This invention is embodied in a cost-optimization device for the layout and construction of a utility-scale photovoltaic (“PV”) power plant. The optimization device employs a set of algorithms designed to find the most cost-effective solution under given conditions. The algorithms are written in computer machine readable code and are highly customizable for the specific tracker equipment requirements and owner/builder/maintainer specifications or preferences.
The preferred optimization device comprises three principal stages (or “units”) of computing: (1) an objective-state unit, (2) an optimum-feasible unit, and (3) a grading unit. In the first stage, the objective-state unit cost-optimizes site grading by orienting a ruling line between a maximum and a minimum pile reveal length for each tracker in the project (the objective-state solution”). When compared to the existing site topography, the ruling line indicates cost-optimized cut and fill locations.
In the second stage, the optimum-feasible unit modifies the objective-state solution to satisfy given geometric constraints of the project. More specifically, the optimum-feasible unit employs a project-wide mesh (“system mesh”) to check whether the objective-state solution complies with the project's geometric restrictions. If not, the optimum-feasible unit applies a marching algorithm to incrementally adjust each ruling line until it finds a solution that minimizes site grading while complying with the project's geometric restrictions (the “optimum-feasible solution”).
In the third stage, the grading unit modifies the optimum-feasible solution to satisfy non-geometric constraints of the project (the “final-state solution”).