Advanced ray tracing to predict the energy yield of bifacial systems with reduced uncertainty - Creek Renewables and PV Lighthouse
www.pvlighthouse.com.au Premium PV software Slide 1 Applying advanced ray tracing to predict the energy yield of bifacial systems with reduced uncertainty PV ModuleTech 2018 Keith McIntosh, Malcolm Abbott and – how rapidly, easily, and accurately the effects can be simulated. Today’s talk: A major advance in simulation that enables us to answer this questions.www.pvlighthouse.com.au Premium PV software Slide 4 Systems investigated • Bifacial. • 1D tracking, NS axis • One-high (1981). NREL Solar Radiation Research Laboratory (SRRL): Baseline Measurement System (BMS); Golden, Colorado (Data); NREL Report No. DA-5500- 56488. http://dx.doi.org/10.5439/1052221www.pvlighthouse.com.au Premium PV software Slide 7 Challenge 2: Spectral variability • Solar spectrum changes throughout day A. Andreas, T. Stoffel; (1981). NREL Solar Radiation Research Laboratory (SRRL): Baseline Measurement System (BMS); Golden, Colorado (Data); NREL Report No. DA-5500- 56488. http://dx.doi.org/10.5439/1052221www.pvlighthouse.com.au Premium PV software Slide 8 Challenge 2: Spectral variability • Solar spectrum changes throughout day – 0.1% for two-high system. • Greater loss on diffuse days. Mismatch loss (due to non-uniform illumination in module)www.pvlighthouse.com.au Premium PV software Slide 29 Evaluate different system configurationswww.pvlighthouse.com.au Premium PV software Slide 30 Accuracy of predicting daily yield Global intensity Non-uniformity in module T amb & wind Direct intensity Spectral dependencies Uncertainty 0% ±11.4% ±18.4% • For our location, albedo, module & system configuration, we define accuracy of daily yield prediction. • Spectral dependencies have the greatest influence. Uncertainty to 95% confidence.www.pvlighthouse.com.au Premium PV software Slide 31 • It can be done: Annual yield solved by ray tracing to micron-level, accounting for – Spectral variability of direct and diffuse light, – Spectral and angular dependencies of ground, torque-tube and module, – Mismatch within a module due to non-uniform illumination. • Solutions currently ~45 mins per configuration. Future: 5mins. • Results allow us to quantify advantages – system configurations, – module features, – simulation assumptions. • Is it worth accounting for spectral variability & mismatch due to non-uniformity? Early days, but yes, it looks that way. Summarywww.pvlighthouse.com.au Premium PV software Slide 32 www.pvlighthouse.com.au WeChat: B7282628 Thank you