Features/csp validation

docs/validation_concentrating_solar_power.rst

@@ -0,0 +1,64 @@
+.. _validation_concentrating_solar_power_label:
+
+Concentrating solar power
+===================================
+
+Scope
+_____
+
+To validate the component 'concentrating_solar_power' irradiance data from the
+location of the Andasol 3 power station for 2019 is used. The direct normal 
+irradiance and the
+horizontal direct irradiance are given. So the two irradiance methods 'normal'
+and 'horizontal' of the component were analysed. The collector is defined by
+parameters, which correspond to the loss method 'Andasol'. Thus, only this
+loss method is analysed. The measured heat output of the collector is
+compared to the calculated heat output.
+
+Method
+_______
+
+The component 'concentrating solar power' is used to calculate the collector heat
+for one year. Therefore the parameters for the collector and the location are
+used. Both irradiance methods result in similar computed heat output. 
+The comparison with measured data is done with data computed for the 'horizontal'
+method. The script of this calculations can be found in the examples of the
+concentrating solar power within the oemof_heat repository.
+The parameters required to model the collector can also be found within this example.  
+
+The script is based on hourly averaged data resulting in 8760 data sets. Due to 
+licencing requirements, the data had to be aggregated to daily summed values.
+This aggregation was done separately and is not performed with the scripts. 
+
+Results
+_______
+Thus, the following diagram shows a comparison of the measured daily collector 
+heat output and the computed heat output for the 365 days of 2019. 
+
+.. 	figure:: _pics/csp_validation.png
+   :width: 70 %
+   :alt: csp_validation.png
+   :align: center
+
+It can be concluded that there is an overall satisfactory agreement between 
+measured and computed data, resulting in a 7% deviation for the annual sum. 
+During time periods with low output measured data is higher that computed values.
+During time periods with high output the opposite has to be noted.
+
+Due to project time restrictions it was not possible to analyse the reasons 
+for the deviations in detail. Although inaccuracies in the mathematical model 
+cannot be ruled out, it has to be noted that operational characteristics have a 
+strong impact on the measured data. For example, there was a maintenance period 
+around day 320 resulting in low measerd thermal output. Also, in periods with 
+high irradiance, it may be possible that the full collector yield cannot be 
+absorbed by the power block and its storage. In those time periods the collector
+yield had to be curtailed by moving the concentrators out of the focal point.
+
+Furthermore, the thermal inertia of the entire collector field is not represented 
+by the computation. This cannot be seen in the daily sums of heat output. But it 
+should be expected that computed heat output starts and ends earlier than measured
+output within the course of a day.  
+
+
+
+

examples/concentrating_solar_power/validation.py

@@ -0,0 +1,165 @@
+# -*- coding: utf-8 -*-
+
+"""
+Date: 17.09.2020
+Author: Franziska Pleissner
+
+This file shows a comparison between the collector heat, which is calculated
+with the oemof-thermal component and the collector heat, which is measured.
+It uses the irradiance data, which are measured at the same spot and the same
+time as the collector heat data.
+"""
+
+############
+# Preamble #
+############
+
+# Import packages
+from oemof.thermal.concentrating_solar_power import csp_precalc
+
+import os
+import pandas as pd
+import matplotlib.pyplot as plt
+from datetime import datetime
+
+
+# Define the used directories
+abs_path = os.path.dirname(os.path.abspath(os.path.join(__file__)))
+results_path = abs_path + '/results/'
+data_path = abs_path + '/data/validation_data/'
+
+# Read parameter values from parameter file
+file_path_param = data_path + 'validation_parameters.csv'
+param_df = pd.read_csv(file_path_param, index_col=1)
+param_value = param_df['value']
+
+# Define some needed parameters
+currentdate = datetime.today().strftime('%Y%m%d')
+number_of_time_steps = int(param_value['number_timesteps'])
+
+# Import weather and demand data
+dataframe = pd.read_csv(
+    data_path + 'validation_timeseries.csv').head(number_of_time_steps)
+dataframe['Date'] = pd.to_datetime(dataframe['Date'])
+dataframe.set_index('Date', inplace=True)
+dataframe = dataframe.asfreq('H')
+
+# Calculate collector data
+collector_precalc_data_normal = csp_precalc(
+    param_value['latitude'],
+    param_value['longitude'],
+    param_value['collector_tilt'],
+    param_value['collector_azimuth'],
+    param_value['cleanliness'],
+    param_value['eta_0'],
+    param_value['c_1'],
+    param_value['c_2'],
+    param_value['inlet_temp_mean'],
+    param_value['outlet_temp_mean'],
+    dataframe['Ambient_temperature_in_degC'],
+    param_value['a_1'],
+    param_value['a_2'],
+    param_value['a_3'],
+    param_value['a_4'],
+    param_value['a_5'],
+    param_value['a_6'],
+    loss_method='Andasol',
+    irradiance_method='normal',
+    dni=dataframe['dni_in_W_per_m2'])
+
+collector_precalc_data_horizontal = csp_precalc(
+    param_value['latitude'],
+    param_value['longitude'],
+    param_value['collector_tilt'],
+    param_value['collector_azimuth'],
+    param_value['cleanliness'],
+    param_value['eta_0'],
+    param_value['c_1'],
+    param_value['c_2'],
+    param_value['inlet_temp_mean'],
+    param_value['outlet_temp_mean'],
+    dataframe['Ambient_temperature_in_degC'],
+    param_value['a_1'],
+    param_value['a_2'],
+    param_value['a_3'],
+    param_value['a_4'],
+    param_value['a_5'],
+    param_value['a_6'],
+    loss_method='Andasol',
+    E_dir_hor=dataframe['E_dir_hor_in_W_per_m2'])
+
+collector_precalc_data_normal.to_csv(
+    results_path + 'validation_precalc_data_normal_{0}.csv'.format(currentdate))
+
+collector_precalc_data_horizontal.to_csv(
+    results_path + 'validation_precalc_data_horizontal_{0}.csv'.format(
+        currentdate))
+
+result_df = pd.DataFrame()
+result_df['method normal'] = collector_precalc_data_normal[
+    'collector_heat']
+result_df['method horizontal'] = collector_precalc_data_horizontal[
+    'collector_heat']
+result_df['Andasol data'] = dataframe[
+    'collector_field_heat_output_in_W_per_m2']
+new_row = pd.Series(
+    data={'method normal': result_df['method normal'].sum(),
+          'method horizontal': result_df['method horizontal'].sum(),
+          'Andasol data': result_df['Andasol data'].sum()},
+    name='Sums')
+result_df_csv = result_df.append(new_row, ignore_index=False)
+result_df_csv.to_csv(results_path + 'validation.csv')
+
+results_summer_1 = result_df[4080:4248]
+results_summer_2 = result_df[4248:4416]
+results_winter = result_df[672:840]
+
+result_df.plot(kind='line', linewidth=2)
+plt.title('specific collector heat output', size=25)
+plt.ylabel('collector heat output [W/m2]', size=25)
+plt.legend(bbox_to_anchor=(1.02, 1), loc='upper left', borderaxespad=0,
+           prop={'size': 25})
+plt.yticks(size=25)
+figure = plt.gcf()
+figure.set_size_inches(30, 15)
+plt.savefig(results_path + 'validation_results_{0}.png'.format(currentdate),
+            dpi=150, bbox_inches='tight')
+
+results_summer_1.plot(kind='line', linewidth=3)
+plt.title('specific collector heat output (20.06-26.06.2019)', size=25)
+plt.ylabel('collector heat output [W/m2]', size=25)
+plt.legend(bbox_to_anchor=(1.02, 1), loc='upper left', borderaxespad=0,
+           prop={'size': 25})
+plt.yticks(size=25)
+figure = plt.gcf()
+figure.set_size_inches(30, 15)
+plt.savefig(
+    results_path + 'validation_results_summer_1_{0}.png'.format(currentdate),
+    dpi=150,
+    bbox_inches='tight')
+
+results_summer_2.plot(kind='line', linewidth=3)
+plt.title('specific collector heat output (27.06-03.07.2019)', size=25)
+plt.ylabel('collector heat output [W/m2]', size=25)
+plt.legend(bbox_to_anchor=(1.02, 1), loc='upper left', borderaxespad=0,
+           prop={'size': 25})
+plt.yticks(size=25)
+figure = plt.gcf()
+figure.set_size_inches(30, 15)
+plt.savefig(
+    results_path + 'validation_results_summer_2_{0}.png'.format(currentdate),
+    dpi=150,
+    bbox_inches='tight')
+
+results_winter.plot(kind='line', linewidth=3)
+plt.title('specific collector heat output (29.01-04.02.2019)', size=25)
+plt.ylabel('collector heat output [W/m2]', size=25)
+plt.legend(bbox_to_anchor=(1.02, 1), loc='upper left', borderaxespad=0,
+           prop={'size': 25})
+plt.yticks(size=25)
+figure = plt.gcf()
+figure.set_size_inches(30, 15)
+plt.savefig(
+    results_path + 'validation_results_winter_{0}.png'.format(currentdate),
+    dpi=150,
+    bbox_inches='tight')