A High-Precision Detection Approach for Catenary Geometry Parameters of Electrical Railway

This paper aims to develop a high-precision detection approach for catenary geometry parameters of electrical railway based on vision technology. The novel idea of the proposed detection method is that, the image coordinates of the spot formed on the contact line by a laser emitter are mapped to the world system to obtain the conductor height and stagger. In the proposed detection system, the key factors that affect the detection accuracy include two components, the localization of the spot and the correction of the detection value. To solve these problems, the genetic particle filter algorithm based on particle swarm optimization (PSO-GAPF) is presented to track and locate the spot in the image. Compared with the standard particle filter, PSO-GAPF can improve the particle deprivation and increase the accuracy of state estimation. In addition, Kalman filter (KF) is used for correcting the detected value of catenary geometry parameters. In this paper, first of all, the model of spot target is established according to the feature distribution of gray color histogram, and the tracking and locating of the spot in the image are realized by PSO-GAPF. Second, the detection formulas of catenary geometry parameters are deduced by utilizing the space coordinate transformation to compute the conductor height and stagger. At last, the correcting of the catenary geometry parameters is achieved by KF. The real-life detection results show that the proposed method has high precision, and can meet the requirements of real-time catenary geometry parameters detection system.