Applying Big Data analytics for energy efficiency.
Perustieteiden korkeakoulu | Master's thesis
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.
International Design Business Management
Master’s Degree Programme in International Design Business Management (IDBM)
7 + 73 + 5
AbstractGlobal energy requirements are continuously increasing. Conventional methods of producing more energy to meet this growth pose a great threat to the environment. CO2 emissions and other bi-products of energy production and distribution processes have dire consequences for the environment. Efficient use of energy is one of the main tools to restrain energy consumption growth without compromising on the customers requirements. Improving energy efficiency requires understanding of the usage patterns and practices. Smart energy grids, pervasive computing, and communication technologies have enabled the stakeholders in the energy industry to collect large amounts of useful and highly granular energy usage data. This data is generated in large volumes and in a variety of different formats depending on its purpose and systems used to collect it. The volume and diversity of data also increase with time.\ All these data characteristics refer to the application of Big Data. This thesis focuses on harnessing the power of Big Data tools and techniques such as MapReduce and Apache Hadoop ecosystem tools to collect, process and analyse energy data and generate insights that can be used to improve energy efficiency. Furthermore, it also includes studying energy efficiency to formulate the use cases, studying Big Data technologies to present a conceptual model for an end-to-end Big Data analytics platform, implementation of a part of the conceptual model with the capacity to handle energy efficiency use cases and performing data analysis to generate useful insights. The analysis was performed on two data sets. The first data set contained hourly consumption of electricity consumed by a set of different buildings. The data was analysed to discover the seasonal and daily usage trends. The analysis also includes the classification of buildings on the basis of energy efficiency while observing the seasonal impacts on this classification. The analysis was used to build a model for segregating the energy inefficient buildings from energy efficient buildings. The second data set contained device level electricity consumption of various home appliances used in an apartment. This data was used to evaluate different prediction models to forecast future consumption on the basis of previous usage. The main purpose of this research is to provide the basis for enabling data driven decision making in organizations working to improve energy efficiency.
SupervisorVartiainen , Matti
Thesis advisorScepanovic, Sanja
big data, energy efficiency, Hadoop, advanced analytics, CIVIS project, classification.