A scalable approach to efficient house power consumption and CO2 management through fuzzy logic

Efficient residential load power management represents, on one hand, a cornerstone towards Smart City Grids while, on the other hand a multifaceted challenge. The identified interplay between multiple parameters concerning both power consumption as well as generation, call for novel design approaches such as Fuzzy Logic, machine learning and AI regarding respective algorithms. Furthermore, aiming to propose highly reliable and practical solutions, respective implementations much be undertaken and evaluated in the context of real-world prominent communication and power control infrastructures. Driven, by such observations this paper presents a complete, efficient and feasible proposal able to tackle respective requirements. Specifically, a comprehensive multimodal Fuzzy Logic Inference (FIS) design is proposed able to incorporate heterogeneous factors pertaining both to residential power consumption and generation. Additionally, proposed designed is implemented in the context of a fully functional KNX based infrastructure monitoring and controlling power consumption of a real residence. Respective performance evaluation, firstly, serve as FIS implementation validation indicating convergence between Matlab based and KNX based performance. Secondly, it highlights the benefits the end-user can anticipate using such a system. Finally, an estimated projection is attempted based on actual performance results and Europe wide statistics concerning residential power consumption as well as CO
₂
omissions.