Local optimization of thermal storage for greenhouses: reduction of energy input and improvement of inner climate

In temperate regions, such as the Mediterranean basin, there is a diurnal excess of energy nearly every day of the year, which is usually dissipated through natural ventilation. However, since suboptimal nighttime temperatures limit productivity of unheated greenhouses for several months a year, extracting the daytime excess energy and reusing it to heat the greenhouse during the night would increase productivity, or at least reduce energy consumption of greenhouses that are heated. This heat extraction would have the additional advantage of reducing ventilation requirement thereby increasing the scope for carbon dioxide fertilization. To achieve this, the performance of the greenhouse as a solar collector has to be maximized by an efficient heat exchanging and heat buffering system. The aim of this research was to define the optimum combination of heat exchange rate, maximum water flow rate of a heat storage buffer and buffer capacity in a commercial green¬house in the Mediterranean region (Sicily, Italy, 37 °N), the cost function being represented by the dose (duration × intensity) of low temperature events. The green¬house temperature was calculated through a previously validated greenhouse climate simulation model, applied to one-year of real local data. The effect of the buffer on the cost function was first calculated for a range of heat exchange values followed by a cost function evaluation for nodes of a pre-selected grid, each node representing a value-pair for the other two buffer defining parameters. In this paper we analyze the trend of the cost function with respect to each parameter of the buffer and how this is affected by the preset tolerance of low temperatures. Furthermore, we discuss a simple method to find an “optimal” configuration of the buffer. Finally, a combination of 3000 m3 ha-1 buffer capacity, 45 m3 h-1 ha-1 maximum water flow rate and an overall heat transfer coefficient of 5 W m-2 K-1 is selected (heat transfer coefficient is defined per m2 greenhouse floor area).