File size: 3,478 Bytes
97ba0dc | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 | # Imports
import pandas as pd
import matplotlib.pyplot as plt
from prophet import Prophet
import gradio as gr
# Load and preprocess data
url = 'https://data.open-power-system-data.org/time_series/2020-10-06/time_series_60min_singleindex.csv'
data = pd.read_csv(url)
df = data[['utc_timestamp', 'DE_load_actual_entsoe_transparency']].copy()
df.rename(columns={'utc_timestamp': 'ds', 'DE_load_actual_entsoe_transparency': 'y'}, inplace=True)
df.dropna(inplace=True)
df['ds'] = pd.to_datetime(df['ds']).dt.tz_localize(None)
# Forecast function with dropdown support
def forecast_energy(days, graph_type):
model = Prophet()
model.fit(df)
future = model.make_future_dataframe(periods=24 * days, freq='H')
forecast = model.predict(future)
forecast_future = forecast.tail(24 * days)
# Paths to store each plot
paths = {}
# Forecast Plot
plt.figure(figsize=(12, 5))
plt.plot(forecast_future['ds'], forecast_future['yhat'], label='Predicted Load (MW)', color='blue')
plt.fill_between(forecast_future['ds'], forecast_future['yhat_lower'], forecast_future['yhat_upper'], color='skyblue', alpha=0.3, label='Confidence Interval')
plt.xticks(rotation=45)
plt.xlabel("Date")
plt.ylabel("Load (MW)")
plt.title(f"Forecasted Load (Next {days} Days)")
plt.legend()
plt.tight_layout()
paths["Forecast Plot"] = f"forecast_plot_{days}.png"
plt.savefig(paths["Forecast Plot"])
plt.close()
# Actual vs Forecast
plt.figure(figsize=(12, 4))
plt.plot(df['ds'].tail(24 * 7), df['y'].tail(24 * 7), label='Actual Load (Past Week)', color='black')
plt.plot(forecast_future['ds'], forecast_future['yhat'], label='Forecast Load', color='green')
plt.xticks(rotation=45)
plt.title("Actual vs Forecast")
plt.xlabel("Date")
plt.ylabel("Load (MW)")
plt.legend()
plt.tight_layout()
paths["Actual vs Forecast"] = f"actual_vs_forecast_{days}.png"
plt.savefig(paths["Actual vs Forecast"])
plt.close()
# Hourly Average
df['hour'] = df['ds'].dt.hour
hourly_avg = df.groupby('hour')['y'].mean()
plt.figure(figsize=(10, 4))
plt.plot(hourly_avg.index, hourly_avg.values, marker='o', linestyle='-', color='orange')
plt.title("Average Load by Hour of Day")
plt.xlabel("Hour")
plt.ylabel("Average Load (MW)")
plt.grid(True)
paths["Hourly Pattern"] = f"hourly_pattern.png"
plt.savefig(paths["Hourly Pattern"])
plt.close()
# Seasonality
seasonality_fig = model.plot_components(forecast)
paths["Seasonality"] = f"seasonality_{days}.png"
seasonality_fig.savefig(paths["Seasonality"])
plt.close()
# Peak demand
peak_row = forecast_future.loc[forecast_future['yhat'].idxmax()]
peak_time = peak_row['ds']
peak_value = round(peak_row['yhat'], 2)
peak_info = f"🔺 Peak Demand Time: {peak_time} — {peak_value} MW"
return paths[graph_type], peak_info
# Gradio Interface
iface = gr.Interface(
fn=forecast_energy,
inputs=[
gr.Radio([3, 7, 14], label="Select Forecast Period (Days)"),
gr.Dropdown(["Forecast Plot", "Actual vs Forecast", "Hourly Pattern", "Seasonality"], label="Select Graph Type")
],
outputs=[
gr.Image(type="filepath", label="Selected Plot"),
gr.Textbox(label="Peak Demand Info")
],
title="Smart Energy Load Forecasting",
description="Choose forecast days and which graph you want to visualize."
)
iface.launch(share=True)
|