Abstract: Some embodiments include a system operable to construct hierarchical training data sets for use with machine-learning for multiple controlled devices. Other embodiments of related systems and methods are also provided.

Abstract: A controller may control a system for extracting liquid water from air comprising a thermal unit, a primary desiccant wheel, and a regeneration fluid path. The controller may comprise a plurality of sensors, a plurality of motors, and a microcontroller coupled to the plurality of sensors and the plurality of motors. The microcontroller may be configured to determine a water extraction efficiency based on at least one signal received from at least one of the plurality of sensors and maximize the water extraction efficiency by adjusting a speed of at least one of the plurality of motors in response to the determined water extraction efficiency. The water extraction efficiency may be a value obtained by multiplying a regeneration fluid flow rate within the regeneration fluid path by an absolute humidity of air on a side of the primary desiccant wheel opposite a side in communication with the thermal unit.