The pantograph-catenary system (PACs) is the only channel for high-speed trains to obtain electric energy, and its state is crucial to the safety of train operation. There are two key indicators to measure the PACs state: the tension state of the catenary dropper and the zig-zag value of the PACs contact point. With the development of deep learning, on-board vision inspection equipment has gradually become an effective means of PACs state inspection. However, since more and more deep learning black boxes are used in the railway industry, the reliability estimation of their decisions has become an unavoidable challenge. As for PACs state detection, there are two other challenges: the lack of defective dropper sample and the difficulty in distinguishing between true and false contact points. To overcome the challenge of defective sample scarcity, this paper proposes to detect dropper defect by the semantic features extracted by a multi-task pantograph-catenary system state detection network (MPCN), which is trained using only normal droppers. Aiming at the contact point tracking, a multiple hypothesis tracking tree (MHTT) algorithm is proposed to track the contact points using the change rule of the contact point sequence. Meanwhile, in the semantic space, domain knowledge is combined with Bayesian Hypothesis testing to monitor the runtime reliability of the MPCN. In this way, the PACs state detection and the MPCN reliability monitoring can be implemented in a unified framework. The effectiveness of our method has been verified in various scenarios of various railway lines.

With the growth of renewable energy, electrified highways can efficiently utilize green energy such as solar and wind for EVs, promoting sustainable transportation and carbon reduction, and accelerating the transition to a greener future. For high-power DC/DC converters in electrified roadways, a lightweight and compact design is crucial, but inductors limit progress. Therefore, this study focuses on the magnetic integration of DC/DC chopping inductors. It first selected and optimized the decoupled magnetic integration form, initial electromagnetic parameters, and core sizes based on circuit topology and device specifications, using core loss and thermal rise models. Then, it determined the optimal winding turns ratio according to the air gap and magnetic resistance ratio, obtaining the final design with insulation considered. The design was verified through finite-element simulation, prototype manufacturing, and testing, and an improved optimization with interleaved parallel control was proposed. Results indicate that magnetic integration reduces the inductor’s volume by 7.93% and footprint by 38.62%, facilitating the lightweight and compact design of relevant magnetic components. With interleaved parallel control, the integrated inductor’s volume can be reduced by 19.74%, significantly decreasing the volume and mass of the chopping inductor.