Manufacturers have faced an increasing need for the development of predictive models that predict mechanical failures and the remaining useful life (RUL) of manufacturing systems or components. Classical model-based or physics-based prognostics often require an in-depth physical understanding of the system of interest to develop closedform mathematical models. However, prior knowledge of system behavior is not always available, especially for complex manufacturing systems and processes. To complement model-based prognostics, data-driven methods have been increasingly applied to machinery prognostics and maintenance management, transforming legacy manufacturing systems into smart manufacturing systems with artificial intelligence. While previous research has demonstrated the effectiveness of data-driven methods, most of these prognostic methods are based on classical machine learning techniques, such as artificial neural networks (ANNs) and support vector regression (SVR). With the rapid advancement in artificial intelligence, various machine learning algorithms have been developed and widely applied in many engineering fields. The objective of this research is to introduce a random forests (RFs)-based prognostic method for tool wear prediction as well as compare the performance of RFs with feed-forward back propagation (FFBP) ANNs and SVR. Specifically, the performance of FFBP ANNs, SVR, and RFs are compared using an experimental data collected from 315 milling tests. Experimental results have shown that RFs can generate more accurate predictions than FFBP ANNs with a single hidden layer and SVR. [DOI: 10.1115/1.4036350]

/pdf/a-comparative-study-on-machine-learning-algorithms-for-smart-rzce0itn13.pdf

A Comparative Study on Machine Learning Algorithms for Smart Manufacturing: Tool Wear Prediction Using Random Forests

A fog computing-based framework for process monitoring and prognosis in cyber-manufacturing

https://hal.archives-ouvertes.fr/hal-00903643/document

Prognostics of PEM fuel cell in a particle filtering framework

Effective machine health monitoring systems are critical to modern manufacturing systems and industries. Among various machine health monitoring approaches, data-driven methods are gaining in popularity due to the development of advanced sensing and data analytic techniques. However, sensory data that is a kind of sequential data can not serve as direct meaningful representations for machine conditions due to its noise, varying length and irregular sampling. A majority of previous models focus on feature extraction/fusion methods that involve expensive human labor and high quality expert knowledge. With the development of deep learning methods in the last few years, representation learning from raw data has been redefined. Among deep learning models, Long Short-Term Memory networks (LSTMs) are able to capture long-term dependencies and model sequential data. Therefore, LSTMs is able to work on the sensory data of machine condition. Here, the first study about a empirical evaluation of LSTMs-based machine health monitoring systems is presented. A real life tool wear test is introduced. Basic and deep LSTMs are designed to predict the actual tool wear based on raw sensory data. The experimental results have shown that our models, especially deep LSTMs, are able to outperform several state-of-arts baseline methods.

Machine health monitoring with LSTM networks

State of the art and taxonomy of prognostics approaches, trends of prognostics applications and open issues towards maturity at different technology readiness levels

Tool failure may result in losses in surface finish and dimensional accuracy of a finished part, or possible damage to the work piece and machine. This paper presents a Fuzzy Neural Network (FNN) which is designed and developed for machinery prognostic monitoring. The FNN is basically a multi-layered fuzzy-rule-based neural network which integrates a fuzzy logic inference into a neural network structure. The fuzzy rules help to speed up the learning process of the complex conventional neural network structure and improve the accuracy in prediction and rate of convergence. A case study for prediction of tool life in a dry milling operation is presented to demonstrate the viability of the proposed FNN for tool condition monitoring. A comparison was made in the case study on prediction performances of different models established with the same set of experimental data. It is shown that the FNN is superior to conventional Multi-Regression Models (MRM), Backpropagation Neural Networks (BPNN) and Radial Basis Function Networks (RBFN) in term of prediction accuracy and BPNN in learning speed.

/pdf/fuzzy-neural-network-modelling-for-tool-wear-estimation-in-115zgersl7.pdf

B. S. Lim

Papers

Fuzzy Neural Network Modelling for Tool Wear Estimation in Dry Milling Operation