Search

The performance and learning speed of the Cascade Correlation neural network (CasCor) may not be optimal because of redundant hidden units' in the cascade architecture and the tuning of connection weights. This study explores the limitations of CasCor and its variants and proposes a novel constructive neural network (CNN). The basic idea is to compute the input connection weights by generating linearly independent hidden units from the orthogonal linear transformation, and the output connection weights by connecting hidden units in a linear relationship to the output units. The work is unique in that few attempts have been made to analytically determine the connection weights on both sides of the network. Experimental work on real energy application problems such as predicting powerplant electrical energy, predicting seismic hazards to prevent fatal accidents and reducing energy consumption by predicting building occupancy detection shows that analytically calculating the connection weights and generating non-redundant hidden units improves the convergence of the network. The proposed CNN is compared with that of the state-of-the-art machine learning algorithms. The work demonstrates that proposed CNN predicts a wide range of applications better than other methods. [ABSTRACT FROM AUTHOR]

This paper presents the classical approximation scheme to investigate the velocity profile associated with the Falkner--Skan boundary-layer problem. Solution of the boundary-layer equation is obtained for a model problem in which the flow field contains a substantial region of strongly reversed flow. The problem investigates the flow of a viscous liquid past a semi-infinite flat plate against an adverse pressure gradient. Optimized results for the dimensionless velocity profiles of reverse wedge flow are presented graphically for different values of wedge angle parameter [beta] taken from 0 [less than or equal to] [beta] [less than or equal to] 2.5. Weighted residual method (WRM) is used for determining the solution of nonlinear boundary-layer problem. Finally, for [beta] = 0 the results of WRM are compared with the results of homotopy perturbation method. [DOI: 10.1115/1.3173763]

In this study, the influence of slip flow and temperature jump on the entropy generation rate are investigated in rectangular microducts. The Knudsen numbers are considered in the range between 0.001 and 0.1, and the aspect ratio lies between 0 and 1. The dimensionless governing equations are solved numerically using Chebyshev spectral collocation method, and the dimensionless velocity and temperature gradients are employed in the entropy generation model. The influences of the dimensionless numbers including Bejan number and irreversibility distribution ratio on the entropy generation rates are investigated and discussed through surface plots and contour diagrams. It is demonstrated that the minimum entropy generation rate exists corresponding to an optimal aspect ratio for each dimensionless number. This minimum entropy generation rate depends upon the nature of dimensionless numbers. [ABSTRACT FROM AUTHOR]

Background: Bangladesh lies in the global thalassemia belt, which has a defined mutational hot-spot in the betaglobin gene. The high carrier frequencies of beta-thalassemia trait and hemoglobin E-trait in Bangladesh necessitate a reliable DNA-based carrier screening approach that could supplement the use of hematological and electrophoretic indices to overcome the barriers of carrier screening. With this view in mind, the study aimed to establish a high resolution melting (HRM) curve-based rapid and reliable mutation screening method targeting the mutational hot-spot of South Asian and Southeast Asian countries that encompasses exon-1 (c.1 - c.92), intron-1 (c.92 + 1 - c.92 + 130) and a portion of exon-2 (c.93 - c.217) of the HBB gene which harbors more than 95% of mutant alleles responsible for beta-thalassemia in Bangladesh. Results: Our HRM approach could successfully differentiate ten beta-globin gene mutations, namely c.79G > A, c.92 + 5G > C, c.126_129delCTTT, c.27_28insG, c.46delT, c.47G > A, c.92G > C, c.92 + 130G > C, c.126delC and c. 135delC in heterozygous states from the wild type alleles, implying the significance of the approach for carrier screening as the first three of these mutations account for ~85% of total mutant alleles in Bangladesh. Moreover, different combinations of compound heterozygous mutations were found to generate melt curves that were distinct from the wild type alleles and from one another. Based on the findings, sixteen reference samples were run in parallel to 41 unknown specimens to perform direct genotyping of the beta-thalassemia specimens using HRM. The HRM-based genotyping of the unknown specimens showed 100% consistency with the sequencing result. Conclusions: Targeting the mutational hot-spot, the HRM approach could be successfully applied for screening of beta-thalassemia carriers in Bangladesh as well as in other countries of South Asia and Southeast Asia. The approach could be a useful supplement of hematological and electrophortic indices in order to avoid false positive and false negative results. [ABSTRACT FROM AUTHOR]