Mathematical Problems in Engineering

AI-Based Condition Monitoring in Manufacturing Systems


Publishing date
01 Nov 2021
Status
Closed
Submission deadline
09 Jul 2021

Lead Editor
Guest Editors

1Wenzhou University, Wenzhou, China

2Amity University, Noida, India

This issue is now closed for submissions.
More articles will be published in the near future.

AI-Based Condition Monitoring in Manufacturing Systems

This issue is now closed for submissions.
More articles will be published in the near future.

Description

Safe operation and performance of manufacturing systems has been a point of concern. Intelligent condition monitoring without interrupting normal machine operations has recently gained great interest for the ever-increasing requirements of unmanned manufacturing systems. Recently, Artificial Intelligence (AI) has revolutionized the modern industry and is widely used for the purpose of intelligent condition monitoring. The AI assists in extracting useful knowledge and making appropriate decisions from measured signal in manufacturing systems.

Although there has been extensive research based on the traditional theory of signal processing and AI methods, there are two opposite research trends in the field of condition monitoring and diagnosis of the manufacturing process. Firstly, due to the development of modern sensing technology and the digitization of the manufacturing process, the manufacturing system is producing a huge amount of heterogeneous “big data”, such as process parameters, monitoring signals, and running historical records. These include vibration, current, image, and textual process data, which enable digitized and networked manufacturing. Therefore, developing advanced signal processing technology and AI-based methods to deal with big data is a hot topic currently, such as nonlinear Kalman filters, particle filters, deep learning, incremental learning, semi-supervised and unsupervised learning. On the contrary, the number of samples is often insufficient in quite a few actual workshops due to the limitations associated with objective conditions, such as limited tool lifetime, which makes the collection of sufficient data under all possible conditions costly and time-consuming. It is difficult to ensure AI models have good generalizability with insufficient samples (“small data”). Although SVM and ELM are suitable for model training with small datasets, the complexity of the manufacturing process can still affect their classification accuracy under these conditions. Therefore, developing appropriate AI-based methods to deal with small data is another topical issue currently, such as transfer learning, few-shot learning, statistical and stochastic process.

The purpose of this Special Issue is to gather state-of-the-art research contributing to recent advances in the field of condition monitoring and diagnosis of the manufacturing process. Original research and review articles are welcome.

Potential topics include but are not limited to the following:

  • Deep learning in condition monitoring
  • Feature extraction in condition monitoring
  • Nonlinear Kalman filters and particle filters in condition monitoring
  • Distributed filtering in condition monitoring
  • Neural networks-based fault diagnosis
  • Distributed fault diagnosis in manufacturing systems
  • Tool condition monitoring
  • Remaining useful life prediction
  • Transfer learning in condition monitoring
  • Few-shot learning in condition monitoring
  • AI-based machining quality prediction
  • Continuous time dynamical systems
  • Fuzzy logic-based condition monitoring
  • Statistical inference method in condition monitoring
  • Semi-supervised and unsupervised learning in condition monitoring
Mathematical Problems in Engineering
 Journal metrics
Acceptance rate27%
Submission to final decision64 days
Acceptance to publication34 days
CiteScore1.800
Journal Citation Indicator0.400
Impact Factor1.305
 Submit

Article of the Year Award: Outstanding research contributions of 2020, as selected by our Chief Editors. Read the winning articles.