With the widespread adoption of wireless communication technologies in modern high-speed rail systems, the Train-to-Ground (T2G) communication system for Electric/Diesel Multiple Units (EMU/DMU) has become essential for train operation monitoring and fault diagnosis. However, this system is increasingly vulnerable to various cyber-physical threats, necessitating more intelligent and adaptive security protection mechanisms. This paper presents an intelligent security defense framework that integrates intrusion detection, risk scoring, and response mechanisms to enhance the security and responsiveness of the T2G communication system. First, feature selection is performed on the TON_IoT dataset to develop a Dream Optimization Algorithm (DOA)-optimized backpropagation neural network (DOA-BPNN) model for efficient anomaly detection. A Bayesian risk scoring module then quantifies detection outcomes and classifies risk levels, improving threat detection accuracy. Finally, a Q-learning-based reinforcement learning (RL) module dynamically selects optimal defense actions based on identified risk levels and attack patterns to mitigate system threats. Experimental results demonstrate improved performance in both multi-class and binary classification tasks compared to conventional methods. The implementation of the Bayesian risk scoring and decision-making modules leads to a 63.56% reduction in system risk scores, confirming the effectiveness and robustness of the proposed approach in an experimental environment.
Axle-box bearings are safety-critical components in high-speed electric multiple units, and their vibration responses are affected by wheel–rail excitation, load variation, and environmental noise. Existing deep-learning diagnosis methods often rely on either raw time-series signals or time-frequency images, which limits their ability to exploit complementary temporal and spectral fault cues. This paper proposes Time-Frequency Dual-Stream Contrastive BearingNet (TFDC-BearingNet). The core idea is that a single vibration signal can be represented in two perceptually complementary forms, a global composite time- and frequency-domain visual view and a local temporal waveform view, and the global representation can serve as a query to guide the selection of fault-discriminative local signal components. Specifically, a visual branch extracts global representations from composite time-domain and frequency-domain images with a pre-trained Vision Transformer, whereas a signal branch uses a multi-scale tokenizer and Transformer encoder to preserve local temporal dependencies from raw sequences. A global-guided local selection module further uses cross-attention and gated fusion to select diagnosis-relevant signal components under visual guidance, and a contrastive objective aligns the two views in a shared semantic space. Experiments on a high-speed train rolling test-rig dataset and the public Southeast University bearing dataset achieve accuracies of 98.78% and 99.95%, respectively. Bidirectional cross-condition experiments on the SEU dataset show target accuracies of 94.77% and 93.46%. Comparative, ablation, cross-condition, noise-robustness, feature-visualization and statistical analyses indicate that the proposed dual-stream alignment and global-guided local selection improve diagnostic accuracy.
In order to realize efficient fault recognition and utilization of the high-speed train (EMU) undercarriage image obtained by detection robot, image alignment technology play an important role. Firstly, research the original image data collection process of the EMU fault detection robot in the practical working site; secondly, perform image enhancement such as gamma transformation and Hanning window adding on the original image; thirdly, compared with the image alignment algorithm based on feature extraction, a Fourier transform-based image alignment algorithm is proposed; finally, the affine processing with global transform parameters acquired from image alignment is given. This paper conducts experimental verification and comparative analysis, and selects a convolutional neural network (CNN) model based on deep residual structure to carry out this experiment. Through comparison with three typical image alignment algorithms based on feature extraction, it shows that the Fourier transform-based image alignment algorithm has significantly improved in the accurate alarm rate, mistaken alarm rate, and omitted alarm rate with strong robustness. Based on the intrinsic subpixel-level precision of domain transformation alignment algorithm, and the parallel image processing method proposed in this paper, the on-site image preprocessing, high-precision alignment, and automatic fault recognition can be realized stably and reliably within a reasonable time.
Reasonably arranging the operation scheme of EMU maintenance line is the key link in the operation organization of EMU Depot. Considering the characteristics of the maintenance line utilization, on the premise that the time of EMU occupying the maintenance line is given, the optimization model of the maintenance line optimization scheme was established, with the compatibility of the maintenance line occupation as the constraint and the balanced utilization of the maintenance line as the optimization goal. Based on the genetic algorithm, the algorithm for solving the model was designed, and the model and algorithm were verified with the actual operation data of a EMU depot.
The application of Train of EMU failures Detection System (TEDS) ensures the running safety of EMU. In order to reduce the manual workload and meet the requirements of fine classification monitoring, we researched the automatic classification technology, designed and implemented the corresponding methods, established the TEDS classification chart of all EMU trains at headquarters, and put forward the automatic calculation operation algorithm using multi-source data. Designed the generation scheme of local chart at the monitoring center, the system application function of automatic dispatching is developed, and the classification monitoring based on train running diagram was realized. The application shows that the method proposed in this paper have reduced the manual workload of work effectively, and the capability of abnormal detection has been enhanced, laid the technical foundation for the effective operation monitoring of EMU.