Search⌘ K
AI Features
Home
Courses
Time Series Analysis with Python

Energy Consumption

Explore how to analyze and forecast energy consumption data using Python. Understand data preprocessing, identify seasonal patterns, and compare SARIMA and Holt-Winters models to make 12-month forecasts. Gain skills to handle real-world utility data effectively.

We'll cover the following...

Context

We'll now analyze energy consumption data from an electric utility from the United States called American Electric Power (AEP). They are present in eleven states and deliver electricity to more than five million customers.

We'll now try to find a good model to forecast energy consumption for the next twelve months.

Let's start by taking a look at the data to gather some basic information.

Python 3.8
import pandas as pd
import matplotlib.pyplot as plt
df = pd.read_csv('AEP.csv')
# Changing the datatype
df["Date"] = pd.to_datetime(df['Date'], format='%Y-%m-%d')
 
# Looking at the dataframe
print(df.head())
print(df.tail())
# Setting the Date as index
df = df.set_index('Date')
# Plotting
fig, ax = plt.subplots(figsize=(7, 4.5), dpi=300)
ax.plot(df)
# Labelling 
plt.xlabel("Date")
plt.ylabel("Megawatts")
plt.title("Energy consumption")
fig.savefig("output/output.png")
plt.close(fig)

First, let's understand the key components of the code.

  • Line 7: We convert our date column to datetime format.

  • Lines 10–11: We print the first five and the last five rows to have an overview of the data.

  • Line 14 ...