The theory of a rigid-perfectly plastic solid when impact loaded was used to relate the dynamic yield strength and MIBANT bend angle of driven steel fasteners. Static yield strength values for pallet nails were found to values (p > 0.0001 be linearly related to the dynamic 2 and R² = 0.950) and hence MIBANT bend angles. Static yield strength values were also related to MIBANT bend angles (p > 0.209 and R² = 0.896) for staples. The Vickers hardness number and carbon content of pallet nails are related to material properties. Vickers hardness number was found to be linearly related to static yield strength(? > 0.0001 and R² = 0.835). Vickers hardness number was also related to and MIBANT bend angle ( P > 0.0001 and R² = 0.845). Carbon content was logarithmically related to static 2 yield strength of hardened nails (P > 0.0001 and R² = C.93) and linearly related to static yield strength of non-hardened nails (P > 0.0001 and R²) Carbon content was related to logarithm of the bend angle of hardened nails (P > 0.0001 and R² = 0.781) and non-hardened nails (P > 0.0081 and R² = 0.486). Problems emanating from the impact test procedure are discussed. Recommendations include utilizing the BRAT tester for impact tests on driven fasteners for wood. / M.S.

In fighter jets there are many different systems that make the plane fly, and one of these is the fuel system consisting of multiple parts including tanks and tubes. During flight, the pressure in these tubes can vary and occasionally pressure spikes can occur. Pressure spikes can be described as high increased pressure at a high rate and decreased pressure at an equally high rate during a limited time span. Depending on the peak pressure and duration of this pressure spike, damage or failure might occur in the tube structure. This is a problem that SAAB is analyzing and wants to find a less conservative approach of analyzing the effect of pressure spikes regarding the structural strength of tubes. In this master’s thesis this is explored. A method for applying pressure spikes to two different tube structures has been created. The pressure spike propagates through the liquid and applies the pressure spike to the tube through Mortar contact. The tube is modelled with shell elements and normal Lagrangian element formulation. The liquid is modelled with solidelements and Arbitrary Lagrangian Eulerian element formulation. The two different tube structures analyzed in this report have the same tube dimensions, a combination of the biggest radius and smallest wall thickness that occur in SAAB’s tubes. The difference between the two is that one structure is straight whilst the other is curved. A parametric study was carried out to analyze at what pressure peak different durations of the pressure spikes would result in a critical effective plastic strain. This critical effective plastic strain was chosen to be 0.08 for the aluminum tube. This is conservative since the fracture strain for the material is 0.12. Results froma static case was compared with the dynamic results from the described method to see how conservative this new method of pressure spike analysis is. The results from the comparison showed that the dynamic method allowed 65% higher pressure peak thanthe static solution. Using a pressure spike with 1ms duration and pressure peak at 19.5MP a in the straight tube compared with the internal pressure of 11.4MP a in the static method for the straight tube. For the curved tube, the dynamic method allowed 90% higher peak pressure for a pressure spike with 1ms duration and peak pressure of 18.5MP a. This pressure spike in the curved tube is compared with the static method for the curved tube which reached critical effective plastic strain at 9.4MP a. For pressure spikes with durations of 20ms in straight tubes, the dynamic and static results are similar. For pressure spikes with durations of 10ms in curved tubes, the dynamic and static results are similar. In these cases, it is reasonable to use the quicker static method instead of the new method of pressure spike analysis. The impulses in both the straight tube and the curved tube cases have a linear relationships with the duration of the pressure spike. For the longer durations the energy needed to reach critical deformation is higher since the affected area is larger. Shorter durations need less energy to reach critical deformation since the affected area is smaller.

Ce mémoire est le résultat d'une recherche purement exploratoire concernant la définition d'une application de réseaux de neurones à base de transistors monoélectroniques (Single-Electron Transistor, SET). Il dresse un portait de l'état de l'art actuel, et met de l'avant la possibilité d'associer les SET avec la technologie actuelle (Field Electron Transistor, FET). La raison de cette association est que les SET peuvent être perçus comme un moyen de changement de paradigme, c'est-à-dire remplacer une fonction CMOS occupant une grande place par un dispositif alternatif présentant de meilleures performances ou équivalentes. Par l'intermédiaire de leurs caractéristiques électriques peu ordinaires au synonyme de"l'effet de blocage de Coulomb", les SET ont le potentiel d'être exploités intelligemment afin de tirer profit sur la consommation énergétique essentiellement. Cette problématique est présentée comme une des propositions alternatives"Beyond CMOS" aux termes de la diminution géométrique des transistors FET à la lumière de l'ITRS. Cette recherche propose d'exposer des circuits électroniques de technologie MOS complétés à l'aide de SET (circuits hybrides) et de montrer que l'on est capable de les remplacer ou les compléter (partiellement) dans des architectures à réseau de neurones. Pour cela, des simulations sous logiciel Cadence Environnement permettront de valider le comportement des circuits sur plusieurs critères tels que la vitesse de réponse et la consommation énergétique, par exemple. En résultat, seront proposées deux architectures à réseaux de neurones de fonctions différentes : une architecture Winner-Take-All et un générateur de spikes en tension. La première étant inspirée d'une publication provenant de GUIMARAES et al., veut démontrer qu'à partir d'une architecture SET existante, il est envisageable de se l'approprier et de l'appliquer aux paramètres des SET du CRN[indice supérieur 2] augmentant donc nos chances de pouvoir les concevoir dans notre groupe de recherche. Le second axe est la simulation d'un circuit capable de générer des signaux à spikes sans perte d'information, ce qui requerrait un nombre considérable de transistors FET sans l'utilisation de SET, mettant donc en valeur la réduction de composants.

The goal of this thesis is to investigate the functional role of the cortical column architecture within some of the existing brain coding theories. Here I focus on the hierarchical models of predictive coding and the 'phase of firing' coding hypothesis. Using an oddball paradigm consisting of a sequence of identical sounds interspersed with rare, unexpected sounds, one can observe a difference between the scalp potentials evoked by oddball and common sounds. This difference has been linked to predictive coding and novelty detection, and Stimulus Specific Adaptation (SSA) has been suggested as a likely substrate at the single neuron level. In order to simultaneously constrain hierarchical models of predictive coding, and so as to investigate the contributions that neural processing within the different cytoarchitectonic layers of the primary auditory cortex (A1) may make to SSA, I simultaneously recorded multi-unit activity and current source density (CSD) profiles across all layers in A1 of the rat in response to standard and oddball tones. Our results suggest that SSA arises at the level of the thalamocortical synapse and is further enhanced in the supragranular layers. The phase of low-frequency Local Field Potentials (LFPs) in primary sensory cortices carries stimulus related information and disambiguates the information about different stimuli evoking similar spike rates. However, it is yet unclear how these informative LFP phase values arise within the laminar organization of cortical columns. To address this issue, I performed CSD recordings in the area A1 of anaesthetized rats during the presentation of complex naturalistic sounds. Information theoretic analysis revealed that most LFP phase information originates from discrete CSD events consisting of strong granular-superficial-layer dipoles, likely triggered by bursts of thalamocortical activation. These events, which occur at rates of 2-4 Hz, reliably reset LFP phases at times of strong network excitation. They therefore provide a useful reference frame to measure neural activity with respect to salient times of stimulus history. CSD events display a diverse, stimulus-dependent morphology: these reflect the outcomes of cortical computations which result in varying extents of activation of infragranular output layers.

The Swing options are the natural outcomes of the increasing uncertainty in the power markets, which came along with the deregulation process triggered by the UK government&rsquo / s action in 1990 to privatize the national electricity supply industry. Since then, the ways of handling the risks in the price generation process have been explored extensively. Producer-consumers of the power market felt confident as they were naturally hedged against the price fluctuations surrounding the large consumers. Companies with high power consumption liabilities on their books demanded tailored financial products that would shelter them from the upside risks while not preventing them from benefiting the low prices. Furthermore, more effective risk management practices are strongly dependent upon the successful parameterization of the underlying stochastic processes, which is also key to the effective pricing of derivatives traded in the market. In this thesis, we refer to the electricity spot price model developed jointly by Hambly, Howison and Kluge ([13]), which incorporates jumps and still maintains the analytical tractability. We also derive the forward curve dynamics implied by the spot price model and explore the effects on the forward curve dynamics of the spikes in spot price. As the main discussion of this thesis, the Grid Approach, which is a generalization of the Trinomial Forest Method, is applied to the electricity Swing options. We investigate the effects of spikes on the per right values of the Swing options with various number of exercise rights, as well as the sensitivities of the model-implied prices to several parameters.

Submitted by Luanna Matias (lua_matias@yahoo.com.br) on 2015-03-11T18:57:33Z No. of bitstreams: 2 Dissertação - Lorena Lopes Ferreira - 2013.pdf: 653277 bytes, checksum: 4c9aa7d38ca40c4e856a2bcb3c40a6f2 (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Approved for entry into archive by Luanna Matias (lua_matias@yahoo.com.br) on 2015-03-11T20:01:36Z (GMT) No. of bitstreams: 2 Dissertação - Lorena Lopes Ferreira - 2013.pdf: 653277 bytes, checksum: 4c9aa7d38ca40c4e856a2bcb3c40a6f2 (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Made available in DSpace on 2015-03-11T20:01:36Z (GMT). No. of bitstreams: 2 Dissertação - Lorena Lopes Ferreira - 2013.pdf: 653277 bytes, checksum: 4c9aa7d38ca40c4e856a2bcb3c40a6f2 (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) Previous issue date: 2013-09-20 / Conselho Nacional de Pesquisa e Desenvolvimento Científico e Tecnológico - CNPq / Ticks are external parasites distributed in the world. Its parasitism causes direct and indirect damage to their hosts. Rhipicephalus microplus is the principal ectoparasite of the livestock causing economic and sanitary implications in the tropics regions. Control strategies are still a challenge and new alternatives are being sought to avoid the use of chemical methods that are reprehensible in relation to animal and human health, and also the environment. It is known that Bos indicus are more resistant to ticks than Bos taurus, while their crosses have a moderate resistance. The olfactory and gustatory senses allow these parasites to seek hosts, mating partners and distinguish between food and harmful substances. The receptors of these sensory organs are present throughout the body and appendages of arthropods, being pore sensilla in the chelicerae important when it comes to ticks taste receptors. Electrophysiology assists in the discovery of active principles that attract or repel these parasites and this will be proven after a behavioral test. This work was performed in order to characterize R. microplus electrophysiological responses to phagostimulants in different concentrations like salts (NaCl and KCl, 10-3 a 10-1M), glucose (10-4 a 10-1M), adenosine triphosphate (ATP, 10-6 a 10-2M), reduced glutathione (GSH, 10-6 a 10-2M), a mixture of substances (MIX, glucose1M+ATP10- 2M+GSH10-3M) and sera of susceptible and resistant cattle to ticks. A single cell electrophysiology technique was used in the taste sensila of R. microplus. Engorged nymphs of R. microplus were collected from naturally infested bovines. Females ticks were fed on rabbits for a period of three to five days. The females were removed and taken directly to the electrophysiology room in order to be prepared for the test. At least 10 females were tested per substance. The shape, frequency and amplitude of spikes were analyzed with the program Autospike. The statistical analysis was performed with the software R applying ANOVA followed by Tukey HSD post-hoc (5%). In relation to the phagostimulants it was observed that the taste receptors of the chelicerae were responsive for all substances tested, having a more intense responses at higher concentrations. The frequency of responses to salts ranged from 11.09 ± 0.98 to 28.50 ± 3.49 spikes/s; to glucose from 25.40 ± 2.33 to 37.15 ± 4.47 spikes/s, to ATP from 17.91 ± 1.28 to 37.30 ± 4.77 spikes/s; to GSH from 19.10 ± 2.84 to 56.6 ± 5.08 spikes/s and the MIX 50.60 ± 4.42 spikes/s. The responses were uni or multicellular depending on the concentration. Sera from resistant cattle xiii triggered a higher frequency of spikes/s (average of 53.48 ± 1.89 spikes/s, ranging from 48.45 ± 4.28 to 61.05 ± 5.14 spikes/s) when compared with the susceptible cattle (average of 40.33 ± 1.63 spikes/s, ranging from 33.30 ± 3.02 to 49.05 ± 5.20 spikes/s). The records analyzed showed that R. microplus was more stimulated by sera from resistant cattle than those from susceptible animals. It is known that in the serum has phagostimulants which probably triggered spikes in both groups. On the other hand, the greater number of spikes to resistant bovine serum might be due to the presence of deterrents compounds. Both sera showed a multicellular response. It was also observed that there were variations in the spikes frequency between animals of the same group according to the characteristic of spikes recorded. It was concluded that R. microplus respond in a dose dependent pattern to the phagostimulants tested. And also that, this tick was able differentiated the sera of susceptible to sera of resistant bovines. / Carrapatos são ectoparasitas obrigatoriamente hematófagos de diversas espécies e estão distribuídos mundialmente. Seu parasitismo acarreta danos diretos e indiretos a seus hospedeiros. Rhipicephalus microplus é o principal ectoparasita da pecuária causando problemas econômicos e sanitários nas regiões tropicais e subtropicais. O seu controle ainda é um desafio e novas alternativas estão sendo buscadas para evitar o uso de métodos químicos que são condenáveis em relação à saúde animal, humana e também ao meio ambiente. Sabe-se que os bovinos Bos indicus são mais resistentes que os Bos taurus, enquanto que seus cruzamentos possuem uma resistência moderada. A olfação e gustação permitem que esses parasitas busquem por hospedeiros, parceiros para cópula e distingam entre alimentos e substâncias nocivas a eles. Há receptores desses sentidos distribuídos por todo o corpo e apêndices dos artrópodes, sendo as porossensilas das quelíceras dos carrapatos importantes quando se refere a receptores gustativos. A eletrofisiologia auxilia na descoberta de compostos químicos que atraem ou repelem esses parasitos. Este trabalho foi feito com intuito de caracterizar as respostas eletrofisiológicas das porossensilas de R. microplus frente a alguns fagoestimulantes em diferentes concentrações como sais (NaCl e KCl, 10-3 a 10-1M), glicose (10-4 a 10-1M), adenosina trifosfato (ATP, 10-6 a 10-2M), glutationa reduzida (GSH, 10-6 a 10-2M) e uma mistura de substâncias (MIX, Glicose1M + ATP10-2M + GSH10-3M). Objetivou ainda avaliar as respostas eletrofisiológicas para soros de bovinos sensíveis (Girolando) e resistentes (Nelore) ao carrapato. Para isso foi utilizada a técnica de registro em sensila única nas porossensilas de fêmeas de R. microplus. Fêmeas alimentadas em coelhos por três a cinco dias foram usadas. Foram testadas no mínimo dez fêmeas por concentração e substância. Após o registro, analisou-se a forma, frequência e amplitude de spikes com o auxílio do programa Autospike. A análise estatística foi feita com o software R aplicando ANOVA seguido de Teste de Tukey com nível mínimo de significância de 5%. As porossensilas foram responsivas a todos os fagoestimulantes testados, mostrando resposta mais intensa nas maiores concentrações. A frequência de spikes/s para os sais variou de 11,09 ± 0,98 a 28,50 ± 3,49; para glicose 25,40 ± 2,33 a 37,15 ± 4,47; para o ATP de 17,91 ± 1,28 a 37,30 ± 4,77; para o GSH de 19,10 ± 2,84 a 56,6 ± 5,08 e o MIX de 50,60 ± 4,42. As respostas foram uni ou multicelulares. Os soros dos bovinos resistentes xi desencadearam uma maior frequência de spikes/s (média de 53,48 ± 1,89 spikes/s, variando 48,45 ± 4,28 a 61,05 ± 5,14 spikes/s) quando comparado com o grupo dos soros dos bovinos sensíveis (média de 40,33 ± 1,63 spikes /s, variando de 33,30 ± 3,02 a 49,05 ± 5,20 spikes/s). Os registros analisados mostraram que as porossensilas de R. microplus foram mais estimuladas pelo soro de animais resistentes do que aqueles de animais sensíveis. A maior frequência de spikes desencadeada pelos soros de bovinos resistentes pode ter ocorrido pela presença de compostos deterrentes além dos fagoestimulantes existentes no soro. Para ambos os soros foi observada uma resposta multicelular. Observou-se também que dentro de um mesmo grupo houve variações em relação à frequência dos spikes, que foi atribuído à resistência dos animais. Concluiu-se que as porossensilas de R. microplus responderam de forma dose dependente aos fagoestimulantes testados e também apresentaram uma maior atividade para soros de bovinos resistentes do que soros de bovinos sensíveis.

Lorsqu'on effectue une tâche, les circuits neuronaux doivent représenter et manipuler des stimuli continus à l'aide de potentiels d'action discrets. On suppose communément que les neurones représentent les quantités continues à l'aide de leur fréquence de décharge et ceci indépendamment les un des autres. Cependant, un tel codage indépendant est inefficace puisqu'il exige la génération d'un très grand nombre de potentiels d'action pour atteindre un certain niveau de précision. Dans ces travaux, on montre que les neurones d'un réseau récurrent peuvent apprendre - à l'aide d'une règle de plasticité locale - à coordonner leurs potentiels d'actions afin de représenter l'information avec une très haute précision tout en déchargeant de façon minimale. La règle d'apprentissage qui agit sur les connexions récurrentes, conduit à un codage efficace en imposant au niveau de chaque neurone un équilibre précis entre excitation et inhibition. Cet équilibre est un phénomène fréquemment observer dans le cerveau et c'est un principe central de notre théorie. On dérive également deux autres règles d'apprentissages biologiquement plausibles qui permettent respectivement au réseau de s'adapter aux statistiques de ses entrées et d'effectuer des transformations complexes et dynamiques sur elles. Finalement, dans ces réseaux, le stochasticité du temps de décharge d'un neurone n'est pas la signature d'un bruit mais au contraire de précision et d'efficacité. Le caractère aléatoire du temps de décharge résulte de la dégénérescence de la représentation. Ceci constitue donc une interprétation radicalement différente et nouvelle de l'irrégularité trouvée dans des trains de potentiels d'actions. / When performing a task, neural circuits must represent and manipulate continuous stimuli using discrete action potentials. It is commonly assumed that neurons represent continuous quantities with their firing rate and this independently from one another. However, such independent coding is very inefficient because it requires the generation of a large number of action potentials in order to achieve a certain level of accuracy. We show that neurons in a spiking recurrent network can learn - using a local plasticity rule - to coordinate their action potentials in order to represent information with high accuracy while discharging minimally. The learning rule that acts on recurrent connections leads to such an efficient coding by imposing a precise balance between excitation and inhibition at the level of each neuron. This balance is a frequently observed phenomenon in the brain and is central in our work. We also derive two biologically plausible learning rules that respectively allows the network to adapt to the statistics of its inputs and to perform complex and dynamic transformations on them. Finally, in these networks, the stochasticity of the spike timing is not a signature of noise but rather of precision and efficiency. In fact, the random nature of the spike times results from the degeneracy of the representation. This constitutes a new and a radically different interpretation of the irregularity found in spike trains.

Les oscillations jouent un rôle de premier plan dans la mise en place des réseaux cérébraux sains et pathologiques. En particulier, au niveau clinique, les activités oscillatoires sont d'une grande importance diagnostique en épilepsie. Par ailleurs, les méthodes non-invasives d'électrophysiologie sont particulièrement adaptées pour la compréhension des réseaux cérébraux à grande échelle. Cependant, la majorité des études en épilepsie a été dirigée vers les pointes intercritiques, qui sont des activités transitoires. Une question qui reste donc en suspens est le lien entre les pointes épileptiques et les activités oscillatoires épileptiques. Cette thèse a visé à résoudre deux problématiques complémentaires autour de cette question. La première problématique est la séparation adéquate entre les activités oscillatoires et transitoires. Il s'agit d'une tâche difficile surtout lors d'un grand chevauchement temporel, qui peut résulter en la contamination d'une activité par l'autre. Nous avons évaluée trois méthodes de filtrage : le filtre FIR (méthode classique), la transformé d'ondelette stationnaire et le filtrage parcimonieux par matching pursuit (MP, basé sur un dictionnaire). Sur des simulations, la SWT a donné de très bons résultats pour la reconstruction des transitoires et le MP pour les oscillations ; de plus, les deux méthodes ont donné un faible taux de faux positifs en détection automatique des oscillations. La SWT et le FIR ont donné les meilleurs résultats de filtrage sur les signaux réels, en particulier lors de la localisation de source. / The Oscillatory activities play a leading role in the development of healthy and pathological brain networks. In particular, at the clinical level, the oscillatory activities are of great importance in the diagnostic of epilepsy. In addition, the non-invasive electrophysiology methods are particularly suitable for understanding the large-scale brain networks. However, most studies in epilepsy have been directed to the interictal spikes, which are transitional activities. One issue that remains unresolved is the relationship between epileptic spikes and epileptic oscillatory activities. This thesis resolves two complementary problems. The first one is the suitable separation between the oscillatory and transitory activity, which is quite sensitive to the presence of the overlap in the time-frequency domain. This can lead to a contamination between the activities. We did evaluate three filtering methods: the FIR (classic methods), the stationary wavelet SWT and the parsimonious filter with the matching pursuit MP. The SWT gave good results in the reconstruction of transient activity and the MP in the reconstruction of oscillatory activity both for simulated data; also they provide a low false positive in automatic detection of oscillatory activity. The SWT and FIR gave the best results on real signals especially for source localization. In the simulated data, the MP is optimal since the atoms of the dictionary resembles to the simulated signals, which isn't guaranteed for real signals. The second problem is the comparison between network connectivity of transient and oscillatory activity, as measured in surface recordings (MEG) and invasive recordings SEEG.

Plusieurs études de simulation in vitro et in vivo ont été réalisées au cours des dernières décennies afin de clarifier les interrelations des sources corticales avec leurs corrélats électrophysiologiques enregistrés sur l’EEG invasif et l’EEG de scalp. L’amplitude des potentiels corticaux, l’étendue de l’aire corticale impliquée par la décharge, de même que la localisation et la géométrie de la source corticale sont des facteurs indépendants qui modulent l’observabilité et la contribution de ces sources sur l’EEG de surface. L’enregistrement simultané et multi-échelle de l’EEG de scalp et intra-crânien (avec des électrodes sous-durales ou profondes) durant l’exploration pré-chirurgicale des patients épileptiques offre une opportunité unique d’explorer cette question fondamentale. Alors que les études précédentes ont considéré essentiellement des sources néocorticales dans le contexte de l’épilepsie du lobe temporal, notre travail s’est intéressé à l’observabilité et la contribution de sources profondes temporales et frontales. Nous avons pu montrer : (1) que les sources épileptiques profondes enregistrées dans les régions temporales médianes et fronto-basales ne sont pas visibles par l’analyse visuelle de routine mais sont détectables après élimination du bruit de fond physiologique généré par les sources corticales de surface sus-jacentes ; (3) que l’amplitude des pointes enregistrées en surface est corrélée avec la surface d’activation corticale de la convexité et avec des ratios élevés d’amplitude pointes/activité de fond / Several in vitro, in vivo, and simulation studies have been performed in the past decades aiming to clarify the interrelations of cortical sources with their scalp and invasive EEG correlates. The amplitude ratio of cortical potentials to their scalp EEG correlates, the extent of the cortical area involved in the discharge, as well as the localization of the cortical source and its geometry, have been each independently linked to the recording of the cortical discharge with scalp electrodes. Simultaneous multiscale EEG recordings with scalp, subdural and depth electrodes, applied in presurgical epilepsy workup, offer an excellent opportunity to address this fundamental issue. Whereas past studies have considered predominantly neocortical sources in the context of temporal lobe epilepsy, the present work addresses deep sources, in mesial temporal and extra-temporal epilepsies. We showed that deep sources, such as those in mesial temporal or fronto-basal regions, are not visible, but are detectable in scalp EEG. Scalp EEG spikes correlate with extensive activation of the cortical convexity and high spike-to-background amplitude ratios

L’EEG de surface permet l'étude spatio-temporelle de l’activité cérébrale avec une résolution temporelle élevée, cependant elle souffre de la forte atténuation du champ électrique propagée par l'os du crâne et de la présence de sources de bruits externes. De ce fait, nous souhaitons exploiter les mesures issues de la Stéréo-EEG (SEEG). Cette modalité consiste en l'introduction d'électrodes d'enregistrement au plus près des générateurs, bénéficiant ainsi d'un rapport signal à bruit bien supérieur à celui observé en EEG. Nous proposons ainsi dans cette thèse une étude de faisabilité de l'imagerie de sources à partir de ces mesures, basée sur une méthode d'inversion de type dipôle équivalent associée à un modèle de propagation à une sphère, capable d'amener à une précision de localisation de l'ordre de quelques millimètres. A partir d'une implantation clinique usuelle de la SEEG, nous évaluons les performances de localisation lorsque différents sous-ensembles de capteurs sont considérés. En présence de bruit réaliste, nous constatons que l'ajout de capteurs lointains peut amener à une dégradation de la précision de localisation. Ces conclusions nous amènent à proposer une approche de sélection des capteurs locaux dans le but d'optimiser la fiabilité des résultats. Les atouts et faiblesses de cette approche sont analysés dans un cadre de simulation réaliste afin d'explorer de façon pertinente les différents paramètres pouvant influer sur la qualité de résolution du problème inverse. Les approches sont appliquées sur des enregistrements SEEG récoltés au CHRU de Nancy afin de confronter les méthodes de localisation proposées à des mesures réelles / The surface EEG makes it possible to study the brain activity with a high temporal resolution, however it suffers from the severe attenuation of the electrical propagation through the skull bone as well as the addition of external artifacts. As an alternative, we would like to exploit the Stereo-EEG (SEEG) recordings, consisting in shaft electrodes implanted in the brain volume in the direct vicinity of the brain generators. These data benefit from a high signal to noise ratio compared to this observed in surface EEG. We propose in this thesis a feasibility study of source imaging from the SEEG, based on an equivalent current dipole inversion method associated with an analytical One-Sphere propagation model, able to bring localization precision of the order of a few millimeters. Using a typical clinical electrode implantation, we evaluate the localization performance when different subsets of sensors are considered. In the presence of realistic noise, we observe that the addition of distant sensors with respect to the source can lead to a degradation of the localization accuracy. These conclusions lead us to propose a local sensor selection approach in order to optimize the reliability of the results. The strengths and weaknesses of this approach are analyzed on a realistic simulation framework, for a relevant exploration of the different parameters impacting on the quality of the SEEG source imaging. The approaches are applied on SEEG recordings collected at the CHRU of Nancy to evaluate their performance when facing real measurements