Solving singular control problems using uniform trigonometrization method

This study investigates the application of a recently developed construct, the uniform trigonometrization method (UTM), to the singular control problems in chemical engineering. The UTM involves minimal modifications to the original problem, thereby generating near-singular control solutions that can be used for conceptual design and serve as an alternate to direct techniques like nested and simultaneous approaches. Eight classical singular control problems with known analytical solutions and three complex singular control problems from chemical engineering domain are solved in this study. The results obtained using the UTM for these problems are found to match well with the literature and are of higher resolution as compared to the results obtained using a direct pseudospectral-based solver. The ability of the UTM to handle complex chemical engineering problems with both singular controls and state path constraints has also been demonstrated in this study.     

A scalable sparse Cholesky based approach for learning high-dimensional covariance matrices in ordered data

Machine Learning - Covariance estimation for high-dimensional datasets is a fundamental problem in machine learning, and has numerous applications. In these high-dimensional settings the number of...     

A Review: Tailor-made Hydrogel Structures (Classifications and Synthesis Parameters)

Hydrogels are being prepared for use in a wide variety of applications ranging from medicines, tissue engineering, superabsorbents, controlled release of drugs & fertilizers, and oil absorbers etc. This review highlights hydrogel structure and their different classifications under various heads. It also discusses various routes to obtain tailormade hydrogels by polymerizing a combination of two or more monomers with proper type of crosslinks in order to obtain desired properties in the resulting hydrogel. Novel hydrogel configurations like microgels and nanogels, slide ring gels, double network hydrogels and nanocomposite gels have also been reviewed.     

Electrical characterization and parameter extraction of organic thin film transistors using two dimensional numerical simulations

In this paper a performance based comparison of top and bottom contact organic thin film transistor (OTFT) device structures, using two dimensional numerical simulations has been carried out. In addition to this, investigations pertaining to the estimation of contact resistance in these OTFTs were also performed. To estimate contact resistance the conventional transmission line method and modified transmission line method (M-TLM) were respectively invoked. Our simulation results clearly indicate that the latter is more accurate in the estimation of contact resistance compared to the conventional method. Furthermore, the M-TLM was used to estimate the gate voltage and film thickness dependence of the contact resistance for the two device structures. The observed results have been explained on the basis of the significantly lowered area of carrier injection and extraction regions, at the source/channel and channel/drain interface respectively, in bottom contact transistor that lead to its inferior performance over the top contact transistor.     

Clustering assisted artificial neural network for handling noisy big data: An application for estimation of parameter in combined mode conduction and radiation heat transfer

A pattern net assisted mapping artificial neural network (PAMANN) model for estimation of parameters in problem with large data (1300 × 121 matrix size) is reported. A pattern net-based multilayer perceptron neural network (MLPNN) model for clustering the data, followed by mapping MLPNN model for mapping the target with the input, is developed as PAMANN model. A heat transfer problem with combined mode conduction and radiation in porous medium is solved numerically, and is called direct model. In the inverse model, a PAMANN model is developed by using data generated through the direct model. The PAMANN model is able to estimate two parameters (extinction coefficient β and convective coupling P₂) after taking temperature profile as input. The model is tested for different number of neurons in hidden layer, and different levels of noise in input data. Twelve different algorithms are explored in training of mapping MLPNN, and compared for performance. Levenberg–Marquardt algorithm is found to estimate the parameters with high accuracy, but took high CPU time. Bayesian regularization is found to consume very high CPU time with moderate accuracy in estimation of parameters. Variations in hidden layer neuron number and noise in input data, were done to analyze the performance of mapping MLPNN with different training algorithms. Algorithms O-Step Secant, conjugate gradient with Polak-Ribiére updates, and conjugate gradient with Fletcher-Reeves updates are able to handle all variations of noise and number of neurons in hidden layer, with good accuracy of estimation and low CPU time consumption. Under high computational resource LM algorithm can be used for all cases. Up to 0.99132 value of regression coefficient is obtained in mapping MLPNN model with 15 hidden neurons, indicating the high accuracy of the model. With the help of PAMANN model, highly accurate (absolute error 1.78%) estimation of parameters is obtained. The model can handle upto 1% noise in input data, while giving accurate results.     

Potential biomass supply for agro-pellet production from agricultural crop residue in Nepal

Nepal, a country rich in biomass, still does not have any commercial pellet production plants and is wasting large amounts of agricultural crop residue. The current study showed that about 5.61 million tonnes (Mt) of biomass in the form of pellets are potentially available from agricultural crop residues. The brick and cement industries could use these agro-pellets. Co-firing of pellets in such industries could play an important role in reducing the import volume of coal and minimize the related environmental loadings.     

Above-ground morphological predictors of rooting success in rooted cuttings of Jatropha curcas L.

Cuttings of 1–3 cm diameter and 45 cm length were collected during the first week of February from branches of previous year’s growth in a mature plantation of Jatropha curcas. The cuttings, without application of any growth regulator, were planted in nursery beds having loam: gravel (1:1 v/v) mixture rooting media. The nursery beds existed inside a polyethylene tunnel where intermittent misting was done. When sprouting percentage had stabilised, sprouted cuttings were removed from the media, and root and shoot characteristics of the cuttings were recorded. The number of roots and root length were found to be significantly correlated (P < 0.01) with one another as well as with sprout length, number of sprouts and number of leaves. The following equations were fitted for prediction of root characteristics of a cutting from of its above-ground characteristics: (i) No. of roots = −0.409 + 0.452 (no. of leaves) + 0.395 (sprout length), and (ii) Root length = 2.656 + 0.206 (no. of leaves) + 0.270 (sprout length); the sprout length and root length are in centimetres in both equations. Thicker cuttings possessed better root and above-ground characteristics.     

Corruption-tolerant bandit learning

We present algorithms for solving multi-armed and linear-contextual bandit tasks in the face of adversarial corruptions in the arm responses. Traditional algorithms for solving these problems assume that nothing but mild, e.g., i.i.d. sub-Gaussian, noise disrupts an otherwise clean estimate of the utility of the arm. This assumption and the resulting approaches can fail catastrophically if there is an observant adversary that corrupts even a small fraction of the responses generated when arms are pulled. To rectify this, we propose algorithms that use recent advances in robust statistical estimation to perform arm selection in polynomial time. Our algorithms are easy to implement and vastly outperform several existing UCB and EXP-style algorithms for stochastic and adversarial multi-armed and linear-contextual bandit problems in wide variety of experimental settings. Our algorithms enjoy minimax-optimal regret bounds, as well as can tolerate an adversary that is allowed to corrupt upto a universally constant fraction of the arms pulled by the algorithm.