Chalmers Open Digital Repository
Not a member yet
26247 research outputs found
Sort by
Sequential Graph-Based Decoding of the Surface Code using a Hybrid Graph and Recurrent Neural Network Model
No full textIn order to achieve reliable quantum computation with noisy qubits, quantum error
correction (QEC) is necessary. Quantum error correcting codes mitigate the inher ent noise in quantum systems by distributing the logical state over several qubits,
thereby introducing redundancy. One such promising code is the surface code. It
encodes the logical qubit using a two-dimensional lattice of physical data qubits
and ancilla qubits. By taking and decoding measurements on the ancilla qubits of
the surface code, one can deduce whether a logical bit- or phase-flip has occurred.
However, this is a complex and potentially time-consuming task. Multiple decoding
algorithms exist, such as the classical minimum-weight perfect matching (MWPM)
decoder. In recent years, data-driven algorithms have been shown to decode the
surface code with a high degree of accuracy. In this thesis, we present a machine
learning approach to decoding the surface code using a combination of graph neural
networks (GNN) and recurrent neural networks (RNN). Specifically, graph represen tations are constructed over a short, sliding time window of syndrome measurement
data. Each representation is processed by a GNN and its output is used as a learned
high-dimensional embedding for an RNN. This enables continuous decoding of mea surement patterns over longer time series. While the decoder is trained on relatively
short syndromes, it is able to generalize for unseen syndromes and longer time se ries, outperforming the classical MWPM algorithm across both short and long time
series. This work opens up a new approach to reliable and potentially fast decoding
of QEC codes
Data-Driven Solutions for Green Production Integrating Resource Efficiency Assessment in Manufacturing Systems
Full text linkIn response to sustainable development efforts by the United Nations 2020 Agenda,
industries are aiming towards more sustainable production. The European Commission
has created the classification system called EU taxonomy, establishing the
definition of sustainability and sustainable activities. Consequently, European manufacturers
are seeking opportunities to reduce their environmental impact, creating
the need to understand how their resources are utilized. Resource efficiency methods
enable the assessment of resource usage but require high data quality and availability,
making the implementation difficult. One of the main challenges for resource
efficiency assessment is data completeness and reliability, especially at a process
level, in tandem with a lack of standardized data collection methods resulting in the
implementation of RE assessment being difficult. Despite the issues, there are still
opportunities and benefits of using already available data in manufacturing systems
with proper indicator selection having data characteristics in mind. This project
aims to leverage available factory data, select indicators based on available data,
and integrate resource efficiency in manufacturing systems to identify opportunities
for greener production with a resource efficiency method in line with the EU
taxonomy. This project showcases a case study implementing resource efficiency
assessment in an automotive plant with an assessment design that includes multiple
methods to be aligned with stakeholder priorities and indicates inefficiencies of
resource usage. A selection method was devised as the project’s core, designed to
be general and adaptable for other cases. However, assessment methods are inherently
different and data quality is a critical factor in implementing them, not only
stakeholder preferences of the company. The study utilized existing data to assess
resource efficiency and proposed automated data handling for future assessments to
streamline the process and reduce the execution time
Evaluating potential supply chains for hydrogen as aviation fuel on the Swedish west coast: A quantitative and qualitative study in how hydrogen can potentially be distributed to airports to supply hydrogen-based aircraft
Full text linkAviation is one of the main polluters within the transport sector and is continuously
growing. To be able to meet environmental goals and at the same time fulfill societal
demands for transport, hydrogen may be used as an alternative fuel to lower emissions,
but research into infrastructure is needed. This thesis uses two methods to
quantitatively and qualitatively investigate the potential of different supply chains
to supply aircraft with green hydrogen, by conducting a techno-economic analysis
and semi-structured interviews with stakeholders. The results of this study show
that a transition to hydrogen in aviation is believed to be feasible by stakeholders,
and synergetic effects along the Swedish west coast are expected. Centrally produced
alternatives are found to be the cheapest alternative early on when volumes
are low, while on-site production of hydrogen emerges as the cheapest alternative
when demand increases. The results can be used to form an understanding of future
airport fuel supply chain planning
Designing a Sustainable Island for Electric Boat Charging
No full textThe increase of electric boats has highlighted the necessity for marine charging
facilities. With numerous marinas lacking the requisite infrastructure to support
a significant influx of electric boats, there is a pressing demand for charging
solutions. This project featured the iterative process of creating an self-sustainable
island that could facilitate the charging needs of electric boats, as well as attract
other visitors. This required extensive research of solutions regarding energy,
waste and water management, different facilities that could generate profit and
costs related to all parts of the island.
The project involved a comprehensive concept development process, research over
current technology and how it could me implemented. Creative methods were
used to generate innovative ideas for the modular island. Through the iterative
process, a final concept wad defined to meet the demands of Volvo Penta and
electrical boat owners. Utilizing CAD models and 3D printing, the team was able
to visualize the final concept, ensuring the design could be implemented.
The final island represented a complete off-grid solution for energy supply, water
management and waste management, meaning it was independent of external
energy sources, water supply and waste management system
VTOL, framtida drönarapplikationer i samhället
No full textThis report presents the development of a VTOL drone prototype, with a particular
focus on mechanical and electrical components, battery selection, wing structure,
and manufacturing methods. A sketch of the developed electrical system shows how
power supply and signal cables are connected. The drone uses a Li-Po battery for
high discharge capacity, enabling sufficient power supply to the motors and flight
computer. The internal structure of the wing is designed to optimize strength and
rigidity, using 3D-printed segments joined with carbon fiber tubes, in figure 0.2 is
a image of the prototype in CAD. To maximize range and efficiency, the drone
quickly transitions from drone mode to airplane mode. The transition is controlled
by adjusting the pusher motor and drone motors’ RPMs based on speed data from
the GPS and pitot tube. Conclusions show that current legislation poses a challenge
for drone deliveries in populated areas, but the wing design significantly improves
the drone’s range and energy efficiency. Future work includes practical flight tests
and software optimization to further enhance the drone’s performance
