KUET Institutional Repository (Khulna University of Engineering & Technology)
Not a member yet
1045 research outputs found
Sort by
Effective Electrodes Position and Features Selection for EEG Based Epilepsy Detection
No full textThis thesis is submitted to the Department of Electrical and Electronic Engineering, Khulna University of Engineering & Technology in partial fulfillment of the requirements for the degree of Master of Science in Electrical and Electronic Engineering, August 2017.Cataloged from PDF Version of Thesis.Includes bibliographical references (pages 63-68).Electroencephalogram (EEG) signal is a representative signal that contains information
about the brain activity, which is the key identifier and used for the detection of epilepsy
since epileptic seizures are caused by a disturbance in the electrophysiological activity of
the brain. The prediction of epileptic seizure from the EEG signal usually requires a detailed
and experienced analysis of EEG data as well as proper collections of epileptic EEG signal
from the effective positions of the scalp. In this thesis, we have introduced a statistical
analysis of EEG signal with the optimized electrodes and features that are capable of
recognizing epileptic seizure with a high degree of accuracy (96.1 %) and helps to provide
automatic detection of epileptic seizure for different ages of epileptic persons. To
accomplish the target research, we extract various epileptic features namely Approximate
Entropy (ApEn), Kolmogorov–Sinai Entropy (KSE), Spectral Entropy (SE), Standard
Deviation (SD), Standard Error (SE), Modified Mean Absolute Value (MMAV), Roll-off
(R), and Zero Crossing (ZC) from the epileptic EEG signal. The k-nearest neighbor (k-NN)
algorithm is used for the classification of epilepsy then regression analysis is used for the
prediction of the epilepsy level at different ages of the patients. Using the statistical
parameters and regression analysis, a prototype mathematical model is proposed which
helps to find the epileptic randomness with respect to age of different subjects. The accuracy
of this prototype equation depends on proper analysis of the dynamic information from the
epileptic EEG signal.Md. Kamrul HasanMaster of Science in Electrical and Electronic Engineerin
Sequentially Renew Weighted Opportunity Cost Based Algorithm in Transportation Problems
No full textThis thesis is submitted to the Department of Mathematics, Khulna University of Engineering & Technology in partial fulfillment of the requirements for the degree of Master of Science in Mathematics, December 2017.Cataloged from PDF Version of Thesis.Includes bibliographical references (pages 37-39).Transportation models are of multidisciplinary fields of interest. In classical transportation approaches, the flow of allocation is controlled by the cost entries and/or manipulation of cost entries – so called Distribution Indicator (DI) or Total Opportunity Cost (TOC). But these DI or TOC tables are formulated by the manipulation of cost entries only. None of them considers demand and/or supply entry to formulate the DI/ TOC table. In this research we have developed Weighted Opportunity Cost (WOC) matrix, which is off course a new idea, for the control of the flow of allocations. It is noted that this WOC matrix is formulated by the manipulation of supply and demand entries along with cost entries as well. In this WOC matrix, the supply and demand entries act as weighted factors. Now it is known that, in Least Cost Matrix method, the flow of allocations are controlled by the least cost entries only and we do not need to change allocation direction in sub-sequence steps. On the other hand in Vogel’s Approximation Method, the flow of allocation is controlled by the DI table and this table is updated after each allocation step. Motivated by this idea, we have reformed the WOC matrix as Sequentially Updated Weighted Opportunity Cost (SUWOC) matrix. The significance difference of these two matrices is that, WOC matrix is invariant through all over the allocation procedures whereas SUWOC matrix is updated in each step of allocation procedures. Note that here update (/invariant) means changed (/unchanged) the weighted opportunity cost of the cells. Finally by incorporating this SUWOC matrix in LCM, we have developed a new approach to find out Initial Feasible Basic Solution of Transportation Problems. Some experiments have been carried out to justify the validity and the effectiveness of the proposed SUWOC-LCM approach. Experimental results have shown that the SUWOC-LCM approach outperforms. Moreover sometime this approach is able to find out optimal solution too.Pushpa AkhtarMaster of Science in Mathematic
Volumetric and Viscometric Studies of Paracetamol in Aqueous Solution of Alcohols at Different Temperatures
No full textThis thesis is submitted to the Department of Chemistry, Khulna University of Engineering & Technology in partial fulfillment of the requirements for the degree of Master of Science in Chemistry, November 2016.Cataloged from PDF Version of Thesis.Includes bibliographical references (pages 99-103)Paracetamol in presence of Water, 80% Water + 20% n-Propanol, 20% Water + 80% n-
Propanol, 80% Water + 20% n-Butanol and 20% Water + 80% n-Butanol were studied
through the measurement of viscosity and density at different temperatures (298.15K to
323.15K) with an interval of 5K. The results were discussed on the basis of structure
making and breaking mechanism of paracetamol in aqueous and aqueous alcohols solution under experimental conditions. The apparent molar volumes increase with the rise of concentration of paracetamol for all the studied systems indicating the structure making interaction for all the studied systems. The limiting apparent molar volume (φv
0) or partial molar volume at infinite dilution of paracetamol are positive and increase when paracetamol content in the solvents increase.
The positive values of transfer apparent molar volume (Δφv)tra suggest the structure
making ability through hydrophilic-hydrophilic interactions between polar groups of
paracetamol and polar groups of alcohols-water. The values of limiting apparent molar
volume expansibilities (Eφ0) are positive and the values of (δEφ0/δT)p are small which
suggest the structure making property in these systems.
Viscosities increase with increasing of paracetamol concentration. The B-coefficients for
paracetamol in the studied systems are positive and thus suggest the presence of solutesolvent interactions or structure making properties. The values of D-coefficient are mainly negative showing weak solute-solute interactions.
The changes of free energies (ΔG#) are increased with the increase of concentration of
paracetamol and the values are positive for all the studied systems. It is also seen that the changes of free energies (ΔG#) of paracetamol in aqueous solutions of n-Propanol and n-Butanol increase very slowly with increasing solute concentration and decrease with increasing temperature. The values of enthalpy of activation (ΔH≠) indicate the interaction presence in solute-solvent through H-bonding. The entropy (ΔS≠) values are positive for all the systems and decrease with increase of paracetamol concentration.
The positive values of change of chemical potential (Δμ1 ≠ - Δμo≠) for all studied show greater contribution per mole of solute to free energy of activation for viscous flow of the
solution and are good agreement with the values of B-coefficient showing structure
making properties between solute and solvent.Shishir Kumar DeyMaster of Science in Chemistr
Study on Eigen Values And Eigen Vectors of Matrices: An Iterative Approach
No full textThis thesis is submitted to the Department of Mathematics, Khulna University of Engineering & Technology in partial fulfillment of the requirements for the degree of Master of Philosophy in Mathematics, May 2016.Cataloged from PDF Version of Thesis.Includes bibliographical references (pages 70-75).There are some smart methods, available in the literature, which are able to find out all Eigen values. But those methods could not find corresponding Eigen vectors simultaneously. Power method and Inverse Power method are able to find out both Eigen-pairs simultaneously. Power method frequently used for finding only largest Eigen value and corresponding Eigen vector. On the other hand Inverse Power method is applied to find out only smallest Eigen value (or desire Eigen value) and corresponding Eigen vector. But Inverse Power method is computationally costly and some time it is unstable for the presence of inverse of the matrix. It is theoretically observed that if all Eigen values are either positive or negative, then without implement of Inverse Power method the modified (using shifting property) Power method is also able to find out smallest Eigen-pair. Here we have proposed Modified Hybrid Iterative Algorithm based on both Power method and Inverse Power method respectively to find out both largest and smallest Eigen-pairs simultaneously. Moreover several lemma regarding the proposed algorithm have been proposed. The proof of each lemma has also been given along with some suitable illustration. Several experiments have been performed to investigate the robustness and effectiveness of the algorithm as well as the lemma. The proposed algorithm is able to find out both (largest and smallest) Eigen-pairs successfully as well as efficiently.Md. Sah AlamMaster of Philosophy in Mathematic
Modified Solution of the Nonlinear Singular Oscillator by Iteration Procedure
No full textThis thesis is submitted to the Department of Mathematics, Khulna University of Engineering & Technology in partial fulfillment of the requirements for the degree of Master of Philosophy in Mathematics, September 2016.Cataloged from PDF Version of Thesis.Includes bibliographical references (pages 45-50).An Iteration method for solving nonlinear oscillatory problems was introduced by R. E. Mickens. Mickens provided a general basis for this method as they are currently used in the calculation of approximations to the periodic solutions of various nonlinear oscillatory differential equations successfully. Latter the method of Iteration has been improved and justified by C. W. Lim, B. S. Wu and R. E. Mickens. It is a widely used technique for handling strong as well as weak nonlinear differential systems with periodic solutions. In this thesis, we have used the truncated Fourier series in each iterative step. The approximate frequencies obtained by the technique shows a good agreement with the exact frequency. Also the obtained solutions are much more accurate than other existing results.M. M. Ayub HossainMaster of Philosophy in Mathematic
Voltammetric Studies of Paracetamol with Some Functional Amines at Different pH Media
No full textThis thesis is submitted to the Department of Chemistry, Khulna University of Engineering & Technology in partial fulfillment of the requirements for the degree of Master of Science in Chemistry, February 2016.Cataloged from PDF Version of Thesis.Includes bibliographical references (pages 140-145).Electroactive paracetamol adducts such as paracetamol-diethylamine and paracetamol-diisopropylamine were synthesized electrochemically. The products has been studied by Cyclic voltammetry (CV), Controlled potential coulometry (CPC), Differential pulse
voltammetry (DPV) and Chronoamperometry (CA) techniques using Glassy carbon (GC),
Gold (Au) and Platinum (Pt) electrodes. The studies have been carried out in variation of
paracetamol, diethylamine and diisopropylamine concentration, buffer solution of different pH, different electrodes and scan rate.
After the insertion of diethylamine and diisopropylamine in paracetamol solution at the
second scan of CV potential, new anodic and cathodic peaks appeared at the negative
potential. In the presence of nucleophiles with paracetamol as Michael acceptors the
oxidation and reduction peak shifted with respect of pure paracetamol. Also the anodic and cathodic peak current changes significantly compared with the pure paracetamol that
indicates the participation of reaction of p-quinoneimine with diethylamine and
diisopropylamine. The products generated from the reaction of paracetamol with
diethylamine and diisopropylamine are assumed to be N-(3-(diethylamino)4-
hydroxyphenyl)acetamide and N-(3-(diisopropylamino)-4-hydroxyphenyl)acetamide
respectively that undergo electron transfer at more negative potentials than the
paracetamol.
The influence of pH of paracetamol in presence of diethylamine and diisopropylamine was studied by varying pH from 3 to 9. For diethylamine and diisopropylamine, it is observed
that at pH 3-5, no new anodic peak appeared after monotonous cycling. In the pH 7, the pdienone endures nucleophilic attack by the amines that voltammetric new anodic peaks A1
and A2 appeared after monotonous cycling. Whereas in pH 9, the voltammogram display a new peak at the first scan of potential. In lieu of diethylamine and diisopropylamine, the
maximum peak current is observed at pH 7. The slopes of the peak potential, E vs pH plot
was determined graphically as the anodic peaks of paracetamol-diethylamine (37 mV/pH
for first anodic peak A3) and paracetamol-diisopropylamine (23 mV/pH for first anodic
peak A3) at 0.1V/s in the cycle. This ascribed that the oxidation reaction of
paracetamol-diethylamine and paracetamol-diisopropylamine adducts proceeded via the
2ei2H process. This also indicates that during the reaction not only electron but also
proton are released from the paracetamol-amine adduct. The reaction was strongly
influenced by the pH as well as concentration of diethylamine and diisopropylamine. The
reaction was mostly favorable in 150mM of diethylamine and 150mM of diisopropylamine
with fixed 2mM of paracetamol at pH 7. The electrochemical reaction of paracetamol-diethylamine
and paracetamol-diisopropylamine adduct are facilitated in neutral media.
The effect of scan rate on cyclic voltammogram of paracetamol in presence of
diethylamine and diisopropylamine were also investigated. The peak currents of both the
anodic and the corresponding cathodic peaks increase with the increase of scan rate. The
nearly proportionality of the anodic and corresponding cathodic peak ascribes that the peak
current of the reactant at each redox reaction is controlled by diffusion process. The
current function, I1,/v"2 of paracetamol-amine derivatives were found to be decreased
exponentially with increasing scan rate. In all the studies of paracetamol-amine derivatives
have been found to undergo electrochemical reaction with ECE (electron transfer-chemical
reaction- electron transfer) type mechanism.
Controlled-potential coulometry was studied of catechol in presence of diethylamine and
diisopropylamine. During the course of coulometry, the appeared peaks height increase
proportionally to the advancement of coulometry, parallel to the decrease in the height of
paracetamol peak. These observations also indicate the participation of reaction of
paracetamol with the studied nucleophiles. The electro-synthesized products originated
from the bulk electrolysis of paracetamol with diethylamine and diisopropylamine were
isolated. The formations of new products were also ascertained by FTIR spectra.Md. Hafizur RahmanMaster of Science in Chemistr