Available at: https://digitalcommons.calpoly.edu/theses/3365
Date of Award
6-2026
Degree Name
MS in Mechanical Engineering
Department/Program
Mechanical Engineering
College
College of Engineering
Advisor
Jacques Belanger
Advisor Department
Mechanical Engineering
Advisor College
College of Engineering
Abstract
Photovoltaic solar farms on uneven terrain present consistent challenges to performance optimization. Standard backtracking algorithms can generate potential losses due to interrow shading caused by sloped terrain as well as a lack of plane of array irradiance that can occur even on level ground. The purpose of this thesis is to use the Gold Tree site as a case study to develop and evaluate different solar tracking control strategies. Specifically, it compares the relative performance of fixed, direct tracking, closed-form backtracking, and global optimization strategies. It is done under a two-dimensional, idealized predictive power model of a representative portion of the site, evaluated across both standard and half-cut cell shading models and under both mechanically limited and unconstrained rotational bounds. Additionally, these strategies cover both the case of a single motor actuating the array as well as the case of individual motors for each row. For single-motor systems, an exhaustive global optimization routine is applied, while particle swarm optimization is used for the higher-dimensional multi-motor configuration. Results show that direct tracking improves daily performance by approximately 32% over a fixed array on average, while single-motor global optimization offers a further 4% improvement over direct tracking, though, solutions found by global optimization require tracker rotational rates that are physically unrealistic, limiting their practical utility without the application of a smoothing scheme. N-motor control schemes outperformed their single-motor counterparts by up to 10% with closed-form backtracking consistently outperforming particle swarm optimization across all configurations. Expanding tracker rotational bounds from ±52° to ±90° offered modest additional gains while the effect of panel configuration on performance varied by control scheme. For the Gold Tree site specifically, refining the existing backtracking algorithm to account for skewed topography is recommended as a low-cost improvement, while transitioning to an N-motor system represents the highest-potential upgrade path.