The increasing complexity of renewable energy systems characterized by multiple energy carriers and local intermittent resources, calls for accurate tools for effective design, operation, and planning. This thesis investigates simulation-based optimization as a tool to support such decision-making processes.

To build a foundation for the proposed method, a background study was conducted on optimization theory in general and on simulation-based optimization with a primary focus on energy systems. Additionally, the functionality of the simulation software used in this thesis, The Illuminator, was explored.

Based on this foundation, a new optimization framework was developed by extending The Illuminator software and through the integration of three algorithms: Particle Swarm Optimization (PSO), Genetic Algorithm (GA), and a gradient-based algorithm (L-BFGS-B). Parallelization was implemented to increase the efficiency of the algorithms. To expand the modeling capability of The Illuminator, several new hydrogen-related component models were developed. The framework was tested across multiple domains by using three distinct scenarios: (1) a hydrogen production facility (hydrogen domain, continuous variables, system design domain), (2) a residential energy hub (electric domain, continuous variables, system operation domain), and (3) an electric vehicle charging station (electric domain, discrete variables, system planning domain).

Among the explored algorithms, Particle Swarm Optimization (PSO) proved to be the most suitable across the three presented scenarios, achieving the lowest average gaps to the best-found solutions in each case (0.107%, 0.363%, and 20.145%, respectively). Parallelization of the population-based algorithms improved the total run time by a factor of almost 5.

The results show that simulation-based optimization is a promising approach for supporting the design, operation, and planning of complex renewable energy systems.

This thesis describes the design and implementation of a LED Driving and Sensing Unit, one of the parts of a UV-C LED sterilization machine for the disinfection of plant seeds. The purpose of this part is to drive LEDs which can destroy different structures of pathogens. While doing this, certain parameters like temperature and ozone generation are being measured.
The report covers many design choices such as the choice to put the LEDs in arrays that are a combination of series and parallel circuits, or how SPI addressable variable resistors were used, in order to have full control over the LEDs and the sensing part. Furthermore, through a MATLAB script, it was discovered how the arrays of LEDs must be placed in specific manner to obtain a uniform radiation pattern. From this, it follows that using two arrays with radii of respectively 17 and 41 mm would work best. Furthermore, a literature research was done on UV-C and on what wavelength is theoretically the best for destroying pathogens on seeds. The focus was set on wavelengths of 255 nm and a combination of 275 nm and 285 nm. Also a 395 nm (UV-A) LED was used to investigate whether pathogens can be awakened from a hibernation state with the help of this type of light.
A boost converter capable of controlling the intensity of the LEDs by changing its gain with the aforementioned variable resistors was designed. Finally, the entire unit was tested successfully. The result of the research is a fully working LED driver module that can sense all parameters that need to be measured, completed with the design of a 3D CAD casing, which has all been manufactured and fabricated. This thesis does not include test results of the disinfection of plant seeds. This information will be included in the supplementary seed testing results report.