A measurement component based on fiber Fabry–Pérot (FP) interferometer integrated into a monitoring system for train axle counting is demonstrated. To better understand the characteristics of the sensors and their suitability for rail transport, two sensors with slightly different sensitivities were placed on the foot of the rail in the middle of the sleepers. During the measurement campaign that was carried out on Slovakia regional railway line, the sensors registered passage of the trainset more than 70 times, and their speeds were in the range of approximately from 15 to 65 km/h. The trainset consists of a diesel locomotive and four wagons with different weights and different wears. The traces were divided into five groups according to speeds, while the actual speed differs from the average one by ±1 km/h. Traces obtained by the sensors were filtered. Subsequently, the phase of the signal was obtained and interpolated, and the derivation of the function was performed. If the maximal value of the first derivation of the function is observed, then the axle (wheel) is identified. In case of good technical condition of the vehicle, it is possible to determine the number of axles regardless of the vehicle speed (in our measurement campaign up to 65 km/h and real operation up to 98 km/h) and sensitivity of the sensor. In addition, the noise level in the trace can tell us something about the vehicle’s technical condition.
This study proposes a Fabry–Perot interferometric system and an associated evaluation method for measuring the weight of moving trains. An optical fiber sensor, comprising a sensing fiber and a supporting structure, is securely bonded to the rail foot. As a train traverses the track, the resulting localized bending induces a change in the sensing fiber’s length, which manifests as a quantifiable phase shift in the interference signal. We developed a physical–mathematical model, based on three Gaussian functions, to describe the temporal change in sensing fiber length caused by the passage of a single bogie. This model enables the determination of a proportionality constant to accurately convert the measured phase change into train weight. Model validation was performed using a train set, including a locomotive and four variably loaded wagons, traveling at 15.47 km/h. This system offers a novel and effective approach for real-time train weight monitoring.