A TOT: tri-optimized-tariff based strategic residential load management with greedy optimization in IEEE33-bus system: a case study with renewable energy penetration

Abstract

The efficiency of a load management system in terms of its energy performance index (EPI) depends on its capacity to enhance the reliability, resilience, and cost effectiveness of the existing system. Artificial intelligence (AI) is crucial in this shift from classical to AI-based power system planning, optimizing renewable energy (RE) and reducing gridstress. On the other hand, proper placement of resources is essential to achieve benefits and reduce transmission losses. Utility sectors of different states has revealed that in certain areas amongst different type of loads, domestic loads accounts for a substantial proportion of energy consumption. Therefore, the present work deals with optimum load scheduling, integration of RE, energy storage (ES) and proposed tri-optimized-tariff (TOT) for prosumers. We have found that the weighted-K-nearest-neighbor (KNN) method excels in selecting features for household appliances and ES scheduling. The composite greedy optimization (CGO) technique outperforms existing methods in optimization. These results demonstrate the efficiency and real-world potential of our model. We have conducted a case study and developed an AI-based strategic-residential-load-managementsystem (SRLMS), which we have tested on the IEEE33 bus system, showing cost effectiveness and improved EPI for prosumers. This work encourages the development of a harmonious relationship between utility-sectors and prosumers.