ROBUST MULTI-OBJECTIVE EVOLUTIONARY OPTIMIZATION TO ALLOW GREENHOUSE PRODUCTION/ENERGY USE TRADEOFFS

The worldwide increase in demand for fresh fruits and vegetables has led to a search for strategies to manage greenhouses in ways that not only meet this demand, but that are also economically viable and environmentally sustainable. A promising technique for managing the greenhouse microclimate is automatic control of mechan-ical systems such as heaters, ventilators, and shade screens. However, conventional open-loop greenhouse controllers are criticized for their excessive use of energy and vagueness of user-adjustable settings. Moreover, due to their vulnerability to external disturbances, unreliability of the final “optimum” solution, and the lack of guaranteed alternatives, closed-loop optimal controllers are only reluctantly accepted in cultivation practice. Our approach is to use a form of evolutionary multi-objective optimization to discover a suite of user-selectable control strategies that optimally balance crop productivity with the financial and environmental costs of greenhouse climate control. Each of the controllers discovered by this approach defines a range of conditions to be maintained depending upon the crop state, external environmental conditions, and resource costs. In this paper, we present the performance of our evolved greenhouse control strategies on a simulated tomato crop, compare them to existing greenhouse control strategies, and discuss the feasibility of implementing these new control strategies in actual greenhouses. In the near future, these control strategies will be validated in an advanced experimental greenhouse currently under construction on the campus of Tongji University in Shanghai, China.