An Automated Approach to Enhance Multiscale Signal Monitoring of Manufacturing Processes

2015 ◽  
Vol 138 (5) ◽  
Author(s):  
Marco Grasso ◽  
Bianca Maria Colosimo
Keyword(s):  
Kernel Estimation ◽  
Optimal Number ◽  
Signal Decomposition ◽  
Data Driven ◽  
Manufacturing Processes ◽  
Density Functions ◽  
Decomposition Approach ◽  
Mode Decomposition ◽  
Manufacturing Applications ◽  
Original Information

Multiscale signal decomposition represents an important step to enhance process monitoring results in many manufacturing applications. Empirical mode decomposition (EMD) is a data driven technique that gained an increasing interest in this framework. However, it usually yields an-over decomposition of the signal, leading to the generation of spurious and meaningless modes and the possible mixing of embedded modes. This study proposes an enhanced signal decomposition approach that synthetizes the original information content into a minimal number of relevant modes via a data-driven and automated procedure. A criterion based on the kernel estimation of density functions is proposed to estimate the dissimilarities between the intrinsic modes generated by the EMD, together with a methodology to automatically determine the optimal number of final modes. The performances of the method are demonstrated by means of simulated signals and real industrial data from a waterjet cutting application.

A Data-driven Method for Identifying Intricate Trend Component Hidden in Measured Signal

2016 ◽  
Vol 15 (02) ◽  
pp. 1650020 ◽  
Author(s):  
X. X. Jiang ◽  
S. M. Li
Keyword(s):  
Euclidean Distance ◽  
Signal Decomposition ◽  
Data Driven ◽  
Measured Signal ◽  
Trend Component ◽  
Mode Function ◽  
Mode Decomposition ◽  
Simulated Time ◽  
Underlying Trend ◽  
Simulated Time Series

Identification of the latent trend component is a vital procedure for further evaluating the measured signal. Many methods have been presented to extract the trend component. However, some essential parameters used in these methods need to be selected depending on much prior knowledge or experience. To avoid the inherent flaws in current methods, we introduce a novel signal decomposition method called variational mode decomposition (VMD) to cope with this issue. Firstly, the parameters involved in VMD method are discussed according to the universal characteristics of trend component. Then, a novel data-driven method based on the iterative VMD is proposed to identify the intricate trend component. Moreover, a criterion called normalized euclidean distance (NED) is presented to evaluate the converging property of the proposed method. Finally, the simulated time series and the collected extensive experimental cases are employed to illustrate the effectiveness of the proposed method. The analysis results show that the proposed method delivers a good performance for the trend component extraction and outperforms the benchmark method, i.e., empirical mode decomposition (EMD)-based method, for which more than one intrinsic mode function (IMF) is regarded as the underlying trend component.

scholarly journals Data-driven Predictive Analysis for Smart Manufacturing Processes Based on a Decomposition Approach

2021 ◽  
Author(s):  
Mohammadhossein Ghahramani
Keyword(s):  
Semiconductor Manufacturing ◽  
Data Driven ◽  
Smart Manufacturing ◽  
Predictive Analysis ◽  
Manufacturing Processes ◽  
Decomposition Approach

<div>The dataset used in this work is obtained from a semiconductor factory, SECOM (Semiconductor Manufacturing) dataset and is publicly available online.</div>

scholarly journals A Novel Hybrid Signal Decomposition Technique for Transfer Learning Based Industrial Fault Diagnosis

2021 ◽  
Vol 5 (4) ◽  
pp. 37-53
Author(s):  
Zurana Mehrin Ruhi ◽  
Sigma Jahan ◽  
Jia Uddin
Keyword(s):  
Deep Learning ◽  
Fault Diagnosis ◽  
Transfer Learning ◽  
Industrial Revolution ◽  
State Of The Art ◽  
Signal Decomposition ◽  
Data Driven ◽  
Detection Model ◽  
Mode Decomposition ◽  
To Come

In the fourth industrial revolution, data-driven intelligent fault diagnosis for industrial purposes serves a crucial role. In contemporary times, although deep learning is a popular approach for fault diagnosis, it requires massive amounts of labelled samples for training, which is arduous to come by in the real world. Our contribution to introduce a novel comprehensive intelligent fault detection model using the Case Western Reserve University dataset is divided into two steps. Firstly, a new hybrid signal decomposition methodology is developed comprising Empirical Mode Decomposition and Variational Mode Decomposition to leverage signal information from both processes for effective feature extraction. Secondly, transfer learning with DenseNet121 is employed to alleviate the constraints of deep learning models. Finally, our proposed novel technique surpassed not only previous outcomes but also generated state-of-the-art outcomes represented via the F1 score.

scholarly journals Data-driven Predictive Analysis for Smart Manufacturing Processes Based on a Decomposition Approach

2021 ◽  
Author(s):  
Mohammadhossein Ghahramani
Keyword(s):  
Semiconductor Manufacturing ◽  
Data Driven ◽  
Smart Manufacturing ◽  
Predictive Analysis ◽  
Manufacturing Processes ◽  
Decomposition Approach

<div>The dataset used in this work is obtained from a semiconductor factory, SECOM (Semiconductor Manufacturing) dataset and is publicly available online.</div>

scholarly journals Quantifying the Impact of a Tsunami on Data-Driven Earthquake Relief Zone Planning in Los Angeles County via Multivariate Spatial Optimization

Geosciences ◽  
2021 ◽  
Vol 11 (2) ◽  
pp. 99 ◽  
Author(s):  
Yueqi Gu ◽  
Orhun Aydin ◽  
Jacqueline Sosa
Keyword(s):  
Los Angeles ◽  
A Priori ◽  
Safety Data ◽  
Spatial Optimization ◽  
Optimal Number ◽  
Multidisciplinary Optimization ◽  
Data Driven ◽  
Fault System ◽  
Design Data ◽  
The Impact

Post-earthquake relief zone planning is a multidisciplinary optimization problem, which required delineating zones that seek to minimize the loss of life and property. In this study, we offer an end-to-end workflow to define relief zone suitability and equitable relief service zones for Los Angeles (LA) County. In particular, we address the impact of a tsunami in the study due to LA’s high spatial complexities in terms of clustering of population along the coastline, and a complicated inland fault system. We design data-driven earthquake relief zones with a wide variety of inputs, including geological features, population, and public safety. Data-driven zones were generated by solving the p-median problem with the Teitz–Bart algorithm without any a priori knowledge of optimal relief zones. We define the metrics to determine the optimal number of relief zones as a part of the proposed workflow. Finally, we measure the impacts of a tsunami in LA County by comparing data-driven relief zone maps for a case with a tsunami and a case without a tsunami. Our results show that the impact of the tsunami on the relief zones can extend up to 160 km inland from the study area.

scholarly journals Identification of Granule Growth Regimes in High Shear Wet Granulation Processes Using a Physics-Constrained Neural Network

Processes ◽  
2021 ◽  
Vol 9 (5) ◽  
pp. 737
Author(s):  
Chaitanya Sampat ◽  
Rohit Ramachandran
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Real Time ◽  
Data Driven ◽  
Wet Granulation ◽  
Manufacturing Processes ◽  
High Shear ◽  
Process Data ◽  
Physical Constraints ◽  
Granule Growth

The digitization of manufacturing processes has led to an increase in the availability of process data, which has enabled the use of data-driven models to predict the outcomes of these manufacturing processes. Data-driven models are instantaneous in simulate and can provide real-time predictions but lack any governing physics within their framework. When process data deviates from original conditions, the predictions from these models may not agree with physical boundaries. In such cases, the use of first-principle-based models to predict process outcomes have proven to be effective but computationally inefficient and cannot be solved in real time. Thus, there remains a need to develop efficient data-driven models with a physical understanding about the process. In this work, we have demonstrate the addition of physics-based boundary conditions constraints to a neural network to improve its predictability for granule density and granule size distribution (GSD) for a high shear granulation process. The physics-constrained neural network (PCNN) was better at predicting granule growth regimes when compared to other neural networks with no physical constraints. When input data that violated physics-based boundaries was provided, the PCNN identified these points more accurately compared to other non-physics constrained neural networks, with an error of <1%. A sensitivity analysis of the PCNN to the input variables was also performed to understand individual effects on the final outputs.

scholarly journals Research on Fault Diagnosis of Gearbox with Improved Variational Mode Decomposition

Sensors ◽  
2018 ◽  
Vol 18 (10) ◽  
pp. 3510 ◽  
Author(s):  
Zhijian Wang ◽  
Junyuan Wang ◽  
Wenhua Du
Keyword(s):  
Fault Diagnosis ◽  
Signal To Noise Ratio ◽  
Optimal Number ◽  
Original Signal ◽  
Variational Mode Decomposition ◽  
Intrinsic Mode Functions ◽  
Multiple Fault ◽  
Mode Decomposition ◽  
Fault Features ◽  
Reconstructed Signal

Variational Mode Decomposition (VMD) can decompose signals into multiple intrinsic mode functions (IMFs). In recent years, VMD has been widely used in fault diagnosis. However, it requires a preset number of decomposition layers K and is sensitive to background noise. Therefore, in order to determine K adaptively, Permutation Entroy Optimization (PEO) is proposed in this paper. This algorithm can adaptively determine the optimal number of decomposition layers K according to the characteristics of the signal to be decomposed. At the same time, in order to solve the sensitivity of VMD to noise, this paper proposes a Modified VMD (MVMD) based on the idea of Noise Aided Data Analysis (NADA). The algorithm first adds the positive and negative white noise to the original signal, and then uses the VMD to decompose it. After repeated cycles, the noise in the original signal will be offset to each other. Then each layer of IMF is integrated with each layer, and the signal is reconstructed according to the results of the integrated mean. MVMD is used for the final decomposition of the reconstructed signal. The algorithm is used to deal with the simulation signals and measured signals of gearbox with multiple fault characteristics. Compared with the decomposition results of EEMD and VMD, it shows that the algorithm can not only improve the signal to noise ratio (SNR) of the signal effectively, but can also extract the multiple fault features of the gear box in the strong noise environment. The effectiveness of this method is verified.

Sign in / Sign up

Export Citation Format

Share Document