Tripura Institute of Technology

About: Tripura Institute of Technology is a based out in . It is known for research contribution in the topics: Electric power system & Renewable energy. The organization has 63 authors who have published 92 publications receiving 510 citations.

Utilization of Electric Vehicle for LFC of Multi-area System Using Wind Driven Optimized Classical Controllers

Abstract: This paper focuses on load frequency control (LFC) of an integrated unequal three area thermal system with the incorporation of electric vehicle (EV) in all three areas. An appropriate generation rate constraint of 3% per minute is considered in each of the three areas. The performance of conventional controllers like Integral (I), Proportional-Integral (PI) and Proportional-Integral-Derivative (PID) controllers are evaluated on the system. A new nature-inspired optimization technique named as Wind Driven Optimization (WDO) technique is utilized for optimizing parameters of the controllers. Comparison between the responses obtained by using I, PI and PID discloses the superior performance of PID controller from the point of view of time of dying out of response, highest deviation and degree of oscillations. The study has been stretched to the application of random loading in Area-1 in place of step load. Also, an analysis is carried out to evaluate the impact of incorporation of EV in the system, which reveals the improvement of system dynamics with the incorporation of EV.

A Novel Cryptographic Approach to Implement Confidentiality Using Point Reflection Property

Abstract: Data Privacy is a prime concern in 21st century. To maintain secrecy of data, many cryptographic algorithms are available. Cryptography is the science or study of the techniques of secret writing, esp. code and also figures frameworks, techniques, and so forth. Cryptography is required so content can be kept secret. Here in our proposed work we have shown a simple point reflection based cryptographic technique for security of data. The user encoded the data using point reflection technique and the receiver can decode the data by applying the process in reverse direction. The encrypted message generated using the mechanism proposed here is quite sturdy and it becomes more difficult for a person to retrieve or predict the original message from the encrypted message. We used PRNG technique by which we can get the random number and also use the LFSR technique for robustness of this technique. Experimental result shows that our method gives a high security and the accuracy during the encoding and decoding process. We narrated the algorithms and explicit formulas to demonstrate that our reflection process support high practical security in cryptographic applications.

Comparative analysis of low power novel encoders for Flash ADC in 45nm technology

Abstract: The electronics world is faced with various challenges when attempting to integrate analog signals with discrete or digital systems. The ADC is not a new concept b y any means, but is still a topic of interest when it comes to technology due to its inevitable necessity in many systems being used today. Power consumption in ICs was always a challenging issue. Most of the technology emphasizes on speed, accuracy, less power consumption. In almost all digital functioning, encoders are mandatory. Decreasing power consumption in an encoder is an area of strong desire. In this paper, we have designed two novel ultra low power encoders for Flash ADC application and also show its comparison with previous encoder available in literature. The work has been done on Tanner tool v15 in 45 nm technology with various DC voltage supplies.

Enhanced Colliding Bodies Optimization Towards the Solution of Economic Load Dispatch Problem

Abstract: This paper proposes application of enhanced colliding bodies optimization algorithm (ECBO) for solving the economic load dispatch problem, which aims to get minimum generation cost through economic scheduling of generating unit. The algorithm is a novel efficient optimization algorithm, and its idea is attained from collisions, which are one-dimensional between two or more bodies. Each agent involved in this collision is shaped as a body having defined mass and defined velocity. To analyze the performance of the proposed method, it is tested on four different test systems consisting of 3, 5, 13, and 18 generating units. Both convex and non-convex fuel cost function have been considered. Numerical results have been compared with various well-known algorithms. A significant improvement in the results has been observed. It has been farther noted that an average 5.1% better results were provided by ECBO in terms of generation cost compared to other algorithms. Moreover, the simulation time is also improved. Besides this, Kruskal-Wallis non-parametric test has been performed.

Optimal Geometric Active Contours: Application to Human Brain Segmentation

Abstract: An efficient Segmentation of lateral ventricles plays a vital role in quantitatively analyzing the global and regional information in magnetic resonance imaging (MRI) of human brain. In this paper, a semi automatic segmentation methodology to support the study of efficient pathologies of the lateral ventricles along with white matter and gray matter of human brain is proposed. The segmentation is executed using an optimal geometric active contour with level set methods. A nominal anatomical knowledge is incorporated into the methodology in order to choose the most probable surfaces of the lateral ventricles of human brain, even if they are disconnected, and to eliminate addition of non ventricle cerebrospinal fluid (CSF) regions. The proposed segmentation method is applied to multislice MRI data and compared with region growing algorithms. The results Dice similarity coefficient 0.955, Jaccard similarity coefficient 0.815 demonstrates the reliability and efficiency.