Grinding Wheel Condition Monitoring With Hidden Markov Model-Based Clustering Methods

Liao, T. Warren; Hua, Guogang; Qu, Jun; Blau, PJ

doi:10.1080/10910340600996175

Cited by 41 publications

(18 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The HMMs are then identified with the parameter set k ¼ (p, A, c, m, R). Once the number and topology of HMM states is chosen as discussed above, the parameters can be iteratively estimated by the Expectation Maximization (EM) algorithm, and then the tool state is estimated by the Viterbi Algorithm (Liao et al, 2006).…”

Section: Hmm For Tool Wear State Classificationmentioning

confidence: 99%

“…Hidden Markov Models (HMMs) have drawn attention in tool condition monitoring lately (Errtunc et al, 2001;Wang et al, 2002;Randall et al, 2003;Baruah and Chinnam, 2005;Liao et al, 2006) due to their excellent representation of local dynamics of non-stationary signals and reasoning property for state estimation (Rabiner, 1989;Li and Douglas, 2003). These studies achieved high estimation accuracy across changing cutting conditions with HMM classification.…”

Section: Introductionmentioning

confidence: 98%

See 1 more Smart Citation

A Comparative Study of Feature Selection for Hidden Markov Model-Based Micro-Milling Tool Wear Monitoring

Zhu

Gao

Wong

2008

Machining Science and Technology

View full text Add to dashboard Cite

& This paper presents a discriminant feature selection approach for hidden Markov model (HMM) modeling of micro-milling tool conditions. The approach is compared with other popular feature selection methods such as principal component analysis (PCA) and automatic relevance determination (ARD) according to their HMM classification rate. In tool condition monitoring (TCM), there are a lot of features that contain redundant information or less sensitive to tool state discrimination. These features are expected to be deleted for less computation and more robust modeling of tool conditions. Fisher's linear discriminant analysis (FDA) is modified for this purpose. The FDA is generally used for classification, and the features are mapped to another space and lose their physical meanings. In the modified discriminant feature selection, the features are selected in the original feature space by maximizing tool state separation and ranked by their separation ability between different tool states. Experimental results from both micro-milling of copper and steel under different working conditions indicate that the FDA is superior to both PCA and ARD for feature selection in HMM's classification. The reasons behind these differences are also discussed.

show abstract

Section: Hmm For Tool Wear State Classificationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 98%

A Comparative Study of Feature Selection for Hidden Markov Model-Based Micro-Milling Tool Wear Monitoring

Zhu

Gao

Wong

2008

Machining Science and Technology

View full text Add to dashboard Cite

show abstract

“…AE has been successfully used in detection of spark and contact in grinding and wheel dimensional characterization [44]. Liao et al [45,46] investigated AE-based grinding wheel condition monitoring, using discrete wavelet decomposition procedure to extract discriminate features from raw AE signals and an adaptive genetic clustering algorithm to classify the wheel state into either sharp or dull. Lee et al [82] discuss the use of acoustic emission (AE) as a monitoring technique at the precision scale for a variety of precision manufacturing processes including grinding, chemical-mechanical planarization (CMP) and ultraprecision diamond turning.…”

Section: Grindingmentioning

confidence: 99%

Quality and Inspection of Machining Operations: Review of Condition Monitoring and CMM Inspection Techniques — 2000 to Present

Roth

Mears

Djurdjanović

et al. 2007

ASME 2007 International Manufacturing Science and Engineering Conference

View full text Add to dashboard Cite

In order to consistently produce quality parts, many aspects of the manufacturing process must be carefully monitored, controlled, and measured. The methods and techniques by which to accomplish these tasks has been the focus of numerous studies in recent years. With the rapid advances in computing technology, the complexity and overhead that can be feasibly incorporated in any developed technique has dramatically improved. Thus, techniques that would have been impractical for implementation just a few years ago can now be realistically applied. This rapid growth has resulted in a wealth of new capabilities for improving part and process quality and reliability. In this paper, overviews of recent advances that apply to machining are presented. Moreover, due to the relative significance of two particular machining aspects, this review focuses specifically on research publications pertaining to using tool condition monitoring and coordinate measurement machines to improve the machining process. Tool condition has a direct effect on part quality and is discussed first. The application of tool condition monitoring as it applies to turning, drilling, milling, and grinding is presented. The subsequent section provides recommendations for future research opportunities. The ensuing section focuses on the use of coordinate measuring machines in conjunction with machining and is subdivided with respect to integration with machining tools, inspection planning and efficiency, advanced controller feedback, machine error compensation, and on-line tool calibration, in that specific order and concludes with recommendations regarding where future needs remain.

show abstract

“…Most frequently a combination of data exploration methods is used for all of the needed procedures. Liao (2010) and Liao et al (2006Liao et al ( , 2007Liao et al ( , 2008 investigated grinding wheel condition monitoring during surface grinding of ceramic materials with a resin-bonded diamond wheel using only the acoustic emission (AE) signal generated by the process. They applied different methods of the AE signal feature extraction and selection in time and frequency domain as well as different wheel state statistical and artificial intelligence classification techniques.…”

Section: Introduction and Related Workmentioning

confidence: 99%

“…The quality of classification expressed as a percent of right classifications varied from 70 to 100 %. The best results (even up to 100 %) were obtained for higher values of the specific material removal rates using the discrete wavelet decomposition of the AE signal and different methods of cluster analysis based on a distance matrix generated with the aid of the hidden Markov model (Liao et al 2006). Liu et al (2005) applied the AE signal measurement and the wavelet packet analysis for its feature extraction to reveal grinding burn while grinding a work-piece made of Inconel 718 with an electro corundum grinding wheel.…”

Section: Introduction and Related Workmentioning

confidence: 99%

The dominance-based rough set approach to cylindrical plunge grinding process diagnosis

Piłacińska

2016

J Intell Manuf

View full text Add to dashboard Cite

The dominance-based rough set approach is proposed as a methodology for plunge grinding process diagnosis. The process is analyzed and next its diagnosis is considered as a multi-criteria decision making problem based on the modelling of relationships between different process states and their symptoms using a set of rules induced from measured process data. The development of the diagnostic system is characterized by three phases. Firstly, the process experimental data is prepared in the form of a decision table. Using selected methods of signal processing, each process running is described by 17 process state features (condition attributes) and 5 criteria evaluating process state and results (decision attributes). The semantic correlation between all the attributes is modelled. Next, the phase of condition attributes selection and knowledge extraction are strictly integrated with the phase of the model evaluation using an iterative approach. After each loop of the iterative feature selection procedure the induction of rules is conducted using the VC-DomLEM algorithm. The classification capability of the induced rules is carried out using the leave-one-out method and a set of measures. The classification accuracy of individual models is in the range of 80.77-98.72 %. The induced set of rules constitutes a classifier for an assessment of new process run cases.

show abstract

Grinding Wheel Condition Monitoring With Hidden Markov Model-Based Clustering Methods

Cited by 41 publications

References 31 publications

A Comparative Study of Feature Selection for Hidden Markov Model-Based Micro-Milling Tool Wear Monitoring

A Comparative Study of Feature Selection for Hidden Markov Model-Based Micro-Milling Tool Wear Monitoring

Quality and Inspection of Machining Operations: Review of Condition Monitoring and CMM Inspection Techniques — 2000 to Present

The dominance-based rough set approach to cylindrical plunge grinding process diagnosis

Contact Info

Product

Resources

About