This page contains my current and previous projects in greater details and with links and files. If you have any inquiries to the projects, feel free to contact me.
The project goal was to be able to make qualitative estimates to accumulation of specific type of valuable minerals on Greenland. The data involved comes from a variety of sources such as:
I constructed a pipeline in python, that could handle the input from any data source, which would then slice the imagery into 10x10 pixel patches, that are then designated their specific geological sample ID. These patches are then loaded into another script, that prepares the data by associating the patches, the sample ID and mineral values into a singular dataframe. Currently I have tested several types of machine learning algorithms to process the data, which are Convolutional Neural Networks and Multi-Layer Perceptron. The algorithm hyper parameters are fine tuned by the use of GridSearch. Tests with pretrained models such as VGG19 also showed promising results. These thesis was presented and defended in June, 2019. The work from my thesis is currently being continued by my supervisor due to be released in a paper in the second quarter of 2020.
Master Thesis 60 ECTS - Full year
Finished - June, 2019
Machine learning - Keras, TensorFlow
Programming - Python
Advanced data visualization
Before the official courses and assembly of the Delphini-1 cubesat began, I was tasked with doing an investigative bachelor thesis in regards to gaining insights to, at that time, unknown elements of the Delphini-1 mission. The thesis contains the following elements:
The thesis quickly revealed the limitation of the satellite due to the lack of an efficient ADCS system. The satellite could be aligned to a certain degree, but cannot pinpoint the camera exactly in three dimensions. However the satellite is a pioneer project for the university, showing the capability of developing a satellite, setting up a control center and being able to retrieve simple pictures from a LEO satellite. The satellite was launched onboard a SpaceX Falcon9 rocket from Florida on the 5th of December 2018. I along with the rest of the assembly and programming team attended the launch from the NASA Causeway in Florida. The satellite was succesfully deployed from its pod on the International Space Station on the 31st of January 2019 and we made contact with a very healthy cubesat on the 5th of February 2019.
As of 2020 the satellite continues to be operational though its orbit is deteriorating. It continues to suprise with new data, and as of recently it have been able to provide information on Earth's shifting magnetic field.
Among the eruptive activity phenomena observed on the Sun, those that threaten human technology the most are flares with associated coronal mass ejections (CMEs) and solar energetic particles (SEPs). Flares with associated CMEs and SEPs are produced by magnetohydrodynamical processes in magnetically active regions (ARs) on the Sun. However, these ARs do not only produce flares with associated CMEs and SEPs, they also lead to flares and CMEs, which are not associated with any other event. In an attempt to distinguish flares with associated CMEs and SEPs from flares and CMEs, which are unassociated with any other event, we investigate the performances of two machine learning algorithms. To achieve this objective, we employ support vector machines (SVMs) and multilayer perceptrons (MLPs) using data from the Space Weather Database of Notification, Knowledge, Information of NASA Space Weather Center, the Geostationary Operational Environmental Satellite, and the Space-Weather Heliospheric and Magnetic Imager Active Region Patches. We show that True Skill Statistics (TSS) and Heidke Skill Scores (HSS) calculated for SVMs are slightly better than those from the MLPs. We also show that the forecasting time frame of 96 hr provides the best results in predicting if a flare will be associated with CMEs and SEPs (TSS = 0.92 ± 0.09 and HSS = 0.92 ± 0.08). Additionally, we obtain the maximum TSS and HSS values of 0.91 ± 0.06 for predicting that a flare will not be associated with CMEs and SEPs for the 36 hr forecast window, while the 108 hr forecast window gives the maximum TSS and HSS values for predicting that CMEs will not be accompanying any events (TSS = HSS = 0.98 ± 0.02).
The Delphini-1 cubesat was made in collaboration with Gomspace, a danish cubesat company. Two courses were created to produce a finished cubesat. The first part consisted of getting to know the details surrounding the different systems and sub systems of the satellite, as well as assembly in a cleanlab and the construction of a groundstation. I was part of the assembly team, putting the different components together, connecting them and conducting tests in the lab. After succesfully assembling of the satellite and groundstation. We conducted several systems test of housekeeping data, imaging and retrieval of the data through the groundstation antenna system.
For the software part, I constructed a satellite tracking system in python, that displays real time information on any orbiting body with a functional TLE. This includes orbital information, local weather information and image retrieval of current position from the Google earth database. The satellite was loaded onboard the International Space Station as a part of the CRS-16 mission and is was launched from the cubesat launcher system, on the 31st of January. On the 5th of February 2019, we established contact with the cubesat.
Established an online portal for primarily danish, but also for international geoscience students. The website consisted of notes, assignments, course material, PhD. papers, articles and an online shop for merchandise. All material were created by students throughout all 5 years of courses at Aarhus University. This vast amount of information were gathered and shared freely for all students, as a means to help creating better research results and making information sharing easier for everyone. At its highest point, I employed 4 students, had several hundred pages and more than 25 gb of material.
The site was shut down in 2018, due to a lack of time within the volunteers group due to approaching bachelor thesis and funding of domain and server space.
Founder, lead designer and content creator
Helping new students
Accessing research data
During the course Astrophysics, we're were tasked with creating a project for communicating astrophysics to highschool students. This lead to the creation of a website that was made to be graphically enticing, yet educational at the same time. The page is setup to provide information on planetary science and the bodies of our solar system, as well as shedding light on the field of exoplanetary exploration.
Website is still active to this day.