SIMULATION-BASED DESIGN AND EVALUATION OF A DUAL-AXIS SOLAR-TRACKING MECHATRONIC SYSTEM FOR A 50-WATT-PEAK PANEL

Abstract

Dual-axis photovoltaic tracking requires coordinated mechanical, electronic, control, and safety subsystems. This simulation-based design study develops an integrated mechatronic architecture for a 50 Wp panel using an Arduino Mega 2560, four light-dependent resistors, two 3 N·m closed-loop NEMA 23 motors, 20:1 worm reducers, 20T:60T timing pulleys, normally closed limit switches, an anemometer, an RTC DS3231, driver alarms, and a hardware emergency stop. The outer loop uses normalized azimuth and elevation light errors, an eight-sample moving-average filter, a deadband equivalent to 0.5°, and dominant-axis pulse scheduling; the inner motor loop uses encoder feedback, and a supervisory state machine manages INIT, HOMING, TRACKING, HOLD, PARK, STOW, and FAULT modes. A deterministic seven-day scenario was evaluated from 08:00 to 17:00, comprising ten hourly reporting points and nine one-hour integration intervals per day (70 reported observations). The environmental and orientation profiles were prescribed for design evaluation rather than sampled from a documented field site. Under these assumptions, the tracked-panel model produced 283.93 Wh/day versus 237.48 Wh/day for the fixed comparator, a descriptive gain of 19.56%. The model-residual combined RMSE was 0.158°, maximum residual error was 0.44°, estimated actuator consumption was 0.495 Wh/day, and estimated net gain was 45.95 Wh/day. These values describe internal model behavior, not validated hardware performance. No physical prototype, independent output-shaft measurement, stochastic uncertainty analysis, or location-specific economic assessment was conducted. The contribution is therefore a system-level integration and verification framework, together with explicit requirements for subsequent prototype and field validation.