Computer-Based Management of Mirror-Washing in Utility-Scale Solar Thermal Plants
BrightSource solar power plants consist of fields of tens of thousands of mirrors, spread across kilometers of open areas. These huge mirrors are in constant motion, reflecting the sun’s rays on to the solar thermal collector. Maintaining high reflectivity of the mirrors is essential for the solar field’s performance, a task that becomes complex when expanded to encompass the solar field’s features. The solution for mirror cleaning must be efficient, cost-effective, time-saving, and easy to maintain for dozens of years. BrightSource has designed and constructed a system of GPS-based mirror washing machines (MWMs) that are controlled and managed by end-to-end software. The system generates optimized cleaning tasks, positions the mirrors, and efficiently controls the navigation and state of the MWMs with their 25-meter-long extendable cranes. All of these actions together provide an optimal mirror cleaning solution. This article describes the BrightSource cleaning control technology, for example, in the Ivanpah project, the world’s largest solar thermal facility. The Ivanpah solar field includes 173,500 heliostats divided among three solar fields. Each heliostat holds two mirrors of approximately 2.5 × 3.5 meters, all of which require periodic cleaning. Specifically, this article addresses issues such as the following: • The mirror washing machine (MWM) types: truck and tractor-based, and their differing usage in the solar field • Designation and choice of the cleaning area • Estimation of the stopping points in the designated area, and association of the mirrors to clean from each stopping point • Cleaning time optimization: stopping point density, order in which to clean heliostats, and heliostat position during cleaning • Heliostat positioning: opening clear corridors through which the MWM can travel, and setting heliostats in cleaning orientations • Receiving and responding to callback messages from the MWMs, such as cleaning progress and machine faults • Working in the real world: resources shared with the power plant, and recovery from system faults