Empowering Sustainable Living Through Edge-AI: A Solar Estimation System for Mindful Energy Consumption and Regenerative Homes

Abstract

Proper assessment of rooftop solar capacity is the central focus of optimization of photovoltaic (PV) system and development of the renewable energy industry. This chapter introduces a hybrid geospatial and machine learning (ML)-based model that proposes highly accurate predictions of rooftop solar energy production. The authors have assembled a system that operates Mapbox Graphics Library to interactive spatial visualizing, Photovoltaic Geographical Information System (PVGIS) Application Programming Interfaces (APIs) that the authors utilize to access past and present solar radiation data sets, and Scikit Learn that the authors use to infer regression-based prediction models such as Random Forest and XGBoost. The authors also created a full stack web app that relies on React.js, TailwindCSS and HTML5 as the front end based on responsiveness and the node.js and express as the back-end services that are scalable. The users can enter parameters of rooftop dimensions, orientation, and shading dynamically and the system estimates potential energy based on data of the solar insolation and trained models. Users get the real-time feedback and relative analytics of cost and environmental impact. The methodology developed with the help of location-specific environmental data and ML will provide exact solar calculators with better predictive ability and much less error margin. In line with contributing to the concept of smart micro energy planning, this work illustrates the pillars of Industry 6.0 that consist of hyper-personalization, intelligent automation, and sustainable practices. This chapter presents the argument of conscious consumption of energy and environmental friendliness that subsequently is consonant with the globalization tendencies of decarbonization and sustainable development agenda.