Wavelet-based identification of delamination defect in cmp (cu-low k) using nonstationary acoustic emission signal

Ganesan, Rajesh; Das, Tapas K.; Sikder, A. K.; Kumar, Ashok

doi:10.1109/tsm.2003.818975

Cited by 35 publications

(6 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Unlike the STFT, in CWT, time and frequency resolution can be independently controlled. 39 Hence in this study, instead of the STFT, a more advanced technique,CWT is applied. The CWT was used in this study to better observe the wear occurring in the cutting tool and prevent any loss of information in the time or frequency domain.…”

Section: Methodsmentioning

confidence: 99%

An industrially viable wavelet long-short term memory-deep multilayer perceptron-based approach to tool condition monitoring considering operational variability

Gudelek

Serin

Özbayoğlu

et al. 2022

Proceedings of the Institution of Mechanical Engineers, Part E:

View full text Add to dashboard Cite

Tool wear is a fundamental aspect of the machining process. Therefore, tool condition monitoring is of paramount importance to ensure part quality and avoid catastrophic damage. Tool wear has a direct correlation with the vibration emanating from the process; however, accurate prediction of tool wear indirectly from the vibration level is difficult because machining parameters such as cutting speed, depth of cut, and feed rate may vary continuously during an operation, depending on tool diameter, geometry, and material. These affect vibration levels as much as wear progress, which demands advanced intelligence that can adapt to variations in cutting conditions. This paper proposes a wavelet long-short term memory (WLSTM)-deep multilayer perceptron (DMLP)-based model, which utilizes the continuous wavelet transform for preprocessing of raw data, long-short term memory (LSTM) for extracting temporal information, and DMLP for regression of the tool wear. First, the model is evaluated by comparing it with other LSTM studies developed using the PHM 2010 dataset in the literature. Afterward, its industrial viability and adaptability performance to variations in cutting speed and tool diameter are assessed with several training scenarios. The results revealed auspicious performance in the proposed architecture’s potential in predicting tool wear under operational variability.

show abstract

Section: Methodsmentioning

confidence: 99%

An industrially viable wavelet long-short term memory-deep multilayer perceptron-based approach to tool condition monitoring considering operational variability

Gudelek

Serin

Özbayoğlu

et al. 2022

Proceedings of the Institution of Mechanical Engineers, Part E:

View full text Add to dashboard Cite

show abstract

“…As shown in Fig. Ganesan et al [38] analyzed the raw AE and COF signals, monitored and recorded during CMP, and filtered them using wavelet-based multiscale analysis and SPRT techniques to detect delamination and end point of copperlow-k system during CMP. The variance sequential probability ratio test (SPRT) method was adopted to analyze and filter the raw data.…”

Section: Advanced Signal Processingmentioning

confidence: 99%

Tribometrology of CMP Process

Gitis¹,

Mudhivarthi²

2007

Microelectronic Applications of Chemical Mechanical Planarization

View full text Add to dashboard Cite

NORM GITIS AND RAGHU MUDHIVARTHI INTRODUCTIONStudies of mechanical and chemical-mechanical surface finishing treatments have long been a part of tribology, a science of friction and wear (material removal). During chemical-mechanical polishing or planarization (CMP), surfaces of both the object being treated (e.g., the semiconductor wafer) and the finishing tool (e.g., the polishing pad), together with the polishing slurry between them, constitute a rather classical tribological system. This is sometimes referred to as a three-body interface [1-3], because it includes two solids in relative motion and the slurry containing abrasive particles at the interface. Modern CMP machines may have an abrasive pad conditioner rubbing the pad during wafer polishing, which keeps one of the surfaces in a state of constant change. A tribological system is typically characterized by classical parameters of the coefficient of friction between the surfaces and by their wear rate.The field of tribology has seen a number of theoretical models for various materials and interfaces [4]. Based on these models, the formulas for the calculation of both friction and wear have been derived. Unfortunately, these equations cannot be used directly by process and equipment design engineers unless the coefficient is known. Indeed, friction and wear properties of a tribological interface depend on macro-, micro-, and nanogeometry of the Microelectronic Applications of Chemical Mechanical Planarization, Edited by Yuzhuo Li

show abstract

“…For example, in biomedical signal processing, source signals are very weak, noisy, and non-stationary [1]. Further, in many other applications such as under water acoustic signal classification [2], speech signal processing [3], feature extraction for acoustic target recognition [4], and analysis of acoustic emission signals [5], the signals of interest are of highly nonstationary nature. Lanka (e-mail: nimesh@ee.pdn.ac.lk, wijayanthanet@ee.pdn.ac.lk,amanthi@ee.pdn.ac.lk,roshangodd@ee.pdn.ac.lk, mpb.ekanayake@ee.pdn.ac.lk, and jan@ee.pdn.ac.lk).…”

Section: Introductionmentioning

confidence: 99%

Extraction and classification of non-stationary acoustic signals via dynamic subspace filtering

Pollwaththage

Nettasinghe

Ratnayake

et al. 2013

2013 IEEE 8th International Conference on Industrial and Information Systems

View full text Add to dashboard Cite

This paper presents a dynamic, subspace based approach for extraction and classification of non-stationary acoustic signals under noisy conditions. The stationary subspace methods commonly used for noise removal take the whole signal into consideration while determining the signal subspace. This, for a non-stationary signal implies that spectral variations that occur through time are not taken into account. Thus, the overall signal subspace formulated will correspond to noise subspaces at certain points of time due to the highly non-stationary nature of the signal, resulting in noise leakage. The method proposed in this paper performs a subspace analysis dynamically at various stages of the observation period, enabling the signal subspace to evolve according to the non-stationary nature of the signal to be extracted. For signal classification problems, this enables the classifier to latch onto details about the non-stationarity in the signal which in turn provides valuable information with regard to the uniqueness of each signal class. Finally, an analysis is conducted on the resultant signals to prove the viability of the proposed methodology for signal extraction and classification problems under heavy noise conditions where the signals in concern are highly non-stationary.

show abstract

Wavelet-based identification of delamination defect in cmp (cu-low k) using nonstationary acoustic emission signal

Cited by 35 publications

References 24 publications

An industrially viable wavelet long-short term memory-deep multilayer perceptron-based approach to tool condition monitoring considering operational variability

An industrially viable wavelet long-short term memory-deep multilayer perceptron-based approach to tool condition monitoring considering operational variability

Tribometrology of CMP Process

Extraction and classification of non-stationary acoustic signals via dynamic subspace filtering

Contact Info

Product

Resources

About