This dissertation presents contact force sensors that are based on an emerging fibre-optic sensing technology, the in-fibre Bragg grating (FBG), for contact force measurements between cartilage surfaces in the human hip. There are two main motivations for force measurement in hips (and other joints). First, there is clinical evidence that suggests excessive force magnitude and duration can cause painful degeneration of joints. Second, insights from ex vivo force measurements during simulated physiologic loading are the basis of the rationale for corrective surgeries meant to halt degeneration and restore proper joint function by restoring natural joint mechanics. The current standard tools for force measurements in joints are force/stress sensitive films. There are problems associated with inserting these films into joints that affect the force/stress measurements. To insert the films, the joint must be dissected of surrounding soft tissues and, ultimately, the joint must be taken apart (disarticulated). Following disarticulation, films are fixed to cartilage surfaces, and the joint is re-assembled so that physiologic loads can be applied. The negative consequence of dissection and disarticulation is that the natural mechanics of the intact joint are permanently lost and, therefore, film measurements do not indicate the actual joint mechanics. Moreover, covering cartilage surfaces with rigid films alters the natural contact mechanics of the joint. The force sensors presented in this dissertation are designed for local force measurement over the region of the optical fibre containing the FBG and address limitations of force/stress sensitive films. The FBG force sensors are extremely small (major diameters ranging from 0.165 mm to 0.24 mm) and can be inserted into joint spaces without dissection of soft tissues and disarticulation thereby allowing the joint to remain intact. Theoretical and experimental results indicate that FBG sensor measurements are less affected by the mechanical properties of cartilage than are film sensors. The sensors presented in this dissertation also address limitations with previous FBG based force sensors and are the first application of FBGs in intact human hips. The sensors are smaller, and therefore less invasive, and insensitive to orientation, axial strain and temperature, unlike other FBG sensors presented in the literature. / Graduate

People with osteoarthritis (OA) suffer from joint degeneration and pain as well as difficulty performing daily activities. Joint contact forces (JCF) are important for understanding individual joint loading, however, these contact force cannot be directly measured without instrumented implants. Musculoskeletal modeling is a tool for estimating JCF without the need for surgery. The results from these models can be very different due to different approaches used in the development of a model that was used for simulation. Therefore, the first purpose of this study was to develop and validate a musculoskeletal model in which lower extremity JCF were calculated at the hip, knee, and ankle in 10 participants with hip OA (H-OA) and 10 healthy control participants using OpenSim 4.0 [simtk.org, 23]. The generic gait2392 model was scaled to participant demographics, then the inverse kinematics (IK) solution and kinetic data were input into the Residual Reduction Algorithm (RRA) to reduce modeling errors. Kinematic solutions from RRA were used in the Computed Muscle Control (CMC) tool to compute muscle forces, then JCF were estimated using the Joint Reaction Analysis tool. Validation included JCF comparisons to published data of similar participant samples during level walking, and movement simulation quality was assessed with residual forces and moments applied at the pelvis, joint reserve actuators, and kinematic tracking errors. The computed JCFs were similar to the overall trends of published JCF results from similar participant samples, however the values of the computed JCFs were anywhere from 0.5 times body weight (BW) to 3BW larger than those in published studies. Simulation quality assessment resulted in low residual forces and moments, and low tracking errors. Most of the reserve actuators were small as well, besides pelvis rotation and hip rotation. The computed JCF were then used in the second portion of this study to determine the effect of group and side on JCF during both the weight acceptance and push-off phases of level walking. It was determined that there was a significant difference in the knee and ankle JCF during the weight acceptance portion of stance phase and at all joints during the push-off phase when comparing the H-OA and control groups on the affected limb. A significant interaction between group and limb was found for the peak hip JCF timing (% stance) during the push-off portion of the stance phase (p=0.009). These results demonstrate that H-OA participants experience an earlier peak hip JCF during propulsion on their affected limb. Based on previous research in OA that has examined spatiotemporal measures, this finding suggests that H-OA participants may use step or stride length changes as a strategy to decrease or limit pain and loading on the affected limb. Knowledge of potential JCF differences in H-OA participants, such as timing of the peaks in either portion of the stance phase, could provide useful insight to clinicians and therapists to make decisions on how to proceed with treatment or rehabilitation programs. / Master of Science / People with osteoarthritis suffer from joint degeneration and pain as well as difficulty performing daily activities, like walking. It is important to understand the forces and loading within individual joints. Musculoskeletal modeling is one way that researchers can estimate these joint contact forces (JCF) without needing a joint replacement implant that can measure these forces. When it comes to modeling simulations, there is a wide variety of results. Therefore, the first purpose of this study was to develop and validate a musculoskeletal model in which JCFs were calculated at the hip, knee, and ankle in 10 participants with hip osteoarthritis and 10 healthy adults. Validation of the model was completed through a comparison between computed results and published data of similar participant samples during level walking. The computed results were similar to the overall trends of published JCF results, however the numerical values themselves were larger than those in published studies. The computed JCFs were then used in the second portion of this study to determine how the two groups and limbs differ during level walking. There was a significant difference in the knee and ankle JCF during the first half of the stance phase and in all joints during the second half of stance when comparing the two groups. The hip osteoarthritis participants also experience an earlier peak hip JCF during the second half of stance phase on their affected limb. This finding suggests that hip osteoarthritis participants may change the way they take a step as a strategy to decrease or limit pain and loading on the affected limb. Knowledge of potential JCF differences, such as timing of the peaks in either portion of the stance phase, could provide useful insight to clinicians and therapists to make decisions on how to proceed with treatment or rehabilitation programs.

Esta tese está dividida em três estudos. No estudo I desta tese, os objetivos foram descrever a cinemática tridimensional do ombro e determinar a contribuição da articulação escápulotorácica no movimento total de elevação. No estudo II, os objetivos foram descrever a atividade eletromiográfica (EMG) e o percentual de cocontração entre os músculos deltóide médio e redondo maior. No estudo III, o objetivo foi determinar o valor máximo da força de contato gleno-umeral e da força muscular. Quarenta sujeitos participaram dos estudos divididos em três grupos: pacientes com prótese total do ombro (PTO), com prótese reversa (PRO) e um grupo controle. Todos realizaram dois exercícios de reabilitação (flexão e elevação no plano escapular) usando diferentes cargas (sem carga externa, halter de 1kg e resistência elástica). A cinemática do ombro foi registrada por meio de um dispositivo eletromagnético de rastreamento, a atividade EMG foi registrada por meio de um eletromiógrafo de 16 canais e o modelo matemático utilizado foi o Delft Shoulder and Elbow Model. As análises estatísticas foram feitas por meio de ANOVAs de dois e três fatores para medidas repetidas. Foi utilizado o teste post hoc de Bonferroni e o nível de significância adotado foi de α < 0,05. Os resultados do estudo I mostram que a escápula contribui mais para o movimento total do ombro em pacientes com prótese em relação aos indivíduos saudáveis e em exercícios realizados com 1 kg e resistência elástica comparados com os exercícios sem carga externa. O ângulo de elevação glenoumeral durante a flexão foi significativamente maior no grupo PTO em comparação ao grupo PRO. O estudo II identificou maior atividade EMG do deltóide médio e posterior no grupo PTO em comparação ao grupo controle e um aumento da atividade EMG do peitoral (parte esternal) no grupo PRO em comparação ao grupo PTO e ao grupo controle. Para os outros músculos (deltóide anterior, redondo maior, peitoral maior - parte clavicular e serrátil anterior) não foram encontradas diferenças significativas entre os grupos. Para todos os músculos, exceto o serrátil anterior, a atividade EMG foi menor nos exercícios sem carga externa quando comparados aos exercícios com 1 kg e resistência elástica. Nenhum efeito principal dos fatores grupo e carga foi encontrado no percentual de cocontração. No estudo III, o grupo controle apresentou maior força de contato gleno-umeral máxima quando comparado ao grupo PTO durante a flexão, mas não foram encontradas diferenças entre os grupos de pacientes nos dois movimentos. A resistência elástica apresentou maiores valores de força de contato gleno-umeral em todos os grupos. O valor máximo da força de todos os músculos analisados variou de 0,32 N a 772 N e o manguito rotador e o deltóide foram os músculos que apresentaram os maiores valores de força em todos os grupos. A presente tese sugere que para uma mesma amplitude de movimento, os pacientes com prótese de ombro apresentarão um movimento escapular maior compensando a perda do movimento gleno-umeral. A cinemática escapular e a atividade EMG destes pacientes foram influenciadas pela implementação de cargas externas, mas não pelo tipo de carga, diferentemente das forças de contato e das forças musculares, nas quais a resistência elástica apresentou maior influência. Além disso, o percentual de cocontração não foi influenciado pelo tipo de prótese e os grupos de pacientes apresentaram menores forças de contato gleno-umeral do que o grupo controle. / This thesis is divided into three studies. In study I, the objectives were to describe shoulder three-dimensional kinematics and to determine the contribution of scapulothoracic motion in total shoulder elevation. In study II, the objectives were to describe the electromyographic activity (EMG) and the percentage of cocontraction between middle deltoid and teres major. In study III, the objective was to determine the maximum glenohumeral contact force and shoulder muscle forces. Forty subjects participated in the studies divided into three groups: patients with total shoulder prosthesis (TSP), patients with reverse shoulder prosthesis (RSP) and a control group. All patients realized two rehabilitation exercises (anteflexion and elevation in the scapular plane) using different loads (without external load, 1 kg dumbbell and elastic resistance). Shoulder kinematics was recorded by means of an electromagnetic tracking device, the EMG activity was recorded through a 16-channel EMG system and the Delft Shoulder and Elbow Model was used. Statistical analyses were performed by means of repeated measures ANOVAs. The Bonferroni post hoc test was used and the adopted significance level was α <0.05. Results of study I showed that the scapula contributes more to the total movement of the shoulder in patients with prosthesis compared to healthy subjects and in exercises performed with 1 kg and elastic resistance compared to exercises without external load. The glenohumeral elevation angle during anteflexion was significantly higher in the TSP group compared to the RSP. The study II identified higher EMG activity of the middle and posterior deltoid in the TSP group compared to the control group and an increase of EMG activity of pectoralis major (sternal part) in the RSP group compared to the TSP and the control group. For the other muscles (anterior deltoid, teres major, pectoralis major - clavicular part and serratus anterior) significant differences were not found among groups. For all muscles, except the serratus anterior, EMG activity was lower during exercises without external load compared to exercises with 1 kg and elastic resistance. No main effect of group and load was found in the percentage of cocontraction. In study III, the control group showed higher maximum glenohumeral contact force when compared to TSP group during anteflexion, but no differences were found between groups of patients during both movements. Elastic resistance exercises showed higher glenohumeral contact force in all groups. The maximum force of all analyzed muscles ranged from 0,32 N to 772 N and the rotator cuff and deltoid muscles presented the highest values in all groups. This thesis suggests that for a same range of motion, patients with shoulder prosthesis present greater scapular motion compensating the loss of glenohumeral motion. Scapular kinematics and EMG activity of these patients were affected by the implementation of external loads, but not by the type of load, unlike the contact forces and muscle forces, in which the elastic resistance had greater influence. Moreover, the percentage of cocontraction was not influenced by the type of prosthesis and patient groups had lower glenohumeral contact forces compared to the control group.

Esta tese está dividida em três estudos. No estudo I desta tese, os objetivos foram descrever a cinemática tridimensional do ombro e determinar a contribuição da articulação escápulotorácica no movimento total de elevação. No estudo II, os objetivos foram descrever a atividade eletromiográfica (EMG) e o percentual de cocontração entre os músculos deltóide médio e redondo maior. No estudo III, o objetivo foi determinar o valor máximo da força de contato gleno-umeral e da força muscular. Quarenta sujeitos participaram dos estudos divididos em três grupos: pacientes com prótese total do ombro (PTO), com prótese reversa (PRO) e um grupo controle. Todos realizaram dois exercícios de reabilitação (flexão e elevação no plano escapular) usando diferentes cargas (sem carga externa, halter de 1kg e resistência elástica). A cinemática do ombro foi registrada por meio de um dispositivo eletromagnético de rastreamento, a atividade EMG foi registrada por meio de um eletromiógrafo de 16 canais e o modelo matemático utilizado foi o Delft Shoulder and Elbow Model. As análises estatísticas foram feitas por meio de ANOVAs de dois e três fatores para medidas repetidas. Foi utilizado o teste post hoc de Bonferroni e o nível de significância adotado foi de α < 0,05. Os resultados do estudo I mostram que a escápula contribui mais para o movimento total do ombro em pacientes com prótese em relação aos indivíduos saudáveis e em exercícios realizados com 1 kg e resistência elástica comparados com os exercícios sem carga externa. O ângulo de elevação glenoumeral durante a flexão foi significativamente maior no grupo PTO em comparação ao grupo PRO. O estudo II identificou maior atividade EMG do deltóide médio e posterior no grupo PTO em comparação ao grupo controle e um aumento da atividade EMG do peitoral (parte esternal) no grupo PRO em comparação ao grupo PTO e ao grupo controle. Para os outros músculos (deltóide anterior, redondo maior, peitoral maior - parte clavicular e serrátil anterior) não foram encontradas diferenças significativas entre os grupos. Para todos os músculos, exceto o serrátil anterior, a atividade EMG foi menor nos exercícios sem carga externa quando comparados aos exercícios com 1 kg e resistência elástica. Nenhum efeito principal dos fatores grupo e carga foi encontrado no percentual de cocontração. No estudo III, o grupo controle apresentou maior força de contato gleno-umeral máxima quando comparado ao grupo PTO durante a flexão, mas não foram encontradas diferenças entre os grupos de pacientes nos dois movimentos. A resistência elástica apresentou maiores valores de força de contato gleno-umeral em todos os grupos. O valor máximo da força de todos os músculos analisados variou de 0,32 N a 772 N e o manguito rotador e o deltóide foram os músculos que apresentaram os maiores valores de força em todos os grupos. A presente tese sugere que para uma mesma amplitude de movimento, os pacientes com prótese de ombro apresentarão um movimento escapular maior compensando a perda do movimento gleno-umeral. A cinemática escapular e a atividade EMG destes pacientes foram influenciadas pela implementação de cargas externas, mas não pelo tipo de carga, diferentemente das forças de contato e das forças musculares, nas quais a resistência elástica apresentou maior influência. Além disso, o percentual de cocontração não foi influenciado pelo tipo de prótese e os grupos de pacientes apresentaram menores forças de contato gleno-umeral do que o grupo controle. / This thesis is divided into three studies. In study I, the objectives were to describe shoulder three-dimensional kinematics and to determine the contribution of scapulothoracic motion in total shoulder elevation. In study II, the objectives were to describe the electromyographic activity (EMG) and the percentage of cocontraction between middle deltoid and teres major. In study III, the objective was to determine the maximum glenohumeral contact force and shoulder muscle forces. Forty subjects participated in the studies divided into three groups: patients with total shoulder prosthesis (TSP), patients with reverse shoulder prosthesis (RSP) and a control group. All patients realized two rehabilitation exercises (anteflexion and elevation in the scapular plane) using different loads (without external load, 1 kg dumbbell and elastic resistance). Shoulder kinematics was recorded by means of an electromagnetic tracking device, the EMG activity was recorded through a 16-channel EMG system and the Delft Shoulder and Elbow Model was used. Statistical analyses were performed by means of repeated measures ANOVAs. The Bonferroni post hoc test was used and the adopted significance level was α <0.05. Results of study I showed that the scapula contributes more to the total movement of the shoulder in patients with prosthesis compared to healthy subjects and in exercises performed with 1 kg and elastic resistance compared to exercises without external load. The glenohumeral elevation angle during anteflexion was significantly higher in the TSP group compared to the RSP. The study II identified higher EMG activity of the middle and posterior deltoid in the TSP group compared to the control group and an increase of EMG activity of pectoralis major (sternal part) in the RSP group compared to the TSP and the control group. For the other muscles (anterior deltoid, teres major, pectoralis major - clavicular part and serratus anterior) significant differences were not found among groups. For all muscles, except the serratus anterior, EMG activity was lower during exercises without external load compared to exercises with 1 kg and elastic resistance. No main effect of group and load was found in the percentage of cocontraction. In study III, the control group showed higher maximum glenohumeral contact force when compared to TSP group during anteflexion, but no differences were found between groups of patients during both movements. Elastic resistance exercises showed higher glenohumeral contact force in all groups. The maximum force of all analyzed muscles ranged from 0,32 N to 772 N and the rotator cuff and deltoid muscles presented the highest values in all groups. This thesis suggests that for a same range of motion, patients with shoulder prosthesis present greater scapular motion compensating the loss of glenohumeral motion. Scapular kinematics and EMG activity of these patients were affected by the implementation of external loads, but not by the type of load, unlike the contact forces and muscle forces, in which the elastic resistance had greater influence. Moreover, the percentage of cocontraction was not influenced by the type of prosthesis and patient groups had lower glenohumeral contact forces compared to the control group.

Esta tese está dividida em três estudos. No estudo I desta tese, os objetivos foram descrever a cinemática tridimensional do ombro e determinar a contribuição da articulação escápulotorácica no movimento total de elevação. No estudo II, os objetivos foram descrever a atividade eletromiográfica (EMG) e o percentual de cocontração entre os músculos deltóide médio e redondo maior. No estudo III, o objetivo foi determinar o valor máximo da força de contato gleno-umeral e da força muscular. Quarenta sujeitos participaram dos estudos divididos em três grupos: pacientes com prótese total do ombro (PTO), com prótese reversa (PRO) e um grupo controle. Todos realizaram dois exercícios de reabilitação (flexão e elevação no plano escapular) usando diferentes cargas (sem carga externa, halter de 1kg e resistência elástica). A cinemática do ombro foi registrada por meio de um dispositivo eletromagnético de rastreamento, a atividade EMG foi registrada por meio de um eletromiógrafo de 16 canais e o modelo matemático utilizado foi o Delft Shoulder and Elbow Model. As análises estatísticas foram feitas por meio de ANOVAs de dois e três fatores para medidas repetidas. Foi utilizado o teste post hoc de Bonferroni e o nível de significância adotado foi de α < 0,05. Os resultados do estudo I mostram que a escápula contribui mais para o movimento total do ombro em pacientes com prótese em relação aos indivíduos saudáveis e em exercícios realizados com 1 kg e resistência elástica comparados com os exercícios sem carga externa. O ângulo de elevação glenoumeral durante a flexão foi significativamente maior no grupo PTO em comparação ao grupo PRO. O estudo II identificou maior atividade EMG do deltóide médio e posterior no grupo PTO em comparação ao grupo controle e um aumento da atividade EMG do peitoral (parte esternal) no grupo PRO em comparação ao grupo PTO e ao grupo controle. Para os outros músculos (deltóide anterior, redondo maior, peitoral maior - parte clavicular e serrátil anterior) não foram encontradas diferenças significativas entre os grupos. Para todos os músculos, exceto o serrátil anterior, a atividade EMG foi menor nos exercícios sem carga externa quando comparados aos exercícios com 1 kg e resistência elástica. Nenhum efeito principal dos fatores grupo e carga foi encontrado no percentual de cocontração. No estudo III, o grupo controle apresentou maior força de contato gleno-umeral máxima quando comparado ao grupo PTO durante a flexão, mas não foram encontradas diferenças entre os grupos de pacientes nos dois movimentos. A resistência elástica apresentou maiores valores de força de contato gleno-umeral em todos os grupos. O valor máximo da força de todos os músculos analisados variou de 0,32 N a 772 N e o manguito rotador e o deltóide foram os músculos que apresentaram os maiores valores de força em todos os grupos. A presente tese sugere que para uma mesma amplitude de movimento, os pacientes com prótese de ombro apresentarão um movimento escapular maior compensando a perda do movimento gleno-umeral. A cinemática escapular e a atividade EMG destes pacientes foram influenciadas pela implementação de cargas externas, mas não pelo tipo de carga, diferentemente das forças de contato e das forças musculares, nas quais a resistência elástica apresentou maior influência. Além disso, o percentual de cocontração não foi influenciado pelo tipo de prótese e os grupos de pacientes apresentaram menores forças de contato gleno-umeral do que o grupo controle. / This thesis is divided into three studies. In study I, the objectives were to describe shoulder three-dimensional kinematics and to determine the contribution of scapulothoracic motion in total shoulder elevation. In study II, the objectives were to describe the electromyographic activity (EMG) and the percentage of cocontraction between middle deltoid and teres major. In study III, the objective was to determine the maximum glenohumeral contact force and shoulder muscle forces. Forty subjects participated in the studies divided into three groups: patients with total shoulder prosthesis (TSP), patients with reverse shoulder prosthesis (RSP) and a control group. All patients realized two rehabilitation exercises (anteflexion and elevation in the scapular plane) using different loads (without external load, 1 kg dumbbell and elastic resistance). Shoulder kinematics was recorded by means of an electromagnetic tracking device, the EMG activity was recorded through a 16-channel EMG system and the Delft Shoulder and Elbow Model was used. Statistical analyses were performed by means of repeated measures ANOVAs. The Bonferroni post hoc test was used and the adopted significance level was α <0.05. Results of study I showed that the scapula contributes more to the total movement of the shoulder in patients with prosthesis compared to healthy subjects and in exercises performed with 1 kg and elastic resistance compared to exercises without external load. The glenohumeral elevation angle during anteflexion was significantly higher in the TSP group compared to the RSP. The study II identified higher EMG activity of the middle and posterior deltoid in the TSP group compared to the control group and an increase of EMG activity of pectoralis major (sternal part) in the RSP group compared to the TSP and the control group. For the other muscles (anterior deltoid, teres major, pectoralis major - clavicular part and serratus anterior) significant differences were not found among groups. For all muscles, except the serratus anterior, EMG activity was lower during exercises without external load compared to exercises with 1 kg and elastic resistance. No main effect of group and load was found in the percentage of cocontraction. In study III, the control group showed higher maximum glenohumeral contact force when compared to TSP group during anteflexion, but no differences were found between groups of patients during both movements. Elastic resistance exercises showed higher glenohumeral contact force in all groups. The maximum force of all analyzed muscles ranged from 0,32 N to 772 N and the rotator cuff and deltoid muscles presented the highest values in all groups. This thesis suggests that for a same range of motion, patients with shoulder prosthesis present greater scapular motion compensating the loss of glenohumeral motion. Scapular kinematics and EMG activity of these patients were affected by the implementation of external loads, but not by the type of load, unlike the contact forces and muscle forces, in which the elastic resistance had greater influence. Moreover, the percentage of cocontraction was not influenced by the type of prosthesis and patient groups had lower glenohumeral contact forces compared to the control group.

The density of single grain states in static granular packings is derived from first principles for an idealized yet fundamental case. This produces the distribution of contact forces P_f(f) in the packing. Because there has been some controversy in the published literature over the exact form of the distribution, this dissertation begins by reviewing the existing empirical observations to resolve those controversies. A method is then developed to analyze Edwards' granular contact force probability functional from first principles. The derivation assumes Edwards' flat measure -- a density of states (DOS) that is uniform within the metastable regions of phase space. A further assumption, supported by physical arguments and empirical evidence, is that contact force correlations arising through the closure of loops of grains may be neglected. Then, maximizing a state-counting entropy results in a transport equation that can be solved numerically. For the present it has been solved using the "Mean Structure Approximation," projecting the DOS across all angular coordinates to more clearly identify its predominant features in the remaining stress coordinates. These features are: (1) the Grain Factor related to grain stability and strong correlation between the contact forces on the same grain, and (2) the Structure Factor related to Newton's third law and strong correlation between neighboring grains. Numerical simulations were then performed for idealized granular packings to permit a direct comparison with the theory, and the data including P_f(f) were found to be in excellent agreement. Where the simulations and theory disagree, it is primarily due to the coordination number Z because the theory assumes Z to be a constant whereas in disordered packings it is not. The form of the empirical DOS is discovered to have an elegant, underlying pattern related to Z. This pattern consists entirely of the functional forms correctly predicted by the theory, but with only slight parameter changes as a function of Z. This produces significant physical insight and suggests how the theory may be generalized in the future.

The focus of this study is on the development of a state-of-the-art single-wheel roller rig for studying contact mechanics and dynamics in railroad applications. The use of indoor-based simulation tools has become a mainstay in vehicle testing for the automotive and railroad industries. In contrast to field-testing, roller rigs offer a controlled laboratory environment that can provide a successful path for obtaining data on the mechanics and dynamics of railway systems for a variety of operating conditions. The idea to develop a laboratory test rig started from the observation that there is a need for better-developed testing fixtures capable of accurately explaining the complex physics of wheel-rail contact toward designing faster, safer, and more efficient railway systems. A review of current roller rigs indicated that many desired functional requirements for studying contact mechanics currently are not available. Thus, the Virginia Tech Railway Technologies Laboratory (RTL) has embarked on a mission to develop a state-of-the-art testing facility that will allow experimental testing of contact mechanics in a dynamic, controlled, and consistent manner. VT roller rig will allow for closely replicating the boundary conditions of railroad wheel-rail contact via actively controlling all the wheel-rail interface degrees of freedom: cant angle, angle of attack, and lateral displacement. Two sophisticated independent drivelines are configured to precisely control the rotational speed of the wheels, and therefore their relative slip or creepage. A novel force measurement system, suitable for steel on steel contact, is configured to precisely measure the contact forces and moments at the contact patch. The control architecture is developed based on the SynqNet data acquisition system offered by Kollmorgen, the motors supplier. SynqNet provides a unified communication protocol between actuators, drives, and data acquisition system, hence eliminating data conversion among them. Various design analysis indicates that the rig successfully meets the set requirements: additional accuracy in measurements, and better control on the design of experiments. The test results show that the rig is capable of conducting various contact mechanics studies aimed for advancing the existing art. Beyond developing the experimental testing fixture for studying contact mechanics, this study provides a comprehensive review of the contact models. It discusses the simplifying assumptions for developing the models, compares the models functionality, and highlights the open areas that require further experimental and theoretical research. In addition, a multi-body dynamic model of the entire rig, using software package SIMPACK, is developed for conducting modal analysis of the rig and evaluating the performance of the rig's components. A MATLAB routine is also developed that provides a benchmark for developing creep curves from measurements of the rig and comparing them with existing creep curves. / Ph. D.

The primary aim of this research was to investigate the bending behaviour of helically wound steel cables of various types (i.e. normal spiral strands, sheathed spiral strands and locked coil cables) in the presence of friction and to propose more efficient computational models for their analysis under combined tension and bending. The proposed model fully takes into account interwire contact forces both in the radial direction (point contact between wires in different layers) and hoop direction (line contact within the wires in the same layer). Extensive theoretical parametric studies have been undertaken on a variety of cable constructions covering a wide range of geometrical and material parameters. Explicit formulations have been developed for the smooth transition of the bending stiffness from no-slip to full slip regimes, as a function of cable curvature. Based on these formulations, it is now possible to calculate the relative displacements of the wires, as well as the tensile, bending and hoop stresses in the individual wires of the cable. Furthermore, bending stiffness of the cable is shown to decrease by a factor of 2 to 16, depending upon the friction coefficient between wires and the type of cable construction. Wherever possible, the theoretical results have been compared with experimental results from the available literature and are found in very good agreement with them. A simple method for the determination of the bending stiffness of large diameter multi-layered cable has been developed. The simplified method is further shown to provide estimates of the bending stiffness which are very close to those calculated by the original theory, allowing hand calculations for an easier use in industry. The proposed formulations have been extended to cater for the effects of external hydrostatic pressure on sheathed spiral strands in deep water applications. These forces are shown to have a great influence on the pattern of interwire contact forces and hence the interlayer slippage between the wires in the strand. Numerical results have been obtained and analysed for three different 127 mm diameter strands with lay angles of 12°, 18° and 24° respectively, experiencing a wide range of external hydrostatic pressures of 0 to 2,000 metres. The significant increase in normal contact force between wires is shown to suppress the slippage of wires in the cable. However, the no-slip and full slip values of the effective bending stiffness of the cable is shown to be independent of the level of hydrostatic pressure. A theoretical model is also proposed for estimating wire kinematics, pattern of interwire slippage, contact forces as well as the flexural rigidity of locked coil cables with outer layers made of shaped wires. In order to validate this model, numerical results are reported for two different locked coil cables. It is shown that the shaped wires in the outer layers of locked coil cables play an important role in the distribution of contact forces, slip initiation and cable unwinding.

Sistemas dinâmicos ferroviários são modelados levando-se em consideração as forças de contato roda trilho. As equações de movimento são fortemente influenciadas pela determinação dessas forças, o que requer o conhecimento da mecânica de contato, ferramenta imprescindível na previsão do comportamento de uma composição ferroviária. O rodeiro é responsável por receber as excitações na forma de irregularidades e imperfeições da via e transferi-la para a suspensão primária. A proposta dessa dissertação consiste em realizar simulações no domínio do tempo, da resposta do comportamento dinâmico do rodeiro ferroviário de 2 graus de liberdade. O modelo matemático equivalente é desenvolvido. A rigidez de contato e a conicidade são linearizados em torno de um ponto de operação, próximo ao centro da via. É apresentada a influência da velocidade na estabilidade para o modelo linearizado e para a conicidade variando de 0.1 a 1.0. Os respectivos modos de vibrar também são apresentados. Em seguida, o modelo matemático não linear do rodeiro ferroviário é gerado por software especializado na criação de sistemas de multicorpos, VAMPIRE, que permite a criação das equações de movimento através da topologia do sistema. O modelo é criado seguindo a proposta do benchmarck apresentada pela International Association of Vehicle System Dynamics (IAVSD). O rodeiro recebe uma força lateral crescente que se estabiliza em 20kN. O rodeiro é excitado com a aplicação de outra força lateral a uma taxa de 50kN/s até o descarrilamento. As forças longitudinais, laterais que aparecem no contato, o ângulo do plano de contato, ângulo de yaw e afastamento lateral são comparados com o benchmarck e outros autores. / Railway dynamics systems are modeled regarding the creep forces between wheel-rail. The equations of movement are strongly influenced by these forces, what requests the knowledge of the theory in mechanical of contact, necessary tool to foresee the behavior of a railway vehicle. The wheelset is responsible by receiving the perturbations of irregularities and transfer them to the primary suspension. A railway vehicle is composed by primary and secondary suspensions. The proposal is based on performing simulations in the time domain, dynamic mresponse of a railway wheelset of two degrees of freedom. The mathematical model is developed. The stiffness of contact and conicity are linearized around an operation point, near to track center. The influence of speed is determined to linearized model and to a range of 0.1 to 1.0 in the conicity. The eigenvectors are presented. In the following, the non linear model of the wheelset is generated using a specialized package of multi body system, VAMPIRE. The package creates equations of movement after the generation of the topology. The non linear mathematical model are defined by the benchmark proposed by International Association of Vehicle System Dynamics (IAVSD). The wheelset is submitted to 20kN at the rail level. In a second case, the wheelset is submitted to a rate of 50kN/s of lateral force until derailing. The determined longitudinal and lateral forces on contact, yaw angle and displacement of the center of gravity of the wheelset, contact angle are compared with the benchmark results and other authors.

Sistemas dinâmicos ferroviários são modelados levando-se em consideração as forças de contato roda trilho. As equações de movimento são fortemente influenciadas pela determinação dessas forças, o que requer o conhecimento da mecânica de contato, ferramenta imprescindível na previsão do comportamento de uma composição ferroviária. O rodeiro é responsável por receber as excitações na forma de irregularidades e imperfeições da via e transferi-la para a suspensão primária. A proposta dessa dissertação consiste em realizar simulações no domínio do tempo, da resposta do comportamento dinâmico do rodeiro ferroviário de 2 graus de liberdade. O modelo matemático equivalente é desenvolvido. A rigidez de contato e a conicidade são linearizados em torno de um ponto de operação, próximo ao centro da via. É apresentada a influência da velocidade na estabilidade para o modelo linearizado e para a conicidade variando de 0.1 a 1.0. Os respectivos modos de vibrar também são apresentados. Em seguida, o modelo matemático não linear do rodeiro ferroviário é gerado por software especializado na criação de sistemas de multicorpos, VAMPIRE, que permite a criação das equações de movimento através da topologia do sistema. O modelo é criado seguindo a proposta do benchmarck apresentada pela International Association of Vehicle System Dynamics (IAVSD). O rodeiro recebe uma força lateral crescente que se estabiliza em 20kN. O rodeiro é excitado com a aplicação de outra força lateral a uma taxa de 50kN/s até o descarrilamento. As forças longitudinais, laterais que aparecem no contato, o ângulo do plano de contato, ângulo de yaw e afastamento lateral são comparados com o benchmarck e outros autores. / Railway dynamics systems are modeled regarding the creep forces between wheel-rail. The equations of movement are strongly influenced by these forces, what requests the knowledge of the theory in mechanical of contact, necessary tool to foresee the behavior of a railway vehicle. The wheelset is responsible by receiving the perturbations of irregularities and transfer them to the primary suspension. A railway vehicle is composed by primary and secondary suspensions. The proposal is based on performing simulations in the time domain, dynamic mresponse of a railway wheelset of two degrees of freedom. The mathematical model is developed. The stiffness of contact and conicity are linearized around an operation point, near to track center. The influence of speed is determined to linearized model and to a range of 0.1 to 1.0 in the conicity. The eigenvectors are presented. In the following, the non linear model of the wheelset is generated using a specialized package of multi body system, VAMPIRE. The package creates equations of movement after the generation of the topology. The non linear mathematical model are defined by the benchmark proposed by International Association of Vehicle System Dynamics (IAVSD). The wheelset is submitted to 20kN at the rail level. In a second case, the wheelset is submitted to a rate of 50kN/s of lateral force until derailing. The determined longitudinal and lateral forces on contact, yaw angle and displacement of the center of gravity of the wheelset, contact angle are compared with the benchmark results and other authors.