Tool wear monitoring using artificial neural network based on extended Kalman filter weight updation with transformed input patterns

Purushothaman, S.

doi:10.1007/s10845-009-0249-y

Cited by 29 publications

(10 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There has been significant interest in tool condition monitoring in the recent past (Byrne et al 1995;Chao and Hwang 1997;Jemielniak et al 1998;Sick 2002;Rehorn et al 2005;Audy 2006;Wang et al 2008;Purushothaman 2009). Various indirect methods for tool condition monitoring (TCM), which use a pattern in sensor data to detect a failure mode (Byrne et al 1995;Scheffer and Heyns 2004;Heyns 2007), have been tried and tested by modeling the correlation between tool wear and sensory signals, namely the cutting force, torque, current, power, vibration, acoustic emission and airborne sound pressure acquired in machining processes.…”

Section: Introductionmentioning

confidence: 99%

Tool wear monitoring and selection of optimum cutting conditions with progressive tool wear effect and input uncertainties

et al. 2009

View full text Add to dashboard Cite

One of the big challenges in machining is replacing the cutting tool at the right time. Carrying on the process with a dull tool may degrade the product quality. However, it may be unnecessary to change the cutting tool if it is still capable of continuing the cutting operation. Both of these cases could increase the production cost. Therefore, an effective tool condition monitoring system may reduce production cost and increase productivity. This paper presents a neural network based sensor fusion model for a tool wear monitoring system in turning operations. A wavelet packet tree approach was used for the analysis of the acquired signals, namely cutting strains in tool holder and motor current, and the extraction of wear-sensitive features. Once a list of possible features had been extracted, the dimension of the input feature space was reduced using principal component analysis. Novel strategies, such as the robustness of the developed ANN models against uncertainty in the input data, and the integration of the monitoring information to an optimization system in order to utilize the progressive tool wear information for selecting the optimum cutting conditions, are proposed and validated in manual turning operations. The approach is simple and flexible enough for online implementation.

show abstract

Section: Introductionmentioning

confidence: 99%

Tool wear monitoring and selection of optimum cutting conditions with progressive tool wear effect and input uncertainties

et al. 2009

View full text Add to dashboard Cite

show abstract

“…11 Another research monitored the condition of the tool in a turning operation by using artificial neural network (ANN) based on extended Kalman filter, finding that the ANN trained with transformed tool wear patterns gave better results in terms of improved classification performance in less iteration, when compared with the results of the ANN trained without transforming the dimensions of the input patterns to a lower dimension. 12 Meanwhile, a study compared the performance of backpropagation neural network (BPNN) and radial basis function (RBF) network in predicting the flank wear of high speed steel drill bits for drilling holes on mild steel and copper workpieces. 13 It was observed that the performance of the RBFN failed to match that of the BPNN when the network complexity and the amount of data available were the constraining factors.…”

Section: Introductionmentioning

confidence: 99%

Investigation on wear prediction model of dental restoration material based on ensemble learning

Zheng

Liu

2014

Materials Research Innovations

View full text Add to dashboard Cite

The wear test was performed with TC4 alloy, by changing the normal bite force, sliding frequency and cycles, in artificial saliva. Taking the two results for testing samples and the others for training samples, the radial basis function (RBF) and multilayer perceptron (MLP) neural network model and least mean square (LMS) and K* model were built for predicting wear loss respectively. Then an ensemble learning model was built which integrated all the single models based on the weight determined by mean absolute error. Compared with the testing results, it was obtained that the error for ensemble learning model was between 3 and 4%. Also, its prediction error rate reduced over 50% for the tenth group data, which embodied good stability and high precision on predicting the wear loss for dental restorative material.

show abstract

“…In TCM the measurement of its orthogonal components often provides the necessary cutting force information as can be found in the work of Jemielniak et al (1998), Purushothaman (2010) and Sharma et al (2008). This work however compares TCM based on orthogonal forces to the one based on measurement of only the unidirectional strain signal.…”

Section: Introductionmentioning

confidence: 99%

Comparing orthogonal force and unidirectional strain component processing for tool condition monitoring

2012

View full text Add to dashboard Cite

Signal processing using orthogonal cutting force components for tool condition monitoring has established itself in literature. In the application of single axis strain sen sors however a linear combination of cutting force compo nents has to be processed in order to monitor tool wear. This situation may arise when a single axis piezoelectric actuator is simultaneously used as an actuator and a sensor, e.g. its vibration control feedback signal exploited for monitoring purposes. The current paper therefore compares processing of a linear combination of cutting force components to the reference case of processing orthogonal components. Recon struction of the dynamic force acting at the tool tip from signals obtained during measurements using a strain gauge instrumented tool holder in a turning process is described. An application of this dynamic force signal was simulated on a ltermodel of that tool holder that would carry a selfsensing actuator. For comparison of the orthogonal and unidirectional force component tool wear monitoring strategies the same timedelay neural network structure has been applied. Wear sensitive features are determined by wavelet packet analysis to provide information for tool wear estimation. The proba bility of a difference less than 5 percentage points between the ank wear estimation errors of above mentioned two process ing strategies is at least 95 %. This suggests the viability of simultaneous monitoring and control by using a selfsensing actuator.

show abstract

Tool wear monitoring using artificial neural network based on extended Kalman filter weight updation with transformed input patterns

Cited by 29 publications

References 35 publications

Tool wear monitoring and selection of optimum cutting conditions with progressive tool wear effect and input uncertainties

Tool wear monitoring and selection of optimum cutting conditions with progressive tool wear effect and input uncertainties

Investigation on wear prediction model of dental restoration material based on ensemble learning

Comparing orthogonal force and unidirectional strain component processing for tool condition monitoring

Contact Info

Product

Resources

About