Projektpraktikum Energiemanagement-Technologien

Projektpraktikum
4 SWS
Module: LV - Detailansicht - TUMonline - Technische Universität München

WS 2024/25 Nr: 0000002497

Description

Participants of the course gain hands-on experience in research projects related to Energy Management.
Possible topics include but not limited to:
- EMS and distributed energy systems development
- Grid simulation
- Energy system optimizations
- Energy forecasting
- Energy Data curation and analytics
Both individual and group work is possible.
A kickoff session will be conducted to introduce the course and topics. Afterwards, students can apply for topics by sending the top three topics and the current transcript of records to e.okoyomon@tum.de.
Regular follow-up meetings with the participants and supervisor are held to aid the participants with the project tasks.
All presentations and reports in English. Advisor↔participant communication language may vary.

Deliverables, important dates, and current topics listed below.

Deliverables

At the end of the semester, participants will have to submit:

a scientific paper (~7 pages per student, submitted as the final report).
a conference-style final presentation, including demo, or proof of concept.

Both of these must be submitted on time and any late work may not be accepted.

Important Dates (Summer Semester 2024)

	Date	Time	Place
Registration Period	30.08.24 - 25.10.24		TUMOnline
Kickoff Meeting	16.10.24	14:00 - 15:00	room N3815
Topic preferences submission	18.10.24.	23:59	email
Topic allocation	23.10.24	19:00	email
Final Report and presentation due	09.02.25	23:59	moodle
Final Presentation	10.02.25	14:00 - 17:00	room N3815

Topics

If you have any more questions about a topic, please contact the advisor/contact directly. For topic preference submissions, contact e.okoyomon@tum.de.

Acronym	Topic description	Required skills	Number of Students	Contact	Time Topic Added
GNN_PF_COMPARE	Implementation and Comparison of SOTA Graph-Models for Power Flow Analysis Research method: Prototyping Research questions: How can we use graph-based data models to do power flow in distribution grids? How do the different SOTA techniques compare to one another? Possible approach: Gather/create training data of various distribution grids (varying in characteristic, perturbation, topology) Read, understand, and implement the models from: A Physics-Informed Graph Attention-based Approach for Power Flow Analysis Solving AC Power Flow with Graph Neural Networks under Realistic Constraints Run Experiments: Model tested on the same grid it was trained on Model tested on perturbation of training grid Model tested on completely unseen grids Compare and interpret results Resources: Power Flow Balancing With Decentralized Graph Neural Networks Reference Paper with published data and code... did something quite similar Graph Neural Networks Stanford Course - Machine Learning with Graphs - Youtube Playlist https://www.youtube.com/playlist?list=PLoROMvodv4rPLKxIpqhjhPgdQy7imNkDn Lecture 6.1-6.3 particularly for GNN overview Stanford Course - Machine Learning with Graphs - Website CS224W \| Home (stanford.edu) Winter 2021 fits the best with the youtube playlist.. can see "Previous Offerings" section PyTorch Geometric Introduction by Example Graph Neural Networks (GNN) using Pytorch Geometric \| Stanford University	Good programming skills in Python Experience / Interest in Machine Learning Interest / Background in Power Systems	2-4 students	Ehimare Okoyomon e.okoyomon@tum.de	10/2024
GRID_INT	Development of Grid Simulation Software Compatibilitiy Interface Research method: Prototyping Research questions: How can input data and results for powerflow or transient simulation be exchanged between different simulation tools? How can we validate the performance of simulation tools based on AC power flow? How can we combine the grid databases of all connected tools and design an interface allowing us to use the combined grid databases in all tools? Possible approach: Setup sample grids in common tools Develop different interaction interfaces between simulation tools Validate the power flows of the transmitted networks so that we can see the transfer worked Combine all interfaces and the databases behind to one Python Library, accessing a common sample grids database for all tools and allowing the setup of all grids on all simulation environments. Resources: Pandapower https://arxiv.org/abs/1709.06743 Andes https://ieeexplore.ieee.org/document/9169830 PowerModelsDistribution.jl https://github.com/lanl-ansi/PowerModelsDistribution.jl OpenDSS https://opendss.epri.com/opendss_documentation.html PowerSystems.jl https://github.com/NREL-Sienna/PowerSystems.jl	Good programming skills in Python and/or Julia Interest in power system modeling	3-4	Sebastian Eichhorn sebastian.eichhorn@tum.de	06/2023
ELECROTHERMAL_DEMAND	Containerized Implementation of Integrated Electrothermal Building Energy Demand Model Research method: Prototyping Research questions: How can electricity and heat demand be concurrently simulated? How can the simulation be packaged into a Docker container and controlled via APIs? Possible approach: Understand state-of-the-art models based on literature Develop efficient Python or Julia code implementing one of the methods Package code in a Docker container and evaluate the implementation via EMS use cases Resources: McKenna, Eoghan; Thomson, Murray (2016): High-resolution stochastic integrated thermal–electrical domestic demand model. In: Applied Energy 165, pp. 445–461. DOI: 10.1016/j.apenergy.2015.12.089.	Good programming skills in Python and/or Julia Interest in energy system modeling	2-3	Christoph Goebel christoph.goebel@tum.de Elgin Kollnig elgin.kollnig@tum.de	10/2024
DIST_LEM_IMPL	Distributed Implementation of Local Electricity Market Research method: Prototyping Research questions: How local electricity markets be simulated? How can such simulations be implemented in a distributed way? Possible approach: Understand LEM mechanisms based on literature Develop efficient Python or Julia code implementing a basic LEM mechanism with participating prosumers using Docker for containerization Use a cluster of raspberry pi minicomputers to simulate the mechanism in a distributed way Create demonstration scenarios, e.g., using OpenTUMFlex Evaluate implementation based on economic (e.g., prosumer cost reduction, social welfare) and technical metrics (e.g., computation time, communication overhead) Resources: Lemlab https://github.com/tum-ewk/lemlab OpenTUMFlex https://github.com/tum-ewk/OpenTUMFlex Bjarghov, Sigurd; Loschenbrand, Markus; Ibn Saif, A. U. N.; Alonso Pedrero, Raquel; Pfeiffer, Christian; Khadem, Shafiuzzaman K. et al. (2021): Developments and Challenges in Local Electricity Markets: A Comprehensive Review. In: IEEE Access 9, S. 58910–58943. DOI: 10.1109/ACCESS.2021.3071830.	Good programming skills in Python and/or Julia Interest in energy system modeling	3-4	Christoph Goebel christoph.goebel@tum.de Ehimare Okoyomon e.okoyomon@tum.de

Seitenhierarchie

Projektpraktikum Energiemanagement-Technologien

Description

Deliverables

Important Dates (Summer Semester 2024)

Topics