In the general frame of process surveillance, principal component analysis (PCA) has been often selected due to its simplicity and ability to capture the linear relations between the stationary process variables. However, the method showed limitations when dealing with industrial data that generally presents non-linear and multiscale features. The approach proposed in this study rests on the modelling using non-linear PCA coupled with artificial neural networks (ANNs) to extract the non-linear inter-correlation between variables and on the wavelet analysis to decompose each sensor signal into a set of coefficients at different scales. The contribution of each variable for each scale is then collected in separated matrices and a non-linear PCA model is constructed for each matrix. The proposed approach is applied to fault detection of pollution parameters affecting the region of Annaba in Algeria. The performance of the approach is then illustrated and compared with those of classic PCA and multiscale PCA (MSPCA).
- Inderscience Publishers
- Non-linear multiscale principal component analysis for fault ...
Emerging Asian Economies Drive the Increase in World Energy Use from 2015 to 2040
World energy consumption is projected to increase by 28% by 2040, according to the International Energy Outlook 2017 (IEO2017), released today by the U.S. Energy Information Administration (EIA). Most of the world`s growth in energy demand is projected to take place in countries outside of the Organization for Economic Cooperation and Development (OECD). China and the other non-OECD Asia nations alone account for more than 60% of the projected increase in world energy demand (Figure 1). "Transportation energy...
Application note issue #5 - Cement Analysis August 2017
Cement analysis using FP program (Fundamental Parameters) Cement is a blend of several minerals. It is critical to control the elemental composition and properties such as strength, setting, time and color. ED-XRF is being used to analyze the raw components, raw meal, clinker, and the final cement product (Na, Mg, Al, Si, S, K, Ca, and Fe in Cement, Clinker, and Raw Meal).XRF is at the heart of the control of the production process in any modern cement works and is central to the ability of the cement maker...
Analysis of precious metals in jewelry
AbstractFor thousands of years, man learned the craft of making jewelry from gold. Gold is a noble metal which does not react with the air and has a shiny yellow color.Over the years, alloying of the gold was necessary to improve its properties since pure gold is too soft to be used for jewelry. Alloying of the gold is usually by low-cost metals such as copper, zinc, silver or nickel. Since different metals have different financial value, which determines the value of the jewelry, it is important to accurately...
Quantitative Analysis of Geological Samples- X-Calibur
Abstract Quantitative and qualitative analysis of geological samples (rocks and river sediments) was performed on Xenemetrix EDXRF Benchtop Spectrometer System, model X-Calibur Objective: Perform XRF measurements on certified geological reference materials to build calibration curves for the major oxides (Na2O, MgO, SiO2, Al2O3, SiO2, P2O5, SO3, K2O, CaO, TiO2, MnO and Fe2O3) and for trace elements Cr, Co, Ni and Pb in these samples. Quantify geological samples of unknown concentrations with respect to...
Quantitative analysis of iron in flour
Abstract A quantitative analysis of Iron in flour via Xenemetrix EDXRF Laboratory Spectrometer System was performed by model EX-3600M. The minimum detection limit for Fe in flour matrix was determined to 0.8 ppm at 3 sigma. Objective To design an easy method by using EDXRF technique for quick and convenient measurements of Fe in flour at the level of tens of ppm. Background EDXRF is a fast and non-destructive technique that can quantify any type of samples in solid, powder or liquid form within a few...