Master's thesis; AI for the optimization of energy systems

Background
The renewable energy sector is increasingly shifting its focus from hardware efficiency to IT systems that optimize energy production, storage, and consumption. For small and medium-sized industries such as farms, this shift presents a challenge: while technologies like solar panels, wind turbines, and batteries are widely deployed, most sites lack intelligent control systems to manage these resources effectively. Existing integrated energy management solutions are often tailored for residential use and do not address the complexity of industrial operations, leaving producers dependent on rigid market structures and limiting the economic benefits of local generation.

RISE and its industrial partners are developing integrated energy management and optimization tools that use AI-based algorithms to manage energy flows at individual sites and, where relevant, across sites. These systems predict energy production and consumption and optimize decisions on electricity purchases and sales, battery charging and discharging, and the control of loads such as heat pumps. The goal is to balance production and consumption, reduce costs, and improve resilience. By integrating both legacy and modern components, this approach enables smarter decision-making without requiring major infrastructure changes.

Description
In your thesis, you will explore and compare optimization techniques such as Mixed Integer Linear Programming (MILP) and Genetic Algorithms (GA) to optimize control of local energy systems; so called energy-communities with a mixture of energy production (solar panels, wind turbines, generators), storage (batteries) and consumption (heat pumps, electrical devices, etc.).

One of the core challenges is to address competing targets: In real world scenarios, the minimization of energy cost is rarely the only target; for instance, temperatures in certain rooms or buildings should be kept in specific intervals, battery degradation needs to be minimized, and up-time of critical components maximized.

On the input side, these systems are fed with predictions for future energy production, energy prices, and the highly use-case specific consumption of energy. The output is a control matrix, determining the control decisions for the integrated components: target temperatures, grid energy purchases, run plans for electrical components, etc.

The central task is to compare the viability of various approaches to the optimization problem on various metrics, such as performance w.r.t. optimization goals, computational costs, and ability to represent relevant constraints in the problem formulation.

We welcome your application to join our team of researchers in the ongoing project "Advancing Industrial Data and Process Integration Infrastructure for Smart Energy Grids (AIPI)" between RISE and industry partners.

Key Responsibilities

• Data handling and processing, to select and prepare the data for analysis
• Literature review on MILPs and GAs for use in energy system optimization
• Implementation and comparison of MILP and GA systems
• Writing and defending master thesis
• Recurrent presentations of project progress

Qualifications
Required skills:

• Good foundational knowledge of Machine Learning, AI & Computer Science techniques
• Good programming skills in Python
• Experience in implementing Machine Learning and related models

Preferred skills:

• Experience with MILP
• Experience with Genetic Algorithms

Terms

• Location: Gothenburg (preferred), Sweden (remote work)
• Time: January to June 2026
• Credits: 30 ECTS
• Compensation: For an approved thesis project worth 30 credits, RISE will pay a compensation of 30,000 SEK.

What We Offer

• Supervision by an experienced team of researchers
• An opportunity to do highly relevant research in an ongoing collaborative project between partners from the energy industry and RISE
• Potential for publication of thesis results in form of a research paper
• A collaborative and stimulating research environment

Supervisors

• Leon René Sütfeld, PhD (RISE)
• Michael Popoff, PhD (RISE)

Welcome with your application
Last day of application; Nov 30, 2025

Contact;