【PVPMC】-阿特斯阳光-赵长瑞-双面HSAT系统的光线跟踪精确阴影因子和失配损失分析

CSIQ NASDAQ Listed SHADING FACTOR MISMATCH LOSS EVALUATION FOR BIFACIAL MODULE WITH VARIOUS RACKING TYPE Changrui ZHAO, Jianwei XIAO, Baohua HE Yuanjie YU, Jean-Nicolas JAUBERT System Technology center 13th PVPMC Workshop, Dec 9th 2019 2Canadian Solar Inc. Modeling methodology introduction02 Shading factor and mismatch loss evaluation03 04 Conclusion Bifacial technology overview01 3Canadian Solar Inc. 2017/5 First commercial talks on bifacial module BiKu 2018/4 Trial production First customer delivery 2018/8 2018/12 2019/1 Canadian Solar driving the bifacial tide 170MWp own bifacial projects completed (China) 300MWp bifacial module milestone Delivering first 250MWp overseas order 2017/12 20kW demo site installed in Changshu 2018/5 2019/11 Bifacial module shipping 900MWp milestone Canadian Solar Energy group: all new projects designed for bifacial 4Canadian Solar Inc. Merging of 4 unique technologies Canadian Solar bifacial modules – BiKu/BiHiKu advantages Worldwide 1st manufacturer to offer double glass bifacial poly module Bifacial cell Double glass Ku dual cell P4 PERC poly Front Rear 144 cells, 992x2022 mm 360 ~ 370 W 157’ cells 144 cells, 1048x2132 mm 400 ~ 410 W 166’ cells BiKu BiHiKu 5Canadian Solar Inc. Canadian Solar bifacial testing sites – Bifacial gains Changshu 24kW BiKu, SAT 1P Albedo ~0.25 Diffuse ratio ~58% 8.9% 12 months SANDIA 11kW BiKu, SAT 1P Albedo ~0.2 5.2% 3 months Wuhai 150kW BiKu, SAT 1P Albedo ~0.28 Diffuse ratio ~54% 6.2% 10 months 4.0% 11 months 7.7% 4 months Mojave Desert 22kW BiKu, SAT 1P Albedo ~0.3 Diffuse ratio ~30% U.S. China BITEC 18kW BiKu, SAT 2P Albedo ~0.2 / 0.6 6.4 - 15.6% 7 months Wuhai 150kW BiKu, Fixed Albedo ~0.27 Diffuse ratio ~54% Changshu 12kW BiKu, SAT 1P Albedo ~0.5516.9%6 months Suzhou 80kW BiKu/BiHiKu, Fixed Albedo ~0.6-0.7/0.25 Diffuse ratio ~58% Darlington ~2MW BiHiKu, SAT 1P Albedo ~0.3 Planning construction Last update: 2019/11/24 Completed On-going Construction CFV 14kW BiHiKu, SAT 1P Albedo ~0.28 5.3% 3 months Boulder 80kW BiKu, SAT 1P Albedo ~0.3 7.1% 7 months 9-12% BiKu-3 month Australia Wuhai 2.8MW BiKu, Fixed Albedo ~0.27 Diffuse ratio ~54% 5.3% BiHiKu-3 month 6.9% 10 months 6Canadian Solar Inc. Modeling methodology introduction02 Shading factor and mismatch loss evaluation03 04 Conclusion Bifacial technology overview01 7Canadian Solar Inc. Modeling methodology introduction – Overview PVmismatch Tool Backwards Ray tracing Mismatch loss @ time t Array IV curve (t) Irradiance profile (t) Methodology pro: • Gives irradiance inhomogeneity in 3D, racking system detailed features shading effect are taken into account. Annual Energy loss With Torque tube Annual Shading factor PVsyst Shading factor PVsyst Mismatch loss factor Electrical Mismatch Loss PVmismatch tool: Open source codes available on GitHub, Sunpower 8Canadian Solar Inc. Modeling methodology introduction – Calculation flow N is the number of hourly data points of the whole year, excluding rear irradiance ≤15W/m2 when calculating mismatch loss. To reduce computation time, representative days and cell rows can be properly selected through the year instead. Different system scenarios + 3D model Location - EPW weather file Tracking angle data Yearly mismatch loss calculation Simulated hourly front and rear irradiances with tube/pier/rail shading Synthetic Irradiance = front irradiance + rear irradiance × bifaciality factor System IV curve for each time stamp → Maximum Power Point (MPP) Cell front I-V STC production data → Random sampling Simulated hourly front and rear irradiances without tube/pier/rail shading Yearly shading factor calculation 𝐘𝐞𝐚𝐫𝐥𝐲𝐬𝐡𝐚𝐝𝐢𝐧𝐠𝐟𝐚𝐜𝐭𝐨𝐫 = 𝟏−(෍ 𝟏 𝑵 𝑮𝒓𝒆𝒂𝒓𝒘𝒊𝒕𝒉𝒔𝒉𝒂𝒅𝒊𝒏𝒈)/(෍ 𝟏 𝑵 𝑮𝒓𝒆𝒂𝒓𝒏𝒐𝒔𝒉𝒂𝒅𝒊𝒏𝒈 DIVA-for-Rhino 𝐘𝐞𝐚𝐫𝐥𝐲𝐦𝐢𝐬𝐦𝐚𝐭𝐜𝐡𝐥𝐨𝐬𝐬 = (෍ 𝟏 𝑵 𝑨𝒓𝒓𝒂𝒚𝒑𝒐𝒘𝒆𝒓)/(෍ 𝟏 𝑵 𝑪𝒆𝒍𝒍𝒔𝑷𝒐𝒘𝒆𝒓 −𝟏 9Canadian Solar Inc. Representative row/module/cells are selected for modeling to reduce modeling time. Modeling methodology introduction – Modeling target surface selection 10Canadian Solar Inc. Modeling methodology introduction – Data process method selection Method B is a time-saving way to provide accurate results 0.0% 2.0% 4.0% 6.0% 8.0% 10.0% 12.0% 14.0% 0 200 400 600 800 1000 1200 Shad ing Fac tor GHI (W/m2) Hourly shading factor comparison Method A Method B Location System Albedo Item Method A(365 days) Method B(36 days) Golmud, CN HSAT, One Portrait 0.6 Shading Factor 7.0% 7.1% Mismatch Loss 0.60% 0.61% -60 -50 -40 -30 -20 -10 0 0 200 400 600 800 1000 1200 Misma tch Powe r Los s ( W) GHI (w/m2) Hourly mismatch loss comparison Method A Methed B 11Canadian Solar Inc. Modeling methodology introduction02 Shading factor and mismatch loss evaluation03 04 Conclusion Bifacial technology overview01 12Canadian Solar Inc. Shading factor and mismatch loss calculation- Various system racking 01 02 0403 System Racking TypeFixed rackingFixed racking-2 Portrait (2P) Fixed racking-3 Landscape (3L) HSAT-2 Portrait (2P) HSAT-1 Portrait (1P) 13Canadian Solar Inc. HSAT 1P: Tube-Module distance impact on shading factor and mismatch loss Shading factor varies from 5.6% to 7.8%, mismatch loss is from 0.24% to 0.94% during albedo range 0.2~0.6. Mismatch loss is more sensitive to tube-module distance than shading factor with Albedo increase Module Height 1.5m Distance 50mm Ground Torque Tube Post 5.6% 7.4% 7.8% 0.0% 2.0% 4.0% 6.0% 8.0% 10.0% Sh ad ing Facto r (Y ear ly) Albedo Shading Factor Under Different Albedo 0.2 0.4 0.6 0.24% 0.50% 0.94% 0.0% 0.2% 0.4% 0.6% 0.8% 1.0% M ism atch Lo ss (Ye ar ly) Albedo Mismatch Loss Under Different Albedo 0.2 0.4 0.6 14Canadian Solar Inc. HSAT 1P: Array height impact on shading factor and mismatch loss The height 2m has similar trend and result with height 1.5m, can increase module backside irradiation, but has no obvious contribution to improvement of module backside irradiance uniformity. 5.6% 7.4% 7.8% 6.5% 7.8% 8.1% 0.0% 2.0% 4.0% 6.0% 8.0% 10.0% Sh ad ing Facto r (Y ear ly) Albedo Shading Factor Under Different Albedo H 1.5m H 2.0m 0.2 0.4 0.6 0.24% 0.50% 0.94% 0.23% 0.47% 0.93% 0.0% 0.2% 0.4% 0.6% 0.8% 1.0% M ism atch Lo ss (Ye ar ly) Albedo Mismatch Loss Under Different Albedo H 1.5m H 2.0m 0.2 0.4 0.6 Distance 50mm Module Height 2m Ground Torque Tube Post Distance 50mm Height 1.5m Module 15Canadian Solar Inc. HSAT 2P: Tube-Module distance and purlin length impact Traditional U shape connection Recommended I shape connection Connection topology Row 1 Row 2 Row 1 Row 2 Distance 100mm System configuration Height 2.35m Torque Tube Module 1 Module 2 Ground Post Distance 50mm Purlin length 2.5mPurlin length 3.5m Gap 150mm 16Canadian Solar Inc. Note: Purlin length 3.5m Distance 100mm without purlin Shading factor is not sensitive to tube-module distance under Albedo≥0.4 condition. Mismatch loss is almost same under different tube-module distance and Albedo. Distance 50mm with purlin Distance 100mm with purlin HSAT 2P: Tube-Module distance impact on shading factor and mismatch loss Rear side irradiance distribution T u b e T u b e T u b e 7.1% 6.9% 7.3% 5.7% 6.9% 7.6% 0.0% 2.0% 4.0% 6.0% 8.0% 10.0% Sh ad ing Facto r ( Y ear ly ) Albedo Shading Factor Under Different Albedo Distance Distance 50mm Distance 100mm 0.2 0.4 0.6 0.18% 0.19% 0.20% 0.22% 0.22% 0.23% 0.00% 0.10% 0.20% 0.30% M ism atch Lo ss (Y ear ) Albedo Mismatch Loss Under Different Albedo Distance Distance 50mm Distance 100mm 0.2 0.4 0.6 17Canadian Solar Inc. Shading factor will increase about 0.8% when purlin length change from 2.5m to 3.5m under different Albedo. Mismatch Loss is not sensitive to purlin length under different albedo condition. No Purlin Purlin length 2.5m Purlin length 3.5m HSAT 2P: Purlin length impact on shading factor and mismatch loss Note: Purlin height 100mm Rear side irradiance distribution T u b e T u b e T u b e 4.9% 6.1% 6.8% 5.7% 6.9% 7.6% 0.0% 2.0% 4.0% 6.0% 8.0% 10.0% Sh ad ing Facto r (Y ear ly) Albedo Shading Factor Under Different Albedo Purlin length Purlin length 2.5m Purlin length 3.5m 0.2 0.4 0.6 0.22% 0.21% 0.23% 0.22% 0.22% 0.23% 0.20% 0.21% 0.22% 0.23% 0.24% 0.25% M ism atch Lo ss (Y ear ly) Albedo Mismatch Loss Under Different Albedo Purlin length Purlin length 2.5m Purlin length 3.5m 0.2 0.4 0.6 18Canadian Solar Inc. HSAT 1P Vs. 2P: Shading factor and mismatch loss result Shading factor: 2 portrait has smaller value than 1 portrait under Albedo range 0.2~0.6. Mismatch loss: 2 portrait has more and more advantage than 1 portrait with increased Albedo, 1P Vs. 2P Height 2.35m Gap 100mm Module 1 Torque Tube Module 2 Ground Height 1.5m Gap 50mm Ground Torque Tube Module 5.6% 7.4% 7.8% 4.9% 6.1% 6.8% 0.0% 2.0% 4.0% 6.0% 8.0% 10.0% Shadi ng Fac to r (Yea rly) Albedo Shading Factor Under Different Albedo 1 Portrait 2 Portrait 0.60.2 0.4 0.24% 0.50% 0.94% 0.22% 0.21% 0.23% 0.0% 0.2% 0.4% 0.6% 0.8% 1.0% M ism atch Lo ss (Ye ar ly) Albedo Mismatch Loss Under Different Albedo 1 Portrait 2 Portrait 0.60.2 0.4 19Canadian Solar Inc. Fixed racking 2P: Shading factor and mismatch loss result Shading factor/Mismatch loss : height 0.5m has similar trend and result with 1.0m under different Albedo. Increased height can obviously boost backside irradiance. Rail Rear side irradiance distribution Height 0.5m/Albedo 0.6 Cell Row: Vertical 48 cells for the analysis Height 1.0m/Albedo 0.6 Bottom Top Top Bottom 17.8% 18.6% 20.0%17.9% 19.4% 20.2% 10.0% 15.0% 20.0% 25.0% Sh ad ing Fact or (Yearly ) Albedo Shading Factor Under Different Albedo Height Height 0.5m Height 1.0m 0.2 0.4 0.6 0.14% 0.16% 0.17%0.16% 0.16% 0.16% 0.0% 0.1% 0.2% 0.3% M ism atch Lo ss (Ye ar ly) Albedo Mismatch Loss Under Different Albedo Height Height 0.5m Height 1.0m 0.2 0.4 0.6 20Canadian Solar Inc. Fixed racking 3L: Shading factor and mismatch loss result Shading factor /Mismatch loss : height 0.5m has higher value than 1.0m under different Albedo. Increased height can obviously increase backside irradiance. Rail Down Rear side irradiance distribution Height 0.5m/Albedo 0.6 Cell Row: Vertical 18 cells for the analysis Height 1.0m/Albedo 0.6 Top Top Down 10.4% 10.3% 10.4% 9.3% 9.6% 8.6% 0.0% 5.0% 10.0% 15.0% Shadi ng Fac to r (Yea rly) Albedo Shading Factor Under Different Albedo Height Height 0.5m Height 1.0m 0.2 0.4 0.6 0.26% 0.32% 0.47% 0.18% 0.20% 0.21% 0.0% 0.1% 0.2% 0.3% 0.4% 0.5% M ism atch Lo ss (Ye ar ly) Albedo Mismatch Loss Under Different Albedo Height Height 0.5m Height 1.0m 0.2 0.4 0.6 Height 21Canadian Solar Inc. Modeling methodology introduction02 Shading factor and mismatch loss evaluation03 04 Conclusion Bifacial technology overview01 22Canadian Solar Inc. Conclusions and Next step Conclusions Next step • Effect of cell/module temperature will be considered into mismatch loss calculation. 5.6% 4.9% 17.8% 8.6%8.1% 7.6% 20.2% 10.4% 0.0% 5.0% 10.0% 15.0% 20.0% 25.0% Sh ad ing facto r (Y ear ly) Shading factor range (Albedo: 0.2~0.6) HSAT 1P HSAT 2P Fixed 2P Fixed 3L 0.23% 0.18% 0.14% 0.18% 0.94% 0.23% 0.17% 0.47% 0.0% 0.2% 0.4% 0.6% 0.8% 1.0% M ism atch loss (Y ear ly) Mismatch loss range (Albedo: 0.2~0.6) HSAT 1P HSAT 2P Fixed 2P Fixed 3L • HSAT 1 portrait: shading factor and mismatch loss is not sensitive to array height, mismatch loss is sensitive to Albedo. • HSAT 2 portrait: backside irradiance is still affected by torque tube, purlin, post, so shading factor can not be 0%, has more advantage than 1 portrait under Albedo≥0.4 condition. • Fixed racking: shading factor is much higher than HSAT system. 23Canadian Solar Inc. INNOVATIONS MAKE THE DIFFERENCE 2019/12 24Canadian Solar Inc. Back-up slides 25Canadian Solar Inc. Various simulation tool comparison: NREL VF Vs. NREL Raytracing Vs. CSI Raytracing NREL Raytracing and CSI Raytracing tool have almost same result for backside irradiance and mismatch loss that are lower than NREL VF result. 0.28% 0.24% 0.23% 0.00% 0.10% 0.20% 0.30% M ism atch loss Mismatch loss comparison based on different tool NREL View Factor NREL Raytracing CSI Raytracing Simulation input: System: HSAT Weather data: EPW/One day July 20th Others: Albedo 0.6 0 500 1000 1500 2000 2500 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Back sid e Irr ad ian ce (Wh/m 2) Row NREL VF NREL Raytracing CSI Raytracing NREL VF_Tool NREL Raytracing_Tool CSI Raytracing_Tool 26Canadian Solar Inc. Effect of Torque-Tube Parameters on Rear-Irradiance and Rear-Shading Loss for Bifacial PV Performance on 1-Axis Tracking Systems Silvana Ayala Peláez, Chris Deline, Joshua S. Stein(Sandia National Labs), Bill Marion, Kevin Anderson (Cypress Creek Renewables), and Matthew Muller 46th IEEE PVSC, June 16-21, 2019, Chicago, Illinois Electrical mismatch loss vs. Mismatch loss factor