The temporal distribution of ridership in metro stations from land-use perspective.

A reasonable land use development around subway stations can balance the utilization rates of the subway system during peak and off-peak hours, thereby enhancing its service levels and operational efficiency. Analyzing the temporal distribution patterns of passenger flow and their influencing factors is crucial for determining the optimum ratio of each land use type surrounding metro stations. Thus, this paper employs principal component analysis (PCA) at first to investigate the temporal distribution of metro ridership, and identify their main patterns and factor loadings. Then, using geographically weighted regression, the study examines the spatial dependencies between the main component proportions and influencing factors, focusing on Xi'an subway stations. The results indicate that the temporal distribution of passenger flow can be decomposed into three principal components: the first representing commuting characteristics, and the second and third representing regulating functions. The overall distribution is a composite of these components in varying proportions. Residential and educational land uses primarily drive morning and evening peak flows, with residential land use in the city center and peripheral areas having a more pronounced effect compared to transitional areas. Conversely, commercial & office, healthcare, and recreational & park land mitigate peak flows and increase off-peak flows. External hub enhances passenger flow throughout the day, while industrial land use has negligible impact.

A Novel Data Augmentation Method for Improving the Accuracy of Insulator Health Diagnosis.

Performing ultrasonic nondestructive testing experiments on insulators and then using machine learning algorithms to classify and identify the signals is an important way to achieve an intelligent diagnosis of insulators. However, in most cases, we can obtain only a limited number of data from the experiments, which is insufficient to meet the requirements for training an effective classification and recognition model. In this paper, we start with an existing data augmentation method called DBA (for dynamic time warping barycenter averaging) and propose a new data enhancement method called AWDBA (adaptive weighting DBA). We first validated the proposed method by synthesizing new data from insulator sample datasets. The results show that the AWDBA proposed in this study has significant advantages relative to DBA in terms of data enhancement. Then, we used AWDBA and two other data augmentation methods to synthetically generate new data on the original dataset of insulators. Moreover, we compared the performance of different machine learning algorithms for insulator health diagnosis on the dataset with and without data augmentation. In the SVM algorithm especially, we propose a new parameter optimization method based on GA (genetic algorithm). The final results show that the use of the data augmentation method can significantly improve the accuracy of insulator defect identification.