Energy awareness through conversational interfaces and prediction on dynamic residential tariff context

• Grant number: 20/05763-6
• Support Opportunities: Research Grants - Innovative Research in Small Business - PIPE
• Duration: March 01, 2021 - November 30, 2021
• Field of knowledge: Interdisciplinary Subjects
• Principal Investigator: Victor Takashi Hayashi
• Grantee: Victor Takashi Hayashi
• Host Company: Victor Takashi Hayashi Tecnologia da Informação Ltda
• CNAE: Distribuição de energia elétrica ; Portais, provedores de conteúdo e outros serviços de informação na internet ; Pesquisa e desenvolvimento experimental em ciências sociais e humanas
• City: Santo André
• Associated scholarship(s): 21/02545-0 - Mobile applications with push notifications by profile, BP.TT

Abstract

The IoT solution described here constitutes a platform with a differentiated architecture in terms of software engineering as it balances the strength and vulnerability aspects of its elements of processing, communication, integration of sensors and IoT actuators, through a focused project availability, fault tolerance, accuracy and usability. Despite the specific scientific and academic terms of software engineering, the biggest highlight of the digital IoT solution, here called Good Energy, is not technical. It is its potential for digital intelligence to support energy consumers to face the digital transformation through sharing, digital inclusion due to the aspect of electricity, necessary for everyone. Consequently, with impacts on the environment and sustainability. The solution could be another elitist circuit for the upper class, by building very sophisticated smart homes. It is not. The solution is based on components and micro services that facilitate the configuration of integration and processing, from the cheapest to the most sophisticated, always aligned with the cost-benefit relation. The basis of the solution is focused on monitoring the consumption of electricity in a home. The solution presents a series of digital connectivity features, creating the possibility of integrating housing units into a digital energy control network, creating the "IoT social network", as the network that integrates and interacts people and things, thus connecting people and companies. Based on this foundation, the Good Energy Project benefits from the scalability of the application on which its business potential is based. That is, instead of identifying Good Energy as an intelligent circuit that takes care of the energy consumption of a home, it should be identified as an intelligent digital node, with functions to support energy management, capable of digitally connecting the units other nodes through local networks and the internet. Thus, the solution presented constitutes an infrastructure for a digital network creating opportunities to guide energy consumption and educate the residents of the housing unit. In addition, the application of intelligent algorithms can optimize consumption for clients and distribution by energy companies, based on precision and real-time collection of information. With digital intelligence, it can expand the functionalities for services with the government, contributing to the so-called smart cities when applied at scale. A great virtue of the project is to obtain an autonomous, inexpensive, fault-tolerant unit, with easy integration with solutions from different hardware and software suppliers. Differentiated, with advanced usability by free text and voice, through integration with current usability devices such as smartphones (Android) and voice assistants (Alexa and Google). In the business model, the metric will be energy savings by consumers, and cost reduction by lessening peak energy consumption for distribution companies. Always based on the real data collected in 30 homes for the technical feasibility study of the platform. (AU)

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

FUJII, TIAGO YUKIO; HAYASHI, VICTOR TAKASHI; ARAKAKI, REGINALDO; RUGGIERO, WILSON VICENTE; BULLA JR, ROMEO; HAYASHI, FABIO HIROTSUGU; KHALIL, KHALIL AHMAD. A Digital Twin Architecture Model Applied with MLOps Techniques to Improve Short-Term Energy Consumption Prediction. MACHINES, v. 10, n. 1, JAN 2022. (20/05763-6)