A numerical study is presented for the fully developed two-dimensional laminar flow of viscous incompressible fluid through a curved square duct for the constant curvature δ = 0.1. In this paper, a spectral-based com...A numerical study is presented for the fully developed two-dimensional laminar flow of viscous incompressible fluid through a curved square duct for the constant curvature δ = 0.1. In this paper, a spectral-based computational algorithm is employed as the principal tool for the simulations, while a Chebyshev polynomial and collocation method as secondary tools. Numerical calculations are carried out over a wide range of the pressure gradient parameter, the Dean number, 100 ≤ Dn ≤ 3000 for the Grashof number, Gr, ranging from 100 to 2000. The outer wall of the duct is treated heated while the inner wall cooled, the top and bottom walls being adiabatic. The main concern of the present study is to find out the unsteady flow behavior i.e. whether the unsteady flow is steady-state, periodic, multi-periodic or chaotic, if Dn or Gr is increased. It is found that the unsteady flow is periodic for Dn = 1000 at Gr = 100 and 500 and at Dn = 2000, Gr = 2000 but steady-state otherwise. It is also found that for large values of Dn, for example Dn = 3000, the unsteady flow undergoes in the scenario “periodic→chaotic→periodic”, if Gr is increased. Typical contours of secondary flow patterns and temperature profiles are also obtained, and it is found that the unsteady flow consists of single-, two-, three- and four-vortex solutions. The present study also shows that there is a strong interaction between the heating-induced buoyancy force and the centrifugal force in a curved square passage that stimulates fluid mixing and consequently enhance heat transfer in the fluid.展开更多
Food security for the 7 billion people on earth requires minimizing crop damage by timely detectionofdiseases.Most deep learningmodels forautomated detectionof diseases in plants suffer fromthe fatal flaw that once te...Food security for the 7 billion people on earth requires minimizing crop damage by timely detectionofdiseases.Most deep learningmodels forautomated detectionof diseases in plants suffer fromthe fatal flaw that once tested on independent data,their performance drops significantly.This work investigates a potential solution to this problem by using segmented image data to train the convolutional neural network(CNN)models.As compared to the F-CNN model trained using full images,S-CNN model trained using segmented imagesmore than doubles in performance to 98.6%accuracy when tested on independent data previously unseen by the models even with 10 disease classes.Not only this,by using tomato plant and target spot disease type as an example,we show that the confidence of self-classification for S-CNN model improves significantly over F-CNN model.This research work brings applicability of automated methods closer to non-experts for timely detection of diseases.展开更多
文摘A numerical study is presented for the fully developed two-dimensional laminar flow of viscous incompressible fluid through a curved square duct for the constant curvature δ = 0.1. In this paper, a spectral-based computational algorithm is employed as the principal tool for the simulations, while a Chebyshev polynomial and collocation method as secondary tools. Numerical calculations are carried out over a wide range of the pressure gradient parameter, the Dean number, 100 ≤ Dn ≤ 3000 for the Grashof number, Gr, ranging from 100 to 2000. The outer wall of the duct is treated heated while the inner wall cooled, the top and bottom walls being adiabatic. The main concern of the present study is to find out the unsteady flow behavior i.e. whether the unsteady flow is steady-state, periodic, multi-periodic or chaotic, if Dn or Gr is increased. It is found that the unsteady flow is periodic for Dn = 1000 at Gr = 100 and 500 and at Dn = 2000, Gr = 2000 but steady-state otherwise. It is also found that for large values of Dn, for example Dn = 3000, the unsteady flow undergoes in the scenario “periodic→chaotic→periodic”, if Gr is increased. Typical contours of secondary flow patterns and temperature profiles are also obtained, and it is found that the unsteady flow consists of single-, two-, three- and four-vortex solutions. The present study also shows that there is a strong interaction between the heating-induced buoyancy force and the centrifugal force in a curved square passage that stimulates fluid mixing and consequently enhance heat transfer in the fluid.
文摘Food security for the 7 billion people on earth requires minimizing crop damage by timely detectionofdiseases.Most deep learningmodels forautomated detectionof diseases in plants suffer fromthe fatal flaw that once tested on independent data,their performance drops significantly.This work investigates a potential solution to this problem by using segmented image data to train the convolutional neural network(CNN)models.As compared to the F-CNN model trained using full images,S-CNN model trained using segmented imagesmore than doubles in performance to 98.6%accuracy when tested on independent data previously unseen by the models even with 10 disease classes.Not only this,by using tomato plant and target spot disease type as an example,we show that the confidence of self-classification for S-CNN model improves significantly over F-CNN model.This research work brings applicability of automated methods closer to non-experts for timely detection of diseases.