Hydrodynamic Study of Swirling Fluidized Bed and the Role of Distributor

The study aims to the hydrodynamic characteristics of swirling fluidized bed, using large particles (Geldart D-type) selected from locally available agricultural produce (coffee beans and black pepper). The important variables considered in the present study include percentage area of opening, angle of air injection and the percentage useful area of the distributor. A total of seven distributors have been designed and fabricated for a bed column of 300 mm, namely single row vane type distributors (15˚ and 20˚ vane angle), inclined hole type distributors (15˚ and 20˚ vane angle) and perforated plate distributors. The useful area of distributor of single row vane type, three now vane-type and inclined hole-type distributors are respectively 64%,91% and 94%. The hydrodynamic parameters considered in the present study include distributor pressure drop, air velocity, minimum fluidizing velocity, bed pressure drop, bed height and the bed behaviour. It has been observed that, in general, the distributor pressure drop decreases with an increase in the percentage area of opening, Further, and increase in the area of opening above 17% will not considerably reduce the distributor pressure drop. In the present study, for the distributor with an area of opening 17%, and corresponding to the maximum measured superficial velocity of 4.33 m/s, the distributor pressure drop obtained was 55.25mm of water. The study on the bed behavior revealed that, in a swirling fluidized bed, once swirl motion starts, the bed pressure drop increases with superficial velocity in the outer region and it decreases in the inner region. This means that, with higher superficial velocity, the air might get by-passed through the inner boundary of the bed (around the cone). So, depending on the process for which the bed is used, the maximum superficial velocity is to be limited to have an optimum bed performance.

Glucoamylase immobilized on montmorillonite: influence of nature of binding on surface properties of clay-support and activity of enzyme

Glucoamylase was immobilized on acid activated montmorillonite clay via two different procedures namely adsorption and covalent binding. The immobilized enzymes were characterized by XRD, NMR and N2 adsorption measurements and the activity of immobilized glucoamylase for starch hydrolysis was determined in a batch reactor. XRD shows intercalation of enzyme into the clay matrix during both immobilization procedures. Intercalation occurs via the side chains of the amino acid residues, the entire polypeptide backbone being situated at the periphery of the clay matrix. 27Al NMR studies revealed the different nature of interaction of enzyme with the support for both immobilization techniques. N2 adsorption measurements indicated a sharp drop in surface area and pore volume for the covalently bound glucoamylase that suggested severe pore blockage. Activity studies were performed in a batch reactor. The adsorbed and covalently bound glucoamylase retained 49% and 66% activity of the free enzyme respectively. They showed enhanced pH and thermal stabilities. The immobilized enzymes also followed Michaelis–Menten kinetics. Km was greater than the free enzyme that was attributed to an effect of immobilization. The immobilized preparations demonstrated increased reusability as well as storage stability.

Acid activated montmorillonite: an efficient immobilization support for improving reusability, storage stability and operational stability of enzymes

Three enzymes, α-amylase, glucoamylase and invertase, were immobilized on acid activated montmorillonite K 10 via two independent techniques, adsorption and covalent binding. The immobilized enzymes were characterized by XRD, N2 adsorption measurements and 27Al MAS-NMR spectroscopy. The XRD patterns showed that all enzymes were intercalated into the clay inter-layer space. The entire protein backbone was situated at the periphery of the clay matrix. Intercalation occurred through the side chains of the amino acid residues. A decrease in surface area and pore volume upon immobilization supported this observation. The extent of intercalation was greater for the covalently bound systems. NMR data showed that tetrahedral Al species were involved during enzyme adsorption whereas octahedral Al was involved during covalent binding. The immobilized enzymes demonstrated enhanced storage stability. While the free enzymes lost all activity within a period of 10 days, the immobilized forms retained appreciable activity even after 30 days of storage. Reusability also improved upon immobilization. Here again, covalently bound enzymes exhibited better characteristics than their adsorbed counterparts. The immobilized enzymes could be successfully used continuously in the packed bed reactor for about 96 hours without much loss in activity. Immobilized glucoamylase demonstrated the best results.

Unconstrained Handwritten Malayalam Character Recognition using Wavelet Transform and Support vector Machine Classifier

This paper presents the application of wavelet processing in the domain of handwritten character recognition. To attain high recognition rate, robust feature extractors and powerful classifiers that are invariant to degree of variability of human writing are needed. The proposed scheme consists of two stages: a feature extraction stage, which is based on Haar wavelet transform and a classification stage that uses support vector machine classifier. Experimental results show that the proposed method is effective

Using Neural Network Classifier Support Vector Machine Regression for the prediction of Melting Point of Drug – like compounds

In our study we use a kernel based classification technique, Support Vector Machine Regression for predicting the Melting Point of Drug – like compounds in terms of Topological Descriptors, Topological Charge Indices, Connectivity Indices and 2D Auto Correlations. The Machine Learning model was designed, trained and tested using a dataset of 100 compounds and it was found that an SVMReg model with RBF Kernel could predict the Melting Point with a mean absolute error 15.5854 and Root Mean Squared Error 19.7576

Impact of process parameters on L-glutaminase production by marine Vibrio costicola in solid state fermentation using polystyrene as an inert support

Process parameters influencing e-glutaminase production by marine Vibrio costicola in solid state fermentation (SSF) using polystyrene as an inert support were optimised. Maximal enzyme yield (157 U/g dry substrate) was obtained at 2% (w/w) t:glutamine, 35°C and pH 7.0 after 24 h. Maltose and potassium dihydrogen phosphate at 1% (w/w) concentration enhanced enzyme yield by 23 and 18%, respectively, while nitrogen sources had an inhibitory effect. Leachate with high specific activity for glutaminase (4.2 U/mg protein) and low viscosity (0-966 Ns/m 2) was recovered from the polystyrene SSF system

Assessing the users’ need for a spatial decision support system of smallholder farming in Kenya

Accurate data of the natural conditions and agricultural systems with a good spatial resolution are a key factor to tackle food insecurity in developing countries. A broad variety of approaches exists to achieve precise data and information about agriculture. One system, especially developed for smallholder agriculture in East Africa, is the Farm Management Handbook of Kenya. It was first published in 1982/83 and fully revised in 2012, now containing 7 volumes. The handbooks contain detailed information on climate, soils, suitable crops and soil care based on scientific research results of the last 30 years. The density of facts leads to time consuming extraction of all necessary information. In this study we analyse the user needs and necessary components of a system for decision support for smallholder farming in Kenya based on a geographical information system (GIS). Required data sources were identified, as well as essential functions of the system. We analysed the results of our survey conducted in 2012 and early 2013 among agricultural officers. The monitoring of user needs and the problem of non-adaptability of an agricultural information system on the level of extension officers in Kenya are the central objectives. The outcomes of the survey suggest the establishment of a decision support tool based on already available open source GIS components. The system should include functionalities to show general information for a specific location and should provide precise recommendations about suitable crops and management options to support agricultural guidance on farm level.

Comparing Support Vector Machines with Gaussian Kernels to Radial Basis Function Classifiers

The Support Vector (SV) machine is a novel type of learning machine, based on statistical learning theory, which contains polynomial classifiers, neural networks, and radial basis function (RBF) networks as special cases. In the RBF case, the SV algorithm automatically determines centers, weights and threshold such as to minimize an upper bound on the expected test error. The present study is devoted to an experimental comparison of these machines with a classical approach, where the centers are determined by $k$--means clustering and the weights are found using error backpropagation. We consider three machines, namely a classical RBF machine, an SV machine with Gaussian kernel, and a hybrid system with the centers determined by the SV method and the weights trained by error backpropagation. Our results show that on the US postal service database of handwritten digits, the SV machine achieves the highest test accuracy, followed by the hybrid approach. The SV approach is thus not only theoretically well--founded, but also superior in a practical application.

Improving Multiclass Text Classification with the Support Vector Machine

We compare Naive Bayes and Support Vector Machines on the task of multiclass text classification. Using a variety of approaches to combine the underlying binary classifiers, we find that SVMs substantially outperform Naive Bayes. We present full multiclass results on two well-known text data sets, including the lowest error to date on both data sets. We develop a new indicator of binary performance to show that the SVM's lower multiclass error is a result of its improved binary performance. Furthermore, we demonstrate and explore the surprising result that one-vs-all classification performs favorably compared to other approaches even though it has no error-correcting properties.

Properties of Support Vector Machines

Support Vector Machines (SVMs) perform pattern recognition between two point classes by finding a decision surface determined by certain points of the training set, termed Support Vectors (SV). This surface, which in some feature space of possibly infinite dimension can be regarded as a hyperplane, is obtained from the solution of a problem of quadratic programming that depends on a regularization parameter. In this paper we study some mathematical properties of support vectors and show that the decision surface can be written as the sum of two orthogonal terms, the first depending only on the margin vectors (which are SVs lying on the margin), the second proportional to the regularization parameter. For almost all values of the parameter, this enables us to predict how the decision surface varies for small parameter changes. In the special but important case of feature space of finite dimension m, we also show that there are at most m+1 margin vectors and observe that m+1 SVs are usually sufficient to fully determine the decision surface. For relatively small m this latter result leads to a consistent reduction of the SV number.

Notes on PCA, Regularization, Sparsity and Support Vector Machines

We derive a new representation for a function as a linear combination of local correlation kernels at optimal sparse locations and discuss its relation to PCA, regularization, sparsity principles and Support Vector Machines. We first review previous results for the approximation of a function from discrete data (Girosi, 1998) in the context of Vapnik"s feature space and dual representation (Vapnik, 1995). We apply them to show 1) that a standard regularization functional with a stabilizer defined in terms of the correlation function induces a regression function in the span of the feature space of classical Principal Components and 2) that there exist a dual representations of the regression function in terms of a regularization network with a kernel equal to a generalized correlation function. We then describe the main observation of the paper: the dual representation in terms of the correlation function can be sparsified using the Support Vector Machines (Vapnik, 1982) technique and this operation is equivalent to sparsify a large dictionary of basis functions adapted to the task, using a variation of Basis Pursuit De-Noising (Chen, Donoho and Saunders, 1995; see also related work by Donahue and Geiger, 1994; Olshausen and Field, 1995; Lewicki and Sejnowski, 1998). In addition to extending the close relations between regularization, Support Vector Machines and sparsity, our work also illuminates and formalizes the LFA concept of Penev and Atick (1996). We discuss the relation between our results, which are about regression, and the different problem of pattern classification.

From Regression to Classification in Support Vector Machines

We study the relation between support vector machines (SVMs) for regression (SVMR) and SVM for classification (SVMC). We show that for a given SVMC solution there exists a SVMR solution which is equivalent for a certain choice of the parameters. In particular our result is that for $epsilon$ sufficiently close to one, the optimal hyperplane and threshold for the SVMC problem with regularization parameter C_c are equal to (1-epsilon)^{- 1} times the optimal hyperplane and threshold for SVMR with regularization parameter C_r = (1-epsilon)C_c. A direct consequence of this result is that SVMC can be seen as a special case of SVMR.

On the Noise Model of Support Vector Machine Regression

Support Vector Machines Regression (SVMR) is a regression technique which has been recently introduced by V. Vapnik and his collaborators (Vapnik, 1995; Vapnik, Golowich and Smola, 1996). In SVMR the goodness of fit is measured not by the usual quadratic loss function (the mean square error), but by a different loss function called Vapnik"s $epsilon$- insensitive loss function, which is similar to the "robust" loss functions introduced by Huber (Huber, 1981). The quadratic loss function is well justified under the assumption of Gaussian additive noise. However, the noise model underlying the choice of Vapnik's loss function is less clear. In this paper the use of Vapnik's loss function is shown to be equivalent to a model of additive and Gaussian noise, where the variance and mean of the Gaussian are random variables. The probability distributions for the variance and mean will be stated explicitly. While this work is presented in the framework of SVMR, it can be extended to justify non-quadratic loss functions in any Maximum Likelihood or Maximum A Posteriori approach. It applies not only to Vapnik's loss function, but to a much broader class of loss functions.