Add tutorial scripts

Author: phumthep

scripts/run_quickstart.py

@@ -0,0 +1,49 @@
+""" This script provides an example of how to run PowNet 2.0.
+"""
+
+import os
+from pownet.core import Simulator
+
+
+def main():
+
+    # --------- User inputs
+
+    input_folder = "..//model_library"
+    output_folder = "..//temptemp"
+
+    model_name = "dummy"
+    model_year = 2016
+
+    # Simulation parameters
+    sim_horizon = 24
+    steps_to_run = 2
+    solver = "gurobi"  # or highs
+
+    # --------- End of user inputs
+
+    # Run the simulation
+    simulator = Simulator(
+        input_folder=input_folder,
+        model_name=model_name,
+        model_year=model_year,
+    )
+
+    simulator.run(
+        sim_horizon=sim_horizon,
+        steps_to_run=steps_to_run,
+        solver=solver,
+    )
+
+    # Write the simulation results
+    if not os.path.exists(output_folder):
+        os.makedirs(output_folder)
+    simulator.write_results(output_folder)
+
+    # Plot the results
+    simulator.plot_fuelmix("bar", output_folder)
+    simulator.plot_unit_status(output_folder)
+
+
+if __name__ == "__main__":
+    main()

scripts/test_arima.py

@@ -1,6 +1,4 @@
-import os
 from pownet.stochastic.solar import SolarTSModel
-from pownet.stochastic.demand import DemandTSModel
 import pandas as pd
 import matplotlib.pyplot as plt
 
@@ -9,62 +7,25 @@
 # Show info
 logging.basicConfig(level=logging.INFO)
 
-# %% Demand
-
-data = pd.read_csv("../temp/hourly_demand_2023.csv")
-
-demand_model = DemandTSModel()
-demand_model.load_data(data)
-exog_vars = ["temp", "rhum", "prcp", "weekend"]
-
-# arima_order, seasonal_order = demand_model.find_best_model(
-#     target_column="demand",
-#     exog_vars=exog_vars,
-# )
+# %% Solar
+data = pd.read_csv("../temp/merra_2019.csv")
 
+solar_model = SolarTSModel()
+solar_model.load_data(data)
 
-demand_model.fit(
-    target_column="demand",
-    exog_vars=exog_vars,
-    arima_order=(1, 0, 1),
-    seasonal_order=(0, 0, 0, 0),
-)
-predictions = demand_model.predict()
+solar_model.fit(target_column="ground_irradiance", arima_order=(2, 1, 2))
+predictions = solar_model.predict()
 
-data.index = pd.to_datetime(data["datetime"])
-exog_data = data[exog_vars].astype(float)
-synthetic = demand_model.get_synthetic(exog_data=exog_data)
+resids = solar_model.pred_residuals
 
+synthetic = solar_model.get_synthetic()
 
-duration = 10
+duration = 30
 # plt.plot(predictions[: 24 * 3], label="Predictions")
 plt.plot(synthetic[: 24 * duration], label="Synthetic")
 plt.plot(
-    data.set_index("datetime")["demand"].iloc[: 24 * duration],
+    data.set_index("datetime")["ground_irradiance"].iloc[: 24 * duration],
     label="Actual",
 )
 plt.legend()
 plt.show()
-
-"""# %% Solar
-data = pd.read_csv("../temp/merra_2019.csv")
-
-solar_model = SolarTSModel()
-solar_model.load_data(data)
-
-solar_model.fit(target_column="ground_irradiance", arima_order=(2, 1, 2))
-predictions = solar_model.predict()
-
-resids = solar_model.pred_residuals"""
-
-# synthetic = solar_model.get_synthetic()
-
-# duration = 30
-# # plt.plot(predictions[: 24 * 3], label="Predictions")
-# plt.plot(synthetic[: 24 * duration], label="Synthetic")
-# plt.plot(
-#     data.set_index("datetime")["ground_irradiance"].iloc[: 24 * duration],
-#     label="Actual",
-# )
-# plt.legend()
-# plt.show()