Integrated Techno-Economic Optimization and Intelligent Decision-Making Framework for Solar PV-Based Agricultural Microgrids.

Abstract

The increasing electricity demand associated with agricultural irrigation has created a need for energy-efficient and economically viable renewable energy systems. This paper presents a techno-economic optimization framework for a solar-powered agricultural microgrid intended to supply irrigation pump loads. The proposed system integrates photovoltaic generation, battery storage, inverter-based conversion, and grid supply. Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Bi-Level Optimization, and Non-dominated Sorting Genetic Algorithm II (NSGA-II) are employed to investigate alternative system configurations. The optimization considers annual operating cost, payback period, grid dependency, and operational feasibility. The obtained results show that PSO provides the minimum annual cost of Rs. 26,091.6 and the shortest payback period of 4.2 years, whereas NSGA-II achieves the lowest grid dependency of 12%. To address the practical problem of selecting an appropriate configuration from competing optimization objectives, an AI-assisted multi-criteria decision-making layer is additionally introduced. The proposed decision layer evaluates annual cost, payback period, grid dependency, and reliability according to different agricultural user priorities. The resulting two-stage framework separates optimization-based solution generation from intelligent configuration selection and provides a flexible basis for future development of adaptive and AI-enabled agricultural energy management systems.