Neural networks, which are simplified models of the biological nervous system, is a massively parallel distributed processing system made up of highly interconnected neural computing elements that have the ability to learn and there by require knowledge and make it available for use. Neural networks are simplified imitations of the central nervous system, and obviously therefore have been motivated by the kind of computing performed by the human brain. The structural constituents of the human brain termed Neurons are the entities, which perform computations such as cognition, logical inference, pattern recognition .etc. Hence the technology which has been built on a simplified imitation of computing by neurons of a brain, has been termed Artificial Neural systems technology or Artificial Neural Networks or simply Neural Networks.
[...] Estimation of mass transfer parameters in fast fluidized beds of fine solids: In this study back-propagation, feed-forward neural networks are applied to estimate mass-transfer parameters in fast fluidized needs of fine solids. These networks are trained to predict mass-transfer rates using measurements of the sublimation rate of coarse naphthalene balls in fast fluidized needs of fine glass beads at several solid-to-gas mass flow rates within the relevant superficial gas-velocity range. When tested to predict the effective diffusivities fro a coarse particle to the bulk of the fast bed of fine solids, trained neural networks calculated the Sherwood number with high accuracy. [...]
[...] However the behavior of a artificial neuron can be captured by a simple model as shown in fig1. the model which forms the basis of artificial neural network x1 w1 SUMMATION UNIT x2 w2 x3 w3 xn wn THRESHOLDING UNIT Fig1: Simple model of an Artificial Neuron Here x1,x2,x3 xn are the N-inputs to the artificial neurons, w1,w2,w3 .wn are the weights attached to the input links. Recollect that a biological receives all inputs through the dendrites, sum them and produces an output if the sum is greater than the threshold value. [...]
[...] The applications of different types of neural networks in Modeling of limestone- SO2 reaction, particle sizing in slurries by reflectance spectroscopy, leak detection in liquefied gas pipelines, mass transfer predictions in fast fluidized beds of fine solids, dynamic process modeling, Optimization of fed batch fermentation process are presented. Nomenclature: SISO-single input/single output MIMO-multi input/multi output ηw –learning rate for weights ηT –learning rate for time constants αw - momentum factor for weights αT - momentum factor time constants. F1- flow rate of acetic acid F2- flow rate of NaOH Fi-inlet flow rate of CSTR Temperature of reactor References: Jayanth K.Bandyopadhyay, Senthil Annamalai, and K.Lal Gauri., “Applications of Artificial Neural Networks in Modeling Limestone-SO2 Reaction”, AIChE journal, Vol.42 2295-2301(1996). [...]
[...] During the period from 1982 until 1986, several seminal publications were published that significantly furthered the potential of Neural Networks. John Hofield(1982,1984) introducing a recurrent neural networks. According to most recent statements(Drefuys 1990), the first authors of the optimization approach for multilayer feed forward system were Bryson( Bryson and Ho 1969) and Kelley (Kelley 1969) who obtained a gradient solution for multistage networks training. Theory: Human brain is one of the most complicated things, which on the hole has been poorly understood. [...]
[...] Fig4: Recurrent Networks Back Propagation: Back propagation is a systematic method of training multilayer artificial neural networks it is built on high mathematical foundation and has very good application paternal. Materials and Methods: Process modeling with recurrent neural networks: SISO system at steady state: The process employed to study the static modeling ability of RNN is the pH CSTR system(fig5). Fig5: pH-CSTR system The training data used here are the steady state relationships of f2 and pH which can be obtained by solving the steady state form of equations. [...]
Online readingwith our online reader
Content validatedby our reading committee