Artifi Cial Neural Networks

One of the main concerns related to air pollution in megacities like the city of Isfahan is the existence of particulate matter. In this study, the model of artificial neural network (ANN) was used to predict the dust emission trend. Applicable information may be obtained from this model to control air pollution and reduce costs. In this model of ANN, the multilayer perceptron (MLP) method and backpropagation learning algorithm were applied. The parameters of wind direction and speed, humidity and temperature are the input and PM2.5 concentration is the output of the output of the model. The correlation coefficient of this model to predict PM2.5 concentration obtained by comparing actual data with simulated data was 0.80. Hence, the model proposed in this study is an efficient approach to analyze environmental issues related to PM2.5 concentration. For evaluating the model efficiency, the analysis of difference among the measured and simulated data was applied and showed the high efficiency of model in predicting particulate matter concentration. It is proved that the ANN model is an appropriate model to be adopted in predicting the air pollutants' measure in cities where there exist the concentration of pollutants and meteorological variables. Predicting PM2.5 concentration may be an efficient method applied in preventing measures like decreasing the greenhouse gas emission and promoting awareness among the effected. In this model the MLP and back propagation algorithm are applied. The source database consists of time series of meteorological variables and PM2.5 concentration collected in various stations in the city of Isfahan. The function of back propagation algorithms is to find correlations between PM2.5 and meteorological variables. These correlations are applicable in obtaining more information on the model to predict PM2.5 concentration. Using artificial neural networks can accurately model the relationship between meteorological, traffic data and PM 2.5 and PM 10 concentrations in the urban environment.

This paper investigates the application of the neural network in a run of mine ore stockpile in Choghart Iron Mine of Iran. While a signifi cant amount of high grade stockpile at Choghart mine, near the open pit, is an environmental hazard, it is also potential source of high grade ores. For future exploitation, determination of stockpile resource tonnage and grade has become an important aspect of this study. In order to achieve this goal, Artifi cial Neural Networks was applied. Initially a Feed-forward Neural Network model was constructed to estimate the iron grade. In this model 52% of samples from stockpile were used for training, 25% for validation and the remaining as the test set. Finally optimal architecture for grade estimation was 3-19-1 and values of R and MSE were 0.8 and 2.36, respectively. The results showed that neural networks offer a valid alternate approach to the problem of stockpile grade estimation, while requiring considerably less knowledge and time.

This paper investigates the asymptotic local power of the the averaged t-test of Im, Pesaran and Shin (2003, IPS hereafter) in the presence of both initial explosive conditions and incidental trends. By utilizing the least squares detrending methods, it is found that the initial condition plays no role in determining the asymptotic local power of the IPS test, a result strikingly different from the finding in Harris et al. (2010), who examined the impact of the initial conditions on local power of IPS test without incidental trends. The paper also presents, via an application of the Fredholm method discussed in Nabeya and Tanaka (1990a, 1990b), the exact asymptotic local power of IPS test, thereby providing theoretical justifications for its lack of asymptotic local power in the neighborhood of unity with the order of N-1/2T-1 while attaining nontrivial power in the neighborhood of unity that shrinks at the rate N-1/4T-1. This latter finding is consistent with Moon et al. (2007) and...

In this paper, as an important problem, the nonlinear equations resulting from the finite volume approximation of the differential equations describing the nanofluid convective heat transfer in a tube are considered to obtain the temperature distribution. One of the numerical algorithms for solving nonlinear equations is the Levenberg-Marquardt in which a system of linear equations must be solved per iteration. In order to prevent the linear system from being ill-conditioned, the damping factor (?) of this algorithm should not be decreased unconditionally. Indeed, it was proposed that a minimum value for ? is set per iteration for obtaining the search direction, where the time complexity per iteration is O(n3) in which n is the number of unknowns (or equations) of the nonlinear system. However, in this case study, this linear system is ill-conditioned. It is shown that the ill conditioning of the linear system due to this case study can be prevented in all iterations by setting an approximate minimum value for ? in the first iteration. This can considerably reduce the computational cost, in other words, this paper introduces an algorithm with a low computational complexity.

This paper presents the results on mixed convection in an inclined lid-driven triangular enclosure filled with water based Cu nanofluid. The enclosure is cooled at the inclined surface and simultaneously heated at the base surface. The vertical wall is adiabatic and moving at a constant speed. The governing equations are solved numerically by the Galerkin finite element method. The effects of parameters such as the Richardson number, nanoparticle volume fraction, and tilt angle on the flow and thermal fields as well as the heat transfer rate of the heated surface are taken into account. The overall heat transfer rate of the heated surface is characterized by the average Nusselt number at the heated surface. It has been observed that the effects of the tilt angle and solid volume fractions are significant on the flow and thermal fields. Besides, an optimum value for the solid volume fraction is found, which results in the maximum heat transfer rate at the considered values of the Richardson numbers. The rate of the increase of the average Nusselt number is slow for higher values of the tilt angle while it is much quicker for lower values of the tilt angle.