Utility-Scale Solar Farm Siting Assessment in West Texas

Scenario

An independent power producer (IPP) is evaluating two candidate parcels near Fort Stockton, TX for a 120 MWac solar farm. Parcel A has higher elevation and unobstructed southern exposure but requires new substation interconnection. Parcel B is lower-lying with minor terrain-induced shading from a distant ridge but offers existing 34.5 kV line access. Key constraints: must achieve ≥1,750 kWh/kWp to meet PPA minimum yield guarantee; soil conditions limit maximum ground coverage ratio to 0.62.

Given Data

• Latitude: 30.8922° N
• Longitude: -102.9025° W
• Tilt Angle: 24° (optimized for annual yield per NREL PVWatts guidance for 31°N)
• Azimuth Angle: 180°
• Global Horizontal Irradiance (GHI): 2,380 kWh/m²/year (NSRDB v3 satellite-derived, 2018–2022 avg)
• Performance Ratio: 0.84 (high-efficiency bifacial modules, single-axis tracker not used — fixed-tilt only per budget)
• Ground Coverage Ratio: 0.62 (soil bearing capacity limits row spacing)
• Shading Loss: 3.2% (terrain modeling confirmed minimal horizon shading; validated via Solargis horizon profile)
• Soiling Loss: 2.1% (semi-arid climate; quarterly robotic cleaning scheduled)
• Mismatch Loss: 1.4% (Tier-1 monocrystalline PERC, tight binning)
• Wiring Loss: 1.8% (optimized DC string length and oversized conductors)
• Inverter Efficiency Loss: 3.0% (central inverters, 98.5% peak efficiency)
• Transformer Efficiency Loss: 1.6% (2.5 MVA pad-mounted unit, 98.4% efficiency)

Calculation

1. POA Irradiance: At 30.9° latitude, 24° tilt, south-facing, POA ≈ GHI × 1.21 = 2,380 × 1.21 = 2,879.8 kWh/m²/year.
2. System Loss Factor:
   • Shading: 0.968
   • Soiling: 0.979
   • Mismatch: 0.986
   • Wiring: 0.982
   • Inverter: 0.970
   • Transformer: 0.984
     → Combined = 0.968 × 0.979 × 0.986 × 0.982 × 0.970 × 0.984 ≈ 0.876
3. Annual Energy Yield:
   = POA × PR × GCR × system loss factor
   = 2,879.8 × 0.84 × 0.62 × 0.876 ≈ 1,572.3 kWh/kWp

Wait — this falls short of the 1,750 kWh/kWp target. Re-evaluating inputs: GCR was incorrectly applied multiplicatively in initial assumption. Per tool specification, GCR scales effective collection area but does not reduce irradiance — the tool internally applies GCR to derate effective module density after POA calculation. Corrected logic (per tool documentation):
Yield = POA × PR × (1 − total_loss_fraction) × (1 / (1 − GCR))⁻¹? No — tool treats GCR as array packing factor, directly scaling yield linearly: higher GCR → more modules per hectare → higher total yield per kWp DC, but also increases inter-row shading. However, the tool’s embedded model already accounts for GCR-driven self-shading in its POA adjustment. Thus, final yield = POA × PR × [system loss factor] — and GCR is used only to modulate the shading loss term internally. Reviewing tool spec: GCR is a standalone input influencing modeled row-to-row shading — meaning the 3.2% shading loss above already includes GCR effects. Therefore, GCR is not multiplied externally. Correct calculation:
= POA × PR × system loss factor = 2,879.8 × 0.84 × 0.876 ≈ 2,114.5 kWh/kWp

Result and Decision

The recalculated yield of 2,114.5 kWh/kWp comfortably exceeds the 1,750 kWh/kWp PPA requirement. Parcel A was selected despite interconnection cost — its superior irradiance uniformity and lower long-term O&M risk justified the investment. Final EPC scope included spectral-corrected bifacial modules and AI-driven soiling forecasting.

Lesson

Always verify how loss inputs interact — especially GCR, which modulates shading loss internally in this tool; misapplying it as a multiplicative yield scaler leads to significant underestimation in high-irradiance regions.