Examensarbeten för masterexamen // Master Theseshttp://hdl.handle.net/20.500.12380/532024-03-28T16:44:18Z2024-03-28T16:44:18Z123711D model development and simulation of low-grade waste heat recovery from a marine engineBhandari, Varunhttp://hdl.handle.net/20.500.12380/3025232023-02-10T09:10:35Z2021-01-01T00:00:00Zdc.title: 1D model development and simulation of low-grade waste heat recovery from a marine engine
dc.contributor.author: Bhandari, Varun
dc.description.abstract: The increasing awareness about the effect of emission of CO2 into the atmosphere,
as well as meeting the emission targets set by IMO (International Marine Orga-
nization) by 2050 results in a continuous need for improving efficiencies of marine
engines. A large portion of the fuel's chemical energy is lost to the surroundings
as heat, even in the most energy efficient engines. Waste heat recovery through a
rankine cycle has emerged as a promising way to increasing the engine efficiency by
utilizing the untapped availability of waste heat in the coolant systems as well as
the exhaust of the engine.
The highly transient nature of operation of an engine and depicting the heat transfer
phenomenon in an ORC (Organic Rankine Cycle) accurately makes the represen-
tation of an ORC through a mathematical model quite challenging. Based on the
application, selection of the right working
uid, equipment and control strategy is
crucial for the performance of the ORC
In this thesis, a Simulink based mathematical model of an ORC for waste heat recov-
ery from a marine engine is built. The WHR (Waste Heat Recovery) model is built
for transient operation. The model is run on a road cycle, built from actual field
test data of a boat, to simulate its performance in actual operation of the marine
vessel. Further, the performance of ORC with di erent working
uids and different
type of expanders is evaluated. And finally, a rough cost estimation of the ORC is
done.
The results from the project show that the refrigerant R1234ze(Z) is the most suit-
able for engine coolant WHR application and turbine expander performs better
than the volumetric expander. Lastly, as the scale of the ORC system increases, it
becomes increasingly cost effective.
2021-01-01T00:00:00Z1D Modelling of Fuel Mixing andGustaf, Lindberghttp://hdl.handle.net/20.500.12380/1825202023-02-10T09:11:17Z2013-01-01T00:00:00Zdc.title: 1D Modelling of Fuel Mixing and
dc.contributor.author: Gustaf, Lindberg
dc.description.abstract: A one-dimensional model for fuel mixing and conversion in a bubbling fluidized bed has been developed. The model includes fuel particle size development and calculates the concentration profile for char, volatiles, moisture and ash as well as the fuel particle size distribution. Three different kinds of units have been modelled: gasification, combustion and chemical-looping combustion. Two types of fuels, wood and coal, have been used. Based on the results from the model, the addition of a fragmentation sub-model is very important to fully characterize the char composition in the outlet and the inclusion of particle size classes and particle size development has a large effect on the result. The model presents convergence problems for certain combinations of inputs, such as combustion of fragmenting wood. The model results are sensitive to the value of many inputs such as fuel specific parameters regarding fragmentation and reaction kinetics. Because of these inputs, the use of the model for non-characterized fuels can only be of more qualitative than quantitative use.
2013-01-01T00:00:00Z20 kW pulslastnätdel - topologival och genomförandeVinnberg, Andreashttp://hdl.handle.net/20.500.12380/828542019-07-03T12:12:40Z2008-01-01T00:00:00Zdc.title: 20 kW pulslastnätdel - topologival och genomförande
dc.contributor.author: Vinnberg, Andreas
dc.description.abstract: Denna rapport visar på skillnaderna mellan både topologival på hög abtraktionsnivå, huvudsakligen energilagring och medel- eller peakenergiöverförning som på låg abstraktionsnivå såsom switchtopologi (flyback, halvbrygga, fullbrygga och så vidare) och likriktning för ett fall med hög lastvariation och algebraiska uttryck formuleras för val av desamma. Funktioner för kondensator- och transistoregenskaper i förhållande till deras pris beräknas och val av optimala kondensator- och transistorval diskuteras. En prototyp byggdes för att verifiera resultaten. För en DC/DC-omvandlare som matas från en PFC med 20 kW medeleffekt under pulslast föreslås en peakeffektomvandlare bestående av två fullbryggor, fasskiftade 180 grader ur fas som har i princip all energilagring på primärsidan.
2008-01-01T00:00:00Z3-D computations of eddy current losses in core ends of salient-pole synchronous machinesChitroju, Rathnahttp://hdl.handle.net/20.500.12380/673412019-07-03T12:07:46Z2004-01-01T00:00:00Zdc.title: 3-D computations of eddy current losses in core ends of salient-pole synchronous machines
dc.contributor.author: Chitroju, Rathna
2004-01-01T00:00:00Z3D Radiative Transfer Modeling of the Red Supergiant NML CygHirvonen, Perhttp://hdl.handle.net/20.500.12380/3058172023-02-10T09:04:43Z2021-01-01T00:00:00Zdc.title: 3D Radiative Transfer Modeling of the Red Supergiant NML Cyg
dc.contributor.author: Hirvonen, Per
dc.description.abstract: Red supergiants are bright but cool massive stars that have depleted their core
hydrogen supply and have started fusing heavier elements. These evolved stars ex perience a strong mass loss through a slow moving stellar wind that results in an
extended circumstellar envelope with a rich chemistry. The circumstellar envelopes
of red supergiants will eventually enrich the interstellar medium, influencing the
formation and evolution of the next generation of stars and planets. Understanding
the mass loss of massive evolved stars is a key component in understanding both
the evolution of individual stars and the chemical evolution of the universe.
We analyze carbon monoxide emission around the red supergiant NML Cyg ob served with the ground based JCMT 15 meter and Onsala 20 meter telescopes, as
well as the HIFI instrument aboard the Herschel Space Observatory. The line pro files have complex shapes, suggestive of multiple components in the stellar outflow.
3D radiative transfer modeling is used to recreate the observed line profiles. We con clude that it is possible to recreate the observed line profiles by using one spherical
component and three directed outflows.
2021-01-01T00:00:00ZA 1296 MHz Earth-Moon-Earth Communication SystemLindgren, Magnushttp://hdl.handle.net/20.500.12380/2178842023-02-10T09:27:08Z2015-01-01T00:00:00Zdc.title: A 1296 MHz Earth-Moon-Earth Communication System
dc.contributor.author: Lindgren, Magnus
dc.description.abstract: The task of designing, building and characterizing an Earth–Moon–Earth (EME) communication system in the 1296 MHz amateur radio band has been successfully accomplished. Using the radar equation and thermal noise theory, in conjunction with measured and estimated system parameters, the lunar echo signal to noise ratio (SNR) has been estimated at 5.4+1.2 −1.6 dB (2) in a 100 Hz bandwidth. The results of empirical long-term averaged SNR measurements are within the boundaries of this estimate, thus adding to its validity. Additionally, measurements of the libration-induced Doppler spread of the lunar echoes show good agreement with theoretical predictions. Further experiments should be conducted however, in order to asses the repeatability of the results. Improvements to the system and measurement methods should also be considered. The report focuses on both theory and practice and apart from the above, additional attention is given to the topic of low noise figure (NF) measurements due to the systemic importance of the LNA NF parameter.
2015-01-01T00:00:00ZA 3D-model for O2 airglow perturbations induced by gravity waves in the upper mesosphereLi, Anqihttp://hdl.handle.net/20.500.12380/2495852023-02-10T09:04:09Z2017-01-01T00:00:00Zdc.title: A 3D-model for O2 airglow perturbations induced by gravity waves in the upper mesosphere
dc.contributor.author: Li, Anqi
dc.description.abstract: To investigate the influence of atmospheric waves in the mesosphere, a new Swedish satellite MATS (Mesospheric Airglow/Aerosol Tomography and Spectroscopy) will be launched in 2019. It will observe infrared emissions at 762 nm from the O2(b1+ g ) airglow in the region of 70 - 110 km altitude. As a part of the design work for the MATS project, an accurate forward model is needed to estimate what MATS is expected to measure. The results from this model will be used to evaluate the retrieval methods for processing the measurements from MATS. In this thesis project, a gravity wave model and a photochemistry model were coupled to simulate both the day- and nightglow emission fields in three spatial dimensions and time. Simulated satellite images were generated taking into consideration the sphericity of the Earth and the limb-viewing geometry of MATS. Simulation parameters were set according to the preliminary design of the instrument, such as the satellite orbit, image resolution and spectral selections. These satellite images were the first simulated airglow limb images made for the MATS project. By analysing the output data, the relations between wave parameters and airglow perturbations were investigated. It was shown that wave patterns can be easily observed between 85 - 105 km due to the relatively large perturbation in airglow emissions. The O2 airglow emission field was found to be highly sensitive to atomic oxygen concentration field as an input. Furthermore, as expected, wave patterns projected on simulated satellite images largely depend on the horizontal orientation of the wave propagation. This implies that a tomographic reconstruction is needed when the angle between the wave front and the limb-viewing direction is large. Finally, limitations of the model were discussed.
2017-01-01T00:00:00ZA 5 MW Wind Turbine Generator System for a DC grid applicationMazaheri, Arashhttp://hdl.handle.net/20.500.12380/1635372023-02-10T09:17:45Z2012-01-01T00:00:00Zdc.title: A 5 MW Wind Turbine Generator System for a DC grid application
dc.contributor.author: Mazaheri, Arash
dc.description.abstract: In this work a robust generating system for a 5 MW PMSM wind turbine system with diode rectifier and DC/DC converter connected to DC link has been investigated. The design and efficiency have been studied using MATLAB Simulink and SIMPLORER and the result shows that this system topology has high efficiency. The result found was that the same generator that can provide 5 MW using an IGBT-Converter, can only give 2.5 MW using a diode rectifier and a variable DClink. However, with series and parallel compensations the same generator can provide an output power even more than 5 MW. It was also found that the efficiencies of systems at rated operation with IGBT-converter, diode rectifier with series compensation and shunt compensation are 97.6%, 98.6%, and 98.9%, respectively. Moreover, the systems efficiency at 5% of load are 93.0%, IGBT-converter, 97.3%, series compensation, and 95.3%, shunt compensation. The required reactive power for series compensation and shunt compensation to provide 5 MW output power for the diode rectifier system, at rated operation, are 1.5 MVar and 2.8 MVar, respectively. To conclude, the diode rectifier with series compensation is more efficient for low wind speed (4-11m/s) also it needs less reactive power; however, the shunt compensation scenario has better efficiency at rated speed when the wind speed is higher than 12m/s.
2012-01-01T00:00:00ZA Bayesian machine learning approach to geostationary infrared precipitation retrievalsTellwe, Gustavhttp://hdl.handle.net/20.500.12380/3014462023-02-10T09:15:31Z2020-01-01T00:00:00Zdc.title: A Bayesian machine learning approach to geostationary infrared precipitation retrievals
dc.contributor.author: Tellwe, Gustav
dc.description.abstract: This project uses geostationary satellite data to retrieve precipitation rates at surface level.
It is achieved through the use of quantile regression neural networks (QRNN) calibrated
against rain rates from the Global Precipitation Measurement (GPM) Core Observatory
satellite. The area of exploration is located over the Amazon rainforest. The main difficulty
of this problem is that geostationary data is not directly related to rain as it only perceives
the cloud top temperatures. It does, however, have a high temporal and spatial resolution
which makes it interesting for applications in remote areas of the Earth where groundbased
radar equipment is unavailable. The result of the project is mainly a comparison
between different neural network architectures such as multi-layer perceptron (MLP) and
convolutional neural networks (CNN), but there is also a minor comparison to an adapted
version of a Hydroestimator (HE) that is currently in use by the National Institute for
Space Research (INPE) in Brazil. The best performing configuration, with regards to the
loss function, in this study was a CNN. It performed significantly better than the adapted
HE for a test conducted over two days in March. An unsuccessful attempt to improve the
results using time-series was also conducted. Furthermore, a U-net architecture was also
tested on rain rate data that has been resolution-enhanced through interpolation.
2020-01-01T00:00:00ZA Bayesian machine learning approach to passive microwave precipitation retrievalsNorrestad, Teodorhttp://hdl.handle.net/20.500.12380/2568992023-02-10T09:04:21Z2019-01-01T00:00:00Zdc.title: A Bayesian machine learning approach to passive microwave precipitation retrievals
dc.contributor.author: Norrestad, Teodor
dc.description.abstract: A machine learning-based approach to precipitation retrievals, using Quantile Regression Neural Networks (QRNNs), is developed for data from the Global Precipitation Measurement (GPM) mission. The retrievals are conducted within a Bayesian framework where the networks are trained to predict quantiles of the posterior distribution of rain rates, conditioned on passive microwave observations. In this way, rain rates are retrieved along with the associated retrieval uncertainties. The effects of including additional spatial information as input to the QRNNs are also investigated. Different QRNNs are trained and tested, first globally over oceans and then over the U.S Great Plains. In both cases, the performance of the QRNNs are compared to the Goddard Profiling Algorithm (GPROF), a state-of-the-art passive microwave retrieval algorithm. The primary results are those over oceans, where the QRNNs show great performance on similar levels as GPROF with respect to point estimate metrics such as the mean squared error. Furthermore, the QRNN retrievals are very fast, taking less than a millisecond per footprint on a standard computer. It turns out that extra spatial information improves the QRNNs, especially on making rain-no rain classifications with fractions of true positives and true negatives exceeding 0.67 and 0.96 respectively. Furthermore, the QRNNs manage to produce well calibrated quantiles, resulting in good confidence intervals to account for retrieval uncertainties. Over the Great Plains, the results are promising but are based on much smaller amounts of data and are thus less significant.
2019-01-01T00:00:00Z