On state and parameter estimation in chaotic systems

State-of-the-art predictions of atmospheric states rely on large-scale numerical models of chaotic systems. This dissertation studies numerical methods for state and parameter estimation in such systems. The motivation comes from weather and climate models and a methodological perspective is adopted. The dissertation comprises three sections: state estimation, parameter estimation and chemical data assimilation with real atmospheric satellite data. In the state estimation part of this dissertation, a new filtering technique based on a combination of ensemble and variational Kalman filtering approaches, is presented, experimented and discussed. This new filter is developed for large-scale Kalman filtering applications. In the parameter estimation part, three different techniques for parameter estimation in chaotic systems are considered. The methods are studied using the parameterized Lorenz 95 system, which is a benchmark model for data assimilation. In addition, a dilemma related to the uniqueness of weather and climate model closure parameters is discussed. In the data-oriented part of this dissertation, data from the Global Ozone Monitoring by Occultation of Stars (GOMOS) satellite instrument are considered and an alternative algorithm to retrieve atmospheric parameters from the measurements is presented. The validation study presents first global comparisons between two unique satellite-borne datasets of vertical profiles of nitrogen trioxide (NO3), retrieved using GOMOS and Stratospheric Aerosol and Gas Experiment III (SAGE III) satellite instruments. The GOMOS NO3 observations are also considered in a chemical state estimation study in order to retrieve stratospheric temperature profiles. The main result of this dissertation is the consideration of likelihood calculations via Kalman filtering outputs. The concept has previously been used together with stochastic differential equations and in time series analysis. In this work, the concept is applied to chaotic dynamical systems and used together with Markov chain Monte Carlo (MCMC) methods for statistical analysis. In particular, this methodology is advocated for use in numerical weather prediction (NWP) and climate model applications. In addition, the concept is shown to be useful in estimating the filter-specific parameters related, e.g., to model error covariance matrix parameters.

Turbulenssimallien soveltuminen syklonien numeeriseen virtauslaskentaan

Tässä työssä tutkittiin miten totuudenmukaisia tuloksia syklonierottimen virtauskentästä saadaan numeerisella laskennalla, kun käytetään eri turbulenssimalleja. Tarkoitus oli myös selvittää yleisesti syklonin toimintaperiaatteita, haasteita sen käytössä sekä syklonin numeerisen virtauslaskennan perusteita. Numeerisen virtauslaskennan teoria selitetään pääpiirteittäin, samoin turbulenssin mallinnus. Työn laskentaosiossa simuloitiin Fluent-ohjelmalla syklonin virtauskenttää kuumalla ilmalla sekä kahdella eri turbulenssimallilla ja verrattiin tuloksia kirjallisuudesta löytyviin mittaustuloksiin. Simuloinnit suoritettiin sekä ajasta riippuvana että ajasta riippumattomana ja kahdella eri laskentahilalla. Simulointien tulokset osoittivat, että RNG k-ε turbulenssimalli ei kykene tuottamaan totuu-denmukaista virtauskenttää. Toisen käytetyn turbulenssimallin, Reynolds-jännitysmallin tulokset vastasivat enemmän mittaustuloksia. Reynolds-jännitysmallia voidaan pitää käyttökelpoisena syklonin simuloinnissa tämän työn ja kirjallisuuden perusteella. Mallissa oli yksinkertaistuksia, esimerkiksi kiinteää ainetta ei otettu huomioon lainkaan.

Linear prediction in optical spectroscopy

A linear prediction procedure is one of the approved numerical methods of signal processing. In the field of optical spectroscopy it is used mainly for extrapolation known parts of an optical signal in order to obtain a longer one or deduce missing signal samples. The first is needed particularly when narrowing spectral lines for the purpose of spectral information extraction. In the present paper the coherent anti-Stokes Raman scattering (CARS) spectra were under investigation. The spectra were significantly distorted by the presence of nonlinear nonresonant background. In addition, line shapes were far from Gaussian/Lorentz profiles. To overcome these disadvantages the maximum entropy method (MEM) for phase spectrum retrieval was used. The obtained broad MEM spectra were further underwent the linear prediction analysis in order to be narrowed.

The supermassive binary black hole system OJ287

Abstract This doctoral thesis concerns the active galactic nucleus (AGN) most often referred to with the catalogue number OJ287. The publications in the thesis present new discoveries of the system in the context of a supermassive binary black hole model. In addition, the introduction discusses general characteristics of the OJ287 system and the physical fundamentals behind these characteristics. The place of OJ287 in the hierarchy of known types of AGN is also discussed. The introduction presents a large selection of fundamental physics required to have a basic understanding of active galactic nuclei, binary black holes, relativistic jets and accretion disks. Particularly the general relativistic nature of the orbits of close binaries of supermassive black holes is explored with some detail. Analytic estimates of some of the general relativistic effects in such a binary are presented, as well as numerical methods to calculate the effects more precisely. It is also shown how these results can be applied to the OJ287 system. The binary orbit model forms the basis for models of the recurring optical outbursts in the OJ287 system. In the introduction, two physical outburst models are presented in some detail and compared. The radiation hydrodynamics of the outbursts are discussed and optical light curve predictions are derived. The precursor outbursts studied in Paper III are also presented, and tied into the model of OJ287. To complete the discussion of the observable features of OJ287, the nature of the relativistic jets in the system, and in active galactic nuclei in general, is discussed. Basic physics of relativistic jets are presented, with additional detail added in the form of helical jet models. The results of Papers II, IV and V concerning the jet of OJ287 are presented, and their relation to other facets of the binary black hole model is discussed. As a whole, the introduction serves as a guide, though terse, for the physics and numerical methods required to successfully understand and simulate a close binary of supermassive black holes. For this purpose, the introduction necessarily combines a large number of both fundamental and specific results from broad disciplines like general relativity and radiation hydrodynamics. With the material included in the introduction, the publications of the thesis, which present new results with a much narrower focus, can be readily understood. Of the publications, Paper I presents newly discovered optical data points for OJ287, detected on archival astronomical plates from the Harvard College Observatory. These data points show the 1900 outburst of OJ287 for the first time. In addition, new data points covering the 1913 outburst allowed the determination of the start of the outburst with more precision than was possible before. These outbursts were then successfully numerically modelled with an N-body simulation of the OJ287 binary and accretion disc. In Paper II, mechanisms for the spin-up of the secondary black hole in OJ287 via interaction with the primary accretion disc and the magnetic fields in the system are discussed. Timescales for spin-up and alignment via both processes are estimated. It is found that the secondary black hole likely has a high spin. Paper III reports a new outburst of OJ287 in March 2013. The outburst was found to be rather similar to the ones reported in 1993 and 2004. All these outbursts happened just before the main outburst season, and are called precursor outbursts. In this paper, a mechanism was proposed for the precursor outbursts, where the secondary black hole collides with a gas cloud in the primary accretion disc corona. From this, estimates of brightness and timescales for the precursor were derived, as well as a prediction of the timing of the next precursor outburst. In Paper IV, observations from the 2004–2006 OJ287 observing program are used to investigate the existence of short periodicities in OJ287. The existence of a _50 day quasiperiodic component is confirmed. In addition, statistically significant 250 day and 3.5 day periods are found. Primary black hole accretion of a spiral density wave in the accretion disc is proposed as the source of the 50 day period, with numerical simulations supporting these results. Lorentz contracted jet re-emission is then proposed as the reason for the 3.5 day timescale. Paper V fits optical observations and mm and cm radio observations of OJ287 with a helical jet model. The jet is found to have a spine–sheath structure, with the sheath having a much lower Lorentz gamma factor than the spine. The sheath opening angle and Lorentz factor, as well as the helical wavelength of the jet are reported for the first time. Tiivistelmä Tässä väitöskirjatutkimuksessa on keskitytty tutkimaan aktiivista galaksiydintä OJ287. Väitöskirjan osana olevat tieteelliset julkaisut esittelevät OJ287-systeemistä saatuja uusia tuloksia kaksoismusta-aukkomallin kontekstissa. Väitöskirjan johdannossa käsitellään OJ287:n yleisiä ominaisuuksia ja niitä fysikaalisia perusilmiöitä, jotka näiden ominaisuuksien taustalla vaikuttavat. Johdanto selvittää myös OJ287-järjestelmän sijoittumisen aktiivisten galaksiytimien hierarkiassa. Johdannossa käydään läpi joitakin perusfysiikan tuloksia, jotka ovat tarpeen aktiivisten galaksiydinten, mustien aukkojen binäärien, relativististen suihkujen ja kertymäkiekkojen ymmärtämiseksi. Kahden toisiaan kiertävän mustan aukon keskinäisen radan suhteellisuusteoreettiset perusteet käydään läpi yksityiskohtaisemmin. Johdannossa esitetään joitakin analyyttisiä tuloksia tällaisessa binäärissä havaittavista suhteellisuusteoreettisista ilmiöistä. Myös numeerisia menetelmiä näiden ilmiöiden tarkempaan laskemiseen esitellään. Tuloksia sovelletaan OJ287-systeemiin, ja verrataan havaintoihin. OJ287:n mustien aukkojen ratamalli muodostaa pohjan systeemin toistuvien optisten purkausten malleille. Johdannossa esitellään yksityiskohtaisemmin kaksi fysikaalista purkausmallia, ja vertaillaan niitä. Purkausten säteilyhydrodynamiikka käydään läpi, ja myös ennusteet purkausten valokäyrille johdetaan. Johdannossa esitellään myös Julkaisussa III johdettu prekursoripurkausten malli, ja osoitetaan sen sopivan yhteen OJ287:n binäärimallin kanssa. Johdanto esittelee myös relativististen suihkujen fysiikkaa sekä OJ287- systeemiin liittyen että aktiivisten galaksiydinten kontekstissa yleisesti. Relativististen suihkujen perusfysiikka esitellään, kuten myös malleja kierteisistä suihkuista. Julkaisujen II, IV ja V OJ287-systeemin suihkuja koskevat tulokset esitellään binäärimallin kontekstissa. Kokonaisuutena johdanto palvelee suppeana oppaana, joka esittelee tarvittavan fysiikan ja tarpeelliset numeeriset menetelmät mustien aukkojen binäärijärjestelmän ymmärtämiseen ja simulointiin. Tätä tarkoitusta varten johdanto yhdistää sekä perustuloksia että joitakin syvällisempiä tuloksia laajoilta fysiikan osa-alueilta kuten suhteellisuusteoriasta ja säteilyhydrodynamiikasta. Johdannon sisältämän materiaalin avulla väitöskirjan julkaisut, ja niiden esittämät tulokset, ovat hyvin ymmärrettävissä. Väitöskirjan julkaisuista ensimmäinen esittelee uusia OJ287-systeemistä saatuja havaintopisteitä, jotka on paikallistettu Harvardin yliopiston observatorion arkiston valokuvauslevyiltä. OJ287:n vuonna 1900 tapahtunut purkaus nähdään ensimmäistä kertaa näissä havaintopisteissä. Uudet havaintopisteet mahdollistivat myös vuoden 1913 purkauksen alun ajoittamisen tarkemmin kuin aiemmin oli mahdollista. Havaitut purkaukset mallinnettiin onnistuneesti simuloimalla OJ287-järjestelmän mustien aukkojen paria ja kertymäkiekkoa. Julkaisussa II käsitellään mekanismeja OJ287:n sekundäärisen mustan aukon spinin kasvamiseen vuorovaikutuksessa primäärin kertymäkiekon ja systeemin magneettikenttien kanssa. Julkaisussa arvioidaan maksimispinin saavuttamisen ja spinin suunnan vakiintumisen aikaskaalat kummallakin mekanismilla. Tutkimuksessa havaitaan sekundäärin spinin olevan todennäköisesti suuri. Julkaisu III esittelee OJ287-systeemissä maaliskuussa 2013 tapahtuneen purkauksen. Purkauksen havaittiin muistuttavan vuosina 1993 ja 2004 tapahtuneita purkauksia, joita kutsutaan yhteisnimityksellä prekursoripurkaus (precursor outburst). Julkaisussa esitellään purkauksen synnylle mekanismi, jossa OJ287-systeemin sekundäärinen musta aukko osuu primäärisen mustan aukon kertymäkiekon koronassa olevaan kaasupilveen. Mekanismin avulla johdetaan arviot prekursoripurkausten kirkkaudelle ja aikaskaalalle. Julkaisussa johdetaan myös ennuste seuraavan prekursoripurkauksen ajankohdalle. Julkaisussa IV käytetään vuosina 2004–2006 kerättyjä havaintoja OJ287- systeemistä lyhyiden jaksollisuuksien etsintään. Julkaisussa varmennetaan systeemissä esiintyvä n. 50 päivän kvasiperiodisuus. Lisäksi tilastollisesti merkittävät 250 päivän ja 3,5 päivän jaksollisuudet havaitaan. Julkaisussa esitetään malli, jossa primäärisen mustan aukon kertymäkiekossa oleva spiraalitiheysaalto aiheuttaa 50 päivän jaksollisuuden. Mallista tehty numeerinen simulaatio tukee tulosta. Systeemin relativistisen suihkun emittoima aikadilatoitunut säteily esitetään aiheuttajaksi 3,5 päivän jaksollisuusaikaskaalalle. Julkaisussa V sovitetaan kierresuihkumalli OJ287-systeemistä tehtyihin optisiin havaintoihin ja millimetri- sekä senttimetriaallonpituuden radiohavaintoihin. Suihkun rakenteen havaitaan olevan kaksijakoinen ja koostuvan ytimestä ja kuoresta. Suihkun kuorella on merkittävästi pienempi Lorentzin gamma-tekijä kuin suihkun ytimellä. Kuoren avautumiskulma ja Lorentztekijä sekä suihkun kierteen aallonpituus raportoidaan julkaisussa ensimmäistä kertaa.

Adaptive Markov chain Monte Carlo and Bayesian filtering for state space models

This thesis is concerned with the state and parameter estimation in state space models. The estimation of states and parameters is an important task when mathematical modeling is applied to many different application areas such as the global positioning systems, target tracking, navigation, brain imaging, spread of infectious diseases, biological processes, telecommunications, audio signal processing, stochastic optimal control, machine learning, and physical systems. In Bayesian settings, the estimation of states or parameters amounts to computation of the posterior probability density function. Except for a very restricted number of models, it is impossible to compute this density function in a closed form. Hence, we need approximation methods. A state estimation problem involves estimating the states (latent variables) that are not directly observed in the output of the system. In this thesis, we use the Kalman filter, extended Kalman filter, Gauss–Hermite filters, and particle filters to estimate the states based on available measurements. Among these filters, particle filters are numerical methods for approximating the filtering distributions of non-linear non-Gaussian state space models via Monte Carlo. The performance of a particle filter heavily depends on the chosen importance distribution. For instance, inappropriate choice of the importance distribution can lead to the failure of convergence of the particle filter algorithm. In this thesis, we analyze the theoretical Lᵖ particle filter convergence with general importance distributions, where p ≥2 is an integer. A parameter estimation problem is considered with inferring the model parameters from measurements. For high-dimensional complex models, estimation of parameters can be done by Markov chain Monte Carlo (MCMC) methods. In its operation, the MCMC method requires the unnormalized posterior distribution of the parameters and a proposal distribution. In this thesis, we show how the posterior density function of the parameters of a state space model can be computed by filtering based methods, where the states are integrated out. This type of computation is then applied to estimate parameters of stochastic differential equations. Furthermore, we compute the partial derivatives of the log-posterior density function and use the hybrid Monte Carlo and scaled conjugate gradient methods to infer the parameters of stochastic differential equations. The computational efficiency of MCMC methods is highly depend on the chosen proposal distribution. A commonly used proposal distribution is Gaussian. In this kind of proposal, the covariance matrix must be well tuned. To tune it, adaptive MCMC methods can be used. In this thesis, we propose a new way of updating the covariance matrix using the variational Bayesian adaptive Kalman filter algorithm.

The solutions and tunneling properties for a linear potential and an inverted harmonic oscillator in an infinite well

In this work we look at two different 1-dimensional quantum systems. The potentials for these systems are a linear potential in an infinite well and an inverted harmonic oscillator in an infinite well. We will solve the Schrödinger equation for both of these systems and get the energy eigenvalues and eigenfunctions. The solutions are obtained by using the boundary conditions and numerical methods. The motivation for our study comes from experimental background. For the linear potential we have two different boundary conditions. The first one is the so called normal boundary condition in which the wave function goes to zero on the edge of the well. The second condition is called derivative boundary condition in which the derivative of the wave function goes to zero on the edge of the well. The actual solutions are Airy functions. In the case of the inverted oscillator the solutions are parabolic cylinder functions and they are solved only using the normal boundary condition. Both of the potentials are compared with the particle in a box solutions. We will also present figures and tables from which we can see how the solutions look like. The similarities and differences with the particle in a box solution are also shown visually. The figures and calculations are done using mathematical software. We will also compare the linear potential to a case where the infinite wall is only on the left side. For this case we will also show graphical information of the different properties. With the inverted harmonic oscillator we will take a closer look at the quantum mechanical tunneling. We present some of the history of the quantum tunneling theory, its developers and finally we show the Feynman path integral theory. This theory enables us to get the instanton solutions. The instanton solutions are a way to look at the tunneling properties of the quantum system. The results are compared with the solutions of the double-well potential which is very similar to our case as a quantum system. The solutions are obtained using the same methods which makes the comparison relatively easy. All in all we consider and go through some of the stages of the quantum theory. We also look at the different ways to interpret the theory. We also present the special functions that are needed in our solutions, and look at the properties and different relations to other special functions. It is essential to notice that it is possible to use different mathematical formalisms to get the desired result. The quantum theory has been built for over one hundred years and it has different approaches. Different aspects make it possible to look at different things.

The solutions and tunneling properties for a linear potential and an inverted harmonic oscillator in an infinite well

In this work we look at two different 1-dimensional quantum systems. The potentials for these systems are a linear potential in an infinite well and an inverted harmonic oscillator in an infinite well. We will solve the Schrödinger equation for both of these systems and get the energy eigenvalues and eigenfunctions. The solutions are obtained by using the boundary conditions and numerical methods. The motivation for our study comes from experimental background. For the linear potential we have two different boundary conditions. The first one is the so called normal boundary condition in which the wave function goes to zero on the edge of the well. The second condition is called derivative boundary condition in which the derivative of the wave function goes to zero on the edge of the well. The actual solutions are Airy functions. In the case of the inverted oscillator the solutions are parabolic cylinder functions and they are solved only using the normal boundary condition. Both of the potentials are compared with the particle in a box solutions. We will also present figures and tables from which we can see how the solutions look like. The similarities and differences with the particle in a box solution are also shown visually. The figures and calculations are done using mathematical software. We will also compare the linear potential to a case where the infinite wall is only on the left side. For this case we will also show graphical information of the different properties. With the inverted harmonic oscillator we will take a closer look at the quantum mechanical tunneling. We present some of the history of the quantum tunneling theory, its developers and finally we show the Feynman path integral theory. This theory enables us to get the instanton solutions. The instanton solutions are a way to look at the tunneling properties of the quantum system. The results are compared with the solutions of the double-well potential which is very similar to our case as a quantum system. The solutions are obtained using the same methods which makes the comparison relatively easy. All in all we consider and go through some of the stages of the quantum theory. We also look at the different ways to interpret the theory. We also present the special functions that are needed in our solutions, and look at the properties and different relations to other special functions. It is essential to notice that it is possible to use different mathematical formalisms to get the desired result. The quantum theory has been built for over one hundred years and it has different approaches. Different aspects make it possible to look at different things.

Methods for Region Based Paper Surface Roughness Measurement

Paperin pinnan karheus on yksi paperin laatukriteereistä. Sitä mitataan fyysisestipaperin pintaa mittaavien laitteiden ja optisten laitteiden avulla. Mittaukset vaativat laboratorioolosuhteita, mutta nopeammille, suoraan linjalla tapahtuville mittauksilla olisi tarvetta paperiteollisuudessa. Paperin pinnan karheus voidaan ilmaista yhtenä näytteelle kohdistuvana karheusarvona. Tässä työssä näyte on jaettu merkitseviin alueisiin, ja jokaiselle alueelle on laskettu erillinen karheusarvo. Karheuden mittaukseen on käytetty useita menetelmiä. Yleisesti hyväksyttyä tilastollista menetelmää on käytetty tässä työssä etäisyysmuunnoksen lisäksi. Paperin pinnan karheudenmittauksessa on ollut tarvetta jakaa analysoitava näyte karheuden perusteella alueisiin. Aluejaon avulla voidaan rajata näytteestä selvästi karheampana esiintyvät alueet. Etäisyysmuunnos tuottaa alueita, joita on analysoitu. Näistä alueista on muodostettu yhtenäisiä alueita erilaisilla segmentointimenetelmillä. PNN -menetelmään (Pairwise Nearest Neighbor) ja naapurialueiden yhdistämiseen perustuvia algoritmeja on käytetty.Alueiden jakamiseen ja yhdistämiseen perustuvaa lähestymistapaa on myös tarkasteltu. Segmentoitujen kuvien validointi on yleensä tapahtunut ihmisen tarkastelemana. Tämän työn lähestymistapa on verrata yleisesti hyväksyttyä tilastollista menetelmää segmentoinnin tuloksiin. Korkea korrelaatio näiden tulosten välillä osoittaa onnistunutta segmentointia. Eri kokeiden tuloksia on verrattu keskenään hypoteesin testauksella. Työssä on analysoitu kahta näytesarjaa, joidenmittaukset on suoritettu OptiTopolla ja profilometrillä. Etäisyysmuunnoksen aloitusparametrit, joita muutettiin kokeiden aikana, olivat aloituspisteiden määrä ja sijainti. Samat parametrimuutokset tehtiin kaikille algoritmeille, joita käytettiin alueiden yhdistämiseen. Etäisyysmuunnoksen jälkeen korrelaatio oli voimakkaampaa profilometrillä mitatuille näytteille kuin OptiTopolla mitatuille näytteille. Segmentoiduilla OptiTopo -näytteillä korrelaatio parantui voimakkaammin kuin profilometrinäytteillä. PNN -menetelmän tuottamilla tuloksilla korrelaatio oli paras.

Theoretical and numerical study of thermomagnetic convection in magnetic fluids

In this thesis, the magnetic field control of convection instabilities and heat and mass transfer processesin magnetic fluids have been investigated by numerical simulations and theoretical considerations. Simulation models based on finite element and finite volume methods have been developed. In addition to standard conservation equations, themagnetic field inside the simulation domain is calculated from Maxwell equations and the necessary terms to take into account for the magnetic body force and magnetic dissipation have been added to the equations governing the fluid motion.Numerical simulations of magnetic fluid convection near the threshold supportedexperimental observations qualitatively. Near the onset of convection the competitive action of thermal and concentration density gradients leads to mostly spatiotemporally chaotic convection with oscillatory and travelling wave regimes, previously observed in binary mixtures and nematic liquid crystals. In many applications of magnetic fluids, the heat and mass transfer processes including the effects of external magnetic fields are of great importance. In addition to magnetic fluids, the concepts and the simulation models used in this study may be applied also to the studies of convective instabilities in ordinary fluids as well as in other binary mixtures and complex fluids.

Level Set Methods in Image Processing

This thesis gives an overview of the use of the level set methods in the field of image science. The similar fast marching method is discussed for comparison, also the narrow band and the particle level set methods are introduced. The level set method is a numerical scheme for representing, deforming and recovering structures in an arbitrary dimensions. It approximates and tracks the moving interfaces, dynamic curves and surfaces. The level set method does not define how and why some boundary is advancing the way it is but simply represents and tracks the boundary. The principal idea of the level set method is to represent the N dimensional boundary in the N+l dimensions. This gives the generality to represent even the complex boundaries. The level set methods can be powerful tools to represent dynamic boundaries, but they can require lot of computing power. Specially the basic level set method have considerable computational burden. This burden can be alleviated with more sophisticated versions of the level set algorithm like the narrow band level set method or with the programmable hardware implementation. Also the parallel approach can be used in suitable applications. It is concluded that these methods can be used in a quite broad range of image applications, like computer vision and graphics, scientific visualization and also to solve problems in computational physics. Level set methods and methods derived and inspired by it will be in the front line of image processing also in the future.

Lossless Compression Methods for Multispectral Images: Development and Comparisons

Main purpose of this thesis is to introduce a new lossless compression algorithm for multispectral images. Proposed algorithm is based on reducing the band ordering problem to the problem of finding a minimum spanning tree in a weighted directed graph, where set of the graph vertices corresponds to multispectral image bands and the arcs’ weights have been computed using a newly invented adaptive linear prediction model. The adaptive prediction model is an extended unification of 2–and 4–neighbour pixel context linear prediction schemes. The algorithm provides individual prediction of each image band using the optimal prediction scheme, defined by the adaptive prediction model and the optimal predicting band suggested by minimum spanning tree. Its efficiency has been compared with respect to the best lossless compression algorithms for multispectral images. Three recently invented algorithms have been considered. Numerical results produced by these algorithms allow concluding that adaptive prediction based algorithm is the best one for lossless compression of multispectral images. Real multispectral data captured from an airplane have been used for the testing.

Drug loading of mesoporous silicon particles

Porous silicon (PSi) is a promising material to be utilized in drug delivery formulations. The release rate of the drug compound can be controlled by changing the pore properties and surface chemistry of PSi. The loading of a poorly soluble drug into mesoporous silicon particles enhances its dissolution in the body. The drug loading is based on adsorption. The attainable maximum loaded amount depends on the properties of the drug compound and the PSi material, and on the process conditions. The loading solvent also essentially affects the adsorption process. The loading of indomethacin into PSi particles with varying surface modification was studied. Solvent mixtures were applied in the loading, and the loaded samples were analyzed with thermal analysis methods. The best degree of loading was obtained using a mixture of dichloromethane and methanol. The drug loads varied from 7.7 w-% to 26.8 w-%. A disturbing factor in the loading experiments was the tendency of indomethacin to form solvates with the solvents applied. In addition, the physical form and stability of indomethacin loaded in PSi and silica particles were studied using Raman spectroscopy. In the case of silica, the presence of crystalline drug as well as the polymorph form can be detected, but the method proved to be not applicable for PSi particles.

Learning Preferences with Kernel-Based Methods

Learning of preference relations has recently received significant attention in machine learning community. It is closely related to the classification and regression analysis and can be reduced to these tasks. However, preference learning involves prediction of ordering of the data points rather than prediction of a single numerical value as in case of regression or a class label as in case of classification. Therefore, studying preference relations within a separate framework facilitates not only better theoretical understanding of the problem, but also motivates development of the efficient algorithms for the task. Preference learning has many applications in domains such as information retrieval, bioinformatics, natural language processing, etc. For example, algorithms that learn to rank are frequently used in search engines for ordering documents retrieved by the query. Preference learning methods have been also applied to collaborative filtering problems for predicting individual customer choices from the vast amount of user generated feedback. In this thesis we propose several algorithms for learning preference relations. These algorithms stem from well founded and robust class of regularized least-squares methods and have many attractive computational properties. In order to improve the performance of our methods, we introduce several non-linear kernel functions. Thus, contribution of this thesis is twofold: kernel functions for structured data that are used to take advantage of various non-vectorial data representations and the preference learning algorithms that are suitable for different tasks, namely efficient learning of preference relations, learning with large amount of training data, and semi-supervised preference learning. Proposed kernel-based algorithms and kernels are applied to the parse ranking task in natural language processing, document ranking in information retrieval, and remote homology detection in bioinformatics domain. Training of kernel-based ranking algorithms can be infeasible when the size of the training set is large. This problem is addressed by proposing a preference learning algorithm whose computation complexity scales linearly with the number of training data points. We also introduce sparse approximation of the algorithm that can be efficiently trained with large amount of data. For situations when small amount of labeled data but a large amount of unlabeled data is available, we propose a co-regularized preference learning algorithm. To conclude, the methods presented in this thesis address not only the problem of the efficient training of the algorithms but also fast regularization parameter selection, multiple output prediction, and cross-validation. Furthermore, proposed algorithms lead to notably better performance in many preference learning tasks considered.

Methods for assessing the sustainability of integrated municipal waste management and

The general striving to bring down the number of municipal landfills and to increase the reuse and recycling of waste-derived materials across the EU supports the debates concerning the feasibility and rationality of waste management systems. Substantial decrease in the volume and mass of landfill-disposed waste flows can be achieved by directing suitable waste fractions to energy recovery. Global fossil energy supplies are becoming more and more valuable and expensive energy sources for the mankind, and efforts to save fossil fuels have been made. Waste-derived fuels offer one potential partial solution to two different problems. First, waste that cannot be feasibly re-used or recycled is utilized in the energy conversion process according to EU’s Waste Hierarchy. Second, fossil fuels can be saved for other purposes than energy, mainly as transport fuels. This thesis presents the principles of assessing the most sustainable system solution for an integrated municipal waste management and energy system. The assessment process includes: · formation of a SISMan (Simple Integrated System Management) model of an integrated system including mass, energy and financial flows, and · formation of a MEFLO (Mass, Energy, Financial, Legislational, Other decisionsupport data) decision matrix according to the selected decision criteria, including essential and optional decision criteria. The methods are described and theoretical examples of the utilization of the methods are presented in the thesis. The assessment process involves the selection of different system alternatives (process alternatives for treatment of different waste fractions) and comparison between the alternatives. The first of the two novelty values of the utilization of the presented methods is the perspective selected for the formation of the SISMan model. Normally waste management and energy systems are operated separately according to the targets and principles set for each system. In the thesis the waste management and energy supply systems are considered as one larger integrated system with one primary target of serving the customers, i.e. citizens, as efficiently as possible in the spirit of sustainable development, including the following requirements: · reasonable overall costs, including waste management costs and energy costs; · minimum environmental burdens caused by the integrated waste management and energy system, taking into account the requirement above; and · social acceptance of the selected waste treatment and energy production methods. The integrated waste management and energy system is described by forming a SISMan model including three different flows of the system: energy, mass and financial flows. By defining the three types of flows for an integrated system, the selected factor results needed in the decision-making process of the selection of waste management treatment processes for different waste fractions can be calculated. The model and its results form a transparent description of the integrated system under discussion. The MEFLO decision matrix has been formed from the results of the SISMan model, combined with additional data, including e.g. environmental restrictions and regional aspects. System alternatives which do not meet the requirements set by legislation can be deleted from the comparisons before any closer numerical considerations. The second novelty value of this thesis is the three-level ranking method for combining the factor results of the MEFLO decision matrix. As a result of the MEFLO decision matrix, a transparent ranking of different system alternatives, including selection of treatment processes for different waste fractions, is achieved. SISMan and MEFLO are methods meant to be utilized in municipal decision-making processes concerning waste management and energy supply as simple, transparent and easyto- understand tools. The methods can be utilized in the assessment of existing systems, and particularly in the planning processes of future regional integrated systems. The principles of SISMan and MEFLO can be utilized also in other environments, where synergies of integrating two (or more) systems can be obtained. The SISMan flow model and the MEFLO decision matrix can be formed with or without any applicable commercial or free-of-charge tool/software. SISMan and MEFLO are not bound to any libraries or data-bases including process information, such as different emission data libraries utilized in life cycle assessments.

Methods for construction and analysis of computational models in systems biology : applications to the modelling of the heat shock response and the self-assembly of intermediate filaments

Systems biology is a new, emerging and rapidly developing, multidisciplinary research field that aims to study biochemical and biological systems from a holistic perspective, with the goal of providing a comprehensive, system- level understanding of cellular behaviour. In this way, it addresses one of the greatest challenges faced by contemporary biology, which is to compre- hend the function of complex biological systems. Systems biology combines various methods that originate from scientific disciplines such as molecu- lar biology, chemistry, engineering sciences, mathematics, computer science and systems theory. Systems biology, unlike “traditional” biology, focuses on high-level concepts such as: network, component, robustness, efficiency, control, regulation, hierarchical design, synchronization, concurrency, and many others. The very terminology of systems biology is “foreign” to “tra- ditional” biology, marks its drastic shift in the research paradigm and it indicates close linkage of systems biology to computer science. One of the basic tools utilized in systems biology is the mathematical modelling of life processes tightly linked to experimental practice. The stud- ies contained in this thesis revolve around a number of challenges commonly encountered in the computational modelling in systems biology. The re- search comprises of the development and application of a broad range of methods originating in the fields of computer science and mathematics for construction and analysis of computational models in systems biology. In particular, the performed research is setup in the context of two biolog- ical phenomena chosen as modelling case studies: 1) the eukaryotic heat shock response and 2) the in vitro self-assembly of intermediate filaments, one of the main constituents of the cytoskeleton. The range of presented approaches spans from heuristic, through numerical and statistical to ana- lytical methods applied in the effort to formally describe and analyse the two biological processes. We notice however, that although applied to cer- tain case studies, the presented methods are not limited to them and can be utilized in the analysis of other biological mechanisms as well as com- plex systems in general. The full range of developed and applied modelling techniques as well as model analysis methodologies constitutes a rich mod- elling framework. Moreover, the presentation of the developed methods, their application to the two case studies and the discussions concerning their potentials and limitations point to the difficulties and challenges one encounters in computational modelling of biological systems. The problems of model identifiability, model comparison, model refinement, model inte- gration and extension, choice of the proper modelling framework and level of abstraction, or the choice of the proper scope of the model run through this thesis.