A comprehensive analytical exploration and customer behaviour analysis of smart home energy consumption data with a practical case study. (November 2022)

• Record Type: Journal Article
• Authors: Purna Prakash, K.; Pavan Kumar, Y.V.; Reddy, Ch. Pradeep; Pradeep, D. John; Flah, Aymen; Alzaed, Ali Nasser; Al Ahamdi, Ahmad Aziz; Ghoneim, Sherif S.M.
• Is Part Of: Energy reports. Volume 8(2022)
• Page Start: 9081
• Page End: 9093
• Publication Date: 2022-11
• Subjects: Customer behaviour analysis -- Energy consumption -- Energy data analytics -- Exploration -- Smart building data -- Smart home data -- Smart meter data; Power resources -- Periodicals; Energy industries -- Periodicals; Power resources; Periodicals; Electronic journals; 621.04205
• DOI: 10.1016/j.egyr.2022.07.043
• Languages: English
• ISSNs: 2352-4847

Abstract

Abstract: Over the years, the automation of traditional power grids has been taking place to overcome the difficulties such as blackouts, outages, demand-side management, load profiling, enhancing customer participation, etc. This automation enables the traditional grids to be transformed into smart grids. Smart homes/buildings are key sub-categories of smart grids. The advanced metering infrastructure connected to them continuously captures and stores the energy consumption data as datasets. Usually, understanding the structure of data and the behaviour of customers from energy consumption datasets is a tedious task. There are some literature works tried to explore various smart home energy consumption datasets as well as investigate customer behaviour, however, most of these methods are complex in implementation. Hence, this paper proposes a simple approach for the comprehensive exploration of the smart home energy consumption dataset. This approach can be used for any similar smart home dataset that contains numerical data. Further, using the exploration results, this paper analyzes the customers' energy consumption behaviour by identifying peak hours in communication and electrical perspectives. To implement the proposed approach, an energy consumption dataset 'Tracebase' is considered as a case study. The exploration of the considered dataset results in 2356 files distributed among various directories. For customer behaviour analysis, the energy consumption data of all 43 appliances (with more than 95 million records) is considered from the "complete" directory of the "Tracebase" dataset. This analysis revealed the peak hours as hour-23 from the communication perspective and hour-9 from the electrical perspective. These represent the customer behaviour in terms of their participation in the power network, which further helps for better grid operations.