The present diploma work investigates visualization of multidimensional data using multilayer neuron networks. The research comprised artificial neuron network (SAMANN) for non – linear projection, and visualization of multidimensional space data to smaller measurement space. The SAMANN network is trained by error backpropagation algorithm, that is, a multilayer perceptron is trained by using gradient descent rule for minimization of accumulated square error. Programme language helped creating a language for SAMANN network realization. Practical researches were carried out, that is, plane visualization of specific data sets complex (vectors). Dependence of visualization accuracy on number of iterations, training speed parameter and number of neurons in hidden layers was established. The following software was used for the work: Microsoft Visual Studio 6.00 C++ (for realization of SAMANN network) and Microsoft Excel 2002 (for visualization of the created programme).

Biological brains exhibit many interesting and complex behaviours. Understanding of the mechanisms behind brain behaviours is critical for continuing advancement in fields of research such as artificial intelligence and medicine. In particular, synchronisation of neuronal firing is associated with both improvements to and degeneration of the brain's performance; increased synchronisation can lead to enhanced information-processing or neurological disorders such as epilepsy and Parkinson's disease. As a result, it is desirable to research under which conditions synchronisation arises in neural networks and the possibility of controlling its prevalence. Stochastic ensembles of FitzHugh-Nagumo elements are used to model neural networks for numerical simulations and bifurcation analysis. The FitzHugh-Nagumo model is employed because of its realistic representation of the flow of sodium and potassium ions in addition to its advantageous property of allowing phase plane dynamics to be observed. Network characteristics such as connectivity, configuration and size are explored to determine their influences on global synchronisation generation in their respective systems. Oscillations in the mean-field are used to detect the presence of synchronisation over a range of coupling strength values. To ensure simulation efficiency, coupling strengths between neurons that are identical and fixed with time are investigated initially. Such networks where the interaction strengths are fixed are referred to as homogeneously coupled. The capacity of controlling and altering behaviours produced by homogeneously coupled networks is assessed through the application of weak and strong delayed feedback independently with various time delays. To imitate learning, the coupling strengths later deviate from one another and evolve with time in networks that are referred to as heterogeneously coupled. The intensity of coupling strength fluctuations and the rate at which coupling strengths converge to a desired mean value are studied to determine their impact upon synchronisation performance. The stochastic delay differential equations governing the numerically simulated networks are then converted into a finite set of deterministic cumulant equations by virtue of the Gaussian approximation method. Cumulant equations for maximal and sub-maximal connectivity are used to generate two-parameter bifurcation diagrams on the noise intensity and coupling strength plane, which provides qualitative agreement with numerical simulations. Analysis of artificial brain networks, in respect to biological brain networks, are discussed in light of recent research in sleep theory.

The goal was to get acquainted with the problems of network elements to describe neural networks that can be used to manage such a feature. The theoretical part deals with the neural networks from their inception to the present. It focuses mainly on the network, witch can be used for management control. These are the two network: Hopfield network and Kohonen network. The practical part deals with the network element model and ist implementation. It contains a practical element model using a neural network, witch is controlled by a network element.

The objective of this thesis is to evaluate the possibility of creating a tool capable of predicting commodity prices. Along with other business strategies, tools and markets analyses for financial and capital markets, this tool should help make the best estimate of future developments on the observed markets. The main market, on which this work is focused, is the agricultural commodities market, namely corn and its related markets. The fundamental basis upon which the arguments in this thesis are built, is the use of artificial intelligence, particularly neural networks. The whole application is presented using a graphical user interface that allows even those with little or no understanding of this field to delve deeper into the interesting area - using modern computer systems to support trading activities.

This work deals with traveling salesman problem (TSP) and examines it’s possibilities to use in business. It is about the optimization of the travel cost, saving time and unnecessary mileage. Part of the work is a program with a GUI written in program MATLAB. Program uses neural networks to calculate the most effective path between places, where the trader has to reach. It’s possible to use the algorithm for many purposes, e.g. distribution of goods, store management, planning of PCBs or rescue services. Program communicates with the Google Maps API server, which provides the actual information of the path.

Towards a sustainable future, more and more powertrains are being electrified today, thus it is important to prevent unwanted failures and secure a reliable operation. Monitoring the internal temperatures of powertrains and keeping them under their thresholds is an important first step. Traditional modeling methods require expert knowledge and complicated modeling. With all the operating information an electric drive can collect nowadays about the whole powertrain, it becomes possible to apply black boxed machine learning to do the temperature estimating job. In this thesis, multiple machine learning algorithms are tested on their ability to estimate temperatures of the rotor fin, the stator winding, the bearing, and the power module case. The tested algorithms range from an ordinary least square to a deep neuron network. For this purpose, about 150 hours of data are recorded by letting the system run under predefined operating conditions. A hyperparameter search is also conducted for each model to find the best configuration. All the algorithms are evaluated by several metrics. It has been found that neuron networks can perform quite well even under fast transient conditions without any expert knowledge. / Mot en hållbar framtid elektrifieras fler och fler drivlinor idag, därför är det viktigt att förhindra oönskade haverier och säkra en tillförlitlig drift. Att övervaka drivlinornas interna temperaturer och hålla dem under sina trösklar är ett viktigt första steg. Traditionella modelleringsmetoder kräver expertkunskap och komplicerad modellering. Med all driftinformation som en elektrisk drivenhet kan samla in nuförtiden om hela drivlinan, blir det möjligt att tillämpa black boxed machine learning för att utföra temperaturuppskattningsjobbet. I den här avhandlingen testas flera maskininlärningsalgoritmer på deras förmåga att uppskatta temperaturer på rotorfenan, statorlindningen, lagret och kraftmodulhuset. De testade algoritmerna sträcker sig från ett vanligt minsta kvadrat till ett djupt neuronnätverk. För detta ändamål registreras cirka 150 timmars data genom att låta systemet köras under fördefinierade driftsförhållanden. En hyperparametersökning görs också för varje modell för att hitta den bästa konfigurationen. Alla algoritmer utvärderas av flera mätvärden. Det har visat sig att neuronnätverk kan fungera ganska bra även under snabba transienta förhållanden utan någon expertkunskap.

Disputats. Rheinische Friedrick-Wilhelms-Universität, 2001.

This thesis, Earth satellites and detection of air andground based activities by Ulf Ekblad of the Physics departmentat the Royal Institute of Technology (KTH), addresses theproblem of detecting military activities in imagery. Examplesof various techniques are presented. In particular, problemsassociated with "novelties" and "changes" in an image arediscussed and various algorithms presented. The imagery usedincludes satellite imagery, aircraft imagery, and photos offlying aircraft. The timely delivery of satellite imagery is limited by thelaws of celestial mechanics. This and other information aspectsof imagery are treated. It is e.g. shown that dozens ofsatellites may be needed if daily observations of a specificsite on Earth are to be conducted from low Earth orbit. New findings from bioinformatics and studies of small mammalvisual systems are used. The Intersecting Cortical Model (ICM),which is a reduced variant of the Pulse-Coupled Neural Network(PCNN), is used on various problems among which are changedetection. Still much more could be learnt from biologicalsystems with respect to pre- and post-processing as well asintermediate processing stages. Simulated satellite imagery is used for determining theresolution limit for detection of tanks. The necessary pixelsize is shown to be around 6 m under the conditions of thissimulation. Difference techniques are also tested on Landsat satelliteimagery with the purpose of detecting underground nuclearexplosions. In particular, it is shown that this can easily bedone with 30 m resolution images, at least in the case studied.Satellite imagery from SPOT is used for detecting undergroundnuclear explosions prior to the detonations, i.e. under certainconditions 10 m resolution images can be used to detectpreparations of underground nuclear explosions. This type ofinformation is important for ensuring the compliance of nucleartest ban treaties. Furthermore, the necessity for havingcomplementary information in order to be able to interpretimages is also shown. Keywords: Remote sensing, reconnaissance, sensor,information acquisition, satellite imagery, image processing,image analysis, change detection, pixel difference, neuronnetwork, cortex model, PCNN, ICM, entanglement, Earthobservation, nuclear explosion, SPOT, Landsat, verification,orbit.

<p>This thesis, Earth satellites and detection of air andground based activities by Ulf Ekblad of the Physics departmentat the Royal Institute of Technology (KTH), addresses theproblem of detecting military activities in imagery. Examplesof various techniques are presented. In particular, problemsassociated with "novelties" and "changes" in an image arediscussed and various algorithms presented. The imagery usedincludes satellite imagery, aircraft imagery, and photos offlying aircraft.</p><p>The timely delivery of satellite imagery is limited by thelaws of celestial mechanics. This and other information aspectsof imagery are treated. It is e.g. shown that dozens ofsatellites may be needed if daily observations of a specificsite on Earth are to be conducted from low Earth orbit.</p><p>New findings from bioinformatics and studies of small mammalvisual systems are used. The Intersecting Cortical Model (ICM),which is a reduced variant of the Pulse-Coupled Neural Network(PCNN), is used on various problems among which are changedetection. Still much more could be learnt from biologicalsystems with respect to pre- and post-processing as well asintermediate processing stages.</p><p>Simulated satellite imagery is used for determining theresolution limit for detection of tanks. The necessary pixelsize is shown to be around 6 m under the conditions of thissimulation.</p><p>Difference techniques are also tested on Landsat satelliteimagery with the purpose of detecting underground nuclearexplosions. In particular, it is shown that this can easily bedone with 30 m resolution images, at least in the case studied.Satellite imagery from SPOT is used for detecting undergroundnuclear explosions prior to the detonations, i.e. under certainconditions 10 m resolution images can be used to detectpreparations of underground nuclear explosions. This type ofinformation is important for ensuring the compliance of nucleartest ban treaties. Furthermore, the necessity for havingcomplementary information in order to be able to interpretimages is also shown.</p><p>Keywords: Remote sensing, reconnaissance, sensor,information acquisition, satellite imagery, image processing,image analysis, change detection, pixel difference, neuronnetwork, cortex model, PCNN, ICM, entanglement, Earthobservation, nuclear explosion, SPOT, Landsat, verification,orbit.</p>

The main task of the Master Thesis is introduction into problems of quality of service in converged networks especially with use of IP protocol version 6. Converged networks are able to transfer different data types - voice, data or multimedia stream. Design of active network unit is realized in Matlab Simulink. Designed model consists of simple network with some computer terminals which are connected with network element - switch. Switch model simulates real traffic of computer terminals, that are sending data to remote users. Packets in switch are sorting by data stream type and QoS. Switch is managed by neuron network. Neuron network reacts to input data and controls switch depending on type of recipient. Switch model can be used in laboratory exercising. Solving this theme needs basic skill in Simulink and theme can be done in one laboratory exercise.