Chapter 2 investigates the discipline of process-based greenhouse modelling. The chapter shows that a considerable number of greenhouse models are being published, and sets out to understand the reasons for the existence of this multitude of models. In Section 2.2, substantial background on the concept of process-based greenhouse modelling is provided, and a general structure of these models is proposed. Subsequently, modelling studies published between 2018 and 2020 are analyzed according to this general structure. The studies are categorized according to their objectives, types of greenhouse they describe, and equipment they consider, and a model inheritance chart is presented, showing how current models are based on earlier works. Moreover, a comparison of modelling validation studies is performed. Based on this analysis, possible reasons for the abundance of greenhouse models are suggested, including a lack of model transparency and code availability, and a belief that model development is in itself a valuable research goal. The chapter ends with several recommendations for the future advancement of the discipline. These include promoting model transparency and availability of source code, and establishing shared datasets and evaluation benchmarks.Chapter 3 presents GreenLight, a process-based model for a greenhouse with a tomato crop, which describes the influence of HPS lamps and LEDs on the climate, crop, and energy use. In this chapter, GreenLight's performance is evaluated by comparing model predictions against data from two greenhouse compartments, one with HPS lamps and one with LEDs. The model is found to have a relative error in predictions of climate and energy use in the range of 1-12%. In order to promote transparency, the model is offered in an open source format at https://github.com/davkat1/GreenLight. In this way, the model is made available for inspection and extension by others.Chapter 4 uses GreenLight to predict the influence of replacing HPS lamps by LEDs in a greenhouse. A wide range of scenarios is considered, including varying climates, from subtropical China to arctic Sweden, and multiple settings for indoor temperature, lamp intensity, lighting duration, and insulation. In all scenarios, LEDs are found to reduce the energy demand for lighting by 40%, but to increase the demand for heating. This results in the total energy saving by transition to LEDs to be in the range of 10-25% for the majority of scenarios considered. An important factor influencing how much energy can be saved by a transition to LEDs is found to be the ratio between the lighting and heating demand of the greenhouse before the transition.
Summary xiiiChapter 5 presents a novel concept for greenhouse climate control: heating a greenhouse by light. Considering the observation that all lamps provide heat as well as light, this chapter suggests that illuminating at high light intensities could eliminate the need to heat the greenhouse by the heating system. This approach could potentially be very efficient, as it utili...