Développement d’un modèle calibré pour la simulation énergétique de serres et analyse des résultats à l’aide d’indicateurs de performance

Abstract

In Quebec, agricultural greenhouse productivity is limited by several factors such as daylight availability and cold temperatures. Competitivity of the local production of vegetables compared to importations from warmer countries is a barrier to food autonomy in this context. Energy consumption, required to maintain optimal growth conditions of greenhouse crops through heating systems or artificial lighting have a great impact on Quebec’s greenhouse sector competitivity. Development of models using building energy performance simulation tools allows to study the thermal profile inside greenhouses and evaluate opportunities to enhance their energy efficiency. The objective of this thesis is to present an approach for developing and calibrating a greenhouse model in Quebec and to demonstrate the importance of using adequate performance indicators to analyse the simulation results according to the context. To fulfil this objective, an existing gothic arch greenhouse was modelled in the building energy performance simulation tool TRNSYS. Simulation results of this model were calibrated by comparing them to data measured in the actual greenhouse for indoor air temperature and humidity as well as available photoactive radiation at the canopy level. The root mean square errors obtained were 1,83° C (R2 = 0,95), 7,78% (R2 = 0,76) and 60,63 μmol.s-1.m-2 (R2 = 0,89), respectively, which shows a good fit between the simulation results and the measured data in the context of unheated greenhouse model simulation. This calibration was confirmed by comparing these results to baseline values reported in the literature. This modelling approach was then used to compare the growth conditions maintained in an unheated gothic arch greenhouse and a comparable partially buried Chinese greenhouse located in Lemieux (46,3°N 72,1°O). The growth conditions were evaluated using adequate performance indicators for an unheated greenhouse context. The thermal growing season was 22 days longer and the growing degree-days for lettuce cultivation were higher inside the Chinese greenhouse than in the gothic arch greenhouse. Daylight availability at the canopy was similar in both modelled greenhouses. However, the available ventilation in both the Chinese greenhouse and gothic arch greenhouse was insufficient to prevent overheating of the greenhouses. This was illustrated by the overheating index values of 244 h in the Chinese greenhouse and 142 h in the gothic arch greenhouse, respectively. The higher occurrence of overheating in the Chinese greenhouse resulted in a lower photothermal index for lettuce which indicates potentially reduced yield for lettuce cultivation despites the generally higher indoor air temperatures observed in this greenhouse. A model was then developed to evaluate the energy consumption and greenhouse gas (GHG) emissions from two heating systems used to regulate air temperature inside a gothic arch greenhouse located in Baie-Comeau (49,2°N 68,2°O). The investigated heating systems were a conventional propane gas unit heater and a datacenter waste heat recovery system. The use of the waste heat recovery system resulted in a 91% reduction in GHG emissions compared to the propane unit heater in Quebec. The waste heat recovery system also consumed 66% less energy (from both electricity and propane sources) for heating of the greenhouse. The maximum electrical power demand of the waste heat recovery system was 36,8 kW. Artificial lighting of the greenhouse using high pressure sodium lamps was required to maintain minimum daylight levels for lettuce cultivation throughout the season. This would consume an additional 23 000 kWh of electricity annually which represented 21,7% of the total energy consumption of the greenhouse equipped with the waste heat recovery system in Baie-Comeau. The artificial lighting of the greenhouse also led to an additional electrical power demand of 30,9 kW.

Date	26 Jan 2022
Original language	French
Awarding Institution	École de technologie supérieure
Supervisor	Danielle Monfet (Supervisor) & Didier Haillot (Co-supervisor)