Fuzzy controlled backpropagation neural network for tool condition monitoring in face milling

Abstract

The performance of a fuzzy controlled backpropagation neural network has been studied to predict the tool wear in a face milling process based on simple process parameters and sensor signal features. The results show the potentiality of the method in comparison to the standard backpropagation neural network and one of its variants. The speed of convergence, accuracy of prediction and total time of system development make fuzzy controlled backpropagation an attractive technique amenable for online tool condition monitoring.

Estimation of tool wear during CNC milling using neural network-based sensor fusion

“…An ANN provides the advantage of non-linear mapping of signal features to tool wear that can be learnt from the training data when explicit mathematical mapping model behind it is of less importance and hence may be considered as a black-box [2,19]. Note that ANNs have been used in TCM earlier as well [3,4,6,7,[10][11][12]15,27], though mostly for the turning process and with a limited number of machining signals.…”

“…But high cost and inconsistency due to variation in illumination have prevented this method from being implemented in the industry. A more economic proposition is to use an indirect method of monitoring tool wear from measured signals (which are affected by tool condition) like cutting force [4,5,8,10,11,12], machine vibration [8], motor load current [26], acoustic emission (AE) from the machining zone [16,17,25] or various combinations of these signals [8]. In application specific domains, this indirect method of monitoring works reasonably well.…”

“…Some of the recent works [11,12] have used cutting force signal and Artificial Neural Network (ANN) in TCM for face milling process. But cutting force based TCM is a costly proposition due to sensor cost and mounting problems.…”

“…Therefore these systems are not guaranteed to work satisfactorily for a semi-intermittent process like grinding or a fully intermittent process like milling. Recent attempts in developing TCM for drilling [13], end milling [9,24], and face milling [11,12] lack the data fusion strategies in the true sense proposed in some innovative works [3,6,23], because a particular signal may not work well for a particular machining process. As for example, use of AE has been reported as not being that helpful for processes like milling or grinding [17,18].…”

Condition monitoring of face milling tool using K-star algorithm and histogram features of vibration signal

“…So there is a need of studying the nature of the signal and their relationship with the tool condition, especially for an intermittent cutting process like face milling. Dutta et al [9] investigated and compared the performance of fuzzy based neural network with the standard back propagation neural network for tool condition monitoring during face milling process using vibration and cutting force signals. They concluded that the proposed method is faster in computational steps and effectively applicable for on-line TCM system.…”