太陽能番茄溫室設計策略之研究

Abstract

近年農業設施導入太陽能模組以提升土地利用與能源產出已成趨勢，但遮蔽配置若不當，可能造成弱光、偏亮與空間不均，進而影響番茄生長與管理。本研究以臺灣番茄溫室為對象，建立以日光積分(DLI)為核心之光環境檢核架構，探討不同地點、遮蔽率、模組類型、排布形式與屋頂結構下之全年光照分布與發電表現，並提出作物優先之配置邏輯。 研究以Rhino與Grasshopper進行幾何建模與參數化控制，搭配Honeybee匯入EPW氣象資料，執行全年8760小時日照輻射模擬；結果由每小時總輻射量換算為光合光子通量密度(PPFD)，並以每24小時累計得到DLI。研究比較三地(臺北、臺中、恆春)，屋頂結構包含雙斜、重複雙斜、單斜與雙重單斜；模組類型包含單晶矽與有機太陽能(OPV)；排布形式包含棋盤式與陣列式；遮蔽率則依模組組別設定為單晶矽20%、25%、30%，OPV 30%、50%。每組情境於栽培平面佈設400測點，以年均DLI、月均DLI、適光月份數、DLI區間比例、達成率(年均DLI落在20-30 mol/m²/day之測點比例)與變異係數(CV)評估光環境，並統計全年發電量作為參考。 結果顯示，未遮蔽高透玻璃對照組年均DLI由北至南提升(臺北 24.7、臺中 28.1、恆春 30.8)，且偏亮比例普遍偏高，恆春DLI >30 達 53%、達成率僅 23%，顯示強日照地區在無遮蔽條件下即具高光風險。導入模組後，遮蔽率為主控因子，直接影響年均 DLI、偏亮比例與達成率；屋頂形式主導陰影分布與空間穩定性，是達成率與 CV 分化的重要來源；排布方式反映控亮與覆蓋穩定之取捨，棋盤式多有利於維持較高達成率與較穩定均勻性，陣列式在部分條件下可降低偏亮，但遮蔽加深時較易引發覆蓋不足。 以作物優先而言，臺北受冬季弱光限制，高遮蔽易導致適光覆蓋不足，宜採用保守遮蔽；臺中光量較充足，策略重點為降偏亮同時維持覆蓋與均勻性；恆春日照最強，遮蔽可作為降低偏亮與熱負荷之必要手段，但需同時控管CV以確保空間差異可管理。整體而言，太陽能模組除發電外亦具光環境調控功能，但弱光地點須以達成率與均勻性為核心避免過度遮蔽。本研究所建流程可作為溫室太陽能整合設計之決策參考。

This study aims to investigate the effects of different roof configurations and solar module shading ratios on the light environment inside tomato greenhouses, with the goal of identifying optimal agrivoltaic strategies that fulfill crop lighting requirements while enhancing solar power generation efficiency. Simulations were conducted using hourly solar radiation data (8,760 hours annually) for three representative locations across Taiwan—Taipei, Taichung, and Hengchun. Geometrical models were constructed for four roof types: double-slope, repeated double-slope, single-slope, and repeated single-slope, to assess their daylighting performance and regional solar potential. The modeling process was carried out using Rhino and Grasshopper as the geometric modeling and parametric control platform, integrated with Honeybee for weather data processing and annual solar radiation simulation. Each simulation was based on EPW weather files corresponding to each location, calculating the hourly global radiation, which was then converted to photosynthetic photon flux density (PPFD), and accumulated every 24 hours to obtain daily light integral (DLI). The simulation covered combinations of location, shading ratio, solar module type (monocrystalline silicon and OPV), arrangement pattern (checkerboard and linear array), and roof structure. A total of 400 sensor points were used to evaluate the spatial distribution of light. For shading configurations, three shading ratios (20%, 25%, and 30%) were applied to monocrystalline silicon modules, and two ratios (30% and 50%) to OPV modules. Simulations were used to assess PPFD and DLI. A multi-layered screening approach was adopted: first, using annual average DLI, coefficient of variation (CV), and spatial ratio of effective DLI (20-30 mol/m²/day) as evaluation criteria; second, evaluating the number of tomato cropping cycles per year based on monthly DLI distributions; third, analyzing hotspot maps to ensure spatial uniformity; and finally, selecting the most suitable greenhouse design configurations for each location. Results indicate that OPV modules, due to their light transmittance, provided higher annual DLI and more uniform light distribution under the same conditions, making them more suitable for designs with uneven lighting or higher shading ratios. Monocrystalline silicon modules demonstrated better energy generation in certain configurations. Among roof structures, the repeated double-slope design was the most consistent in balancing north-south solar asymmetry, enhancing uniformity and plantable space. The checkerboard arrangement effectively mitigated striped shading patterns caused by module arrays and improved light distribution balance. Regional design recommendations include limiting shading ratios to below 25% in Taipei to ensure crop lighting needs, utilizing flexible configurations in Taichung due to moderate climate, and adopting higher shading ratios in sun-rich Hengchun to optimize solar energy harvesting. This study establishes a parametric simulation and evaluation workflow encompassing climate data processing, shading modeling, PPFD - DLI conversion, spatial light analysis, energy estimation, and visual representation. This framework provides a scientific and comparative decision-making tool for greenhouse design, promoting the integration of smart agriculture and green energy development in Taiwan.