Chatter detection in milling machines by neural network classification and feature selection

Abstract

In modern industry, milling is an important tool when a high material removal rate is required. Chatter detection in this situation is a crucial step for improving surface quality and reducing both noise and rapid wear of the cutting tool. This paper proposes a new methodology for the chatter detection in computer numerical control milling machines. This methodology is based on vibratory signal analysis and artificial intelligence. The methodology consists of five major steps:(1) data acquisition, (2) signal processing, (3) features generation, (4) features selection and (5) classification. As chatter components occur around system resonance frequencies, a multiband resonance filtering method is proposed at the processing step. The process is then followed by envelope analysis. This allows the signal-to-noise ratio to be increased and the sensitivity of generated features to be increased. Extracted features are then ranked based on their entropy in which only best features are selected and presented to the system for classification. At the classification step, the selected features are classified into two classes: stable and unstable utilizing neural networks. Two neural network approaches, radial basis function and multi-layer perceptrons, are tested. The developed approach is applied for chatter detection in a Huron K2X10 milling machine. This approach is tested on a milling machine at different depths of cut and various rotational speeds. Discussions are made and the results confirm the accuracy of the proposed methodology.

Spatial modeling of PM2.5 concentrations with a multifactoral radial basis function neural network

“…While performances of geostatistic interpolation, remote sensing retrieval, and air dispersion modeling are greatly influenced by the well-defined physical-chemical mechanisms of PM 2.5 pollution which have not been clearly disclosed so far, RBF neural network, due to the prominent self-learning ability and high simulation accuracy, has recently shown to be suitable for exploring the undefined relationships between PM 2.5 concentration and its influencing factors (Fang et al 2011;Lamraoui et al 2013;Zhao and Hasan 2013;Losser et al 2014). In studies that rely on RBF neural networks, variables were excessively limited to traffic emission, air pollutants' concentration, and meteorological factors such as temperature and pressure (Pérez et al 2000;Fang et al 2011;Zheng and Shang 2013;Losser et al 2014).…”

Experimental validation of gas leak detection in screw thread connections of galvanized pipe based on acoustic emission and neural network

“…Some studies combined SVM and signal processing technology to identify leaks, such as SVM and local mean decomposition method, fuzzy SVM, and wavelet packet denoising . Some researchers reported that ANN has better performance in dealing with the complex nonlinear problems in AE‐based leak detection . Pressure at inlet and outlet and AE signals inside the pipe were used to train the ANN‐based leak classifier.…”

“…24,25 Some researchers reported that ANN has better performance in dealing with the complex nonlinear problems in AE-based leak detection. 26 Pressure at inlet and outlet 27 and AE signals inside the pipe 28 were used to train the ANN-based leak classifier. Further, momentum term and self-adaptive learning rate were adopted to solve the local optimal convergence of back propagation (BP) neural network.…”

Evaluación De La Estabilidad Dinámica Del Fresado a Través Del Análisis De Perfiles De Rugosidad

“…La detección del fenómeno de las vibraciones regenerativas es un aspecto muy importante a considerar en la obtención de una buena calidad superficial en la pieza, la reducción del ruido y la disminución de desgaste de la herramienta de corte (Lamraoui et al, 2015). Ko y Shaw (2009) afirman que las vibraciones regenerativas que afectan el sistema permiten determinar el efecto regenerativo en la estabilidad dinámica del sistema y diferenciarlo del efecto de las vibraciones forzadas producidas por la frecuencia de entrada de dientes.…”