Periodic Reporting for period 1 - vPERFORM (Developing advanced vibration performance assessment for new generation of lightweight pedestrian structures using motion platform and virtual reality environments)
Période du rapport: 2020-10-01 au 2022-09-30
An ever-increasing use of building materials with high strength-to-weight ratios has led to development of slender and lightweight structures with dazzling structural forms, especially in case of landmark public structures such as footbridges, walkways and corridors between buildings, at airports and shopping centres. Such a trend is fuelled by the urgent need to improve sustainability through minimising use of materials in construction. It follows that structures are more sensitive to human-generated dynamic loading than ever before and their design is largely governed by vibration serviceability limit state. Assessment of structural vibration induced by human walking is currently based on loading models obtained on rigid surfaces. It is largely unknown how footstep forces would be modified due to human-structure interaction if the surfaces are vibrating. Pedestrians start interacting with vibrating structures under certain and still largely unknown conditions resulting in vibration-dependent dynamic force and unacceptably large errors in predictions of the actual vibration response. This has a serious consequence in misclassification of the bridge as being on the brink of “unacceptable structure” instead of being classified in a more favourable category of “providing mean comfort”. In a climate of fast urbanisation, only if we develop understanding of how people interact with vibrating supporting surfaces, we will be able to satisfy demand for vibration serviceable structures at minimum cost and material use and then achieve net-zero carbon emissions.
The project, vPERFORM, aims to transform the current design practice by developing reliable predictive models of vibration performance of lightweight pedestrian structures and for the first time to establish how vertical vibration influences pedestrian walking and resulting dynamic force. Human walking behaviours and ground reaction forces on vibrating surfaces are being examined to reveal the influence of the platform-to-human component of the human-structure dynamic interaction. The output of vPERFORM will advance vibration serviceability design methods for pedestrian structures to replace the existing oversimplified approach to vibration assessment, pave the way for the adoption of high strength-weight ratio materials such as timber, and finally make structures more serviceable and greener.
The project, vPERFORM, aims to transform the current design practice by developing reliable predictive models of vibration performance of lightweight pedestrian structures and for the first time to establish how vertical vibration influences pedestrian walking and resulting dynamic force. Human walking behaviours and ground reaction forces on vibrating surfaces are being examined to reveal the influence of the platform-to-human component of the human-structure dynamic interaction. The output of vPERFORM will advance vibration serviceability design methods for pedestrian structures to replace the existing oversimplified approach to vibration assessment, pave the way for the adoption of high strength-weight ratio materials such as timber, and finally make structures more serviceable and greener.
Trial tests were carried out right at VSimulators (VSim) at the University of Exeter from the beginning of the project in November 2020. As a newly built testing facility, a few technique issues needed to be solved before preparing a formal testing program. Experimental data obtained from trial tests were analysed. Based on the trial tests, a formal test plan was developed in August 2021. Ethics application was approved in September 2021. Volunteers were then recruited from students and staff at the University of Exeter.
By using VSim, I investigated the influence of a wide range of combinations of vibration amplitudes, frequencies and durations. High-fidelity data was collected by using a motion capture system and force plates. In addition, it was the first time that the metabolic cost was introduced to human-structure dynamic interaction study, which was beyond the scope of the original DoA. Additional tests were carried out at an outdoor FRP footbridge, which is used to validate the findings obtained from VSim testing.
To date, an analysis framework has been established by using MATLAB and test results are being prepared for upcoming publications. Initial results show that an increasing level of vibration results in a significant increase in step-to-step variability for most parameters of human gait like step frequency. It implies the current design method based on deterministic force models can lead to large errors. Furthermore, extra self-excited forces associated with the vibrating surface were observed, which should be included in future force models. In addition, some notable adjustments can also be found in gait parameters such as heel rise, knee flexion and step width.
In parallel, based on the previous work by my supervisor and her collaborators, I evaluated various bipedal models for walkers, which were originally developed in the research field of biomechanics and have been identified as potential candidates for modelling pedestrians in structural engineering applications. The next step is for walking on a vibrating surface and the experimental data obtained on VSim platform and the FRP footbridge will be used for validations.
vPERFORM can estimate the likelihood of the pedestrian-structure interaction (PSI) occurrence in any new structure and enable calculation of the vibration response in those cases when the PSI presents. These outputs will constitute required refinement of the existing design process that still applies the force representing walking on rigid surfaces onto the structure dynamics model to calculate the vibration response and compare it with pre-defined vibration limit criterion. At last, I will produce design charts of parameter space in which the interaction is likely to occur, which will empower structural engineers to consider the PSI modelling only when needed and avoid making unnecessary provisions (and associated monetary costs) for vibration control devices.
By using VSim, I investigated the influence of a wide range of combinations of vibration amplitudes, frequencies and durations. High-fidelity data was collected by using a motion capture system and force plates. In addition, it was the first time that the metabolic cost was introduced to human-structure dynamic interaction study, which was beyond the scope of the original DoA. Additional tests were carried out at an outdoor FRP footbridge, which is used to validate the findings obtained from VSim testing.
To date, an analysis framework has been established by using MATLAB and test results are being prepared for upcoming publications. Initial results show that an increasing level of vibration results in a significant increase in step-to-step variability for most parameters of human gait like step frequency. It implies the current design method based on deterministic force models can lead to large errors. Furthermore, extra self-excited forces associated with the vibrating surface were observed, which should be included in future force models. In addition, some notable adjustments can also be found in gait parameters such as heel rise, knee flexion and step width.
In parallel, based on the previous work by my supervisor and her collaborators, I evaluated various bipedal models for walkers, which were originally developed in the research field of biomechanics and have been identified as potential candidates for modelling pedestrians in structural engineering applications. The next step is for walking on a vibrating surface and the experimental data obtained on VSim platform and the FRP footbridge will be used for validations.
vPERFORM can estimate the likelihood of the pedestrian-structure interaction (PSI) occurrence in any new structure and enable calculation of the vibration response in those cases when the PSI presents. These outputs will constitute required refinement of the existing design process that still applies the force representing walking on rigid surfaces onto the structure dynamics model to calculate the vibration response and compare it with pre-defined vibration limit criterion. At last, I will produce design charts of parameter space in which the interaction is likely to occur, which will empower structural engineers to consider the PSI modelling only when needed and avoid making unnecessary provisions (and associated monetary costs) for vibration control devices.
This fellowship reinforces European leadership in modelling and design of vibration serviceable structures and brings it to the next level at the time when the topic attracts increasing interest worldwide (particularly from China, Canada and the USA). Although the project served for the design purpose in the structural engineering context, a multidisciplinary approach, by combining human motion science and physiology, has been implemented. Beyond the scope of the original proposal, metabolic cost was measured to reveal the energy cost of human walking on vibrating surfaces. A long continuous vibration (5 min) was studied. This is the first time that metabolic measurement was introduced to the field of human-structure interactions. The test results not only provide new evidence for changes in human walking patterns, but also help researchers in physiology understand the energy cost of human walking in vibration perturbed environments. The data can be used to study human movement, balance control or perturbation therapy.
This research leads the way in introducing more realistic, multi-sensory, factors in assessment of serviceability and comfort in civil engineering infrastructure. It will also open new avenues for development of performance-based vibration serviceability design that makes better and more sustainable use of construction materials. Recently, the necessity of sustainable building construction has led to more focus being directed towards renewable and bio-based low-carbon building materials such as timber. Although tremendous benefits can be brought to the environment and human well-being by using timber as a building material, timber floors have an unavoidable issue of vibration serviceability due to a high strength-to-weight ratio of timber, which typically governs timber design. Unfortunately, the guidance on vibration design of timber floors, especially mass timber floors, has much room for improvement. The output of vPERFROM will contribute to development of design methods for mass timber floors, pave the way for the adoption of timber, and eventually make buildings greener.
This research leads the way in introducing more realistic, multi-sensory, factors in assessment of serviceability and comfort in civil engineering infrastructure. It will also open new avenues for development of performance-based vibration serviceability design that makes better and more sustainable use of construction materials. Recently, the necessity of sustainable building construction has led to more focus being directed towards renewable and bio-based low-carbon building materials such as timber. Although tremendous benefits can be brought to the environment and human well-being by using timber as a building material, timber floors have an unavoidable issue of vibration serviceability due to a high strength-to-weight ratio of timber, which typically governs timber design. Unfortunately, the guidance on vibration design of timber floors, especially mass timber floors, has much room for improvement. The output of vPERFROM will contribute to development of design methods for mass timber floors, pave the way for the adoption of timber, and eventually make buildings greener.