Incipient fault detection and diagnosis (FDD) is a key technology for enhancing safety and reliability of high-speed trains. This paper develops a real-time incipient FDD method named deep principal component analysis (DPCA) for electrical drive in high-speed trains. This method can effectively detect incipient faults in electrical drive before they develop into faults or failures. This scheme adopting multivariate statistics is composed of multiple data processing layers to extract more accurate signal features of electrical drive, which exhibits several salient advantages: 1) It can establish precise data models containing both systematic and noise information of electrical drive, which are helpful for incipient fault detection; 2) the incipient faults are described by multicharacteristics which can improve the fault diagnosis ability; 3) it can be easily implemented even if the system models and parameters of electrical drive are unknown. The effectiveness and feasibility of the proposed FDD scheme are authenticated via a mathematical analysis and validated via two experiments. Results of two experiments show that the missing alarm rate and detection delay by using the proposed DPCA-based FDD method are less than 10% and 0.68 s, respectively. In comparison with the standard PCA-based FDD method, the proposed DPCA-based FDD method can show its superiorities by the detailed performance comparisons.

In this paper, a novel adaptive fault-tolerant sliding-mode control scheme is proposed for high-speed trains, where the longitudinal dynamical model is focused, and the disturbances and actuator faults are considered. Considering the disturbances in traction force generated by the traction system, a dynamic model with actuator uncertainties modeled as input distribution matrix uncertainty is established. Then, a new sliding-mode controller with design conditions is proposed for the healthy train system, which can drive the tracking error dynamical system to a predesigned sliding surface in finite time and maintain the sliding motion on it thereafter. In order to deal with the actuator uncertainties and unknown faults simultaneously, the adaptive technique is combined with the fault-tolerant sliding-mode control design together to guarantee that the asymptotical convergence of the tracking errors is achieved. Furthermore, the proposed adaptive fault-tolerant sliding-mode control scheme is extended to the cases of the actuator uncertainties with unknown bounds and the unparameterized actuator faults. Finally, the case studies on a real train dynamic model are presented to explain the developed fault-tolerant control scheme. The simulation results show the effectiveness and feasibility of the proposed method.

Considering the challenge of deep mining of root causes in machine failures, a knowledge aggregation fault diagnosis (KAFD) model is proposed, in which the graph convolutional network (GCN) GraphSAGE is improved and introduced into the knowledge graph (KG)-based fault diagnosis. Historical maintenance data of machines is used to construct a fault phenomenon-FBG, which is then combined with the fault diagnosis knowledge graph (FDKG) to form a collaborative FDKG. A single-layer knowledge aggregation network (KAN) that incorporates sensitivity factors and configures different types of GCN aggregators is constructed in the proposed KAFD. Based on deep neighbor aggregation operations on collaborative FDKG, KAFD obtained by stacking multiple KANs, can capture the higher order structural information and semantic information, which results in the multihop reasoning, improvement of the rationality and diversity of fault cause tracing. The KAFD is experimentally validated through two fault diagnosis datasets, which are constructed by the maintenance data of an industrial enterprise, and the results demonstrate the excellent performance.

This article proposes a distributed adaptive fault-tolerant formation–containment control with prescribed performance for heterogeneous multiagent systems (MASs) consisting of multiple unmanned aerial vehicles (UAVs) and unmanned ground vehicles (UGVs) in the presence of actuator faults. First, utilizing the neighborhood formation error information, the distributed fault-tolerant formation control strategy is developed for the trajectory dynamics of each UAV to achieve the formation tracking, that is, all UAVs track the virtual leader and perform the prespecified formation configuration. Then, the adaptive fault-tolerant containment algorithm, independent of the positions of the leaders, is proposed to guarantee the UGVs converge to the convex hull formed by the leader UAVs. The adaptive estimation scheme is constructed to compensate for the unknown system parameters and actuator loss-of-effectiveness and bias faults. The formation–containment tracking performance is analyzed based on Lyapunov theory with the synchronization errors satisfying the prescribed performance. A simulation example based on UAVs–UGVs systems is adopted to verify the effectiveness of the proposed control strategy.

This paper proposes a stator-winding incipient shorted-turn fault detection method for the traction motors used in China high-speed railways. First, a mathematical description for incipient shorted-turn faults is given from the quantitative point of view to preset the fault detectability requirement. Then, an interval sliding mode observer is proposed to deal with the uncertainties caused by measuring errors from motor speed sensors. The active robust residual generator and the corresponding passive robust threshold generator are proposed based on this particularly designed observer. Furthermore, design parameters are optimized to satisfy the fault detectability requirement. This developed technique is applied to an electrical traction motor to verify its effectiveness and practicability.