Synergism of high and low level systems for the efficient management of greenhouses

Abstract

The advantages of using artificial intelligence (AI) decision support tools in synergism with low level process controllers or schedulers are investigated in this paper. The development of a modern control and management system for greenhouses used recent advances in software design, and development tools, to provide an open system for rapid program development. To effectively integrate expert system applications in a control and management system, an environment was built that supports all required interfaces between AI applications and the greenhouse management system (GMS). This environment incorporates a native fuzzy knowledge based system (KBS) and a number of procedural control functions, in the GMS, that can effectively interact. The programmable logic controller (PLC) houses all well-known control function blocks, in library form, callable to implement various control loop designs. Functions that have not been foreseen in the PLC control library can be instantly implemented using the open KBS system. The innovative addition of integral initial conditions on a proportional-integral-derivative (PID) controller, for repetitive load switching applications, is an example, demonstrated in this paper. The usefulness of other control blocks such as a self-adjusting Smith predictor is also tested for a real application of a mixing process with long dead time. Synergism of fuzzy decisions and fuzzy controllers, at the supervisory level, with low level process regulators provide adaptive systems, which can optimize both long-term objectives and the short time dynamic responses.