materials for dc workshop

dc.md

@@ -0,0 +1,47 @@
+# DC data sharing
+
+## Creating Sites, Observations, Forecasts
+
+* Create a Site (weather station)
+  * Boulder: 40 N, -105 E, 1650 m
+* Create a Site (power plant)
+  * Boulder Power Plant: 10 MW AC, 14 MW DC, tracking, .4 gcr
+* Create an Observation.
+  * AC power, hourly average, beginning
+* View sample data, upload boulder_cloudy_aug_sept.csv
+  * Note: procedure must be repeated for each variable. Tedious, but avoids need to agree upon and parse column names
+* Create a Forecast
+  * AC power, hourly average, beginning
+* upload boulder_fx_demo_aug_sept.csv
+
+## Data sharing
+
+For this exercise you need two accounts. Here we use demo and holmgren.
+For now we must use uuids to share data, but email support will come soon.
+
+* demo: 4f738239-e632-11e9-ab45-52540015d5ce
+* holmgren: 26653f79-e62d-11e9-ab45-52540015d5ce
+
+* Goal: demo@solarforecastarbiter.org wants to share its AC power data with a forecast provider (holmgren@email.arizona.edu), receive a forecast from the forecast provider, and generate a report comparing to its own forecast (created previously).
+* Show roles page
+* In a second incognito window, log in as holmgren@email.arizona.edu, show available sites -- cannot see demo’s
+* From demo window, create role for sharing sites, obs metadata, values
+* Use “all *” permissions for simplicity, but note this can be made much more fine grained
+* Assign role to forecast provider account
+* From holmgren window, create forecast, upload forecast
+  * AC power, hourly average, beginning
+  * upload boulder_fx_holmgren_aug_sept.csv
+* Note that demo could have created a forecast metadata object
+* Share new holmgren forecast metadata and data with the demo account
+* From demo account, see that there is now a new forecast
+* From demo create report comparing demo forecast and holmgren forecast to demo obs
+
+## Data sharing on MIDC Arizona OASIS
+
+* From holmgren window, create OASIS forecast, upload forecast
+  * issue time of day 7Z, run length 1 day, lead time 1 day
+* Premade uploaded forecast is a bias corrected version of GFS forecast
+* Note that demo could have created a forecast metadata object
+* If not already done, share new holmgren forecast metadata and data with the demo account
+* From demo account, see that there is now a new forecast
+* Create report comparing reference forecast and holmgren forecast to reference obs

generate_ac_obs.py

@@ -0,0 +1,56 @@
+import numpy as np
+import pandas as pd
+
+from solarforecastarbiter import datamodel
+from solarforecastarbiter import pvmodel
+
+from pvlib.location import Location
+
+
+# define metadata
+modeling_params = datamodel.SingleAxisModelingParameters(
+    ac_capacity=10, dc_capacity=14, temperature_coefficient=-0.004,
+    dc_loss_factor=0, ac_loss_factor=0,
+    axis_tilt=0, axis_azimuth=0, ground_coverage_ratio=.4,
+    backtrack=True, max_rotation_angle=45)
+
+metadata = datamodel.SolarPowerPlant(
+    name='Boulder Power Plant', latitude=40., longitude=-105.,
+    elevation=1650., timezone='America/Denver', provider='Sandia',
+    modeling_parameters=modeling_params)
+
+times = pd.date_range(
+    start='20190801', end='20191001', freq='5min', closed='left')
+
+solpos = pvmodel.calculate_solar_position(
+    metadata.latitude, metadata.longitude, metadata.elevation, times)
+cs = pvmodel.calculate_clearsky(
+    metadata.latitude, metadata.longitude, metadata.elevation,
+    solpos['apparent_zenith'])
+ac_clear = pvmodel.irradiance_to_power(
+    modeling_params, solpos['apparent_zenith'], solpos['azimuth'],
+    cs['ghi'], cs['dni'], cs['dhi'])
+
+ac_clear_1h = ac_clear.resample('1h').mean()
+ac_clear_1h = pd.DataFrame({'value': ac_clear_1h, 'quality_flag': 0})
+ac_clear_1h.to_csv('boulder_clear_aug_sept.csv', index_label='timestamp')
+
+# make it cloudy, but apply clipping
+random = np.random.random(size=len(times)) + 0.5
+random = random.clip(max=1)
+ac = ac_clear * random
+
+ac_1h = ac.resample('1h').mean()
+ac_1h = pd.DataFrame({'value': ac_1h, 'quality_flag': 0})
+ac_1h.to_csv('boulder_cloudy_aug_sept.csv', index_label='timestamp')
+
+fx_demo = ac_1h['value'].shift(1)
+fx_demo.to_csv('boulder_fx_demo_aug_sept.csv',
+               header=['value'], index_label='timestamp')
+
+csi = ac / ac_clear
+csi = csi.fillna(0)  # account for divide by 0
+ac_csi = csi.shift(12) * ac_clear
+holmgren_demo = ac_csi.resample('1h').mean()
+holmgren_demo.to_csv('boulder_fx_holmgren_aug_sept.csv',
+                     header=['value'], index_label='timestamp')