Centre for Infrastructure Systems Engineering (CISE)
Infrastructure Systems Engineering covers activities across a wide range of length scales aimed at sustainable infrastructure delivery and management. Research encompasses the characterisation and modelling of damage processes through asset life-cycles, the simulation of complex structural systems under cyclic, dynamic and extreme loads, and the performance assessment and management of large infrastructure networks.
- Innovative construction materials and their durability
- Reliability assessment of large infrastructure systems
- Asset performance & management of potable water networks
- Advanced FE modelling & testing of structural systems under static/dynamic/fatigue loading
- Structural glass & glass facades
- Robustness & progressive collapse of structures
- Structural health monitoring and damage assessment
Dynamic and extreme loads
Sudden removal of columns caused by a blast or impact introduces instability in the building frame. Adjacent floors lose their bearing abruptly. This instability propagates and can be seen as a wave of buckling columns that propagates outwards and later upwards. Building instability resulting from an immediate column removal requires dynamics analysis. This work is part of a larger effort led by Dr Szyniszewski to understand the behavior of complex phenomena through the use of high-fidelity simulations. More information can be found on Dr Szyniszewski's website.
Civil infrastructure monitoring and analytics (CIMA)
Digital technologies, including sensing technologies, telecommunication engineering, and data science, have enormous potential to transform the construction industry. Monitoring data analytics, whether physics-based or data driven, can improve maintenance efficiency and optimise asset life. At Surrey, we have so far investigated a range of issues related to the application of such techniques in metallic bridges, pipelines and other structures, bringing together expertise in materials and structures, as well as statistics, informatics and decision theory. Our collaboration with various industry partners enables us to analyse real data from critical infrastructure assets and to influence the development of industry guidelines on inspection and maintenance.
Dr Ying Wang has recently been awarded an EPSRC grant: An integrated physics-based and data-driven approach to structural condition identification, which aims to develop an integrated algorithm to create a reliable and effective approach for structural health monitoring, which can find different applications.
- Prof Gerry Parke
- Prof Marios Chryssanthopoulos
- Dr Alex Hagen-Zanker
- Dr Boulent Imam
- Dr Juan Sagaseta
- Dr Stergios Mitoulis
- Dr Mike Mulheron
- Prof Paul Smith
- Dr David Jesson
- Prof Mark Lawson
- Dr Ying Wang
Prof Dimitra Achillopoulou
Offshore wind turbines (OWT) have become hugely important to modern societies as they strive to meet their future targets of sustainable and environmentally friendly energy utilization. With the advantages OWT turbines offer with respect to the foregoing, comes the challenges of the ends of their initial design lives, fast approaching as well as an insufficient track record. This inadvertently will lead to unanticipated breakdowns in need of unplanned maintenance, which are carried out at huge costs as well as unnecessary downtime. Also, the nature of forces (i.e. cyclic and dynamic) that an OWT experiences as well as the environment (both sea and supporting soil) where it operates, make it difficult to accurately capture, ab-initio, the conditions it may be exposed to, while in operation. These thus warrant a monitoring system capable of keeping track of the conditions of OWT systems at regular intervals or better still, in real time. Other researchers have developed monitoring methods targeting specific damage types and limited to the first two damage detection levels. In this sense, this research embarks on the development of calibrated FE models to be used for the assessment of the long term effect(s) of vibration on the structural integrity of OWT systems and sets the stage towards the automation of the monitoring procedures (i.e. Digital Twins) for this purpose. This strategy involves building Finite Element Models of OWTs; Experimental modal testing of OWTs; Updating of the Finite Element Models and finally, linking the updated FE models with their physical counterparts. With the developed FE model, the structural states of the physical OWTs can be interrogated after long term load applications for the identification of damage in the OWT structure
The project is developing a generic Bayesian decision analysis framework to quantify the expected value of information from inspections on a structure. The influence of selected system effects on the analysis is being investigated. These include:
o Dependencies in Degradation Modelling
Fatigue is the primary damage mechanism that is considered. Bayesian data analysis is being used to fit predictive models that coherently and consistently account for variability within and dependency between model parameters. In addition, the effect of dependency between uncertain inputs is also being considered.
o Improved Characterisation of Imperfect Information
The project is developing improved methods of characterising inspection activities in terms of the precision, bias and relevance of the information that they provide, as well as the risks and expected costs of completion. This will allow a more comprehensive evaluation regarding the extent to which they facilitate improved risk management strategy.
o Spatial Dependencies
Where dependencies can be quantified between structural condition at different locations on a structure, the project will quantify the additional value of information that results from any inference (updating) that can be completed at locations other than those directly inspected.
Urban growth and the associated reduction in surface permeability affects the hydrological cycle and can increase flood risk. Understanding past and future patterns of urban growth is therefore an important aspect of flood risk management. Earth observation data is widely used to classify urban land. However, when classified maps are used to characterise change over time, the uncertainty of the classification can hinder meaningful analysis. This study aims to improve the mapping and analysis of patterns of urban growth, by using and developing methods for classification and pattern analysis that exploit the longitudinal nature of time series of earth observation imagery and are thereby less sensitive to inherent uncertainty.
Digital technologies, including sensing technologies, data transmission technologies, and data science, have enormous potential to transform the construction industry. Monitoring data analytics can help to improve the maintenance efficiency and optimise asset life.
New machine learning algorithms will be investigated in order to find a suitable algorithm that will help in analysing the data obtained from civil infrastructure. The main objective of the project will be the interpretation of the sensing data acquired from the structural health monitoring system installed on civil infrastructure systems such as metallic bridges, pipelines, and offshore wind turbines. Machine learning algorithm will help to provide an assessment on the condition of the infrastructure and assist in the decision-making for several aspects, for example, predictive maintenance and life extension.
Experimental works in the laboratory will be conducted for the validation of the proposed algorithm. In addition, further investigation of real structures will be conducted, through the collaboration with industry/asset owners.
Globally, concrete is the most widely used man-made construction material. This is due to its combination of flexibility of form, wide availability, and balance of cost and strength when compared to alternative materials (P. Kumar Mehta et al, 2006). The low tensile strength and strain capacity are some of the undesirable properties of concrete as they can cause failure to occur shortly after the formation of the first crack (Amir Alani et al, 2013). These properties of mass concrete have led to the use of steel reinforcement (rebars) to improve its tensile properties. However, the use of steel reinforcement carries durability implications such as the corrosion of the embedded steel among others (Per Jahren et al, 2014). An alternative method of concrete reinforcement has been the introduction of short-fibre into the concrete, examples include steel, polypropylene and glass fibres. The addition of these fibres to concrete influences the manner in which cracks develop (Amir Alani et al, 2013) as well as increasing the strain capacity at peak load, and provides significant energy absorption in the post-peak portion of the load vs. deflection curve (L.N.Vairagade et al, 2015).
A number of research studies have looked at how the introduction and dosage of fibres in a concrete mix affect mechanical properties, and there exists a body of guidance documents on the use of short-fibre steel, polymer and glass reinforced concrete. However, only a limited amount of research has considered the effect of mixing different fibre types to form mixed fibre composites. To date there has been no coherent study of the “hybrid” effect of using combinations of short-fibre reinforcement with significantly different stiffness and strength in order to identify combinations that yield the best properties (toughness, bond strength, etc.) and the potential structural benefits of such hybrid composites. Hence, an investigation into the benefits of hybrid, short-fibre reinforced concrete and its use for structural advantage needs to be conducted. The research focus on short fibres thus requires that these materials be optimized for use in large engineering components and structures to commercial and engineering advantage. To adequately perform structural element optimization, extensive material testing will be conducted comparing the performance of plain concrete and concrete containing short fibres such as steel, polypropylene, glass, etc. In addition to this, testing of large elements will be conducted to investigate component property. In terms of the material properties of the concrete the samples tested will be of the scale 10–20 cm whereas a scale of about 1–3 m will be used for structural elements. The experimental results will be compared with the output of finite element models to gain a further understanding of the structural implications of short fibre additions.
The outcome of this study will provide a better understanding of the structural performance of short-fibres and their potential for cost optimization in construction through a reduced requirement for link reinforcement, a faster construction cycle and also a better structural performance