@lnmiit.ac.in
Assistant Professor, Dept. of CSE
The LNM Institute of Information Technology Jaipur
Artificial Intelligence, Autonomous Agent and Multiagent Systems, Intelligent Transportation System, Internet of Vehicles, Distributed Systems.
Scopus Publications
Poulami Dalapati and Saurabh Kumar
CRC Press
Kaushik Paul, Poulami Dalapati, and Niranjan Kumar
Springer Science and Business Media LLC
Kaushik Paul, Niranjan Kumar, and Poulami Dalapati
Springer Science and Business Media LLC
Kaushik Paul and Poulami Dalapati
Springer Singapore
Poulami Dalapati and Kaushik Paul
Springer Singapore
Kaushik Paul*, , Niranjan Kumar, Poulami Dalapati, , and
Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP
The power system congestion is treated as a vital issue in the restructured topology of the power system. The analysis of appropriate technique to control congestion is of preeminent interest. This paper proposes a congestion controlling scheme with the optimal placement and sizing of the Distributed Generation (DG) so as to ensure an optimal power flow in the power system network. A multi-objective framework is formulated for the proposed approach considering the operating cost, Voltage Stability Index (VSI) and the system losses. A hybrid optimization technique is proposed involving Improved Genetic Algorithm (IGA) and Bat Algorithm (BA) to optimize the objectives proposed in this research. The efficiency of the proposed methodology is verified using IEEE 33 and 69 bus systems. A comparative analysis is established between the outcomes obtained with hybrid IGA-BA and Particle Swarm Optimization (PSO) technique. The output obtained clarifies that by combining IGA and BA, greater efficiency is achieved compared to the PSO algorithm output.
Poulami Dalapati, Abhijeet Padhy, Bhawana Mishra, Animesh Dutta, and Swapan Bhattacharya
Informa UK Limited
Abstract Advancement in intelligent transportation systems with complex operations requires autonomous planning and management to avoid collisions in day-to-day traffic. As failures and/or inadequacy in traffic safety systems are life-critical, such collisions must be detected and resolved in an efficient way to manage continuously rising traffic. In this paper, we address different types of collision scenarios along with their early detection and resolution techniques in a complex railway system. In order to handle collisions dynamically, a novel agent-based solution approach is proposed using the idea of max-sum algorithm, where each agent communicates and cooperates with others to generate a good feasible solution that keeps the system safe, i.e. collision free. We implement the proposed mechanism in Java Agent DEvelopment Framework (JADE). The simulation results are evaluated with exhaustive experiments and compared with different existing methods to show the efficiency of our proposed approach.
Poulami Dalapati, Piyush Agarwal, Animesh Dutta, and Swapan Bhattacharya
Informa UK Limited
ABSTRACT This paper addresses the issues concerning resource allocation and process scheduling in a dynamic environment, where resources are distributed and availability of them is uncertain. In this context, we introduce a new multi-agent-based resource allocation and process scheduling approach, where agents communicate and cooperate among themselves to produce an optimal schedule. A distributed constraint optimization problem-based model in accordance with Markov Decision Process is proposed in this regard. We overcome the hardship of existing centralized approach and our technique optimizes not only the process completion delay but also the number of resources being idle, which is much more beneficial. Apart from the theoretical approach, we take a case study in its practical application domain to validate our claim. Analysis and experimental results show that this proposed method outperforms the state-of-the-art methods and bridges the gap between theory and its applications.
Poulami Dalapati, Arambam James Singh, and Animesh Dutta
IEEE
Disaster management in railway network is an important issue. It requires to minimize negative impact and also fast, efficient recovery from the disturbances. The main challenge here is that, the effect of inconvenience spreads out very fast in time and space. It takes noticeable amount of time to get back everything in the previous situation. This paper proposes a multi agent based algorithmic approach for disaster handling in Railway Network. This takes care of fast response to get total number of affected trains in a fast and efficient manner. We propose few algorithms to handle this situation and simulate it using JADE (Java Agent Development Framework) platform. Finally we take a case study and compare our proposed method with an existing manual technique.
Poulami Dalapati, Arambam James Singh, Animesh Dutta, and Swapan Bhattacharya
IEEE
This paper proposes a multi agent based timetable scheduling algorithm for railway system which handles the in-between time delay of the newly introduced train. The delay management indeed optimizes the total journey time, hence increases the total utility of the whole railway system as well. Here we show that schedule generated by our proposed algorithm is the most optimized schedule. It is done by using the notion of DCOP(Distributed Constraint Optimization Problem), where we define some metric to analyze the system to achieve our goals. We use JADE(Java Agent DEvelopment Framework) platforms to simulate our work and test it using a small network. We also take a small case study to compare our proposed work with the existing one and the results are therefore presented.
Arambam James Singh, Poulami Dalapati, and Animesh Dutta
Springer International Publishing