A comparative evaluation of neural networks and hidden Markov models for monitoring turning tool wear

Abstract

Condition monitoring of machine tool inserts is important for increasing the reliability and quality of machining operations. Various methods have been proposed for effective tool condition monitoring (TCM), and currently it is generally accepted that the indirect sensor-based approach is the best practical solution to reliable TCM. Furthermore, in recent years, neural networks (NNs) have been shown to model successfully, the complex relationships between input feature sets of sensor signals and tool wear data. NNs have several properties that make them ideal for effectively handling noisy and even incomplete data sets. There are several NN paradigms which can be combined to model static and dynamic systems. Another powerful method of modeling noisy dynamic systems is by using hidden Markov models (HMMs), which are commonly employed in modern speech-recognition systems. The use of HMMs for TCM was recently proposed in the literature. Though the results of these studies were quite promising, no comparative results of competing methods such as NNs are currently available. This paper is aimed at presenting a comparative evaluation of the performance of NNs and HMMs for a TCM application. The methods are employed on exactly the same data sets obtained from an industrial turning operation.The advantages and disadvantages of both methods are described, which will assist the condition-monitoring community to choose a modeling method for other applications.

A Review of Indirect Tool Condition Monitoring Systems and Decision-Making Methods in Turning: Critical Analysis and Trends

“…The structure of hidden Markov model is shown in Figure 14 . Scheffer et al [ 60 ] implemented neural networks and hidden Markov models as a comparative study in monitoring of TW. In turning of aluminum alloy, a comprehensive comparison was performed, and neural network was found as capable of performing continuous estimations; however, if the problem to solve is defined well, hidden Markov properly estimates TW.…”

A probabilistic-based study on fused direct and indirect methods for tracking tool flank wear of Rene-108, nickel-based alloy

“…It was shown that the ease of training HMM with the Baum–Welch algorithm is the major advantage of the HMM over a neural network. 22 However, the drawback is that the HMM only works for classification purposes and cannot be used for continuous tracking of tool wear. For tracking a continuous state of a system, stochastic-based filters such as Kalman filter and particle filter were proposed.…”

Adaptive-VDHMM for prognostics in tool condition monitoring

“…HMMs have been implemented for health-state estimation in quite a number of literatures [3][4][6][7][8][9][10][11]. Although they offer a better connection between the model structure and the physical process than the more commonly used multilayer perceptron, there are two limitations for applying the conventional HMMs to TCM applications.…”

“…In the literature, the same working condition is selected for establishing and verifying the model [8][9][10]. However, in the industry, there will be a range of cutting conditions for machining a workpiece.…”