📋 Case Study
Levelized Cost of Energy (LCOE) Analysis in Challenging Environments
Highly variable solar irradiance due to frequent coastal fog (camanchaca), extreme diurnal temperature swings (−2°C to 45°C), abrasive dust causing accelerated PV soiling and turbine wear, and limited grid interconnection—requiring LCOE modeling that accounts for non-standard degradation, O&M escalation, and fuel logistics risk.
🏗️ Project Overview
Off-grid hybrid power system for a remote copper mining operation in the Atacama Desert, Chile; 42 MW peak capacity (25 MW solar PV, 12 MW wind, 5 MW battery storage + diesel backup); 20-year project lifetime.
🎯 Challenge
Highly variable solar irradiance due to frequent coastal fog (camanchaca), extreme diurnal temperature swings (−2°C to 45°C), abrasive dust causing accelerated PV soiling and turbine wear, and limited grid interconnection—requiring LCOE modeling that accounts for non-standard degradation, O&M escalation, and fuel logistics risk.
🔧 Design Approach
Integrated probabilistic LCOE framework combining Monte Carlo simulation for weather uncertainty, site-specific degradation curves (validated via 18-month on-site sensor array), and stochastic diesel fuel price and transport cost modeling; validated against IRENA’s LCOE guidelines and adapted per IEEE 1547-2018 for islanded microgrid dispatch constraints.
📐 Design Diagram
AI-generated project design illustration
📐 Key Calculations
LCOE base case
LCOE = (Σ_t (CAPEX_t + OPEX_t + Fuel_t) / (1+r)^t) / (Σ_t (E_gen,t / (1+r)^t))
Result: USD 0.138/kWh
Baseline benchmark reflecting nominal assumptions; reveals sensitivity to fuel cost volatility in isolated locations.
Soiling-adjusted energy yield loss
Annual yield loss = ∫[0,T] (1 − exp(−k·dust_accumulation_rate·t)) · GHI_t dt × system efficiency
Result: 11.2% average annual reduction in PV output
Directly increases effective LCOE by 9.4%—underscores need for dynamic cleaning schedules and anti-soiling coating ROI analysis.
Probabilistic LCOE 90th percentile
LCOE_P90 = 90th percentile of 10,000 Monte Carlo LCOE simulations incorporating irradiance, wind speed, and fuel price uncertainties
Result: USD 0.169/kWh
Critical for financial covenant compliance—used by lenders to set debt service coverage ratio (DSCR) thresholds under worst-case operational stress.
📊 Results
Metrics: LCOE base case: USD 0.138/kWh, LCOE P90: USD 0.169/kWh, Diesel displacement rate: 73%, Levelized O&M cost: USD 0.021/kWh
The hybrid design achieved 22% lower LCOE than diesel-only alternative and met internal hurdle rate (8% real IRR) across 94% of Monte Carlo scenarios; robustness confirmed via ±15% CAPEX and ±25% fuel price stress tests.
💡 Lessons Learned
- •Site-specific environmental degradation models are non-negotiable inputs—not scalable from temperate-region benchmarks.
- •LCOE must be coupled with dispatch-aware energy time-series modeling to avoid overestimating firm capacity in intermittency-constrained environments.
✅ Key Takeaways
- 1In off-grid industrial applications, LCOE is not a single value but a risk-weighted distribution—decision-making must anchor to P80–P90 bounds, not mean estimates.