Evaluation of transducer signature selections on machine learning performance in cutting tool wear prognosis

Sun, I-Chun; Cheng, Ren-Chi; Chen, Kuo-Shen

doi:10.1007/s00170-021-08526-w

Cited by 12 publications

(2 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A lot of ML-based regression algorithms have been established for predicting cutting tool remaining useful lifespan (RUL) and few failures. Different techniques using the entire spectrum of the AI tools such as classification tools [5,18,23], decision trees [10,11,13,14], artificial neural networks [12,21,31,40], autoencoders [3,15,28,30]. In summary of the literature review, most of the researchers employed supervised learning methods which yield potential results without any doubt however they limit themselves to classify data as supervised and prone to type II errors in case of lack of feature engineering.…”

Section: Introductionmentioning

confidence: 99%

Milling cutter fault diagnosis using unsupervised learning on small data: A robust and autonomous framework

Patange,

Soman,

Pardeshi

et al. 2024

Eksploatacja i Niezawodność – Maintenance and Reliability

View full text Add to dashboard Cite

Tool condition affects the tolerances and the energy consumption and hence needs to be monitored. Artificial intelligence based data-driven techniques for tool condition determination are proposed. Unfortunately, the data-driven techniques are data-hungry. This paper proposes a methodology for classification based on unsupervised learning using limited unlabeled training data. The work presents a multi-class classification problem for the tool condition monitoring. The principal component analysis (PCA) is employed for dimensionality reduction and the principal components (PCs) are used as input for classification using k-means clustering. New collected data is then projected on the PC space, and classified using the clusters from the training. The methodology has been applied for classification of tool faults in 6 classes. The use of limited input parameters from the user makes the method ideal for monitoring a large number of machines with minimal intervention. Furthermore, due to the small amount of data needed for the training, the method has the potential to be transferable.

show abstract

Section: Introductionmentioning

confidence: 99%

Milling cutter fault diagnosis using unsupervised learning on small data: A robust and autonomous framework

Patange,

Soman,

Pardeshi

et al. 2024

Eksploatacja i Niezawodność – Maintenance and Reliability

View full text Add to dashboard Cite

show abstract

“…A smart factory is a concept of connecting all kinds of equipment, products, as well as the stocks or orders in factories through internet and monitoring the status of production flow and machine tools by instantly analyzing the obtained data during manufacturing for achieving cost-effective and quality assurance results [1,2] and related demonstration systems such as mMS4.0 [3] of BOSCH Rexroth have been developed for the purpose of process emulation. The entire system can be seen as a miniature of smart factory resulted from integrating technologies related to smart manufacturing, including robot applications, mechatronics [4,5], sensor fusion, status monitoring, and machine learning related work [6].…”

Section: Introductionmentioning

confidence: 99%

A Framework on Analysis and Integration of Robot Manipulators for Smart Factory Applications

Han

Yang

Chen

et al. 2023

The Proceedings of the 11th IIAE International Conference on Industrial Application Engineering 2023

View full text Add to dashboard Cite

A framework of industrial robot centering on smart factory applications is presented. The scenarios are realized by implementing dynamic modelling, vision guidance, sensor networking, status monitoring, multi-machine integration, and human-robot interaction. For dynamic modelling of robot manipulators, a 3D simulation interface and the corresponding model are built. For integrating robot manipulator with vision system, an indirect calibration method is conducted to set up a reference frame for both the robot and the vision system and a vision guided robot system is achieved. For sensor networking and status monitoring, sensor fusion is studied and a dashboard is built for integrating internal information from robot controllers with external data. To achieve multi-machine integration and human-robot interaction, IoT platforms are utilized for creating the communications among different equipment. Furthermore, a demonstration involving the integration of a robot arm and an automated guided vehicle is realized for object transportation. In addition, external sensors are integrated to create a safe human-machine environment. It is believed that the methodologies developed can be applied to fulfill smart factory in a pragmatic way in the future.

show abstract

A Review: Sensors Used in Tool Wear Monitoring and Prediction

Unal¹,

Deveci²,

Özbayoğlu

2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Evaluation of transducer signature selections on machine learning performance in cutting tool wear prognosis

Cited by 12 publications

References 33 publications

Milling cutter fault diagnosis using unsupervised learning on small data: A robust and autonomous framework

Milling cutter fault diagnosis using unsupervised learning on small data: A robust and autonomous framework

A Framework on Analysis and Integration of Robot Manipulators for Smart Factory Applications

A Review: Sensors Used in Tool Wear Monitoring and Prediction

Contact Info

Product

Resources

About