DRES: Data recovery for condition monitoring to enhance system reliability

The system reliability depends heavily on the sensed condition data which are mainly collected by various types of sensors. The missing or faulty condition data can result in wrong decision-making or lead to system fault. To realize data integrity for system condition monitoring, one data-driven framework for recovering condition data is proposed in this article. The proposed model is combined by mutual information and Multivariable Linear Regression (MLR). The correlations among condition monitoring data sets are firstly analysed by mutual information. Then, MLR is utilized to recover condition monitoring data. A case study of aircraft engine condition monitoring data sets which are offered by National Aeronautics and Space Administration Ames Research Center is carried out to evaluate the performance of the data-driven framework.