34 S. Gonen et al. Fig. 4.1 Layout of the Bridge and the sensor setup for OMA (dimensions in mm) from train crossing data are compared to those obtained using ambient and free vibrations. The results call attention to the importance of bridge-train interaction and the practical aspects of using different data types. Further, the importance of the structure’s higher modes of vibration on its dynamic response is highlighted. 4.2 The Railway Bridge The railway bridge investigated in this study, referred to as “the Bridge” hereafter, is a set of twin bridges that house one track each. It has a unique structural system as the extensions of the bridge from the abutments were designed as cantilevers without any contact between the bridge deck and the abutment. The bridge under scrutiny has five spans and a total length of 48.6 m. Outer spans of the bridge rest on elastic bearings 4.5 m away from the bridge’s ends, and two reinforced concrete piers support the mid-spans of the bridge. The piers are located at the 18.2 m and 37.0 m of the bridge. They have a 1.4 m diameter and a pier cap directly connected to the reinforced concrete bridge deck, which is 0.5 m thick. The layout of the Bridge is presented in Fig. 4.1. 4.3 Instrumentation and Measurements The instrumentation deployed on the bridge consists of a 20-bit low noise low-power data acquisition system and five triaxial MEMS digital accelerometers. The accelerometers were placed at each bridge span, as shown in Fig. 4.1, for a 24-hour period. Data were collected continuously from the five accelerometers during this 24-hour period at a sampling frequency of 250 Hz. During the measurements, both ambient and train-induced vibrations from 50 train crossings were recorded. Several factors influence the dynamic response of railway bridges. Among those related directly to the loading are the speed and axle load of the train, axle spacing, and stiffness and damping properties of the axles. These affect the dynamic amplification of the bridge response and maximum vertical accelerations of the bridge, thus influencing the bridge’s safety and serviceability. However, in this study, there is limited information about the train properties. Among the three types of trains, the express trains are 106 m long and weigh 218 tons. They cross the bridge at higher speeds than the other trains. Passenger trains come in various lengths and weights, ranging from 108 m and 218 tons to 201 m and 411 tons. Similarly, the freight trains’ length and weight vary between 359 and 527 m and 549 and 2412 tons. They cross the bridge with the lowest speed. An example of train crossing data for each type of train crossing is presented in Fig. 4.2 by concatenating data obtained from Sensor number 7. As mentioned earlier, the information about the train properties and speeds is limited. Thus, it is not possible to infer the forcing frequencies of the train loading using the well-known formula of f =nV/L , where n is a positive integer, V is the train speed, and Lis the distance between the centers of two consecutive carriages. The distribution of the maximum accelerations recorded at each sensor in the instrumentation setup is depicted in Fig. 4.3. In the figure, each type of train is distinguished by a different marker. While there is no clear trend associated with the train type, freight trains seem to cause higher maximum accelerations at each of the sensors. Figure 4.3 shows that most of the sensors are very well within the limit of 0.35 g set forth by [14]. However, for some train crossings, the maximum recorded maximum accelerations approach the limit of 0.35 g, even exceeding this limit for one case at the ends of the bridge (Sensor 7). This observation suggests that the dynamic behavior of the Bridge is very close to the limits, and a long-term measurement campaign will provide invaluable information that can be further used for approval of the bridge for the current and new train loads.
RkJQdWJsaXNoZXIy MTMzNzEzMQ==