One of the main characteristics of a smart grid is the availability of large volumes of data, for example, gained from sensors. This data can be utilized as a tool to estimate the state of the system as a whole or any component within the system. In order to achieve actionable information from the variety of data that is available from smart grids, it is important to use the correct mathematical and signal processing tools. Furthermore, the future smart grid is expected to have high levels of reliability. This can be achieved by integrating the condition monitoring systems with maintenance management, wherein the focus is shifted from corrective maintenance to predictive condition-based maintenance.

This chapter introduces the concept of reliability-centered asset management (RCAM). The RCAM approach provides the possibility of both qualitative and quantitative analysis toward optimal maintenance strategy. Furthermore, various issues with condition monitoring in smart grids have been discussed along with some literature that suggest possible solutions for these issues. Finally, a detailed case study of a data-based condition monitoring method based on artificial neural network is presented to demonstrate one of the many possibilities to use data from various measurement systems to reach actionable decisions.