Time series forecasting refers to the prediction of future values based on historical, time-ordered data. Patterns such as trends, seasonal effects, and recurring fluctuations observed in the past are used to forecast future developments.

This tutorial demonstrates how machine learning can be applied to predict values over time. Using electricity prices as an example, it shows how forecasts can be translated into concrete optimization decisions. The optimization process is realistically simulated over several days, and the results are systematically evaluated and analyzed.

The tutorial is aimed at developers, data analysts, and technically interested users. It is suitable for anyone who wants to understand how artificial intelligence can be used for process optimization. Basic knowledge of Python is helpful, but no domain-specific knowledge of the energy sector is required.

Contents

• Introduction to time series forecasting
• Application of the presented approaches using a practical example of electricity price forecasting

Benefits for Participants

• You will gain a clear understanding of how AI can be used to optimize complex processes
• You will be guided through an exemplary workflow from data and forecasts to automated decision-making
• The concepts can be transferred to many other use cases, as the focus is on methodology rather than the energy market itself

With our AI tutorials, we aim to make artificial intelligence tangible and accessible. Depending on prior knowledge, a tutorial can take up to one day to complete. The online format allows you to manage your time flexibly according to your needs. Detailed explanations guide you step by step through the tutorial and its data. In addition, you will receive instructions on which free software tools to install in order to complete the tutorial.

AI Tutorial  
Time series forecasting

► Start the tutorial now

This e-learning is freely accessible to everyone. No registration or sign-up is required. Like all our AI tutorials, it is available free of charge.