System Reliability
System reliability is an important research part in our lab, which denotes the calculation and optimization of the reliability to properly predict and make maintenance plans for a specified system such as series system, parallel system, series-parallel system, multi-state k-out-of-n systems and other complex networks. Indexes applied in our research include failure rate, mean time to failure, availability, network service reliability and so on. On the foundation of the system reliability mentioned above, we deeply combine the knowledge of operation research, optimization and statistics. Our research can be divided into several parts:
• Formulating standard regulations for system reliability indexes,
• Modelling, evaluation, prediction and optimization of complex network system reliability,
• Optimizing operation and maintenance policies for complex systems,
• Emergency rescue for complex network.
Our research has also conducted in-depth cooperation with many enterprises, such as Huawei, high-speed rail, CGN, Volkswagen and other well-known companies.
Prognostics and Health Management（PHM）
Prognostics and health management (PHM) is a multifaceted discipline that protects the integrity of components, products, and systems of systems by avoiding unanticipated problems that can lead to performance deficiencies and adverse effects on safety. More specifically, prognostics is the process of predicting a system’s remaining useful life (RUL). By estimating the progression of a fault given the current degree of degradation, the load history, and the anticipated future operational and environmental conditions, PHM can predict when a product or system will no longer perform its intended function within the desired specifications. Health management is the process of decision-making and implementing actions based on the estimate of the state of health (SOH) derived from health monitoring and expected future use of the systems. PHM techniques have advanced and matured considerably since 2008. In the Internet of Things (IoT) era, the dramatic increase of sensors, data rates, and communication capabilities continue to drive the complexity of PHM applications to new levels. Our research interests include:
• Fault diagnosis of complex industrial systems, such as high-speed trains, wind turbines, etc.
• Weakly-supervised anomaly detection
• Life cycle management of key components such as battery and hollow shaft.
Quality Control / Management
With recent advances in data acquisition technology, large amounts of quality-related data streams from multiple sensors characterizing specific processes have become available. It is highly significant to monitoring these data stream for quality control and management. However, traditional statistical process control (SPC) and statistical process monitoring (SPM) methods fail to deal with these complex data streams due to their big volume and complicated structure. Hence, our research aims to promote innovative approaches for handling such complex data that arise in the area of SPC/SPM. Our research interests include:
• Detecting anomalies for high-dimensional (HD) data, e.g., the multivariate functional data
• Monitoring functional data from multiple sources accounting for its high heterogeneity
• Statistical modeling for spatially correlated functional data
• Modeling and monitoring for mixed-type data considering the complex correlations between the mixed variables