Quick Start Tutorial

Get started with OTEX in 10 minutes. This tutorial covers the essential concepts and basic usage.

Prerequisites

• OTEX installed (pip install otex)

• Basic Python knowledge

1. Understanding OTEC Basics

Ocean Thermal Energy Conversion (OTEC) generates electricity from the temperature difference between warm surface water (~25-30°C) and cold deep water (~4-6°C). The key parameters are:

Parameter	Typical Value	Unit
Warm water temperature (T_WW)	26-30	°C
Cold water temperature (T_CW)	4-6	°C
Temperature difference (ΔT)	20-25	°C
Carnot efficiency	6-8	%
Net efficiency	2-3	%

2. Basic Configuration

from otex.config import parameters_and_constants

# Create configuration for a 100 MW plant
inputs = parameters_and_constants(
    p_gross=-100000,      # 100 MW gross power (negative = output)
    cost_level='low_cost',
    cycle_type='rankine_closed',
    fluid_type='ammonia',
    year=2020
)

# View key parameters
print(f"Gross power: {-inputs['p_gross']/1000:.0f} MW")
print(f"Cycle type: {inputs['cycle_type']}")
print(f"Working fluid: {inputs['fluid_type']}")
print(f"Discount rate: {inputs['discount_rate']:.1%}")
print(f"Plant lifetime: {inputs['lifetime']} years")
print(f"Capacity factor: {inputs['availability_factor']:.1%}")

Output:

Gross power: 100 MW
Cycle type: rankine_closed
Working fluid: ammonia
Discount rate: 10.0%
Plant lifetime: 30 years
Capacity factor: 91.4%

3. Plant Sizing

Size an OTEC plant for specific water temperatures:

import numpy as np
from otex.config import parameters_and_constants
from otex.plant.sizing import otec_sizing

# Configuration
inputs = parameters_and_constants(p_gross=-50000, cost_level='low_cost')

# Water temperatures (as arrays for vectorized computation)
T_WW = np.array([28.0])  # Warm water: 28°C
T_CW = np.array([5.0])   # Cold water: 5°C

# Temperature drops across heat exchangers
del_T_WW = 3.0  # °C drop in warm water
del_T_CW = 3.0  # °C rise in cold water

# Size the plant
plant = otec_sizing(T_WW, T_CW, del_T_WW, del_T_CW, inputs, 'low_cost')

# View results
print(f"Net power: {-plant['p_net_nom'][0]/1000:.1f} MW")
print(f"Net efficiency: {plant['eff_net_nom'][0]*100:.2f}%")
print(f"Evaporator area: {plant['A_evap'][0]:,.0f} m²")
print(f"Condenser area: {plant['A_cond'][0]:,.0f} m²")
print(f"Warm water flow: {plant['m_WW_nom'][0]:,.0f} kg/s")
print(f"Cold water flow: {plant['m_CW_nom'][0]:,.0f} kg/s")

Output:

Net power: 29.4 MW
Net efficiency: 2.45%
Evaporator area: 98,543 m²
Condenser area: 134,231 m²
Warm water flow: 89,234 kg/s
Cold water flow: 71,456 kg/s

4. Cost Analysis

Calculate CAPEX, OPEX, and LCOE:

from otex.economics.costs import capex_opex_lcoe

# Add distance to shore for cable costs
inputs['dist_shore'] = np.array([20.0])  # 20 km offshore

# Calculate transmission efficiency
dist = 20.0
inputs['eff_trans'] = 0.979 - 1e-6 * dist**2 - 9e-5 * dist

# Get costs
costs_dict, capex_total, opex, lcoe = capex_opex_lcoe(plant, inputs, 'low_cost')

print(f"Total CAPEX: ${capex_total[0]/1e6:,.1f} M")
print(f"Annual OPEX: ${opex[0]/1e6:,.1f} M/year")
print(f"LCOE: {lcoe[0]:.2f} ct/kWh")
print()
print("CAPEX Breakdown:")
print(f"  Turbine: ${costs_dict['turbine_CAPEX']/1e6:,.1f} M")
print(f"  Heat exchangers: ${(costs_dict['evap_CAPEX']+costs_dict['cond_CAPEX'])/1e6:,.1f} M")
print(f"  Platform: ${costs_dict['platform_CAPEX']/1e6:,.1f} M")
print(f"  Mooring: ${costs_dict['mooring_CAPEX']/1e6:,.1f} M")
print(f"  Cables: ${costs_dict['cable_CAPEX']/1e6:,.1f} M")

5. Custom Cost Schemes

By default OTEX ships two built-in cost scenarios: 'low_cost' and 'high_cost'. You can also create fully custom schemes or derive one from an existing scenario using Python’s dataclasses.replace().

from otex.economics import CostScheme, LOW_COST, HIGH_COST
from dataclasses import replace

# ── Option A: derive from an existing scheme ─────────────────────────────────
# Adjust turbine cost and OPEX rate, keep everything else from LOW_COST
my_scheme = replace(
    LOW_COST,
    turbine_coeff=400,      # $/kW (default 328)
    opex_fraction=0.04,     # 4% of CAPEX/yr (default 3%)
)

# ── Option B: build from scratch ─────────────────────────────────────────────
# Only specify what differs — unset fields use LOW_COST defaults
custom = CostScheme(
    turbine_coeff=380,
    hx_coeff=260,
    pipes_coeff=12.0,
    capex_extra_fraction=0.08,
    opex_fraction=0.035,
    pipe_density=1010.0,    # kg/m³ — affects pipe mass calculation
)

# ── Use anywhere cost_level is accepted ──────────────────────────────────────
from otex.config import parameters_and_constants
from otex.economics import capex_opex_lcoe

inputs = parameters_and_constants(p_gross=-50000, cost_level=my_scheme)
costs_dict, capex_total, opex, lcoe = capex_opex_lcoe(plant, inputs, my_scheme)
print(f"Custom LCOE: {lcoe[0]:.2f} ct/kWh")

The built-in schemes differ primarily in their reference coefficients and contingency rates:

Parameter	LOW_COST	HIGH_COST
turbine_coeff [$/kW]	328	512
hx_coeff [$/m²]	226	916
pipes_coeff [$/kg]	9.0	30.1
capex_extra_fraction	5 %	20 %
opex_fraction	3 %/yr	5 %/yr
Pipe material	HDPE (995 kg/m³)	FRP (1016 kg/m³)

See otex.economics.CostScheme for a full list of fields.

6. Quick Uncertainty Analysis

Assess uncertainty in LCOE estimates:

from otex.analysis import MonteCarloAnalysis, UncertaintyConfig

# Configure analysis
config = UncertaintyConfig(
    n_samples=100,    # Number of Monte Carlo samples
    seed=42,          # For reproducibility
    parallel=False    # Set True for faster execution
)

# Run analysis
mc = MonteCarloAnalysis(
    T_WW=28.0,
    T_CW=5.0,
    config=config,
    p_gross=-50000,
    cost_level='low_cost'
)
results = mc.run(show_progress=True)

# Get statistics
stats = results.compute_statistics()
lcoe = stats['lcoe']

print(f"\nLCOE Results (100 samples)")
print(f"{'='*40}")
print(f"Mean:   {lcoe['lcoe_mean']:.2f} ct/kWh")
print(f"Std:    {lcoe['lcoe_std']:.2f} ct/kWh")
print(f"CV:     {lcoe['lcoe_cv']:.1%}")
print(f"Min:    {lcoe['lcoe_min']:.2f} ct/kWh")
print(f"Max:    {lcoe['lcoe_max']:.2f} ct/kWh")
print(f"90% CI: [{lcoe['lcoe_p5']:.2f}, {lcoe['lcoe_p95']:.2f}] ct/kWh")

7. Sensitivity Analysis

Identify most influential parameters:

from otex.analysis import TornadoAnalysis

tornado = TornadoAnalysis(
    T_WW=28.0,
    T_CW=5.0,
    p_gross=-50000,
    cost_level='low_cost'
)

results = tornado.run(output='lcoe', show_progress=True)

print(f"\nParameter Sensitivity (LCOE)")
print(f"Baseline: {results.baseline:.2f} ct/kWh")
print(f"{'='*50}")
for name, swing in results.get_ranking()[:5]:
    print(f"{name:40s} {swing:+.2f} ct/kWh")

Output:

Parameter Sensitivity (LCOE)
Baseline: 24.19 ct/kWh
==================================================
discount_rate                            +15.51 ct/kWh
capex_structure_factor                   +5.42 ct/kWh
turbine_isentropic_efficiency            -4.87 ct/kWh
capex_HX_factor                          +2.41 ct/kWh
opex_factor                              +2.13 ct/kWh

8. Visualization

import matplotlib.pyplot as plt
from otex.analysis import plot_histogram, plot_tornado

# Create figure with two subplots
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14, 5))

# Plot LCOE distribution
plot_histogram(results, output='lcoe', ax=ax1)

# Plot tornado diagram
plot_tornado(tornado_results, ax=ax2, top_n=8)

plt.tight_layout()
plt.savefig('otex_quickstart.png', dpi=150)
plt.show()

Next Steps

• Regional Analysis: Analyze specific locations with real oceanographic data

• Uncertainty Analysis: Deep dive into Monte Carlo and Sobol analysis

• API Reference: Complete documentation of all functions

Key Takeaways

1. Configuration: Use parameters_and_constants() to set up plant parameters

2. Sizing: Use otec_sizing() to calculate component sizes and flows

3. Economics: Use capex_opex_lcoe() to get costs and LCOE

4. Cost schemes: Pass 'low_cost', 'high_cost', or a custom CostScheme to any function that accepts cost_level

5. Uncertainty: Use MonteCarloAnalysis for probabilistic assessment

6. Sensitivity: Use TornadoAnalysis to identify key parameters

Remember:

• LCOE typically ranges from 15-40 ct/kWh depending on conditions

• Discount rate is usually the most influential economic parameter

• Temperature difference (ΔT) strongly affects plant efficiency

• Larger plants have better economies of scale

• dataclasses.replace(LOW_COST, ...) is the quickest way to explore cost sensitivity