An Online Classification Method for Fault Diagnosis of Railway Turnouts.

Railway turnout system is a key infrastructure to railway safety and efficiency. However, it is prone to failure in the field. Therefore, many railway departments have adopted a monitoring system to monitor the operation status of turnouts. With monitoring data collected, many researchers have proposed different fault-diagnosis methods. However, many of the existing methods cannot realize real-time updating or deal with new fault types. This paper-based on imbalanced data-proposes a Bayes-based online turnout fault-diagnosis method, which realizes incremental learning and scalable fault recognition. First, the basic conceptions of the turnout system are introduced. Next, the feature extraction and processing of the imbalanced monitoring data are introduced. Then, an online diagnosis method based on Bayesian incremental learning and scalable fault recognition is proposed, followed by the experiment with filed data from Guangzhou Railway. The results show that the scalable fault-recognition method can reach an accuracy of 99.11%, and the training time of the Bayesian incremental learning model reduces 29.97% without decreasing the accuracy, which demonstrates the high accuracy, adaptability and efficiency of the proposed model, of great significance for labor-saving, timely maintenance and further, safety and efficiency of railway transportation.

Surrogate-based optimization with adaptive sampling for microfluidic concentration gradient generator design.

This paper presents a surrogate-based optimization (SBO) method with adaptive sampling for designing microfluidic concentration gradient generators (µCGGs) to meet prescribed concentration gradients (CGs). An efficient physics-based component model (PBCM) is used to generate data for Kriging-based surrogate model construction. In a comparative analysis, various combinations of regression and correlation models in Kriging, and different adaptive sampling (infill) techniques are inspected to enhance model accuracy and optimization efficiency. The results show that the first-order polynomial regression and the Gaussian correlation models together form the most accurate model, and the lower bound (LB) infill strategy in general allows the most efficient global optimum search. The CGs generated by optimum designs match very well with prescribed CGs, and the discrepancy is less than 12% even with an inherent limitation of the µCGG. It is also found that SBO with adaptive sampling enables much more efficient and accurate design than random sampling-based surrogate modeling and optimization, and is more robust than the gradient-based optimization for searching the global optimum.