐of‐freedom interference of involved factors.…”

Section: Discussionmentioning

confidence: 99%

“…[ 16 ] However, the methods can barely prevent flaw formation before it happens, since the vision or feedback outlier data that arose were both caused by the flaw initiation. [ 16 ] Additionally, to obtain a proper training bank of vision or machine feedback data experimentally is also challenging due to the unavoidable human/equipment errors and high degree‐of‐freedom interference of involved factors. The presence of the errors in a training bank can significantly impact the identification accuracy of machine learning programs.…”

Section: Discussionmentioning

confidence: 99%

“…In previous efforts regarding FSW/P quality control, AI programs were also applied with some sensing techniques for correlating the detected information with desired outputs without clear logic connections in human sense. In the literature, some existing suggestions were focused on the AI assisted monitoring of the temperature distribution, [ 15 ] welding machine feedback, [ 16 ] and processed surface imaging. [ 17 ] All of the factors were necessary to include in a monitoring loop for a better production quality.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Numerically Trained Ultrasound AI for Monitoring Tool Degradation

Jin

Wang

Fox

et al. 2022

Advanced Intelligent Systems

View full text Add to dashboard Cite

Monitoring tool degradation during manufacturing can ensure product accuracy and reliability. However, due to variations in degradation conditions and complexity in signal analysis, effective and broadly applicable monitoring is still challenging to achieve. Herein, a novel monitoring method using ultrasound signals augmented with a numerically trained machine learning technique is reported to monitor the wear condition of friction stir welding and processing tools. Ultrasonic signals travel axially inside the tools, and even minor tool wear will change the time and amplitude of the reflected signal. An artificial intelligence (AI) algorithm is selected as a suitable referee to identify the small variations in the tool conditions based on the reflected ultrasound signals. To properly train the AI referee, a human‐error‐free data bank using numerical simulation is generated. The simulation models the experimental conditions with high fidelity and can provide comparable ultrasound signals. As a result, the trained AI model can recognize the tool wear from real experiments with subwavelength accuracy prediction of the worn amount on the tool pins.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Numerically Trained Ultrasound AI for Monitoring Tool Degradation

Jin

Wang

Fox

et al. 2022

Advanced Intelligent Systems

View full text Add to dashboard Cite

show abstract

“…By nature, the detail coefficients are sparse, which is the primary reason for its huge range of applications in the field of compressive sensing. Recent studies in FSW have utilized the wavelet coefficients extracted by DWT for real-time monitoring and control of the process and monitoring of the tool quality [1,30]. In this process, it is crucial to choose a suitable mother wavelet for a particular signal as different mother wavelets may produce different results.…”

Section: Methodsmentioning

confidence: 99%

Weld defect localization in friction stir welding process

et al. 2020

Self Cite

View full text Add to dashboard Cite

The article attempts to detect the defects in friction stir welding (FSW) process by analyzing the signal acquired during welding. The said welding technique utilizes pressure and heat developed by the usage of a non-consumable tool. Thus, the axial force signal carries a lot of information about the physical process, and hence, it could be used to identify the weld defects. Signal analysis has been performed by using wavelet-based techniques. Before this analysis, a methodology has been followed to select the best mother wavelets suitable for the signal. The results of defect identification have been validated by mapping the processed signal with the actual weld quality.

show abstract

“…Few works reported the weld defect monitoring by using time-frequency wavelet analysis of axial force or torque signal in FSW of alluminium alloys. 27,28 However, there is a scanty work on the process monitoring of FSW of polycarbonate sheets by using axial force along or torque signals. In order to bridge these gaps, the present work focuses on two aspects: (a) a detailed experimental investigation, and (b) sensor based process monitoring, followed by development of emperical model for prediction of joint strength.…”

Section: Introductionmentioning

confidence: 99%

Multi sensor based strategies for accurate prediction of friction stir welding of polycarbonate sheets

Sahu

Mishra

Pal

et al. 2020

Proceedings of the Institution of Mechanical Engineers, Part C:

Self Cite

View full text Add to dashboard Cite

High performance thermoplastic is often used in automotive and aerospace sectors. Friction stir welding (FSW) has been a suitable method for joining these thermoplastic materials. However, the process characteristics vary dynamically, even at same parametric conditions due to process disturbances. While quite a few studies have been reported in experimental investigation, the usage of sensors for the process monitoring during welding has not been explored. Therefore, the present paper focuses on various sensor-based approaches on process monitoring. The study includes an in-depth experimental investigation for joining of polycarbonate sheets using full factorial design of experiments. Three different tool pin profiles like cylindrical, square and triangular have been selected, whereas tool rotational speed and traverse speed are considered as primary process variables. Various statistical time domain features of the force and torque signals starting from average, its deviation and root mean square as well as thermal cycle based weld peak temperature and its cooling rate were found to have a strong correlation with weld bead profile. The axial force and thermal cycles were found to be the major indicators of joint strength efficiency rather than torque as per regression models developed by using response surface methodology. Finally, an attempt has been made for improving the joint strength predictability by using various sensor-based approaches. This work can be used potentially to real time welding process monitoring in industrial sectors.

show abstract

Real time monitoring and control of friction stir welding process using multiple sensors

Cited by 63 publications

References 13 publications

Numerically Trained Ultrasound AI for Monitoring Tool Degradation

Numerically Trained Ultrasound AI for Monitoring Tool Degradation

Weld defect localization in friction stir welding process

Multi sensor based strategies for accurate prediction of friction stir welding of polycarbonate sheets

Contact Info

Product

Resources

About