scispace - formally typeset
Search or ask a question
Institution

Chittagong University of Engineering & Technology

EducationChittagong, Bangladesh
About: Chittagong University of Engineering & Technology is a education organization based out in Chittagong, Bangladesh. It is known for research contribution in the topics: Renewable energy & Dielectric. The organization has 1200 authors who have published 1444 publications receiving 10418 citations. The organization is also known as: Engineering College, Chittagong & Bangladesh Institute of Technology, Chittagong.


Papers
More filters
Proceedings ArticleDOI
01 Dec 2015
TL;DR: A framework to recognize handwritten Bangla words in real time considering different writing styles is proposed and Evaluation result for various writing styles reveals that the propose framework can recognize Bangla handwritten words with 77% accuracy.
Abstract: Handwriting recognition is one of the most important ways to ease the handling of information between man and machine. Online handwriting recognition can be a very attractive method when people feel inconvenient using keyboards to handle information with computing devices. The most complicated task associated with online Bangla handwritten recognition is to separate the adjacent characters and vowel signs from one another within a Bangla word. This problem becomes more complicated due to the variations of writing style of individuals. In this paper, we propose a framework to recognize handwritten Bangla words in real time considering different writing styles. We used fuzzy linguistic rules in order to recognize Bangla handwritten words. Evaluation result for various writing styles reveals that the propose framework can recognize Bangla handwritten words with 77% accuracy.

10 citations

Proceedings ArticleDOI
01 Sep 2016
TL;DR: In this article, the potentiality of Molybdenum Sulphide (MoS2) as buffer has been investigated to explore the higher performance of CZTS solar cells.
Abstract: The absorber layer of CZTS solar cell is a compound semiconductor which has favourable optical and electrical properties. Researchers are highly interest to investigate the CZTS solar cells for its earth abundant, non-toxic and low cost feature. Consequently, the absorber layer of CIGS is replaced to CZTS absorber layer. In this research work, the potentiality of Molybdenum Sulphide (MoS2) as buffer has been investigated to explore the higher performance of CZTS solar cells. SCAPS 2802 simulator was used to evaluate the performance of CZTS solar cells with prospective MoS2 buffer layer. The possibility of ultra-thin CZTS solar cells was examined and the higher efficiency of 17.03% (Jsc = 29.42 mA/cm2, Voc = 1.01 V and FF = 0.574) was achieved for 1 μm thickness of absorber layer. In addition, the proposed CZTS solar cells had better thermal stability at higher operating temperature also.

10 citations

Book ChapterDOI
14 Dec 2020
Abstract: The amount of textual data generation has increased enormously due to the effortless access of the Internet and the evolution of various web 2.0 applications. These textual data productions resulted because of, the people express their opinions, emotions or sentiment regarding any service or product through tweets, posts, and reviews. Sentiment analysis concerns the computational technique of determining and classifying views stated in a piece of text, primarily to identify whether the author’s attitude toward a specific issue is positive, negative, or neutral. The impact of customer review is significant to perceive the customer attitude towards a restaurant. Thus, the automatic detection of sentiment from reviews is advantageous for the restaurant owners, service providers as well as customers to make their decisions or services more satisfactory. This paper proposes, a deep learning-based technique (i.e., BiLSTM) to classify the restaurant reviews expressed in Bengali into positive and negative polarities. A corpus consists of 8435 reviews is constructed to evaluate the proposed technique. The results of the evaluation on test dataset shows that BiLSTM technique provides the highest accuracy of 91.35% compared to other existing techniques.

10 citations

Journal ArticleDOI
TL;DR: In this paper, a polycrystalline sample of ceramic oxide Ni0.27Cu0.10Zn0.63Fe2O4 was prepared by the solid state reaction method and the sintered sample was well polished to remove any oxide layer formed during sintering and two surfaces of the pellet were coated with a silver paste as a contact material.
Abstract: Abstract A polycrystalline sample of ceramic oxide Ni0.27Cu0.10Zn0.63Fe2O4 was prepared by the solid state reaction method. The sintered sample was well polished to remove any oxide layer formed during sintering and the two surfaces of the pellet were coated with a silver paste as a contact material. Among dielectric properties, complex dielectric constant (ε* = εʹ - jεʺ), loss tangent (tanδ) and ac conductivity (σac) in the frequency range of 20 Hz to 2 MHz were analyzed in the temperature range of 303 to 498 K using a Wayne Kerr impedance analyzer (model No. 6500B). The experimental results indicate that ε, εʺ, tanδ and σac decrease with an increase in frequency and increase with increasing temperature. The transition temperature, as obtained from dispersion curve of εʹ, shifts towards higher temperature with an increase in frequency. The variation of dielectric properties with frequency and temperature shows the dispersion behavior which is explained in the light of Maxwell-Wagner type of interfacial polarization in accordance with the Koop’s phenomenological theory. The frequency dependent conductivity results satisfy the Jonscher’s power law, σT(ω) = σ(o)+Aωn, and the results show the occurrence of two types of conduction process at elevated temperature: (i) low frequency conductivity, due to long-range ordering (frequency independent, region I), (ii) mid frequency conductivity at the grain boundaries (region II, dispersion) and (iii) high frequency conductivity at the grain interior due to the short-range hopping mechanism (frequency independent plateau, region III).

10 citations

Proceedings ArticleDOI
08 Dec 2014
TL;DR: Output/input voltage feedback and boost inductor average current tracking method is used for controlling the regulator and the average inductor current tracking control shows acceptable regulation of the converter output voltage.
Abstract: A two bi-directional switch based three-phase boost AC voltage regulator is presented in this paper. The operational principle of the proposed three-phase regulator is a direct consequence of using two three-phase bi-directional switches in the boost converter topology. The topology has reduced number of switches for a three-phase AC voltage regulator having output voltage regulation, input current shaping and power factor improvement capability. It provides opportunity for simple control technique to be adopted. In this paper output/input voltage feedback and boost inductor average current tracking method is used for controlling the regulator. Reduced number of switches allows simple control circuit, less switching loss, reliable operation and low cost. Steady state analysis and simulation results of close loop control of the proposed converter are presented in this paper. The average inductor current tracking control shows acceptable regulation of the converter output voltage. During sudden variation of load or input voltage the controller adjusts the duty cycle to recover to the desired output voltage. Experimental investigation of the converter without any feedback control is also described.

10 citations


Authors

Showing all 1219 results

Network Information
Related Institutions (5)
Bangladesh University of Engineering and Technology
7.6K papers, 83.9K citations

89% related

University of Dhaka
9.8K papers, 136.4K citations

83% related

Tomsk Polytechnic University
13.2K papers, 103.7K citations

79% related

Universiti Malaysia Pahang
9.5K papers, 104.4K citations

78% related

University of Engineering and Technology, Lahore
7.9K papers, 82.3K citations

77% related

Performance
Metrics
No. of papers from the Institution in previous years
YearPapers
20234
202240
2021243
2020241
2019228
2018119