PSI - Issue 17

João Morais et al. / Procedia Structural Integrity 17 (2019) 448–455

451

João Morais et al. / Structural Integrity Procedia 00 (2019) 000 – 000

4

between the vehicle chassis and wheel shafts and to measure the vehicle dynamics and interactions with the tracks. The system also includes two velocity sensors, installed on front and rear shaft. These sensors, developed in LNEC, are based on a magnetic proximity sensor, made by SICK (reference MM18), sensing the passage of eight magnets embedded in a stainless steel ring attached to each shaft. The signal generated by these velocity sensors complements the information obtained by a GPS system installed on the vehicle, to allow proper geo-referencing of the recorded data. The data acquisition system used (reference ARK-2150F plus Q.Station) is capable of acquiring 26 simultaneous channels, at a sampling frequency of 1000 Hz, which corresponds to a spatial resolution of 20 mm for a vehicle speed of 20 m/s. After the system requirements were established and the solution that would be used was developed, the different components were acquired. Preliminary system calibrations of the measurement chains were performed, under laboratory conditions, to ensure that the quality of the acquired information complied with project specifications. These tests confirmed the proper functioning of the entire system, in terms of signal generation by instrumentation, data acquisition and communication between the two subsystems. The first data processing performed by the computer application developed was also successful. For a full description of the proposed system and the preliminary calibration tests in laboratory conditions, refer to the paper by Santos et al. (2018) In the context of this project, a self-propelled rail vehicle was adapted to house the system previously described. The adaptation procedure consisted on mounting the instrumentation in exterior and interior locations of the vehicle, construction of fastening and protection elements for said instrumentation and assembling other complementary components, such as wiring harnesses and technical cabinets/interface boxes. Given the almost symmetrical form of the vehicle, it was considered adequate and even advantageous to divide the data acquisition and data collection subsystem by the two vehicle cabins, in a distributed architecture configuration. Fig. 2 shows some examples of the installed devices. 3. Prototype system implementation

c)

b)

d)

a)

Fig. 2. Instrumentation used on the prototype vehicle: a) ±500 g accelerometer installed on a wheel box cover; b) velocity sensor installed on a wheels shaft; c) laser displacement transducer installed over a wheels shaft; d) ±4 g accelerometer installed in the vehicle cabins.

For this task, it was necessary to design, assemble and mount several mechanical interfaces on the vehicle, establish cable paths for the instrumentation and communication wiring, including the opening of passage holes between the outside and the inside of the vehicle cabins, fix interface boxes and cabinets, among other tasks (Fig. 3). In order to ensure proper installation of the system, taking into account the specifics related to the instrumentation and to the vehicle and its operation, a close collaboration between the various elements of LNEC and Mota-Engil was fundamental. Given that a significant part of the instrumentation and wiring was mounted in the under belly of the vehicle, these tasks had to be carried out in a railway vehicle pit, in a Mota-Engil maintenance yard.