An intelligent landscaping framework for net-zero energy smart cities: A green infrastructure approach

Utility providers acknowledge the significance and indispensability of demand response (DR) initiatives in curtailing the escalating generation expenses linked to growing energy needs. The advent of sophisticated data and communication arrangements has paved the way for cost -centric DR programs as a practical means to regulate loads in intelligent landscaping design of microgrids (MGs). Individuals in residential settings are increasingly adopting unpredictable RESs, such as photovoltaic (PV). This research investigates a unified landscaping framework aimed at addressing issues related to the scheduling or commitment of loads in households involving RESs, regardless of the tariff structure applied. Decision -making problems influenced by uncertainty find effective resolution through reinforcement learning (RL). The study puts forth an RL-powered resolution to load commitment challenges in intelligent MGs. An inventive facet of the research lies in the formulation of a comprehensive framework based on unscented transform (UT) that factors in user contentment, unpredictable renewable energy, and tariff considerations. Through simulated assessments, the proposed framework is evaluated for its efficacy and adaptability. An in-depth analysis of the methods' performance is presented using a residential user scenario featuring devices with and without schedulable characteristics, in addition to a PV resource.