应用光线追踪方法建模双面电站系统 范例与验证-薄中南 Jefferson Bor-Fraunhofer ISE
© Fraunhofer ISE ©Fraunhofer ISE / Photo: Guido Kirsch 薄中南 Jefferson Bor 弗朗霍夫太阳 能 研究 所 ISE Special thanks to Dr. C. Reise, J. Kang, Fraunhofer ISE M.Chiodetti, EDF Fraunhofer Institute for Solar Energy Systems ISE PVPMC, Weihai, 05.12.2018 Ray tracing methodology for Bifacial System modeling: Exemplar and Validation 应用光线追踪方法建模双面电站系统: 范例与验证 photo provided by NEOEN © Fraunhofer ISE 2 议程 AGENDA 1. 双面建模 的挑战 2. 光 线 跟踪法 如何运 作? 3. 实际应用 范例 4. 模型验证 5. 结论与展 望 1. Main Challenges for Bifacial Simulation 2. How does Ray Tracing Method Works? 3. Project and Test Exemplars 4. Validation of the Model Performance 5. Conclusion and Perspectives Expected development of monofacial and bifacial modules share of the world market Source: ITRPV report 20171 © Fraunhofer ISE 3 双面的效率与发电 Bifacial Efficiency and Power 一个直观的想法 A Straight Forward Thinking Yield ~ G(front) * eff(front) + G(rear) * eff(rear) = G(front) * eff(front) + G(rear) * eff(front) * BF STC value optical gain bifaciality factor 18% … 22% 5% … 50% 75% … 95% these factors determine bifacial gain module property system property module property laboratory software laboratory © Fraunhofer ISE 4 影响双面增益的因子 Factors affecting bifacial gain 支架几何 the mounting geometry 高度 height 倾角 module tilt 间距 row-to-row distances (GCR) 地面反射率与 不 均匀 度the ground albedo and its homogeneity 支架结构 the mounting structure 组件双面特性 Bifacial characteristics of module 许多因素并不影响单面系统的模拟 Mono-facial system simulation is not affected by most of these factors!! © Fraunhofer ISE 5 双面模拟的挑战 Main challenges for bifacial simulation 背面辐照不均 匀 Non-uniformly distributed radiation on the rear side 周 遭物体 都有不同 的反射 或散射特 性 All surrounding objects have different patterns of reflection and scattering Other adaptation compared to mono-facial model Electrical modelling for the two-side irradiance Modified thermal behaviour Winter Summer Measured irradiance from 3 pyranometers mounted on the rear side of a bifacial system © Fraunhofer ISE 6 双面系统的建模 Modeling of bifacial system 主要的两种模型 Two dominating method for modeling: View factor method 光线跟踪法 Ray tracing method JCT, ECN, CEA INES, ISC KONSTANZ, REC, SERIS, Sandia, NREL, University of Iowa, RTWH,, Enel, Pvsyst, Polysun, University of Nevada, Fraunhofer CSP, ZHAW, Univertsity of Stuttgart, KAUST… [1] [1] Source: J. Libal et al., bifi PV workshop October 26, 2017 Fraunhofer ISE, EDF [1] , SERIES Singapore © Fraunhofer ISE 7 The View Factors Method View Factor : fraction of radiation from s1, s2 (or more) that hits the m surface 优点 Pros 较 容 易应用 Can be easily implemented 运算快 速 Low computing time 缺点 Cons 复 杂 的形状 较难考 虑 Complex geometries are difficult to be considered 支架结构 与外物 遮 蔽等细节 较难整 合 Mounting structure and shading objects are hard to be integrated 使用者 多, 结果不 错 Mainly applied, decent result EdF R&D 2014 © Fraunhofer ISE 8 光线跟踪法 The Ray Tracing Method 反向光线跟踪 计 算 Backward ray tracing calculation © Fraunhofer ISE 9 光线跟踪法 The Ray Tracing Method 光 线 发射后-每束 光开始 经历: Light begins to travel back – each beam will: 1. 碰 到 障碍物 Hit obstacles …根 据 周遭 环境 depending on surrounding 2. 镜 面 反射或 散射 Specular reflect or scatter …根 据 材料 表面depending on material surface 3. 最 终 到达天 空 (半球顶) Hit the sky (hemisphere) 2D示意图 为 了简单 表现光 线跟踪 法概念 图中简 化光束 数量 2D illustration The amount of beams is reduced to better conceptualized the ray tracing method © Fraunhofer ISE 10 光线跟踪法 The Ray Tracing Method 光 线 发射后-每束 光开始 经历: Light begins to travel back – each beam will: 1. 碰 到 障碍物 Hit obstacles …根 据 周遭 环境 depending on surrounding 2. 镜 面 反射或 散射 Specular reflect or scatter …根 据 材料 表面depending on material surface 3. 最 终 到达天 空 (半球顶) Hit the sky (hemisphere) 2D示意图 为 了简单 表现光 线跟踪 法概念 图中简 化光束 数量 2D illustration The amount of beams is reduced to better conceptualized the ray tracing method © Fraunhofer ISE 11 光线跟踪法 The Ray Tracing Method 光 线 发射后-每束 光开始 经历: Light begins to travel back – each beam will: 1. 碰 到 障碍物 Hit obstacles …根 据 周遭 环境 depending on surrounding 2. 镜 面 反射或 散射 Specular reflect or scatter …根 据 材料 表面depending on material surface 3. 最 终 到达天 空 (半球顶) Hit the sky (hemisphere) 2D示意图 为 了简单 表现光 线跟踪 法概念 图中简 化光束 数量 2D illustration The amount of beams is reduced to better conceptualized the ray tracing method © Fraunhofer ISE 12 Model and Sky (9 O‘Clock at 21. September in Freiburg) 按角度解析的辐照值 Angle-resolved Irradiance Value 辐照密度场 Radiation Density Fields © Fraunhofer ISE 13 正面 Field of view, front side 按角度解析的辐照值 Angle-resolved Irradiance Value 辐照密度场 Radiation Density Fields © Fraunhofer ISE 14 正面 Field of view, front side 按角度解析的辐照值 Angle-resolved Irradiance Value 辐照密度场 Radiation Density Fields © Fraunhofer ISE 15 背面 Field of view, rear side 按角度解析的辐照值 Angle-resolved Irradiance Value 辐照密度场 Radiation Density Fields © Fraunhofer ISE 16 光线跟踪法 The Ray Tracing Method 想象成: 背 面 每一 观察点 有一个 假想的 传感器 Just image: Virtual Sensor on each defined point 求全部点 平均 Average of all points 每个点会射出1024条光线 每 一 次 的 反 射后， 每一条 会再变 成至多1024 条… 举例: 3次反射后我们 总共 会 有10 1 2 条光束… 1024 Beams sent out from each point After each reflection, one beam will become 1024 beams… Ex: after 3 reflections, we will get max. 10 1 2 beams… © Fraunhofer ISE 17 光线跟踪法 The Ray Tracing Method 优点 Pros 贴 近 真实的 背面辐 射 Realistic rear side irradiance 3D 的 不 均匀 分布 Considering inhomogeneity in 3D 可 以 重建细 节( 结构, 空间 分布, 物体) Possible to build up all configurations (structure, spacing, objects) 可以整 合CAD软件 Possible to feed-in with CAD software 缺点 Cons 较 不 易整合 进现有 模型 Less easy to implement 计 算 较为耗 时 Computing time consuming © Fraunhofer ISE 18 光 线 跟踪法应 用实例 Appliance Exemplar with ISE Ray Tracing Model 85MW 约旦 85MW, Jordan 85MW, Jordan 2302 kWh/m² (avg. 2008-2017) DNI: 1738 kWh/m² DHI: 564 kWh/m² Bright sand – Abeldo: ~30% Bifacial module, 76% bifaciality factor Horizontal single axis tracker Row distance: 13m (GCR 31%) Height: 2.8 meter Dark rocks and soil will be removed to expose the white sand Source: : Soltec, https://soltec.com/soltec-unveils-the-sf7-tracker-that-achieves-higher-yield-and-lower-cost/ © Fraunhofer ISE 19 光 线 跟踪法应 用实例 Appliance Exemplar with ISE Ray Tracing Model 85MW 约旦 85MW, Jordan © Fraunhofer ISE 20 6 am 12 pm 15 pm 光 线 跟踪法应 用实例 Appliance Exemplar with ISE Ray Tracing Model 日变化, 85MW 约旦 Daily Change 85MW, Jordan © Fraunhofer ISE 21 6 am 12 pm 15 pm 不同阴影密度: 不同反射辐照比例 Different density of shadow means: different amount of reflected irradiance 光 线 跟踪法应 用实例 Appliance Exemplar with ISE Ray Tracing Model 日变化, 85MW 约旦 Daily Change 85MW, Jordan © Fraunhofer ISE 22 越靠近边缘越亮: 散射光线较多 Closed to the edge is brighter means: More diffuse light incoming 光 线 跟踪法应 用实例 Appliance Exemplar with ISE Ray Tracing Model 不均匀度85MW 约旦 Inhomogeneity 85MW, Jordan © Fraunhofer ISE 23 Specific Yield: 2,703 kWh/kWp 光线增益 BGopt = Grear / Gfront = 10.9% 组件增益 BGmod = ( Grear * bf ) / Gfront = 8.3% 系统增益 BGsys = Erear / Efront = ( Ebifi – Emono ) / Emono = 7.1% 光 线 跟踪法应 用实例 Appliance Exemplar with ISE Ray Tracing Model 结果85MW 约旦 Result 85MW, Jordan © Fraunhofer ISE 24 Specific Yield: 2,703 kWh/kWp 光线增益 BGopt = Grear / Gfront = 10.9% 组件增益 BGmod = ( Grear * bf ) / Gfront = 8.3% 系统增益 BGsys = Erear / Efront = ( Ebifi – Emono ) / Emono = 7.1% 组 件 特性 Module characteristic 系 统几何 设计 System geometry 电 气设计 Electrical design 光 线 跟踪法应 用实例 Appliance Exemplar with ISE Ray Tracing Model 结果85MW 约旦 Result 85MW, Jordan © Fraunhofer ISE 27 模型验证 Model Validation 与法国电力共同合作项目 Joint project with EDF Results revealed on Sep. 26 th 2018 at EU PVSEC, Belgium 法电与ISE 的不同光伏系统 PV Installations from EDF and ISE Source: M. Chiodetti et al., EU PVSEC, Belgium, Sep. 26th 2018 © Fraunhofer ISE 28 不同水平验证 On different level: 辐照 Irradiance 发电参数 Electrical value 监测数据包含 Monitoring at 水平与反射辐 照 GHI front and Albedo (horizontal) 环境与组件温 度 Ambient and module temperature 直流测电流电 压 功率Current, voltage and power on DC side 数据筛选 Data Filtering Source: M. Chiodetti et al., EU PVSEC, Belgium, Sep. 26th 2018 模型验证 Model Validation © Fraunhofer ISE 29 模型验证,法国:7个月 Model Validation, France: 7 Months Irradiance measurement vs simulation Winter Summer Source: M. Chiodetti et al., EU PVSEC, Belgium, Sep. 26th 2018 © Fraunhofer ISE 30 模型验证,法国:7个月 Model Validation, France: 7 Months Power measurement vs simulation n-PERT modules fixed-tilt 30° with 50% GCR Albedo: 30% Source: M. Chiodetti et al., EU PVSEC, Belgium, Sep. 26th 2018 © Fraunhofer ISE 31 模型验证,加州:6个月 Model Validation, California: 6 Months Power measurement vs simulation p-PERC modules Single axis tracker, with 35% GCR Albedo: 32% Source: M. Chiodetti et al., EU PVSEC, Belgium, Sep. 26th 2018 © Fraunhofer ISE 32 结论 Summary 双面系统 会有更 高 的发电量 发电模型 可以有 贴 近单面模 型的准 确 率 均 方 根值低 于7% 平均 偏差 ±4% 在解读双 面增益 时 很重要的 一定要 了 解数据背 后的边 界 条件 Bifacial PV systems may achieve high performance Simulation may reach similar accuracy as with monofacial systems Root Mean Square Errors below 7% Mean Bias Errors ±4% When interpreting the Bifacial gain, always pay attention to the boundary condition © Fraunhofer ISE 33 结论 Summary 更 进 一步研 究 Further Investigation 对比view factor 模型 Comparison with models based on view factor theory 验 证 更多电 站数据 Validation with more real system performance data 模型加 强 Model improvement 反 照 率的变 动 Albedo variability 双 面 的电气 模型 Bifacial electrical model 热建模 Thermal model 长 期 双面组 件衰减 Long-term bifacial module degradation © Fraunhofer ISE 34 谢谢您的 参 与! Thank you for your attention Fraunhofer Institute for Solar Energy Systems ISE Jefferson Bor www.ise.fraunhofer.de jefferson.bor@ise.fraunhofer.de
solarbe文库
