(PDF) Neural network radiative transfer

Added on - 24 Apr 2021

  • 12

    Pages

  • 3941

    Words

  • 12

    Views

  • 0

    Downloads

Trusted by +2 million users,
1000+ happy students everyday
Showing pages 1 to 4 of 12 pages
CHARLES DARWIN UNIVERSITYELECTRICAL ENGINEERINGENG 720 INTERIM REPORTTITLE:INVERSION OF THE RADIATIVE TRANSFER EQUATION USING MACHINELEARNING TECHNIQUESSTUDENT NAMESTUDENT REGISTRATION NUMBERDATE OF SUBMISSION
LITERATURE REVIEWThe optical properties discuss the effects on light signals when they are incident on a solid or anyother form of material. The incident light is either reflected, absorbed or transmitted such thatmaterials can be optically classified as transparent, translucent, or opaque depending on whathappens to the incident light. The metal materials have a fine succession of energy states wherethe near-surface electrons absorb visible light. In other instance, shiny metal reflects the light atthe same frequency as that of the incident signal. The reflectivity is such that,Reflectivity=IRIO;0.950.90The light is refracted when it is transmitted through various different materials. Many of theelectronic properties of materials, information on the bonding, and material composition wasdiscovered using spectroscopy. It provides the evidence on the energy levels in atoms, evidencefor energy bands and band-gaps, and photoelectric effect[ CITATION Lia11 \l 1033 ].If the materialis not perfectly transparent, the intensity decreases exponentially with distance. Absorption oremissions of light by the materials is due to excitation or relaxation of the electronics in theatoms. Some of the organic or carbon-based materials consist of molecules which are relativelyweakly connected to other molecules. The absorption spectrum is dominated by the absorptionsdue to the molecules themselves. When the incident illumination is of a wavelength that excitesthe different modes of the different frequencies, the illumination tends to be absorbed. Thetechnique allows us to measure concentrations of different gas species in the atmosphere. Byfitting spectra of known gases to the measured atmospheric spectra, one can figure out thequantities of each of the gases.Dispersion of light occurs when the white light is transmitted through the material at differentwavelengths and then arrives at the other end at different times. The speed of the wave dependson the wavelength and the white light is a combination of light signals at different wavelengths.Another optical property is luminescence which describes the re-emission of pre-absorbedradiative energy. The type of incident radiation determines the type of luminescence so that thereis photoluminescence, electroluminescence, or even cathodoluminescence. All these opticalproperties are demonstrated in the use of fiber optical cables which enable data transmissionthrough light signals. It allows for digital data transmission and caters for attenuation using1
repeaters on the way. The loss in the cable is important because it determines the maximumuninterrupted length of the fiber. Losses depend on the wavelength of light and the purity of thematerial. For the very high-purity materials, Rayleigh scattering is very common. The Rayleighscattering results from minute local density variations which are present in the liquid glass due toBrownian motion and become frozen into the solid. The total internal reflection at the boundaryof the fiber acts as a light signal guide around bends and curves in the fiber. The refraction angledepends on the wavelength where the shorter wavelengths are refracted more than the longerones, and the identity of the material. The diffraction of light waves is a source of color and thematerials are having stacked layers with distances in the visible range of light.Spectroscopy seeks to study the light obtained from a given object using an instrument thatspreads out the light to make a range of electromagnetic energy separated by wavelength ordifferent colors. Neural networks have been widely used in implementation of spectroscopy. Acase study was carried out where the artificial neural networks were used to extract opticalproperties of the marine organisms from their Mueller scattering matrix[CITATION Hul \p 6 \l1033 ]. The neural network was trained using the Mie calculations for a range of sizedistributions.The neural networks are used in different applications as they learn by example orduring training as opposed to the functions that require complex algorithms and curve fittingmethods to implement. The neural network is based on a set of training and testing data. Theexperimenter or researcher ought to prepare the database that will provide a well characterizedtraining data for the neural network[CITATION JLa07 \l 1033 ].The training and test data used in theimplementation of neural networks is setup in terms of matrices.The artificial neural networks are applied in the optical properties of nano-semiconductors. Thenanomaterials depict improved attributes of atoms or molecules which may exist in a bulk stateas a result of the constituent matter. There are a number of statistical experiments that are used innanoscience and technology to come up with conclusions on the desired properties of acompound or item. However, each experiment demonstrates its own effects of variables and thegoals of optimizing the desired properties. There is no mathematical model that can be used as astandard for the nanotechnology to demonstrate the relationship between the input parametersunder study and the attributes of the designed nanomaterials. The use of artificial neuralnetworks allows the system to learn from the experimental information and results before2
coming up with a model. Some of the key nanomaterial processes that have been implementedsuccessfully using the artificial neural networks are:(i)Modelling the growth rate of a nanomaterial(ii)Modelling the grain and particle size of a component(iii)Modelling of the photocatalytic properties and magnetic properties of components(iv)Modelling of the oxidation, kinetic and emulsion stability of nanomaterials(v)Modelling the effects of parameters in ball milling processing, sol-gel synthesis, andspray reaction synthesis of the materials.As highlighted earlier, to perform the modelling of the processes for the nanomaterials, it isimportant to divide the experimental data obtained into training and testing data. In the case ofNano processing, the experimental data is very expensive to obtain due to high cost ofexperiment hence the best models need to be obtained from the available experimental data. Theacquisition of experimental data for training and testing is very crucial to avoid underfitting andoverfitting problems which result from the use of less and excessive experimental datarespectively[ CITATION Tab11 \l 1033 ]. The two extreme cases result in networks that do notrepresent the true model of a given nano process. The right amount of data will improve theneural network’s predictability and the testing will produce accurate results or output.The modelof the neural networks required up to 75 percent of the experimental data to be used as trainingdata and the 25 percent as testing data. The data used as the test data is usually normalized beforeit’s used in the network. One of the common neural networks adopted in nano processing is themulti-layer feed forward backpropagation network[ CITATION Tab11 \l 1033 ]. Another approachis the use of the recurrent networks where each layer uses its own output as one of the inputs ofthat layer. In neural network, the first layer is the input layer, the middle layer is the hidden layerthat performs the modelling, and the third layer is the output layer that presents the results of themodelling using the test data.The most efficient model is one where all the neurons of the feed-forward network or those of the recurrent network have an effect on the output of thesystem[ CITATION Tab11 \l 1033 ].The Neural networks have found their application in industrial processes[ CITATION Juk15 \l1033 ]. They are considered as highly parallel dynamical systems that can performtransformations by means of their state response to their input information. The transformations3
desklib-logo
You’re reading a preview
Preview Documents

To View Complete Document

Click the button to download
Subscribe to our plans

Download This Document