Structural Health Monitoring (SHM), used for identifying physical and chemical changes in material properties, represents an innovative approach for the early detection of susceptibility to deterioration, such as chloride-induced corrosion. Despite the common practice of embedding sensors within concrete to monitor its material properties, the construction industry faces challenges related to what is known as the ‘Sensor Data Dilemma’ and the ‘SHM Cost Constraint’. The research reported in this thesis raises critical questions, such as how to effectively utilise large amounts of sensor data for obtaining valuable insights on the material characteristics as well as for predicting the performance of concrete in service environments. Until these questions are addressed, there is a substantial barrier to using SHM systems for durability monitoring and identifying maintenance strategies. In addition, the significant initial investment and ongoing maintenance expenses associated with SHM systems can deter their adoption, despite their long-term benefits.
Therefore, to address these challenges, this study introduces a novel Automated Clustering-based Piecewise Linear Regression – Structural Health Monitoring (ACPLR-SHM) methodology for analysing electrical resistance and temperature data obtained using sensors embedded in concrete. Such data can then be used to obtain the steady-state condition of the resistance and thereby to determine the diffusion coefficient and hence to predict the performance of concrete. The methodology was applied to sensor data from three high performance concrete blocks placed at an exposure site and from a bridge in Northern Ireland (the Abercorn Bridge), thereby addressing the 'Sensor Data Dilemma'.
The results from these case studies showed that the ACPLR-SHM methodology not only automates the analysis process but also extracts meaningful insights from the raw resistance data. The values of the diffusion coefficient thus determined from the resistance values were compared using a novel 5-category framework for the durability of concrete, signifying the robustness of the methodology in assessing the performance of concrete. Further, the results from the Abercorn Bridge case study highlighted the potential of the SHM and ACPLR-SHM methodology to identify early-age issues, such as water leakage after repairs, which might remain undetected until significant deterioration occurs or until a visual inspection is carried out. This not only signifies the practical application of the ACPLR-SHM methodology in a real-world setting but also may lead to a reduction in unnecessary major repairs and long-term costs.
To address the major challenge of high initial cost of using the SHM systems for monitoring the performance of concrete structures in service, a financial methodology was developed. It incorporated life cycle cost analysis (LCCA) and compared the costs and benefits of various SHM systems to traditional inspection, maintenance and repairs implemented on a 100-metre span concrete bridge. The findings highlighted the strategic selection of the SHM system based on project scale to optimise cost-effectiveness and safety. For instance, Smart Wireless Electrical Sensors available with advanced data analysis are the most cost-effective option for large-scale projects. The results also showed that implementation of SHM and detailed SHM analysis leads to considerable savings, especially in large-scale projects, and enhances safety by minimising the risk of accidents and operational disruptions, thereby addressing the ‘SHM Cost Constraint’.
Overall, it can be concluded that the adoption of the ACPLR-SHM methodology, complemented by a strategic financial approach through LCCA, demonstrates a significant advance in SHM for assessing the durability. This methodology can be implemented for analysing both short-term and long-term resistance data from any type of concrete under any exposure condition. It can provide a more reliable prediction of concrete performance and facilitate cost-effective maintenance strategies. Thus, this study addresses the critical industry challenge of using sensors and sensor data for valuable performance insights and economic benefits, thereby contributing to the sustainability and resilience of concrete structures.
