Lcoe branch

docs/examples/financial/lcoe_sam_validation.py

@@ -0,0 +1,193 @@
+"""
+LCOE Calculation
+================
+
+Example of an LCOE calculation for a utility-scale site in Albuquerque, NM
+using the approach implemented by NREL's SAM software
+"""
+
+# This example shows usage of pvlib's lcoe calculation with
+# :py:meth:`pvlib.financial.lcoe`, :py:meth:`pvlib.financial.wacc`,
+# :py:meth:`pvlib.financial.nominal_to_real`, and
+# :py:meth:`pvlib.financial.crf` to generate a Series of annual cost and
+# production data, and a real LCOE. TMY GHI, DNI, and DHI irradiance data
+# for Albuquerque is loaded from the NSRDB and
+# :py:meth:`pvlib.location.get_solarposition` is used with
+# :py:meth:`pvlib.irradiance.get_total_irradiance`to calculate POA
+# irradiance. DC energy production is calculated with
+# :py:meth:`pvlib.pvsystem.pvwatts_dc` to get annual AC power output.
+# Capital cost is calculated using the FCR method described here:
+# http://samrepo.nrelcloud.org/help/index.html?fin_lcoefcr.htm with real
+# discount rates. Construction interest is assumed to be zero. Input values
+# with an asterisk were sourced from NREL's ATB projections for a residential
+# system in 2022 with moderate
+# technological advancement or the set of financial assumptions under which
+# NREL produces the ATB. Monthly POA output [kWh/m^2], annual AC output [kWh],
+# and LCOE should match values calculated by SAM.
+
+import numpy as np
+import pandas as pd
+import datetime
+from pvlib import location
+from pvlib import irradiance
+from pvlib import temperature
+from pvlib import pvsystem
+from pvlib import inverter
+from pvlib import financial
+# from .conftest import DATA_DIR
+
+from pvlib.tests.conftest import DATA_DIR
+
+# Get annual AC output
+
+# Installed DC capacity [W] (total is 1 MW)
+installed_dc = 1000000
+
+# Surface tilt and azimuth
+tilt, surface_azimuth = 30, 180
+
+# Set Albuquerque as location
+lat, lon, elev = 35.054942, -106.540485, 1657.8
+loc = location.Location(lat, lon, altitude=elev)
+
+# Albuquerque TMY data from NSRDB
+data_file = DATA_DIR / 'albuquerque_tmy.csv'
+data = pd.read_csv(data_file, skiprows=[0, 1])
+
+# Set DatetimeIndex for data
+data.set_index(pd.DatetimeIndex(data[['Month', 'Day', 'Hour', 'Minute']]
+                                .apply(lambda x:
+                                       datetime.datetime(2022, x['Month'],
+                                                         x['Day'],
+                                                         x['Hour'],
+                                                         x['Minute']),
+                                       axis=1)), inplace=True)
+
+# loc.get_solarposition() assumes UTC unless times is localized
+# but Albuquerque is in Etc/GMT-7
+temp = loc.get_solarposition(times=pd.date_range(start=data.index[0]
+                                                 + pd.Timedelta(7, 'h'),
+                                                 end=data.index[-1]
+                                                 + pd.Timedelta(7, 'h'),
+                                                 freq='1H'))
+# Shift index back to align with Etc/GMT-7
+solar_position = temp.set_index(temp.index.shift(periods=-7, freq='1H'))
+
+# Get POA and apply AOI modifier to direct and diffuse components
+poa_irrad = irradiance.get_total_irradiance(
+    surface_tilt=tilt, surface_azimuth=surface_azimuth,
+    dni=data['DNI'], ghi=data['GHI'], dhi=data['DHI'],
+    solar_zenith=solar_position['zenith'],
+    solar_azimuth=solar_position['azimuth'],
+    albedo=data['Surface Albedo'])['poa_global']
+
+# Calulate and display daily/monthly stats
+daily_ghi = data['GHI'].groupby(data.index.map(lambda x: x.date())).sum().\
+    mean()/1000
+daily_dhi = data['DHI'].groupby(data.index.map(lambda x: x.date())).sum().\
+    mean()/1000
+daily_dni = data['DNI'].groupby(data.index.map(lambda x: x.date())).sum().\
+    mean()/1000
+monthly_poa = poa_irrad.groupby(poa_irrad.index.map(lambda x:
+                                                    x.date().month)).\
+    sum()/1000
+
+print('Daily average GHI is ' + str(np.round(daily_ghi, 3)) + ' kWh/m^2')
+print('Daily average DHI is ' + str(np.round(daily_dhi, 3)) + ' kWh/m^2')
+print('Daily average DNI is ' + str(np.round(daily_dni, 2)) + ' kWh/m^2')
+print('Monthly POA averages [kWh/m^2]:')
+print(monthly_poa)
+
+# Get system losses
+losses = pvsystem.pvwatts_losses()/100
+
+# Get cell temperature
+cell_temp = temperature.pvsyst_cell(poa_irrad, data['Temperature'],
+                                    data['Wind Speed'])
+
+# Get hourly DC output using PVWatts [W DC]
+dc_power = pvsystem.pvwatts_dc(poa_irrad, cell_temp, installed_dc,
+                               -0.0037) * (1 - losses)
+
+# Get hourly AC output using PVWatts [W AC]
+ac_power = inverter.pvwatts(dc_power, installed_dc/1.1)
+
+# Check that AC power data is evenly spaced over hour increments
+if ~np.all(np.unique(np.diff(ac_power.index)/np.timedelta64(1, 'h')) == 1):
+    raise ValueError
+
+# Riemann-sum to get annual AC output [kWh]
+annual_ac_output = ac_power.sum()/1000
+print('Annual AC output is ' + str(np.round(annual_ac_output, 2)) + ' kWh')
+
+# Period of financial analysis
+n = 20
+
+# Assume system degradation rate is 1%
+sdr = 0.01
+
+# Apply degradation rate to AC production over analysis period
+production = np.array([annual_ac_output*(1 - sdr)**i for i in range(n)])
+
+# Total installed capital costs [$] *
+capex = 1119.82*installed_dc/1000
+
+# Fixed O&M costs [$] *
+fixed_op_cost = np.full(n, 19.95*installed_dc/1000)
+
+# Inflation rate *
+inflation_r = 0.025
+
+# Nominal interest rate *
+interest_r = 0.04
+
+# Real interest rate
+rint = financial.nominal_to_real(interest_r, inflation_r)
+
+# Nominal internal rate of return *
+irr = 0.0775
+
+# Real internal rate of return
+rroe = financial.nominal_to_real(irr, inflation_r)
+
+# Fraction of capital cost covered by a loan *
+loan_frac = 0.518577595078774
+
+# Effective tax rate *
+tax_r = 0.2574
+
+# Real weighted average cost of capital
+my_wacc = financial.wacc(loan_frac, rroe, rint, inflation_r, tax_r)
+print('Real weighted average cost of capital is ' + str(np.round(my_wacc, 5)))
+
+# Depreciation schedule
+dep_sch = pd.Series([20, 32, 19.2, 11.52, 11.52, 5.76])/100
+
+# Present value of depreciation
+pvdep = np.sum([dep_sch.at[j]/((1 + my_wacc)*(1 + inflation_r))**(j+1)
+                for j in range(len(dep_sch))])
+
+# Project financing factor
+pff = (1 - (tax_r*pvdep))/(1 - tax_r)
+
+# Construction financing factor
+cff = 1
+
+# Capital recovery factor
+my_crf = financial.crf(my_wacc, n)
+
+# Fixed charge rate
+fcr = my_crf*pff*cff
+print('Fixed charge rate is ' + str(np.round(fcr, 5)))
+
+debt_tenor = n
+
+# Annuity (annual payment) on total capital cost [$]
+cap_cost = np.full(n, capex*fcr)
+print('Annual payment on capital cost is $' + str(np.round(cap_cost[0], 2)))
+
+# Call lcoe()
+my_lcoe = financial.lcoe(production=production, cap_cost=cap_cost,
+                         fixed_om=fixed_op_cost)
+
+print('Real LCOE = ' + str(my_lcoe) + str(' cents/kWh'))

docs/examples/financial/simple_lcoe_calculator.py

@@ -0,0 +1,65 @@
+"""
+LCOE Calculation
+================
+
+Example of an LCOE calculation using an approach implemented in NREL's "Simple
+LCOE Calculator", accessible at http://www.nrel.gov/analysis/tech-lcoe.html
+"""
+# %%
+# This example shows basic usage of pvlib's lcoe calculation with
+# :py:meth:`pvlib.financial.lcoe` and :py:meth:`pvlib.financial.crf`.
+# The example shown here will generate a Series of annual cost and production
+# data, and a numerical LCOE. To be comparable with NREL's implemenation,
+# this example adheres to the following assumptions: that energy production
+# and O&M costs are constant and that the entire project is financed with a
+# loan. Input values for CAPEX, capacity factor, and O&M were sourced from
+# NREL's ATB for a residential system in 2022 located in Resource Class 5
+# with moderate technological advancement. The discount rate is set to the
+# value recommended in NREL's implementation.
+
+import numpy as np
+import pandas as pd
+from pvlib import financial
+
+# Analysis period
+n = 20
+
+# Capacity factor
+cf = 0.15357857
+
+# Constant annual energy production
+energy = np.full(n, cf*8760)
+
+# Real discount rate
+discount_rate = 0.03
+
+# Capital recovery factor
+my_crf = financial.crf(discount_rate, n)
+
+# CAPEX
+capex = 2443.45
+
+# Fraction of capital cost
+loan_frac = 1
+
+# Annual capital costs
+cap_cost = np.array([capex*loan_frac*my_crf for i in range(n)])
+
+# Constant annual O&M
+fixed_om = pd.Series(data=[26.98 for j in range(n)])
+
+# Put data in table and display
+table = pd.DataFrame(columns=['Production [kWh/kW]', 'Capital cost [$/kW]',
+                              'O&M [$/kW]'])
+table['Production [kWh/kW]'] = energy
+table['Capital cost [$/kW]'] = cap_cost
+table['O&M [$/kW]'] = fixed_om
+table.index.name = 'Year'
+table
+
+# %%
+# Get LCOE
+
+my_lcoe = financial.lcoe(production=energy, cap_cost=cap_cost,
+                         fixed_om=fixed_om)
+print('LCOE = ' + str(my_lcoe) + str(' cents/kWh'))

docs/sphinx/source/reference/financial.rst

@@ -0,0 +1,13 @@
+.. currentmodule:: pvlib
+
+Financial
+=========
+
+.. autosummary::
+   :toctree: generated/
+
+   financial.lcoe
+   financial.crf
+   financial.nominal_to_real
+   financial.real_to_nominal
+   financial.wacc

docs/sphinx/source/reference/index.rst

@@ -20,3 +20,4 @@ API reference
    bifacial
    scaling
    location
+   financial

docs/sphinx/source/whatsnew/v0.10.2.rst

@@ -24,6 +24,11 @@ Enhancements
 * :py:class:`~pvlib.pvsystem.PVSystem` objects with a single
   :py:class:`~pvlib.pvsystem.Array` can now be created without wrapping the
   ``Array`` in a list first. (:issue:`1831`, :pull:`1854`)
+* Added :py:mod:`pvlib.financial` module with functions
+  :py:func:`~pvlib.financial.lcoe`, :py:func:`~pvlib.financial.crf`,
+  :py:func:`~pvlib.financial.nominal_to_real`,
+  :py:func:`~pvlib.financial.real_to_nominal`, and
+  :py:func:`~pvlib.financial.wacc` (:pull:`1687`)
 
 
 Bug fixes
@@ -84,3 +89,4 @@ Contributors
 * Will Holmgren (:ghuser:`wholmgren`)
 * Mark Mikofski (:ghuser:`mikofski`)
 * Kevin Anderson (:ghuser:`kandersolar`)
+* Emma Cooper (:ghuser:`eccoope`)

docs/sphinx/source/whatsnew/v0.9.5.rst

@@ -37,6 +37,7 @@ Enhancements
 * Use `Horner's Method <https://en.wikipedia.org/wiki/Horner%27s_method>`_
   to evaluate polynomials in :py:func:`~pvlib.irradiance.disc`, may
   decrease runtime by 20%. (:issue:`1180`, :pull:`1183`)
+
 
 Bug fixes
 ~~~~~~~~~
@@ -65,6 +66,8 @@ Testing
 Documentation
 ~~~~~~~~~~~~~
 * Remove LGTM.com integration. (:issue:`1550`, :pull:`1651`)
+* Added two examples demonstrating how :py:mod:`pvlib.financial` can
+  be used (:pull:`1687`)
 
 Benchmarking
 ~~~~~~~~~~~~~