A low-cost partial Hardware-in-the-Loop virtual platform for dual-axis photovoltaic solar tracking control

Abstract

Abstract This paper presents a low-cost virtual platform for evaluating dual-axis photovoltaic solar trackers, assisted by control algorithms, using a partial Hardware-in-the-Loop (signal-in-the-loop) approach. A computer-aided design-derived mechanical model is built in SolidWorks® and executed in Matlab-Simulink®/Simscape Multibody, while an Arduino® interface provides real-time interaction and controlled disturbance injection. The mechanical tracker is simulated in Simulink/Simscape, while the only physical hardware is the Arduino® interface, which is used for real-time disturbance induction and signal acquisition. Four tests were performed using a 15 h astronomical trajectory (54 000 s): undisturbed tracking, disturbed tracking without control, and disturbed tracking with PI control (including a zoomed window). Disturbance magnitudes are reported as the equivalent angular deviations observed between the setpoint and the measured tracker angles. Without control, perturbations produced sustained deviation and an unusable trajectory. With PI control, the platform quantified disturbance rejection using standard tracking metrics. Maximum transient errors were |e_max| ≈ 27° (elevation) and |e_max|≈ 33° (azimuth); root mean square errors were 15.46° and 17.51°, respectively, with residual errors on the order of 10^− 3 degrees and settling times (±1° band) of 4.98 s and 5.06 s. The results show that the proposed PHIL virtual environment supports repeatable, disturbance-oriented controller tuning and early-stage design decisions aligned with renewable energy efficiency improvements. Future work will focus on automating disturbance profiles, mapping tracking errors to harvested energy, and extending the platform toward a Digital Twin with synchronized data and real-time parameter updates.