Advanced HPP Model
Evaluating the performance of a hybrid power plant using HyDesign
HyDesign is an open-source tool for design and optimization of utility scale wind-solar-storage based hybrid power plants.
In this notebook we will evaluate a hybrid power plant design in a specific location.
A hybrid power plant design consists on selecting the following parameters:
Wind Turbine design:
Clearance [m] (
clearance): Height from the ground to rotor tip at lowest posstion. This parameter controls the wind turbine hub height given a rotor radius:clearance = HH - R.Specific power of the wind turbine [MW/m2] (
sp): Defined assp = p_rated / ( pi * R^2 ), it controls the turbine rotor size for a given rated power. Turbines with lower specific power produce more power at lower wind speeds, but are more expensive.Rated power of the wind turbine [MW] (
p_rated)
Wind Plant design:
Number of wind turbines in the wind plant [-] (
Nwt)Wind power installation density [MW/km2] (
wind_MW_per_km2): This parameter controls how closely spaced are the turbines, which in turns affect how much wake losses are present.
PV Plant design:
Solar plant power capacity [MW] (
solar_MW)Surface tilt [deg] (
surface_tilt)Surface azimuth [deg] (
surface_azimuth)DC-AC ratio [-] (
solar_DCAC): This parameter controls how much over-planting of PV (in DC power) is connected to the inverters. It is common practice in PV design to havesolar_DCAC = 1.5.
Battery Storage design:
Battery power [MW] (
b_P)Battery energy capacity in hours [MWh] (
b_E_h): Battery storage capacity in hours of full battery power (b_E = b_E_h * b_P).Cost of battery power fluctuations in peak price ratio [-] (
cost_of_batt_degr): This parameter controls how much penalty is given to do ramps in battery power in the HPP operation.
Imports
Install hydesign if needed. Import basic libraries. Import HPP model assembly class. Import the examples file path.
[2]:
# Install hydesign if needed
import importlib
if not importlib.util.find_spec("hydesign"):
!pip install git+https://gitlab.windenergy.dtu.dk/TOPFARM/hydesign.git
[3]:
import os
import time
import yaml
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from hydesign.assembly.hpp_assembly import hpp_model
from hydesign.examples import examples_filepath
Specifying the site
Hydesign, provides example data from several sites in India and Europe.
The site coordinates (longitude, latitude, and altitude) are given in examples_sites.csv.
[4]:
examples_sites = pd.read_csv(f'{examples_filepath}examples_sites.csv', index_col=0, sep=';')
examples_sites
[4]:
| case | name | longitude | latitude | altitude | input_ts_fn | sim_pars_fn | price_fn | price_col | H2_demand_col | Unnamed: 11 | input_HA_ts_fn | price_up_ts | price_dwn_ts | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | India | Indian_site_good_wind | 77.500226 | 8.334294 | 679.803454 | India/GWA2/input_ts_Indian_site_good_wind.csv | India/hpp_pars.yml | India/Indian_elec_price_t.csv | Price | India/H2_demand.csv | NaN | NaN | NaN | NaN |
| 1 | India | Indian_site_good_solar | 68.542204 | 23.542099 | 29.883557 | India/GWA2/input_ts_Indian_site_good_solar.csv | India/hpp_pars.yml | India/Indian_elec_price_t.csv | Price | India/H2_demand.csv | NaN | NaN | NaN | NaN |
| 2 | India | Indian_site_bad_solar_bad_wind | 77.916878 | 17.292316 | 627.424643 | India/GWA2/input_ts_Indian_site_bad_solar_bad_... | India/hpp_pars.yml | India/Indian_elec_price_t.csv | Price | India/H2_demand.csv | NaN | NaN | NaN | NaN |
| 3 | Europe | France_good_solar | 4.229736 | 44.422011 | 204.000000 | Europe/GWA2/input_ts_France_good_solar.csv | Europe/hpp_pars.yml | Europe/2030-EL_PRICE.csv | FR_R | Europe/H2_demand.csv | NaN | NaN | NaN | NaN |
| 4 | Europe | France_good_wind | -0.864258 | 48.744116 | 302.000000 | Europe/GWA2/input_ts_France_good_wind.csv | Europe/hpp_pars.yml | Europe/2030-EL_PRICE.csv | FR_R | Europe/H2_demand.csv | NaN | NaN | NaN | NaN |
| 5 | Europe | France_bad_solar_n_wind | 2.167969 | 47.428087 | 140.000000 | Europe/GWA2/input_ts_France_bad_solar_n_wind.csv | Europe/hpp_pars.yml | Europe/2030-EL_PRICE.csv | FR_R | Europe/H2_demand.csv | NaN | NaN | NaN | NaN |
| 6 | Europe | Germany_bad_solar_n_wind | 10.766602 | 49.310798 | 442.000000 | Europe/GWA2/input_ts_Germany_bad_solar_n_wind.csv | Europe/hpp_pars.yml | Europe/2030-EL_PRICE.csv | DE_ME | Europe/H2_demand.csv | NaN | NaN | NaN | NaN |
| 7 | Europe | Germany_good_wind | 7.873535 | 53.287111 | 5.000000 | Europe/GWA2/input_ts_Germany_good_wind.csv | Europe/hpp_pars.yml | Europe/2030-EL_PRICE.csv | DE_NW | Europe/H2_demand.csv | NaN | NaN | NaN | NaN |
| 8 | Europe | Denmark_good_solar | 11.813965 | 55.397760 | 42.000000 | Europe/GWA2/input_ts_Denmark_good_solar.csv | Europe/hpp_pars.yml | Europe/2030-EL_PRICE.csv | DK_E | Europe/H2_demand.csv | NaN | NaN | NaN | NaN |
| 9 | Europe | Denmark_good_wind | 8.594398 | 56.227322 | 85.000000 | Europe/GWA2/input_ts_Denmark_good_wind.csv | Europe/hpp_pars.yml | Europe/2030-EL_PRICE.csv | DK_W | Europe/H2_demand.csv | NaN | NaN | NaN | NaN |
| 10 | Europe | Denmark_offshore | 7.906111 | 55.529722 | 85.000000 | Europe/GWA2/input_ts_Denmark_offshore.csv | Europe/hpp_pars_offshore.yml | Europe/2030-EL_PRICE.csv | DK_W | Europe/H2_demand.csv | NaN | NaN | NaN | NaN |
| 11 | Europe | Denmark_good_wind_BM | 8.594398 | 56.227322 | 85.000000 | Europe/GWA2_BM/input_ts_Denmark_good_wind_DA.csv | Europe/hpp_pars.yml | Europe/2030-EL_PRICE.csv | DK_W | Europe/H2_demand.csv | NaN | Europe/GWA2_BM/input_ts_Denmark_good_wind_HA.csv | Europe/BM_Prices/Up_reg_price.csv | Europe/BM_Prices/Down_reg_price.csv |
| 12 | Europe | Denmark_hybridization_wind_Norhede_Hjortmose | 8.366400 | 56.095400 | 16.662000 | Europe/GWA2/input_ts_Denmark_hybridization_win... | Europe/hpp_pars_Hjortmose.yml | Europe/2030-EL_PRICE.csv | DK_W | Europe/H2_demand.csv | NaN | NaN | NaN | NaN |
| 13 | Europe | Denmark_hybridization_solar_Langelinie | 11.290641 | 54.717469 | 0.042000 | Europe/GWA2/input_ts_Denmark_hybridization_sol... | Europe/hpp_pars_Langelinie.yml | Europe/2030-EL_PRICE.csv | DK_E | Europe/H2_demand.csv | NaN | NaN | NaN | NaN |
[5]:
name = 'Denmark_good_wind'
ex_site = examples_sites.loc[examples_sites.name == name]
longitude = ex_site['longitude'].values[0]
latitude = ex_site['latitude'].values[0]
altitude = ex_site['altitude'].values[0]
input_ts_fn = examples_filepath+ex_site['input_ts_fn'].values[0]
sim_pars_fn = examples_filepath+ex_site['sim_pars_fn'].values[0]
Initializing the HPP model
Initialize the HPP model (hpp_model class) with the coordinates and the necessary input files.
[7]:
hpp = hpp_model(
latitude=latitude,
longitude=longitude,
altitude=altitude,
num_batteries = 5,
work_dir = './',
sim_pars_fn = sim_pars_fn,
input_ts_fn = input_ts_fn,
)
Fixed parameters on the site
-------------------------------
longitude = 8.594398
latitude = 56.227322
altitude = 85.0
Evaluating the HPP model
[8]:
start = time.time()
clearance = 10
sp = 350
p_rated = 5
Nwt = 70
wind_MW_per_km2 = 7
solar_MW = 100
surface_tilt = 50
surface_azimuth = 180
solar_DCAC = 1.5
b_P = 50
b_E_h = 3
cost_of_batt_degr = 10
x = [clearance, sp, p_rated, Nwt, wind_MW_per_km2, \
solar_MW, surface_tilt, surface_azimuth, solar_DCAC, \
b_P, b_E_h , cost_of_batt_degr]
outs = hpp.evaluate(*x)
hpp.print_design(x, outs)
end = time.time()
print(f'exec. time [min]:', (end - start)/60 )
Design:
---------------
clearance [m]: 10.000
sp [W/m2]: 350.000
p_rated [MW]: 5.000
Nwt: 70.000
wind_MW_per_km2 [MW/km2]: 7.000
solar_MW [MW]: 100.000
surface_tilt [deg]: 50.000
surface_azimuth [deg]: 180.000
DC_AC_ratio: 1.500
b_P [MW]: 50.000
b_E_h [h]: 3.000
cost_of_battery_P_fluct_in_peak_price_ratio: 10.000
NPV_over_CAPEX: 0.209
NPV [MEuro]: 79.771
IRR: 0.078
LCOE [Euro/MWh]: 47.404
Revenues [MEuro]: 27.206
CAPEX [MEuro]: 382.117
OPEX [MEuro]: 6.040
Wind CAPEX [MEuro]: 267.506
Wind OPEX [MEuro]: 5.365
PV CAPEX [MEuro]: 33.500
PV OPEX [MEuro]: 0.675
Batt CAPEX [MEuro]: 15.128
Batt OPEX [MEuro]: 0.000
Shared CAPEX [MEuro]: 65.982
Shared Opex [MEuro]: 0.000
penalty lifetime [MEuro]: 0.000
AEP [GWh]: 734.430
GUF: 0.279
grid [MW]: 300.000
wind [MW]: 350.000
solar [MW]: 100.000
Battery Energy [MWh]: 150.000
Battery Power [MW]: 50.000
Total curtailment [GWh]: 339.217
Total curtailment with deg [GWh]: 153.804
Awpp [km2]: 50.000
Apvp [km2]: 1.226
Plant area [km2]: 50.000
Rotor diam [m]: 134.867
Hub height [m]: 77.434
Number of batteries used in lifetime: 2.000
Break-even PPA price [Euro/MWh]: 31.474
Capacity factor wind [-]: 0.202
exec. time [min]: 0.14858049154281616
[9]:
b_E_SOC_t = hpp.prob.get_val('ems.b_E_SOC_t')
b_t = hpp.prob.get_val('ems.b_t')
price_t = hpp.prob.get_val('ems.price_t')
wind_t = hpp.prob.get_val('ems.wind_t')
solar_t = hpp.prob.get_val('ems.solar_t')
hpp_t = hpp.prob.get_val('ems.hpp_t')
hpp_curt_t = hpp.prob.get_val('ems.hpp_curt_t')
grid_MW = hpp.prob.get_val('ems.G_MW')
n_days_plot = 14
plt.figure(figsize=[12,4])
plt.plot(price_t[:24*n_days_plot], label='price')
plt.plot(b_E_SOC_t[:24*n_days_plot], label='SoC [MWh]')
plt.plot(b_t[:24*n_days_plot], label='Battery P [MW]')
plt.xlabel('time [hours]')
plt.legend(loc='upper center', bbox_to_anchor=(0.5, 1.15),
ncol=3, fancybox=0, shadow=0)
plt.figure(figsize=[12,4])
plt.plot(wind_t[:24*n_days_plot], label='wind')
plt.plot(solar_t[:24*n_days_plot], label='PV')
plt.plot(hpp_t[:24*n_days_plot], label='HPP')
plt.plot(hpp_curt_t[:24*n_days_plot], label='HPP curtailed')
plt.axhline(grid_MW, label='Grid MW', color='k')
plt.xlabel('time [hours]')
plt.ylabel('Power [MW]')
plt.legend(loc='upper center', bbox_to_anchor=(0.5, 1.15),
ncol=5, fancybox=0, shadow=0)
[9]:
<matplotlib.legend.Legend at 0x231a3f27b10>
[10]:
results_1year = {'Electricity Price':price_t[:8760], 'Wind Power': wind_t[:8760],'Solar Power': solar_t[:8760],'HPP Power': hpp_t[:8760],'Curtailed Power': hpp_curt_t[:8760] ,'Battery SOC': b_E_SOC_t[:8760],'Battery Power': b_t[:8760]}
df = pd.DataFrame(results_1year)
df.to_csv('EMS_out.csv')
[11]:
design_df = pd.DataFrame(columns = ['clearance [m]',
'sp [W/m2]',
'p_rated [MW]',
'Nwt',
'wind_MW_per_km2 [MW/km2]',
'solar_MW [MW]',
'surface_tilt [deg]',
'surface_azimuth [deg]',
'DC_AC_ratio',
'b_P [MW]',
'b_E_h [h]',
'cost_of_battery_P_fluct_in_peak_price_ratio',
'NPV_over_CAPEX',
'NPV [MEuro]',
'IRR',
'LCOE [Euro/MWh]',
'Revenues [MEuro]',
'CAPEX [MEuro]',
'OPEX [MEuro]',
'Wind CAPEX [MEuro]',
'Wind OPEX [MEuro]',
'PV CAPEX [MEuro]',
'PV OPEX [MEuro]',
'Batt CAPEX [MEuro]',
'Batt OPEX [MEuro]',
'Shared CAPEX [MEuro]',
'Shared OPEX [MEuro]',
'penalty lifetime [MEuro]',
'AEP [GWh]',
'GUF',
'grid [MW]',
'wind [MW]',
'solar [MW]',
'Battery Energy [MWh]',
'Battery Power [MW]',
'Total curtailment [GWh]',
'Total curtailment with deg [GWh]',
'Awpp [km2]',
'Apvp [km2]',
'Plant area [km2]',
'Rotor diam [m]',
'Hub height [m]',
'Number of batteries used in lifetime',
'Break-even PPA price [Euro/MWh]',
'Capacity factor wind [-]',
'Execution time [min]',
] , index=range(1))
design_df.iloc[0] = list(x) + list(outs) + [(end - start)/60]
design_df = design_df.transpose()
design_df.to_csv('output.csv')