Energy management using novel nature inspired optimization algorithms

Abstract

Today the growing energy consumption greatly influences the socio-economic progress across the world. However, such enormous power demand negatively impacts the environment context by incessant usage of conventional primary sources (coal, fossil fuels) which further leads to huge carbon emission. Hence, many developing countries have encouraged the clean energy production for its huge abundance, sustainability, low-emission together with minimizing the reliance on traditional fossil fuels and promote energy savings to attain long-term development. Moreover, the insertion of clean energy sources mitigates the weakness of centralized energy system includes environmental impacts, line losses and high costs. Nevertheless, the reliance on various environmental and geographical indicators has brought significant planning and control issues. Thus, the integration of green energy with efficient microgrid technology has become the holistic solution to address these challenges. The low-power microgrid framework delivers enormous possibilities of improving system efficacy and reliability by incorporating various distributed resources (conventional, renewable) and controllable loads effectively by working either in an isolated or grid-connected way. Further, the stochastic and volatile nature of clean sources with limited power availability, specifically during isolated operation, creates a major issue in managing the essential generation-demand ratio for increased system resilience. Hence, this research mainly focuses on solving the generation-demand problem in microgrids using distributed energy management framework to minimize the operational cost. The integration of clean, renewable sources and microgrids indeed requires a comprehensive Energy Management System (EMS) that primarily targeted at resource scheduling, design, optimization, and control. In addition, energy management with an optimization algorithm is critical for the microgrid platform to effectively oversee the cluster of various sources to fulfill the demand economically. Mainly, the energy management problem is typically newlineformulated as a non-linear, complex model with two different centralized and distributed schemes. The centralized architecture demands high computational ability to solve the optimization problems and impinges on consumer privacy. Alternatively, most existing literature on distributed strategies presume that all demand and generation are interconnected to a single bus which overlooks the main power networks as well as accompanying system and power flow constraints newline