A hot-box solar cooker designed for domestic applications had its thermal performance initially evaluated from experimental measurements, and then numerically using artificial neural networks (ANNs). Ambient temperature, temperature of the cooker and that of a known volume of water were measured as functions of incident solar radiation intensity and time of day. The cooker output power and efficiency, and the exergy output and efficiency computed from experimental measurements, constituted the four cooker thermal performance parameters. Experimental results show that the cooker has peak energy output of 21.3 W and peak energy efficiency of about 12 %. Peak exergy out was 1.4 %, while peak exergy efficiency was about 0.9 %. ANNs were then employed to estimate the four cooker performance parameters, for arbitrary inputs. A feed-forward neural network based on the Levenberg-Marquardt backpropagation algorithm was developed that used the five experimentally measured variables as inputs, while the four computed thermal performance parameters were the targets. Statistical parameters were used to rate the performance of ANNs in estimating the four experimental thermal performance parameters. Linear regressions were used to find relationships between ANN estimated and experimental energy/exergy outputs, and efficiencies. Results show good correlations between ANN predictions and experimentally measured cooker performance parameters. Correlation coefficients for all the four thermal parameters were very close to unity, with relatively low root-mean square errors (rmse) of only up to ±0.20. Peak mean absolute errors (mae) were about 0.10 W for cooker power, while mean absolute percentage errors (mape) were in the range of 0.57−1.25 %. High values of the correlation coefficients and relatively low error values show that the ANN model developed here successfully estimates cooker power, energy efficiency, exergy and exergy efficiencies with some good degree of accuracy.
The present study reports the histopathological, haematological and biochemical changes in broiler chicken naturally infected with Eimeria acervulina. A polymerase chain reaction (PCR) was used to identify this species in 225 gut samples of broiler chicken from different farms in North Eastern (NE) region of India. Postmortem examination revealed greyish white transeversely elongated area on the serous surface, oedema, together with necrosis and sloughing of intestinal epithelium.Haematological changes included a decrease in haemoglobin,(Hb) and packed cell volume. The value of mean corpuscular haemoglobin concentration (MCHC), on the otherhand increased slightly. Biochemical changes showed a significant increase in the level of glucose, cholesterol, Alaninine amino transferase (ALT), Asparase amino transferase (AST) and Alkaline phosphatase. The PCR assay was based on internal transcribed spacer (ITS1) region of the ribosomal DNA ofEimeria sp which has shown interspecies variation that enables to differentiate the species. The six isolates of Eimeria acervulina obtained were sequenced and a phylogenetic tree wasprepared. The sequences of the six isolates were searched for matching with isolates available in the Gen Bank database for sequence similarity using nucleotide blast. The sequence analysis showed that the newly isolates of E. acervulina had 99% Similarity with isolate of Turkey origin.
In this paper Improved Grasshopper Optimization Algorithm (IGSD) and Electric Field Algorithm (EFA) is designed to solve the reactive power problem. Real power loss reduction is the key objective in this work. Grasshopper Optimization Algorithm has been hybridized with self –adaptive differential algorithm. Social communication, gravity force and wind advection are basics for Grasshopper movement. In self –adaptive differential algorithm mutation, cross over and selection are the main operators. Proposed IGSD algorithm increases the exploration competence and population diversity will be maintained in last phase of the iterations. Then In this paper Electric Field Algorithm (EFA) is projected to solve the problem. Based on the coulomb law of electrostatic force proposed algorithm has been modelled. In the proposed Electric Field Algorithm (EFA) electrostatic attractive force has been considered and in that highly charged particle, (best particles) attract the particles which possess low charges sequentially there will be movement in the exploration space. Proposed Improved Grasshopper Optimization Algorithm (IGSD) and Electric Field Algorithm (EFA) is tested in IEEE 30, bus system- real power loss minimization, voltage deviation minimization, and voltage stability index enhancement has been attained. Then the Proposed Improved Grasshopper Optimization Algorithm (IGSD) and Electric Field Algorithm (EFA) has been tested in standard IEEE 14, 30, 57, 118, 300 bus test systems without considering the voltage stability index. Projected algorithms reduced the power loss effectively and control variables are within the limits.
The extraction of robust vocal features, immersed in different ambient noises, in a noisy crowded environment, is one of the fundamental challenging issues in Speech Processing. In this paper, a new feature extraction method, named Fractional Root Coefficient (FrRC), has been introduced for speech recognition application in noisy environments. The proposed FrRC method is based on the short-time Fourier transform (STFT) and the Root function. Since the selection of fractional and root transform coefficients for the proper analysis of multicomponent signals, such as speech are still being actively researched, hence, in the FrRC method, the optimal alpha and gamma parameters are obtained respectively for the fractional Fourier transform and the root function, using Metaheuristic algorithms. The TIMIT and Noisex-92 databases are utilised to evaluate the amount of robustness and speech recognition accuracy of the recognition system. The simulation results indicate high robustness and high recognition accuracy of the novel proposed FrRC method compared to other functional extraction methods in severe noisy environments. Also, by the analytical and mathematical method, the efficiency of the proposed FrRC method is verified.
This paper involves a simulation study to predict the maximum and minimum eigenvalues of a random matrix whose elements are coming from binomial distribution. In the first study, we fix the number of trials n and for different values of the probability of success p we generate 100 random matrices of order 10×10 in MATLAB. Next, we will plot a graph between mean eigenvalue (maximum and minimum separately) and p that obtain the equation of the best curve fit using MS-EXCEL. \nIn the second study, the same procedure is repeated except that here we fix pandvary n. Both these studies are repeated for random matrices of 5×5 and it is observed that the regression equations for predicting the maximum or minimum value of eigenvalue does not get affected much by reducing the order of the random matrix. Additionally, we also give a theoretical analysis of predicting the range of the sum of all the eigenvalues of a diagonalizable random matrix with the help its trace and Chebyshev’s inequality.
The interaction of dye and surfactant is studied in ethanol-water mixture spectrophotometrically. The wide range of interaction has been determined by obtaining the binding constant, partition coefficient and standard Gibb�s free energy between anionic dye methyl red (MR) and cationic surfactant dodecyl-trimethyl ammonium bromide (DTAB) in volume fraction of ethanol of 0.1, 0.2, 0.3 and 0.4 respectively. The concentration of DTAB are within ? 0.12 M to 0.0001 M and the concentration of MR is ? 0.0003 M.