Railway turnouts require high-performance condition monitoring to prevent disastrous railway accidents. In industrial practice, turnouts monitoring is usually done by railway workers who visually inspect the operating current curves. This results in huge labor cost, and is prone to human mistakes. Thus automating the process of turnouts monitoring via fault detection algorithms is imperative. The available turnout field data bring three difficulties to fault detection: (1) large amounts of data do not have any labels; (2) data collected in normal condition have multiple unknown modes; and (3) there are only a small number of samples in some modes. To address these difficulties, this paper develops a novel unsupervised fault detection method by using deep autoencoders, which is composed of an unknown modes mining stage and a multimode fault detection stage. First, unknown modes are identified through clustering and employing engineer expertise. Then, an ensemble monitoring model, consisting of local monitoring models developed with individual fault-free modes and a global monitoring model developed by merging the data in all fault-free modes, is proposed to improve the overall fault detection performance. Additionally, to construct local models for the modes with a small number of samples, a one-class transfer learning algorithm is presented. In online monitoring, the decision of a newly arrived sample exploits both local models and the global model. Using both the simulated turnout data and the field data collected from a high-speed railway in China, the efficacy and robustness of the proposed approach is demonstrated by comparisons with other methods.

In this article, we study the challenging few-shot fault diagnosis (FSFD) problem where limited faulty samples are available. Metric-based meta-learning methods have been a prevalent approach toward FSFD; however, most of them rely on learning a generalized distance metric and fall short of leveraging intraclass and interclass distribution information. To this end, we develop a novel reweighted regularized prototypical network to improve the performance of FSFD, where an intraclass reweighting strategy is proposed to reduce the influence of noise and outliers and obtain stable estimations of fault prototypes. In addition, a novel balance-enforcing regularization (BER) is proposed to hedge against the between-class imbalance and improve the discrimination capability. These two remedies help to reduce the intraclass difference and enlarge the interclass difference via episodic training. In this way, an improved metric space and a better diagnostic performance can be attained in a few-shot learning context. Case studies on the Tennessee Eastman benchmark process and a real-world railway turnout dataset demonstrate that the proposed FSFD approach compares favorably against state-of-the-art methodologies with desirable diagnostic performance.

A method for fault detection (FD) of networked control systems with unknown network-induced delay and unknown input based on eigendecomposition, adaptive evaluation and adaptive threshold is proposed in this paper. In the method, a known structure matrix of network-induced delay which is supposed to be smaller than one sampling period is extracted by eigendecomposition and an observer based residual generator is constructed. Then an adaptive evaluation function is proposed to ensure the optimal robustness to network-induced delay and unknown input at the evaluation stage based on the Parity Space approach. Finally, based on the estimation of the maximal influence of unknown random delay and unknown input on the evaluation function, adaptive threshold is designed to enhance robustness of the FD system and reduce the miss detection rate as much as possible.

Fault detection (FD) of networked control systems (NCS) with limited communication is addressed in this article. In periodic communication sequence-based NCS, at each time instant only parts of the actuators and sensors are allowed to communicate with the central FD system, which leads to incomplete plant information at the central station. In this article, a parity space based multi-solution FD scheme is proposed based on different information structures of the incomplete plant information. In this multi-solution scheme, each solution is an independent residual generator and has different properties. It is proved that there is a tradeoff between fast FD and detection performance, i.e. the solution with a smaller FD delay has a poorer fault sensitivity and thus a larger miss detection rate, whereas the one with a higher fault sensitivity has a larger FD delay. The FD scheme can be realised by diagnostic observers as well, which have more design freedom. Simulation results are also given to illustrate the results obtained.