Optimal Control of Dynamic Operation in Flexible Power Plants through MPC and AI

Institut
Lehrstuhl für Energiesysteme
Typ
Bachelorarbeit / Semesterarbeit / Masterarbeit /
Inhalt
theoretisch /  
Beschreibung

Combined-cycle power plants amongst others have a generation balancing role to compensate for the uncertainties
in renewable generation, such that the reliability of supply is guaranteed. This role entails within it fast start-up and
shutdown times, and fast load gradient requirements for these units to introduce a substantial flexibility in operation.
In this work, the focus is to improve and optimize the operational flexibility of combined-cycle units through Model
Predictive Control (MPC) and Artificial Intelligence (AI). An already simulated dynamic plant model will be provided
in APROS ® along with an implemented MPC algorithm. The task is to fine tune the parameters of the MPC using
AI methodologies, in order to achieve the desired flexible behaviour during load changes. Several scenarios will be
considered and the added value of MPC and AI in terms of increasing the flexibility of dynamic operation of the unit
will be evaluated.
The work will be carried on at the Chair of Energy Systems. Knowledge in MPC and good coding skills are advantageous.

Voraussetzungen

• Knowledge of APROS® is an advantage (or comparable process simulation software).
• Background in thermal power plants,
• Background in control theory and Model Predictive Control is an advantage,
• Background in AI is required,
• Good coding skills in Matlab and/or Python,
• Strong motive and analytical skills.

Möglicher Beginn
sofort
Kontakt
Pouya Mahdavipour, M.Sc.
Raum: 3712
Tel.: 089 289 16312
pouya.mahdavipourtum.de
Ausschreibung