Show simple item record

dc.contributor.advisorAlam, Md. Golam Rabiul
dc.contributor.authorMoti, Md Mahraj Murshalin Al
dc.contributor.authorUddin, Rafsan Shartaj
dc.contributor.authorAnik, Abdul Hai
dc.contributor.authorSaleh, Tanzim Bin
dc.date.accessioned2021-09-04T05:19:45Z
dc.date.available2021-09-04T05:19:45Z
dc.date.copyright2021
dc.date.issued2021
dc.identifier.otherID 17101301
dc.identifier.otherID 17101311
dc.identifier.otherID 17101312
dc.identifier.otherID 17101310
dc.identifier.urihttp://hdl.handle.net/10361/14968
dc.descriptionThis thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2021.en_US
dc.descriptionCataloged from PDF version of thesis.
dc.descriptionIncludes bibliographical references (pages 48-51).
dc.description.abstractElectricity is deeply integrated into both our modern society and the economy. However, with our ever-growing society and increasing demand for electricity, the scarcity of resources is deeply felt through load shedding in most third world countries. Moreover, since most of the world depends on electricity systems built around more than 60 years ago, they are becoming increasingly inefficient and fail to solve the problems of modern-day global challenges. A Smart grid is an intelligent electricity network that allows efficient and optimal electricity distribution from source to consumers through smart integration of power technologies, information, and telecommunication through the existing system. The current system is a one-way interaction that only supplies electricity to consumers. That limits the ability to respond to the ever-changing and rising demands of society. However, smart grids allow the exchange of electricity and information between producers and customers. A smart home will communicate with the grid and allow consumers to manage electricity usage through a smart meter efficiently, and that will also efficiently manage electricity bills. Inside a smart home, the Home Area Network (HAN), will integrate all smart appliances into one energy management system so that these appliances can adjust the run schedule to lessen the demand on electricity at peak times, therefore, lowering bills. Reinforcement learning and a decentralized local market through block-chain can be used for electricity load and price forecasting. It is possible to fine-tune parameters to increase overall distribution and performance through efficient feature selection and feature extraction methods. The use of block-chain will connect prosumers and suppliers in a secure and decentralized system that will be used to forecast usage and bills. Also, through the use of reinforcement learning techniques and the block-chain’s information, it will be possible to analyze prosumer behavior. So, the integration of block-chain and smart grids will increase flexibility and scalability, leading to an overall optimized system.en_US
dc.description.statementofresponsibilityMd Mahraj Murshalin Al Moti
dc.description.statementofresponsibilityRafsan Shartaj Uddin
dc.description.statementofresponsibilityAbdul Hai Anik
dc.description.statementofresponsibilityTanzim Bin Saleh
dc.format.extent51 pages
dc.language.isoenen_US
dc.publisherBrac Universityen_US
dc.rightsBrac University theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission.
dc.subjectSmart Griden_US
dc.subjectBlock-chainen_US
dc.subjectPrice Forecastingen_US
dc.subjectElectricity demand and supplyen_US
dc.subjectSmart Meteren_US
dc.subjectReinforcement Learningen_US
dc.subject.lcshReinforcement learning.
dc.titleReinforcement learning based electricity price forecasting in Blockchain based smart grid environmenten_US
dc.typeThesisen_US
dc.contributor.departmentDepartment of Computer Science and Engineering, Brac University
dc.description.degreeB. Computer Science


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record