Performance Optimization and Net Yield Evaluation of a Fuzzy Logic-Driven Single vs. Dual-Axis Solar Tracker for 100 kW PV Systems under Lucknow's Climatic Profile

Abstract

The present paper is an extensive study on performance optimization and techno-economic net yield of a 100 kW grid-tied photovoltaic (PV) system operating under the unique meteorological conditions of Lucknow, Uttar Pradesh, India. Continuous parasitic motor tracking under highly diffuse, non-directional and scattered light has a negative impact on the net energy balance of conventional tracking systems in the extended monsoon season and severe winter morning fog. To eliminate these losses, an AI based optimization technique is used, intelligent Mamdani-type Fuzzy Logic Controller (FLC) was designed and implemented via a Python programming framework. The controller continuously samples the hourly real-time Sky Clearness Index K_t and solar angle of incidence θ to dynamically govern mechanical tracker actuation states. Using pvlib-python and scikit-fuzzy libraries, three primary configurations were simulated and contrasted over a full annualized cycle: a static Fixed-Tilt System (FTS), a Single-Axis Tracking System (SATS), and a Dual-Axis Tracking System (DATS). The definitive analytical results prove that while standard DATS achieves marginally higher gross yield, the FLC-driven SATS delivers the optimized net yield matrix by eliminating up to 42.6% of redundant parasitic motor tracking rotations on overcast days. Consequently, this optimization drastically reduces the Levelized Cost of Electricity (LCOE) and shortens the lifecycle investment payback loop to an optimal 5.1 years under regional climatic realities.