A Novel Health Indicator for the Polygonal Wear of the High-Speed Train Wheels Based on the Wheel Profile Data

Abstract

SUMMARY & CONCLUSIONSThe wheel polygonal wear refers to the phenomenon that the actual periodic radial deviates from the ideal round shape of the wheel’s circumference. This would introduce excessive vibration between wheels and tracks, leading to severe damages to the wheelsets and even causing derailment. A large number of research works have been devoted to investigating the evolution of wheel polygonal wear during high-speed train operation. The high-frequency flexible resonance of the bogie system and the braking force have been identified as the main root causes of wheel polygonal wear. However, the mechanism underpinning wheel polygonal wear is still not fully understood.As the state-of-the-art practice, during the wheel repair process, a specially designed polygon detection device is utilized to measure the wheel profile, concentrating on the polygonal wear orders and degrees of the wheels [1]. These wheel profile data are then utilized to determine whether to reprofile the wheel into the best possible roundness and concentricity. More specifically, various parameters like maximal radial runout and degrees of polygonal wear are often calculated from the wheel profile data, according to the European Standards Committee and the International Railway Union [2].In the Chinese High-speed train industry, equivalent conicity and maximal radial runout are often calculated from the wheel profile data and then utilized as the health indicators to decide whether to reprofile a wheel [3]. By definition, the equivalent conicity is the tangent of the cone angle of a wheelset, which indicates the lateral vibration of the trains [3]. According to relative research, the equivalent conicity is the optimal estimation of the contact between wheels and rails on a straight section of track and in curves of large radius [4]. The maximal radial runout is the largest difference between the radius of the wheel and the distance from the wheel center to any point on the wheel tread, which indicates vertical vibration of the trains [5]. The maximal radial run-out is also a critical geometric tolerance of the wheel considered in standards [6], which reflects the degree of vertical vibrations of the wheels [5]. However, these two indicators can not illustrate the wheel polygonal wear fully. Wheel polygonal wear of order as high as 24 has been reported to cause severe impact between wheels and rails [7]. Therefore, it is necessary to put forward a new indicator for the severity of the wheel polygonal wear of the trains.This paper reports a systemic study on the health indicators for wheel polygonal wear during high-speed train operation. Several issues are identified in the health indicators currently utilized for maintenance decision-making. It is shown that neither the equivalent conicity nor the maximum of radial runout can differentiate the health conditions of the wheels before and after being reprofiled. Specifically, equivalent conicity cannot reflect the periodicity of the wheel. The maximal wheel radial runout only indicates the general wear of the wheel, providing limited information on the severity of the polygonal wear over different orders. These findings highlight the weakness of the current maintenance practice, underlining the need for an improved health indicator to characterize the health condition of wheels.To address the above issues, a novel health indicator is proposed to effectively represent the wheel health condition. The approach is based on the energy distribution along the different polygonal orders of the wheel that indicates the severity of polygonal wear. This provides a fundamental advantage over the current health indicators that only partially utilizes the information of the profile data. Specifically, the entire polygonal orders are divided into six closed intervals, namely [1, 5], [6, 10], [11, 15], [16, 24] and [25, 40]. This follows the field experience of the maintenance crews that the severity of polygonal wear is mainly represented by the high orders [16, 40], especially within the interval [16, 24]. Whereas the orders in the interval [1, 5] are considered as the least significant. Then the energy content is computed and examined within each interval, respectively. The results suggest that most energy is concentrated in the interval [1, 5], and the corresponding energy distribution between wheels before and after being reprofiled almost always overlap. On the other hand, the energy distribution within other intervals exhibits satisfactory separation between wheels before and after being reprofiled. Finally, it is discovered that a distribution analysis of the energy within the interval [6, 40] would indicate a clear separation between wheels before and after being reprofiled, which would be utilized as an effective health indicator for the maintenance of high-speed train wheels. The effectiveness of our proposal is validated on the wheel profile data of a fleet of operating high-speed trains in the year 2019. The results of this research can be used to verify the effectiveness of the wheel reprofiling process and ultimately improve the efficiency of the maintenance activities of high-speed train wheels.