Examensarbeten för masterexamen

Browse

Senast publicerade

Visar 1 - 5 av 651
  • Post
    Quantum Routing using Value-Based Reinforcement Learning
    (2023) OPPERUD, MIKKEL; Chalmers tekniska högskola / Institutionen för fysik; Chalmers University of Technology / Department of Physics; Granath, Mats
    This thesis addresses the Quantum routing problem through the implementation of a reinforcement learning algorithm. Quantum routing is the problem of making quantum circuits executable on a quantum computer with limits of connectivity which requires requires swapping information between qubits. A value-based variant of the Q-learning algorithm, coupled with deep convolutional neural networks, was employed to optimize the routing process in a grid topology environment. The environment allowed the agent to place and remove swaps and to "pull back" any immediately executable qubits. The reward scheme was designed to optimize for a shortened circuit depth with the first layers of swaps not counted, thus solving the Quantum routing and placement problem concurrently. The study focused on smaller grid sizes of 3x2, 3x3, and 3x4. Due to time constraints we were not fully able to adequately access the performance of the model and were only able to achieve solutions for smaller models, while the results for the larger ones (3x3 and 4x3) were lackluster. For larger grid sizes our analysis on multiple hyper-parameters revealed a better understanding for the reasons for this, suggesting possible remedies. In conclusion, while the algorithm encountered issues during the experiment, these obstacles present opportunities for future improvement and refinement. This research provides a foundation for future studies in the realm of Quantum routing, highlighting potential avenues for enhanced algorithm performance.
  • Post
    Masked Prediction of Time Series data Using Novel Machine Learning Models
    (2024) Gopalakrishnan, Dinesh Krishnan; Chalmers tekniska högskola / Institutionen för fysik; Chalmers University of Technology / Department of Physics; Granath, Mats; Ganesan, Deepak Guru
    We are living in a time where data is the new oil, where the industries are data driven. This change was expedited by the enormous amount of data that we are producing every second and the increase in computational power. At present the automotive sector is thriving in this data-driven model by their ubiquitous need for common and industrial purposes and the various data they are collecting to improve the sector as a whole. We can now see an automobile as not just a mechanical prod uct but as a robot on wheels. Along with this data-driven model, the electrification of automobiles has revolutionized the industry. In the electrification process, the battery module is one of the key components that power the systems. To do this we must analyse the data from the battery modules for its efficient usage. However, due to certain hardware issues or if the vehicle is out of range and it cannot update the data we might lose data. This loss of data can obstruct the efficient usage of the data in machine learning models to optimize the system. There are several methods to impute missing data, for example, there are statisti cal methods such as the Auto-regressive methods, which are limited by their time and the high cost of their computations. This thesis focuses on this problem and designing a neural network model for masked prediction of the Time series data. In this thesis, a Transformer Network is implemented for the masked prediction of the missing time series data. In this thesis, we have built the machine learning model from scratch after weighing several factors. The data on which the model is trained is generated by the vehicle collected. This was led by pre-processing, later following the selection of the model. The model developed here is a variation of the transformer model, called the Time Series Transformer(TST), which predicts the missing values in the time series data. This model is then evaluated with suitable metrics by the model and the problem statement. The thesis aims to predict the missing values to improve the quality of the data collected and its quality usage to improve the performance of the vehicle.
  • Post
    Quantum Routing using Value-Based Reinforcement Learning
    (2023) Opperud, Mikkel; Chalmers tekniska högskola / Institutionen för fysik; Chalmers University of Technology / Department of Physics; Granath, Mats
    This thesis addresses the Quantum routing problem through the implementation of a reinforcement learning algorithm. Quantum routing is the problem of making quantum circuits executable on a quantum computer with limits of connectivity which requires requires swapping information between qubits. A value-based vari ant of the Q-learning algorithm, coupled with deep convolutional neural networks, was employed to optimize the routing process in a grid topology environment. The environment allowed the agent to place and remove swaps and to "pull back" any immediately executable qubits. The reward scheme was designed to optimize for a shortened circuit depth with the first layers of swaps not counted, thus solving the Quantum routing and placement problem concurrently. The study focused on smaller grid sizes of 3x2, 3x3, and 3x4. Due to time constraints we were not fully able to adequately access the performance of the model and were only able to achieve solutions for smaller models, while the results for the larger ones (3x3 and 4x3) were lackluster. For larger grid sizes our analysis on multiple hyper-parameters revealed a better understanding for the reasons for this, suggesting possible reme dies. In conclusion, while the algorithm encountered issues during the experiment, these obstacles present opportunities for future improvement and refinement. This research provides a foundation for future studies in the realm of Quantum routing, highlighting potential avenues for enhanced algorithm performance.
  • Post
    Evaluation of cooling system in an industrial fuel cell setup by effectively managing exhaust water
    (2024) Bandara, Janitha; Karthikeyan, Umashankar; Chalmers tekniska högskola / Institutionen för fysik; Chalmers University of Technology / Department of Physics; Wickman, Björn; Tholandar, Fredrick
    Fuel cells are an emerging portable energy source known for their high energy density and ability to produce clean, sustainable energy. Unlike traditional combustion processes that directly burn fuels to release energy, fuel cells harness energy from the reaction between the fuel and the oxidizer while producing minimal pollutants and greenhouse gases. When it comes to the industrial scale, a more significant drawback is managing the exhaust water. One primary application is the capture and reuse of exhaust water for cooling purposes within the fuel cell system. By utilizing the waste heat generated during the electrochemical reaction, the exhaust water can serve as a cooling agent, reducing the need for external cooling systems and enhancing the overall energy efficiency of the fuel cell. This study aims to investigate various aspects of fuel cell systems, such as quality analysis of exhaust water samples from Volvo Penta fuel cell setup, a literature survey on various applications that can be used for exhaust water, and developing 1D and 3D models for two of the identified applications. The first application is to vaporize the water using a chimney/muffler, which is most suitable for mobile applications such as trucks and marine applications. Chimney size is optimized through the Matlab Simulink model. The other application is proposed to use water as a cooling agent for a radiator setup which can be used for both stationary and mobile applications. CFD analysis is done to simulate and optimize the setup using Creo-ANSYS by considering water storage, spraying patterns, and system dynamics. Results are shown that there is a 4.8 percent increase in overall efficiency. As a summary, this report will try to clarify how to improve the industrial fuel cell setups, which can be considered a feasible alternative to substitute conventional mobility methods.
  • Post
    Searching for Vector Wave Dark Matter with Levitated Magnetomechanics Analysis of a Hypothetical Direct Detection Experiment Using Levitating Superconductive Test Objects
    (2023) Anduri, Måns; Chalmers tekniska högskola / Institutionen för fysik; Chalmers University of Technology / Department of Physics; Catena, Riccardo; Catena, Riccardo
    This master’s thesis investigates the sensitivity of a hypothetical direct detection experiment using levitated magnetomechanics for a vector bosonic wave dark mat ter candidate. The sensitivity studies done is for masses between 1.24 · 10−13 eV and 4.14 · 10−12 eV and with four different possible background terms, where the background is assumed to be thermally based. The vector boson is described with the Lagrangian, L = − 1 4 FµνF µν− 1 2m2 DMAµAµ+gAµnγ¯ µn, and couples to the baryon number minus the lepton number, thus yielding an EP-violating force on a charge neutral test object. Since it is ultralight it can be seen as a classical wave. To measure the sensitivity, the exclusion and discovery limits for the coupling constant, g, are asymptotically derived with the Asimov data set for 95% confidence interval and the 5σ level respectively. The limits are plotted against the mass scanned over and compared with the discovery limit for an optomechanical experiment for the same dark matter candidate. It is shown that the sensitivity for the most optimistic background is comparable with said experiment.