Implementation and analysis of vibration measurements obtained from monitoring the Magdeburg water bridge

Transcription

1 Implementation and analysis of vibration measurements obtained from monitoring the Magdeburg water bridge B. Resnik 1 and Y. Ribakov 2 1 BeuthHS Berlin, University of Applied Sciences, Berlin, Germany 2 Ariel University Center of Samaria, Israel Abstract Scientific cooperation with the BeuthHS Berlin, Germany and the Ariel University Center of Samaria has led to several research projects in the field of structural health monitoring. Within the past few years this cooperation has been extended to the field of ambient vibration monitoring, which is based on the dynamic characteristics analysis of large span engineering constructions like bridges. This method as part of structural health monitoring has gained in importance during the last years and has been applied in various problematic domains. Information about changes in structural dynamic characteristics can be detected by using originally developed accelerometers. Commercially available geophysical measurement equipment as well as corresponding data processing programs are expensive and also not suitable for these purposes. The paper presents a novel measurement system based on low-cost accelerometers that nevertheless perform measurements with high accuracy as well as originally developed software for results interpretation. The basic measurement process and the necessary data analysis are exemplified on the Magdeburg Water Bridge, Germany, which is a unique structure from importance, safety and other viewpoints. Numerous measurements performed at this structure in the last years have confirmed that the developed measurement system is well suited for the control and the secure interpretation of dynamic deformations arising due to natural loads such as wind or traffic. It demonstrates the efficiency of the proposed method and proves its accuracy for health monitoring of complex structures. Keywords: Structural health monitoring, ambient vibration, dynamic characteristics, accelerometer 1. Measuring system for vibration monitoring Structural Health Monitoring (SHM) of large engineering structures such as bridges has gained in importance during the last years and has been applied in various problematic domains. Geodetic monitoring is one part of SHM (Structural Health Monitoring), which investigates long-term deformations. The field of Vibration Monitoring is based on the analysis of vibration characteristics of large engineering structures such as bridges (see Fujino et al. 2010, Lynch et. al 2006, Wang et al. 2006, Wenzel 2009). In contrast to Geodetic Monitoring Systems, which are in essence restricted onto the acquisition and analysis of geometric deformations and dislocations of the structure, Vibration Monitoring can immediately detect changes of structural integrity and even determine cause and location of an occurred malfunction. The fundamentals of this approach are based on the unique dynamic characteristics that are a derivative of the equation of motion and can be interpreted as a vibration signature. Knowledge and analysis of the current natural frequencies can lead to fast and reliable conclusions about the condition of the structure. A single measurement sensor for the investigation of ambient oscillations costs thousands of Euros. Commercially prevailing geophysical measurement systems and corresponding processing programs are not only expensive; they are also not very suitable for educational

2 purposes (especially in the areas of engineering surveying). Simple sensors which are on the market for about ten years now are nowadays used in mechanical engineering and wind turbine constructions. The price of them however is only a fraction of the special measurement systems. But they are not suitable for geodetic measurements without alteration and new software. Due to this fact, the BeuthHS has recently started to develop and to use low-cost systems for SHM as well as the analysis of vibration characteristics. The developed measurement system, in contrast, consists of multiple simple sensors used by mechanical engineering. The designed device features have memory card slots and possess an internal battery that lasts for about 8 hours under maximum energy consumption, which makes it applicable even in challenging work environments. The waterproof housing and the temperature resistant components further ensure permanent usability under nearly all weather conditions. In addition, real time data transfer to a computer can be achieved through wireless LAN and USB connections. Ultimately, a pre-implemented high-pass filter offers the possibility to reduce offsets and drift characteristics in the datasets. All mentioned features enable a real-time controlling. Hence, the operator is able to react to possible problems by changing the respective measurement setup, if necessary. The unprocessed data can simply be recorded for extensive post-processing, if needed. We propose a software solution for this purpose developed for the specific problems in the data processing of several accelerometer sensors. Nonetheless, this software is an intuitively operable tool for data analysis and evaluation by an inexperienced user. Fig. 1 illustrates the introduced system. (a) (b) (c) Fig. 1. USB sensor with laptop (a), wireless LAN sensor (b) and software (c) Prior to the actual measurement of the bridge its construction has to be analyzed in terms of mode shape and natural frequencies to obtain optimal sensor locations. For large and complex structures finite element analysis should be conducted to achieve an optimal acquisition of the dynamic behaviour. In dynamic measurements, the sensors are positioned as a rule along a span axis and are then connected to a field computer. As a next step, the distances between the measurements points are recorded, which enable for further analysis. For the selected ambient windows of the time series frequency spectrum are calculated according to the Fast Fourier Transform algorithm (FFT). In order to present the results in multiple spectrums at the same time, a so-called Trendcard is used. These graphics visualize multiple signals in the area of frequency simultaneously and in relation to their position in the construction. To increase the clarity of the visualization, a two-dimensional depiction (frequency and length of a construction) as represented in Fig. 2 is used

3 Fig. 2. Concept of measurement and analysis 2. Practical example Numerous test measurements in the last years have confirmed that the developed measurement system is well suited for the control and the secure interpretation of dynamic deformations of constructions arising because of natural sources such as wind or traffic. After some educational projects, which showed very successful results, the authors concluded that this system can also be applied for commercial usage due to its very economical price/performance ratio. In this regard, engineering companies from Germany, Russia and Armenia realized some commercial projects. The basic measurement process and the necessary data analysis are exemplified in on the Magdeburg Water Bridge (Germany), which has been chosen due to its very clearly perceptible vibration behavior. The Magdeburg Water Bridge is a navigable aqueduct in Germany, opened in October 2003 and part of the Magdeburg crossing of waterways (Fig. 3). It connects the Elbe- Havel Canal to the Mittellandkanal, crossing over the Elbe river. It is notable for being the longest navigable aqueduct in the world, with a total length of 918 metres. The water bridge connects Berlin s inland harbour network with the ports along the Rhine river. The aqueduct's tough structure incorporates 24,000 tons of steel and 68,000 cubic meters of concrete. The bridge has a steel superstructure. Fig. 3. Magdeburg water bridge (Germany) Finite element analysis allows breaking up a complex structure into simple parts and analyzing their dynamic behavior, separately. With the help of special computer programs the deformation

4 behavior of the whole structure can be finally investigated. However, a very exact determination is often not possible with this method because of it is too difficult to model the behavior of number bolt joints, minor elements like rails, deformation joints and etc. The measurement system can also be used for the continual monitoring of very important structures. In this case, the possible change of oscillation parameters during the measurement process, which can arise due to factors such as temperature fluctuations and traffic, has to be taken into account. This fact was measured also on the Water Bridge. A lot of measurement were carried out in the last years during traffic and shall include sufficient measurements for subsequent analysis. The measurement system was installed during several hours in the middle of the bridge (point 3 in Fig. 4) and thereby the changes of high frequency deformations in relation to traffic were investigated. In order to compare natural frequencies by the continual monitoring in have been taken interval of many hours into consideration, are combined to a 2D contour plot. The frequencies are plotted on the x-axis against the sensor locations over the time with interpolated spectrums with interval of about 0.2 h. In spectrum obtained from point 1 (Fig. 5) the natural frequencies at approximately 1.2 und 1.7 Hz can be identified considerably. Other measurements points show similar frequency pictures. Fig. 4. Schematic representation of the River crossing with measuring points Fig. 5. Vertical natural frequencies for Point 1 To analyze the simultaneous measurements at several points the results must be converted into a trend map too. As an example the frequency spectrum for the vertical acceleration depending on the location of the sensor along the control line (Points 1 5 in Fig. 4). Fig. 6 represents the same number of signals in the frequency range without transport measurements (left) and during typical Ship passage (right). Evaluation appears clear differences between results. Thus, the natural frequencies of 1.2 and 3.5 Hz appear only for the current traffic phenomenon

5 Fig. 6. Vertical natural frequencies for the points 1-5 Conclusion In everyday situations we often analyze oscillations without noticing and use them for the control of different processes. Emerging defects can often be detected early through the change of the oscillation spectrum. Thus, a quality test can be achieved during the manufacturing process (roofing tiles, crockery, etc.). The field of Vibration Monitoring is based on the analysis of vibration characteristics of large engineering constructions A wireless low-cost monitoring system for vibration monitoring of structures as well as necessary steps for data acquisition, analysis and interpretation has been presented in this paper. Furthermore a proper data acquisition is fundamental for signal processing and the resulting quality of the natural frequencies. A previous planning of the measurement configuration is essential and on-site verification of the signal in terms of sections of ambient vibration is recommended. Further work will focus on the evaluation of mode shapes and damping characteristics as well as the evaluation of displacement values derived from acceleration measurements. References Fujino, Y., D. M. Siringoringo, T. Nagayama, and D. Su (2010), Control, simulation and monitoring of bridge vibration Japan s recent development and practice, IABSE-JSCE Joint Conference on Advances in Bridge Engineering-II, Lynch, J. P., Y. Wang, K. J. Loh, J.-H. Yi, and C.-B. Yun (2006), Performance monitoring of the Geumdang Bridge using a dense network of high-resolution wireless sensors, Smart Materials and Structures, 15(6), Wang, Y., K. J. Loh, J. P. Lynch, M. Fraser, K. Law, and A. Elgamal (2006), Vibration Monitoring of the Voigt Bridge using Wired and Wireless Monitoring Systems, Proceeding of 4th China-Japan-US Symposium on Structural Control and Monitoring. Wenzel, H. (2009), Health Monitoring of Bridges, Wiley, Chichester