Back Propagation Wavelet Neural Network Based Prediction of Drill Wear from Thrust Force

Abstract

<p class="MsoNormal" style="margin: 0cm 0cm 4pt; line-height: 12pt; mso-line-height-rule: exactly;"><span style="font-size: 10pt; mso-bidi-font-size: 11.0pt;" lang="EN-US"><span style="font-family: Times New Roman;">The fast monitoring of tool wears by using various Cutting signals and the prediction models developed rapidly in recent years. Comparatively, various wear forecast models based on artificial neural networks (ANN) perform much better in accuracy and speediness than the conventional prediction models. Combining the prominent dynamic properties of back propagation neural network (BPNN) with the enhanced ability of a wavelet neural network (WNN) in mapping nonlinear functions, a Back propagation wavelet neural network (BPWNN) is newly established to perform prominent prediction of drill wear. In this work, a multilayer BPWNN with wavelet algorithm has been applied to predict the average wear of a K10 carbide drill bit for drilling on a high silicon aluminum work piece. Mean value of the thrust force, cutting torque, and drilling depth, spindle speed and feed-rate are inputs to the network, and drill wear is the output. Drilling experiments have been carried out over a wide range of cutting conditions and the effects of drill wear, cutting conditions (spindle speed, drilling depth and feed-rate) on the thrust force and cutting torque have been investigated. Performance of BPWNN has proved to be satisfactory by experimental result. The accuracy of the prediction of drill wear using BPWNN is found to be better than using BPNN, and that BPWNN can learn the pattern faster compared to BPNN and could be used advantageously in online drill wear monitoring and prediction. </span></span></p

Estimating high precision hole diameters of aerospace alloys using artificial intelligence systems: a comparative analysis of different techniques

“…As reported in the literature, components of machining forces have presented a good correlation with geometrical errors [5,7], burr size [25] and tool wear and tool failure [31][32][33], so the lowest error in estimation diameters verified when FZ was used input in MLP network in this study contributed to support such correlation. The results in Fig.…”

“…In drilling operation, the artificial neural networks have been employed in monitoring of drill wear with aid of sensors that can acquire many signals. The various types of signals employed include those produced during machining and obtained by dynamometer loads [7], electrical current obtained by the application of Hall-effect sensors on electric motors [8], vibrations [9] and also a combination of these and other sensors such as accelerometers and acoustic emission sensors [10]. The status of the wear is analysed with base on input variables such as cutting speed, feed rate, drill diameter, drill geometry among others.…”

Correcting geometric deviations of CNC Machine-Tools: An approach with Artificial Neural Networks

“…Due to these facts, we can infer that the processed BP models got stuck at a local minimum and the generated models were not able to solve the problem in hand. BP algorithm limitations have already been discussed by a number authors as Schalkoff [40], Jian et al [45], Özel and Davim [46] and Yang et al [25].…”

“…Tandon and El-Mounayri [23] studied the cutting force for developed a model to predict parameters for end milling. On the other hand, Panda et al [24] and Yang et al [25] used ANN for prediction of drills wear out through data capturing by sensors, measuring and controlling of machining forces. Briceno et al [26] developed an ANN with back-propagation (BP) training to estimate milling parameters, Zuperl and Cus [27] applied a similar network to predict the three cutting forces (Fx, Fy, and Fz) in the machining of molds.…”

Data processing from manufacturing systems to decision support systems: propositions of alternative design approaches

“…Another kind of DSS refers to the field of artificial intelligence via fuzzy logic (Mok, 2009) (Garavalli, 1999) or artificial neural networks. The applications are wide, from the definition of a sensors fault tolerant control (Magdy, 2009) to the prediction of drill wear from thrust force and cutting torque signals (Yang et al, 2009). These systems need a learning phase, generally obtained with an observation of the states of the system along the production (Pierreval, 1992).…”