An integrated supervised learning solution for monitoring process mean vector

When an out-of-control condition is detected by a control chart, a search begins to identify and eliminate the source(s) of the signal. Identification of the time when a process first changed is an important step in root cause analysis which helps a process engineer to eliminate the source(s) of assignable cause effectively. The time when a change takes place in the process is referred to as the change point. In multivariate environment, since there is more than one variable involved, then root cause analysis is relatively harder compared to the case of univariate because it is not clear exactly which variable has contributed to the out-of-control condition and in what direction its mean has shifted. Hence, a procedure that identifies the change point, performs diagnostic analysis, and specifies the direction of the shift in the mean of the contributing variable(s) all simultaneously could help to conduct root cause analysis effectively. Although different multivariate methods exist in the literature that allow to either estimate change point in the process mean vector or identify the contributing variables leading to the out-of-control condition, but in this research, an integrated supervised learning solution is proposed, which helps to (1) detect of an out-of-control condition, (2) identify the change point leading to shift in the mean vector, (3) specify the variable(s) contributing to the out-of-condition, and (4) identify the direction of the shift in the mean of each contributing variable simultaneously. A real case study is used to evaluate and compare the performance of the proposed integrated approach to existing methods in the literature.     

Simultaneous polarographic determination of isoniazid and rifampicin by differential pulse polarography method and support vector regression

A differential pulse polarography (DPP) for the simultaneous determination of isoniazid and rifampicin was proposed. Under optimum experimental conditions (pH = 7, scan rate = 10 mV/s, pulse amplitude = −50 mV), serious overlapping polarographic peaks were observed in the mixture of these compounds. In this study, support vector regression (SVR) was applied to modeling the overlapped polarograms. Furthermore, a comparison was made between the performance of SVR and partial least square (PLS) on data set. The experimental calibration matrix was designed with 30 mixtures of these compounds. Calibration graphs were linear in the range of 6 × 10⁻⁸-10⁻⁴ and 10⁻⁷-10⁻⁴ M for isoniazid and rifampicin, respectively. The results demonstrated that SVR is a good well-performing alternative for the analysis and modeling of DPP data than the commonly applied PLS technique.     

Structure,texture and surface acidity studies of a series of mixed zinc–aluminum (60–90 molar % Al) phosphate catalysts

A series of mixed zinc–aluminum phosphate (ZnAlP) catalysts containing 40–90 aluminum molar % were synthesized by a coprecipitation method and characterized by nitrogen adsorption–desorption, X‐ray diffraction, FTIR spectroscopy, thermogravimetric analysis (TGA), differential thermal analysis (DTA), diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) and temperature programmed desorption (TPD) of ammonia. The presence of aluminum greatly affected the surface properties of Zn₃(PO₄)₂ by delaying the crystallization process of Zn₃(PO₄)₂. All amorphous samples were shown to be mesoporous and they contained two types of aluminum surface hydroxyl groups and one type of phosphorus hydroxyl group, as shown by DRIFT spectra. The specific surface area and the acidity of ZnAlP increased on increasing the aluminum content. On the other hand, a great difference in the texture and the concentration of surface acid sites was found by changing the precipitating agent and calcination temperature. Thus these factors also play an important role in the final properties of these catalysts. © 2001 Society of Chemical Industry     

Intelligent Human-Machine Interaction Based on Dynamic Bayesian Networks Probabilistic Intention Recognition

In this article, a novel human–machine interaction based on the machine intention recognition of the human is presented. This work is motivated by the desire that intelligent machines as robots imitate human–human interaction, that is to minimize the need for classical direct human–machine interface and communication. A philosophical and technical background for intention recognition is discussed. Here, the intention–action–state scenario is modified and modeled by Dynamic Bayesian Networks to facilitate for probabilistic intention inference. The recognized intention, then, drives the interactive behavior of the machine such that it complies with the human intention in light of the real state of the world. An illustrative example of a human commanding a mobile robot remotely is given and discussed in details.     

Analysis of a dephased complex-coupled distributed-feedback laser by the power-series method

A novel power-series method to solve the coupled-wave equations is introduced. The method is used to calculate the threshold gain margins of a complex-coupled distributed-feedback laser as functions of the ratio of gain coupling to index coupling (|kappa(g)|/|kappa(n)|) and of the phase difference between the index and the gain gratings. For coupling coefficient |kappa|l < ., the laser shows a mode degeneracy at specific values of the ratio |kappa(g)|/|kappa(n)| for cleaved facets. At phase differences pi/2 and 3pi/2 between the gain and the index gratings, an antireflection-coated complex-coupled laser becomes multimode, and a different mode starts to lase. The effect of facet reflectivity (both magnitude and phase) on the gain margin of a complex-coupled DFB laser is also investigated. Although the gain margin varies slowly with the magnitude of the facet's reflectivity, it shows large variations as a function of the phase. Spatial hole burning was found to be minimum at phase difference npi, n =, ..., and maximum at phase differences pi/2 and 3pi/2.     

Threshold analysis of a chirped-grating distributed-feedback laser with the power series method

A new and efficient method for analyzing a chirped-grating distributed-feedback (DFB) laser is presented. We show that coupled-wave equations can be solved by use of the power series method. The single-mode gain margin of a linearly chirped-grating DFB laser is calculated for different chirping factors and coupling constants. We found that clearly optimum chirping exists for which the single-mode gain margin is maximum. The gain margins were also calculated for different positions of the cavity center. The effect of facet reflectivities and their phases on the gain margin was investigated. We found that the gain margin is maximum and the spatial hole burning is minimum for the cavity center at the middle of the laser cavity.