Showing papers by "Sir Padampat Singhania University published in 2022"

Machine Learning Techniques for Music Genre Classification

Abstract: Music plays a significant role in one’s life. The quantity of music released daily is increasing tremendously on Internet platforms like Soundcloud and Spotify. Music brings like-minded people together. Classification has been a challenging task in the field of music retrieval (MIR). Categorization of music can help explain some of the exciting problems such as creating song references, finding related songs, and finding communities that will love that particular song. The aim of our research is to find the best machine learning algorithm that predicts the type of songs using machine learning methods. This paper provides a comparative analysis of the varied machine learning methods used in the field of genre planning.

Microgrid Energy Management Strategy for Low Dynamics Load Applications Using Simulink

Abstract: This paper presents an energy management strategy for microgrid consisting of photovoltaic, wind energy conversion system, battery, fuel cell, and electrolyzer system. Rule-based energy management strategy is designed to meet efficiently the slow varying residential load demand of rural households within the individual source limits. Simulation is carried out using MATLAB. Simulation results show that the developed energy management strategy is able to meet the instantaneous load demand in varying irradiance, wind speed, and load conditions keeping the battery state of charge and electrolyzer hydrogen pressure in allowable limits.

Lean-SE: Framework Combining Lean Thinking with the SDLC Process

Abstract: Software development process has evolved with the objective to develop methodologies that would adapt to the changing nature of software. Even software has evolved from a scientific equipment to a ubiquitous device. This ubiquitous nature demands it to be more user centric. Developing user-centric products requires strong and continuous customer collaboration and ensuring that no product is developed without a purpose to the user. This paper presents a framework combining lean thinking activities within the SDLC processes models with an aim to strengthen the analysis phase. This framework can be integrated within the waterfall and agile model during the analysis phase to better understand the requirements and transform them into features for the development team. Integration may make the analysis and design phase a bit stretched out, but the resulting requirement will be specific and clear and aid to mitigate changes, improve the end product and user satisfaction.

Analysis of Electromagnetic Interference in Solar Photovoltaic Grid System

Abstract: Electromagnetic interference (EMI) generated in grid-connected solar photovoltaic (SPV) system is addressed in this research paper. The major emphasis has been given on the issues related to generate EMI magnitude due to PV panel capacitance to earth, Common Mode (CM) interference due to switching of inverters, and the length of DC cable in medium power grid-connected SPV system. The CM EMI is more dominant as compared to Differential Mode (DM) EMI. The simulation results of the generated EMI at 50 kW grid-connected SPV installed on the roof-top are shown. The results show that the conducted EMI generated by inverters changes its behavior under various circumstances. The generated EMI is compared against the CISPR 22 standard of electromagnetic compatibility. Switching control techniques are tested in this system. The output performance of these strategies is also discussed in this paper.

Analysis of Energy Efficiency in Hierarchically Structured Homogeneous Routing Protocols for Green WSN

Abstract: The world of IoT has witness huge expansion in the past few decades, which has brought our focus on the energy conservation aspect of wireless sensor networks. WSN constitutes of small sensing units which are responsible to sense the event, compute its sensed data and communicate it to the receiving base station for further processing and decision making. Sensors are extensively used for various applications ranging from environmental monitoring to health care and to surveillance. Green WSN focuses to optimize the amount of energy used by these tiny sensors. These sensors sustain in the network for long period of time on limited power supply of battery. Therefore, managing energy decay of sensors for increasing network lifespan is an underlying matter of concern in these sensor networks. Efficient routing is a prominent technique that targets on developing a Green WSN by reducing energy used in communication. Furthermore, routing protocols which are based on network structure are majorly categorized into flat, hierarchical and location-oriented routing protocols, among which hierarchically structured homogeneous routing protocols are best-suited solution for reducing energy utilization of WSN, as it avoids unnecessary and repetitive data transmission. In this paper, we have analyzed and created a Green WSN using Leach, Heed, Pegasis and tree-based clustering protocols that deals in energy efficiency and simulation is done in MATLAB 2019b. The results exhibit that TBC protocol surpasses Leach, Heed and Pegasis protocol with respect to the number of packets addressed to the receiving base station, average residual energy and the round through which the first node dies, hereby increasing the lifespan of the network.

Study of Radiant Cooling System with Parallel Desiccant Based Dedicated Outdoor Air System with Solar Regeneration

Abstract: Radiant cooling system (RCS), along with dedicated outdoor air system (DOAS) is an appropriate system to enhance energy efficiency and maintaining better thermal comfort conditions compared to the compression based system. In general, DOAS uses a fan coil system to dehumidify the air which consumes a significant amount of energy. This chapter describes desiccant wheel based dehumidification system for DOAS using the solar energy for the desiccant regeneration. Different dehumidification techniques were discussed in the chapter using the simulation study conducted on the software (EnergyPlus). A reference medium scale office building was considered for simulation study. Impact of RCS operated by compression and absorption based system and coupled with desiccant assisted DOAS was designed for the office building and involvement of solar energy towards the annual energy consumption were investigated. The comparative analysis shows the energy saving potential and feasibility of absorption system for RCS against the compression system. The simulation study was conducted for two climatic conditions i.e. warm-humid and hot-dry climate. The simulation results realize the energy saving potential of later case up to 10.91% and 28.7% for warm-humid and hot-dry conditions, respectively. Next, the contribution of solar energy can be ensured up to 99%.