System health management of safety critical systems using artificial neural networks

Safety critical systems are the systems which defines the safety and monitors the system performance based on some safety criterions. A system health management (SHM) system monitors the safety critical system while it is in operation, and thus is able to detect faults, as soon as they occur. In this paper we propose an approach to determine the system health using artificial neural network. Unmanned Aerial Vehicle is used as a case study to demonstrate the SHM approach using ANN. Artificial Neural Networks (ANN) is one of the powerful technique which helps to predict the functionality modes of a UAV and can also be used to classify the data. The system can be implemented as two important stages as neural network predictor and classifier. The MATLAB R2014. A neural network tool box is using here because it can easily meet our requirements of prediction and classification of neural networks. Even there are various critical sub systems are available in UAV safety critical systems, the considering sub system is sensor subsystem. At the output stage, by observing the performance of classifier, we obtain the idea about whether the system is healthy or not.