A Visual Model-Based Evaluation Framework of Cloud-Based Prognostics and Health Management

Nowadays, systems for various smart city industries such as city power plants, public transportation and integrated modular avionics are becoming increasingly complicated, with most of components being embedded and geologically distributed. In order to ensure the availability of target systems, prognostic and health management develops as an important system to assess the current health state and estimate the future health state based on modern Industrial Internet of Things (IIoT) and cloud computing technologies. The early design of PHM systems requires system engineering methodology to ensure the robustness and effectiveness of systems. However, as the complexity grows, there may be some misunderstanding in specification that are not obvious and it is hard to perform verification and evaluation in a more intuitive way. In order to enhance the intuitiveness and reduce the misunderstanding in early development stage, we present a visual model-based framework to simulate and evaluate the cloud-based prognostics and health management systems. The framework proposes a three-abstraction-layer hierarchical architecture to represent distributed data sources and cloud-based PHM service center. The design of the framework is based on system modeling language and allows flexible algorithm implementations of function modules. The simulation of framework is visualized through the animated behavior diagrams. Furthermore, we propose a performance evaluation and measurement method combining quality of maintenance analysis and cost-benefit analysis for PHM systems.