Chapter 6 An Ontological Approach to Structural Health Monitoring George P. Tsialiamanis, David J. Wagg, I. Antoniadou, and K. Worden Abstract In the current work, an ontological framework for structural health monitoring (SHM) is discussed. Ontologies are used in disciplines like knowledge engineering and natural language processing, but their structure and goals also fit the purposes of SHM. In SHM projects – as in all projects – many problems arise during knowledge sharing and application. Ontologies can deal with these problems and at the same time have more benefits for SHM processes, as their modularity may assist in extending and transferring knowledge. An SHM-specific ontology is constructed here and described; It contains many objects that can be used in the procedure of monitoring structures. The ontology can also be used as a database to store data acquired, but also serves as a knowledge-base for the current discipline’s algorithms and methods. Further, having close connections to object-oriented programming, ontologies straightforwardly facilitate software development and reusability of their components. Certainly, the ontology can be used to save time during the application of SHM, but also can be applied to improve performance of existing methods, by finding within the ontology the best algorithm to fit the purpose of each method. Keywords Structural health monitoring (SHM) · Ontologies · Database · Knowledge-base · Knowledge engineering 6.1 Introduction In modern societies, everyday activities are becoming increasingly dependent on structural and mechanical systems. These systems have a design life which is heavily dependent on the external conditions that they will be subjected to throughout their operation. Since it is critical that they survive their design while remaining operational and safe, a framework to ensure both operability and safety is required. With this purpose in mind, structural health monitoring (SHM) can be employed. SHM refers to the process of implementing a damage detection strategy for aerospace, civil or mechanical engineering infrastructure [1]. Application of SHM to systems can be partitioned into the following steps [2]: (1) observation of the system during its operation, (2) data acquisition from the system and, (3) extraction of features that are sensitive to damage and determine the current state of system’s health. The steps of SHM can be implemented in many ways. More specifically, the final two steps of feature extraction and current state evaluation have been implemented with various methods that come from different scientific disciplines like signal processing, machine learning, physical modelling, etc. All these methods interact with each other and have their advantages and disadvantages. Being motivated by this and by the fact that in many projects problems arise from poor communication between various project members and difficulty in knowledge sharing [3], this paper is proposing a connecting framework for these components and for sharing knowledge within the SHM process. The proposed framework here is an ontological one. Ontologies are used in many fields including computer science, semantics and natural language processing. The Ontology’s purpose of sharing knowledge, developing software modules and interoperability between different projects fits the needs of SHM and can be exploited to improve performance of such applications. In the current work, an SHM ontology is constructed and discussed in the context of using it to better understand its component functions: using it as a database, to apply SHM techniques more efficiently and to implement software according to the ontology. Although applicable to SHM in a broad sense, an ontology could be particularly useful in a Population-based SHM (PBSHM) setting [4–8], where the goal is to develop general inference tools across a population. Here an ontology would G. P. Tsialiamanis ( ) · D. J. Wagg · I. Antoniadou · K. Worden Dynamics Research Group, Department of Mechanical Engineering, University of Sheffield, Sheffield, UK e-mail: g.tsialiamanis@sheffield.ac.uk; david.wagg@sheffield.ac.uk; i.antoniadou@sheffield.ac.uk; k.worden@sheffield.ac.uk © The Society for Experimental Mechanics, Inc. 2021 B. Dilworth (ed.), Topics in Modal Analysis & Testing, Volume 8, Conference Proceedings of the Society for Experimental Mechanics Series, https://doi.org/10.1007/978-3-030-47717-2_6 51
RkJQdWJsaXNoZXIy MTMzNzEzMQ==