| Accurate measurement of particle size distribution,optical constants,and particle morphology parameters is very important in many fields,such as aerosol measurement,combustion diagnosis,and nanoparticle preparation.Elastic light scattering is a commonly used particle size measurement method and an accurate characterization tool of particle size distribution.In the past,the research based on the light scattering method tends to study the characteristic parameters of spherical particles,but less on nonspherical particles.In general,in addition to spherical particles,ellipsoidal,cylindrical,and agglomerated particles also widely exist in real life.Therefore,it is very important to accurately measure the size distribution,optical constants,and morphology parameters of non-spherical particles.When the measured particles become random in shape,the inversion process based on the light scattering method becomes more complex.At the same time,a more advanced solution is needed to retrieve the size distribution,morphology,and optical constants of spherical and non-spherical particles from light scattering signals.The main work of the study can be divided into the following aspects.After the introductory chapter one,a total of five different approaches are introduced for achieving the purpose of our research objectives.In this study,the core researches are segmented into five sections started from chapter two to chapter six.In chapter two of section 2.1,different signal combinations are used to retrieve the size distribution and the optical constant of spherical particles.Multiple slab thicknesses transmittance,multi-angle scattering intensity,and a combination of multi-angle multi-wavelengths scattering intensity along with multiple slab thicknesses transmittance at multiple-wavelengths have been used to retrieve particle size distribution(PSD)and complex refractive index by an inverse simulation.The scattering and transmittance intensities are calculated by solving the radiative transfer(RTE)equation and the Lambert-Beer law,respectively.The retrieval accuracy of PSD is tested using all those signals,and the inverse simulation is run by using improved quantum particle swarm optimization(IQPSO)algorithm.Retrieval results show good agreement with the original values.When more information is used as a form of combined input signals,retrieval results attribute good accuracy.The use of multiple slab thicknesses can eliminate the dependency of complex refractive index on wavelengths variation for measuring transmittance signal.We conduct a detailed study to identify the signal sensitivity on slab thickness.Effects of the angular positions on signal sensitivity are also examined.The robustness of the IQPSO algorithm is tested at different measurement noise and the retrieval results found satisfactory.Next in chapter two of section 2.2,we identified the particles' average diameter by a particle size analyzer and compare it with our suggested inverse numerical approach.After that,in chapter two of section 2.3,a numerical study investigates the feasibility of simultaneous retrieval of the particle size distribution(PSD)and optical constants of spheroids by optical spectroscopy.In this simulation,the particles are considered spheroids.The aspect ratio of an oblate and prolate spheroid is set as 0.8 and 1.2respectively.At this constant aspect ratio,the particle's orientation has been changed by altering its major semi-axis.Two continuous wavelength lasers are employed to irradiate the particle samples.Multi-angle and multi-wavelengths elastic forward scattering intensity and the spectral collimated transmittance are employed to measure signals.For forward scattering,the spheroid is effectively replaced by a sphere of an approximated radius,and the modified Mie theory is employed to calculate the scattering intensity.For the collimated transmittance,the extinction efficiency of the non-spherical particle is measured based on the extended anomalous diffraction approximation.The Log-Normal distribution(L-N)is used to get the volume frequency distribution of the particles and the inverse process is done by using the improved quantum particle swarm optimization.Two different sets of optical constants(i.e.complex refractive index),the semi-major axis of non-spherical particle,and discrete rate are retrieved by the inverse simulation.The results show that the proposed spectroscopic technique can retrieve PSD and optical constants of non-spherical particles simultaneously within the tolerable error limit of less than 10%.In chapter three,bimodal aerosol size distribution is estimated from aerosol optical thickness using a commercial optical depth measuring instrument named Microtopes II Sun photometer.An attractive and repulsive particle swarm optimization algorithm(ARPSO),which is an advanced derivative of particle swarm optimization(PSO),is used to solve an inverse problem.Based on the operating principle of the measuring instrument,a detailed study regarding the set and the number of optimum wavelength selections are carried out for accurate inversion of the target parameters based on log-normal(L-N)bimodal aerosol size distribution(ASDs)function.A parallel optimum wavelength set selection technique is adopted by integrating a simple algorithm before the main loop of the inverse model.The inverse accuracy and the efficiency of the algorithm are studied by changing the number of selected wavelengths and the swarm size.The result depicts that,at an optimum wavelength set along with the number of wavelengths and swarm size,ASDs of dust like particle shows good agreement with the original distribution.It is found that,even at 5% measurement noise added in the system,the proposed model can retrieve all target parameters keeping relative error less than 10%.In chapter four,multiple wavelengths elastic light scattering spectroscopy is employed for the identification of primary particle averages diameter of aggregates in the system.From studied literature,the experimental test principle of the Rayleigh-DebyeGans(RDG)theory for light scattering by fractal aggregates is adopted as a theoretical model for calculating the value of the Rayleigh ratio under multiple wavelengths.It is used further to calculate signals for the inverse simulation.A log-normal(L-N)distribution function is used to identify the particle size distribution(PSD)of primary particles present in the aggregates system.A probability density function-based ant colony optimization algorithm(PDF-ACO)is employed to retrieve target parameters from the searching space by inverse simulation.All prerequisite parameters for inverse simulation are set closer or similar to the reference experimental parameters for the forward problem.After the theoretical inversion based on the reference experiment model,the retrieval results show good agreement with the measured experimental values found in the literature study.The increments of multiple wavelengths enhance the retrieval accuracy of target parameters.The robustness of the PDF-ACO algorithm could successfully limit the retrieval errors to less than 10% even though at a high noise present in the system.In chapter five,angular resolved elastic light scattering signals are employed to retrieve the spheroidal and cylindrical particle size distribution(PSD)using the T-matrix method.The interesting features of this simulation are that,to retrieve the size distribution of the particles,the simulation does not need any initial information about the number density of the particles per unit volume.Moreover,both cylindrical and spheroidal PSD can be retrieved by using this same simulation.A probability density function-based ant colony optimization(PDF-ACO)algorithm is used to solve the inverse problem to retrieve the target parameters.For the direct problem,the intensity of the scattering particles is measured at seven different angular positions and the ratio of the sum of this scattering intensity to the scattering intensity at a reference angle is used to fit the objective function of the inverse problem.In regards to retrieving parameters,the robustness of the PDF-ACO algorithm shows good agreement with the true parameters.Even with 5% measurement noise,the retrieval accuracy can be achieved within the tolerable error limit of less than 10%.Finally,in chapter six,an experimental model is suggested and integrated with a numerical scheme for simultaneous identification of arbitrary shape particle size distribution and its degree of the non-sphericity present in a system.In this spectroscopic instrumentation,the transmittance intensity of light is measured by calculating the bulk absorption coefficient of randomly oriented arbitrarily shaped particles by adopting parameterized anomalous diffraction theory(PADT).The effects of reflection and refraction of arbitrary shapes particles are also considered.The multi-wavelength transmittance signals(estimated and measured value)are used to fit the objective functions for the identification of target parameters.The unique feature of this simulation is the simultaneous identification of the joint function particle size distribution(PSD)and the shape perimeter of arbitrary particles using the joint log-normal distribution function.A probability density function-based ant colony optimization(PDF-ACO)algorithm is used for the inverse simulation.The parameters retrieved by the multi-wavelength transmittance signals show good agreement with the set original values of target parameters.The robustness of the PDF-ACO algorithm could successfully keep the retrieval errors of the estimated parameters within the tolerable limit(error<10%)even at high noise in the system. |
PDF Full Text Request