A5082222647- Railway Engineering and Dynamics
- Advanced Sensor and Control Systems
- Fuzzy Logic and Control Systems
- Advanced Algorithms and Applications
- Railway Systems and Energy Efficiency
- Industrial Technology and Control Systems
- Advanced Measurement and Detection Methods
- Vehicle License Plate Recognition
- Risk and Safety Analysis
- Electrical Contact Performance and Analysis
- Machine Fault Diagnosis Techniques
- Evaluation Methods in Various Fields
- Traffic Prediction and Management Techniques
- Evaluation and Optimization Models
- Gear and Bearing Dynamics Analysis
- Civil and Geotechnical Engineering Research
- Transportation Planning and Optimization
- Advanced Computational Techniques and Applications
- Surface Roughness and Optical Measurements
- Metaheuristic Optimization Algorithms Research
- Neural Networks and Applications
- Multi-Criteria Decision Making
- Elevator Systems and Control
- Fault Detection and Control Systems
- Industrial Vision Systems and Defect Detection
Nanjing University of Science and Technology
2015-2025
Shanxi Agricultural University
2024
Nanjing University of Aeronautics and Astronautics
2024
Huazhong University of Science and Technology
2021
University of Nottingham
2020
Nanjing University
2011-2013
Beijing Jiaotong University
2008-2012
University of Science and Technology of China
2006
China Academy of Railway Sciences
2003-2004
The online detection of wheelset size has important implications for ensuring the safety railway operation and decreasing maintenance costs. Based on laser displacement sensors (LDS), a novel system wheel is proposed using only six two-dimensional LDS two one-dimensional LDS. calculation principles tread profile diameter are given, as well calibration method. Errors induced by wheel-rail vibration, misalignment, sensor noise, S-shape running, differential also analyzed. After implementation,...
This paper focuses on how to minimize the total passenger travel time cost by computing and adjusting skip-sop patterns with given time-varying origin-to-destination demand matrices. A bi-objective nonlinear integer programming model linear constraints is proposed precisely formulate operating under minute-dependent from different origin–destination pairs. The implemented using genetic algorithm idea point optimization solvers, we show its effectiveness real world instance of Guangzhou Metro Line 8.
The pantograph is one of the most critical electrical equipment in electric locomotive as it transmits power from contact wire to locomotive. Condition monitoring important ensure safe and consistent operation a train. However, condition difficult due complex lighting situation background environment around pantograph. In this paper, we propose an image-based online system for wear attitude. First, working principle introduced. Second, evaluation method sliding plate state presented. curves...
In the process of imbalanced dataset classification, traditional oversampling technologies do not take into account distribution characteristics data sets, which easily results in low classification accuracy positive samples. order to improve recognition rate samples, algorithm combining K-means clustering with MAHAKIL is proposed this article. Firstly, used divide samples K clusters. The each cluster are then divided two initial populations according magnitude Mahalanobis distance. Finally,...
The rail train number is the only sign information of vehicle. It very important to use image processing method locate improve efficiency management. Therefore, an automatic positioning framework for numbers developed. First, aiming at problem low contrast between and vehicle body caused by insufficient illumination in a tunnel indoor, improved single-scale Retinex algorithm based on brightness control used enhance image, which provides basis subsequent positioning. Then, location speed-up...
An approach to construct multiple Pareto-optimal fuzzy systems based on NSGA-II is proposed in this paper. First, order obtain a good initial system, modified clustering algorithm used identify the antecedents of while consequents are designed separately reduce computational burden. Second, Pareto multi-objective genetic and interpretability-driven simplification techniques evolve system iteratively with three objectives: precision performance, number rules sets. Resultantly, Pareto- optimal...
In order to improve the efficiency of wheel size detection and measurement accuracy, an online detecting technology based on 2D laser displacement sensors is proposed. The are installed both sides track, then scan tread detected region dynamically. Next, use coordinate transformation data fusion preprocess collected finally calculate flange height, width diameter. system was site measure wheel-set verify feasibility system. Comparing results with manual results, experiment show that can...
In this paper, the vibration acceleration signal obtained by rail corrugation simulation is analyzed based on variational mode decomposition (VMD) method. The envelope entropy of decomposed used as index identification, track spectrum an initial threshold, and combined with smooth pseudo Wigner-Ville distribution (SPWVD) time-frequency analysis method, amplitude variation different wave depth studied wavelength position are identified. Finally, taking Guangzhou Metro Line 8 example, data...
The automatic depth control electrohydraulic system of a certain minesweeping tank is complex nonlinear system, and it difficult for the linear model obtained by first principle method to represent intrinsic characteristics such system. This paper proposes an approach construct accurate with RBF neural network trained genetic algorithm‐based technique. In order improve accuracy designed model, algorithm used optimize centers network. maximum distance measure adopted determine widths radial...
In order to timely grasp the wheelset size information and ensure safe operation of train, a method forecasting based on particle swarm optimization (PSO)-support vector machine(SVM) model is proposed. The fast global search ability PSO used optimize parameters SVM, prediction accuracy SVM improved. Taking wheel diameter city rail train as research object, PSO-SVM established, value compared with actual explain effect model. results show that correlation coefficient between predicted reaches...