1 Introduction. F. Augusztinovicz 1, P. Forián Szabó 2, P. Fiala 1, A. B. Nagy 1, F. Márki 1 and Z. Horváth PDF

Description
Design, construction and verification measurements of vibration isolation of the new Budapest metro line M4, with special regard to structure-borne noise levels in surrounding buildings F. Augusztinovicz

Please download to get full document.

View again

of 16
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information
Category:

Calendars

Publish on:

Views: 29 | Pages: 16

Extension: PDF | Download: 0

Share
Transcript
Design, construction and verification measurements of vibration isolation of the new Budapest metro line M4, with special regard to structure-borne noise levels in surrounding buildings F. Augusztinovicz 1, P. Forián Szabó 2, P. Fiala 1, A. B. Nagy 1, F. Márki 1 and Z. Horváth 3 1 Budapest University of Technology and Economics Dept. of Networked Systems and Services H-1117 Budapest, Magyar tudósok krt. 2., Hungary 2 FSzP Ltd. H-1013 Budapest, Lánchíd u CDM Hungary Ltd. H-2011 Budakalász, Ciklámen u. 43. Abstract In order to keep the structure-borne noise level caused by the underground rail traffic below the maximum allowable values, vibration limit curves were defined in the planning phase of the new line M4, derived from a hybrid calculation method. The used model consisted of four essential elements: tunnel-soil interaction, propagation in the soil, soil-building interaction and re-radiation of sound from building structures, excited by accelerations measured in the tunnel of two other existing metro lines. The results of the model calculations were verified against measured transfer functions for some typical soil-building combinations. Based on these the contractor of the track designed and applied resilient rail pads below the sleepers throughout the whole line, and floating slabtrack along app. 25 % of the track. The line was put into normal operation in March 2014, and results of the first control measurements have shown that the design and construction of the track meets the requirements, resulting in more than 12 dba noise reduction. 1 Introduction Although Budapest s fourth metro line was already in the air in the early 1970s, plans for M4 only emerged in earnest in 1995, the foundation stone was laid in 2006 and the new, green line was eventually put into operation in March It connects Kelenföldi train station in Buda s south-west with Keleti train station in Pest, serving 10 stops along the 7.4km long track running underground. The new track is actually just the first phase of the line which is planned to be longer in both directions. Except for the first year of operation when a supervisor will be present, the trains of type Alstom Metropolis will be driverless. The trace is relatively shallow: most stops were built in depth between 15,6 and 24 m, and M4 crosses the existing line M3 from above. The tunnels approach the foundation of some dwelling houses, and run under public facilities which are especially sensitive to noise or vibration, such as an acoustical and an analytical laboratory, as well as the large orchestral studio of the Hungarian Radio. This has meant as from the beginning of planning that special attention must be paid to environmental problems such as vibrations and ground-borne noise. 3397 3398 PROCEEDINGS OF ISMA2014 INCLUDING USD2014 In the first phase of the trace design it was still possible to keep off places which are most sensitive from the acoustical point of view in appropriate distance, e.g. the studios of the Hungarian Radio. These investigations [2] were performed at the time of environmental impact assessment, when the selection and correction of trace was still sufficiently flexible. Nevertheless, dwelling houses in the downtown could not be by-passed. In Hungary maximum noise limits in form of equivalent noise levels are prescribed for daytime and night periods [1]. In order to ensure L eq and a eq conformity for day and night even in case of unfavourable background noise and vibration levels, we have calculated the maximum permitted passby noise and vibration values. These values, denoted by immission levels below and to be met everywhere above the metro line irrespective of building structure, soil type and distance between the tunnel and building foundation, are the following: for passby noise L A,F,max = 42 db, for passby vibration: a w,f,max = 18 mm/s 2 (weighted acceleration according to ISO , rms value integrated by Fast time constant). (One has to note at this point that these values are above the sensation threshold. Unfortunately, the Hungarian system of environmental limit values is not too strict and thereby permits a certain level of disturbance.). When the tender of the track construction was invited, a number of parameters such as the train type, structure and trace of the tunnel, the surrounding soil type, buildings above the line and their distance from the tunnel was already fixed and given. In the tender call the bidders were requested to trim the vibration isolation of the track such that the immission limit values will be met for any building, and the conformity had to be proved in the offer dependably. Considering the large number of all possible variation of the parameters, this task would have been very difficult to meet by the applicants. On the other hand, the applications writer was in difficult position too, because he had to evaluate the correctness of 6 to 8 offers supported by varied and complex vibroacoustic analyses in a short time. The applications writer has decided therefore to have an extended investigation performed himself. As a result of these preparation works, the immission limit values were brought into the tunnel : the constructor of the track had to ensure to keep the vibration levels on the wall of the tunnel under the maximum permitted limit values, called tunnel-wall emission limit values, only. Although the applicants were still allowed to use immission limit values in their offer, all of them chose the simplified emission calculation method in their applications. The investigations related to the vibration isolation of the track and reported in this publication are concentrated on three major steps of the project: definition of the requirements, a detailed analysis of the selected track system, and the results of the first control measurements performed in the tunnel and in apartments above critical parts of the line. 2 Specification of the tunnel-wall emission limit curves 2.1 Principle of the specification method Figure 1. shows a schematic representation of the generation of structure-borne noise components of underground rail traffic. In order to define maximum vibration values which will result in noises which will just meet the noise limits prescribed by environmental legislation, the real-life excitation of the rolling trains, the transmission from the rails to the tunnel and from the tunnel to the buildings, as well as the noise radiation inside the building had to be determined. This procedure was hindered by many uncertainties and even unknowns, at least at the time of the track design phase of the project (in ). The workabout, a hybrid measurement-and-calculation procedure developed and applied to come to a reasonably accurate vibration specification, consisted of the following steps: It was assumed in the first phase of investigations that the vibration load of the new, Alstom Metropolis trains will be similar to the load of the earlier generation of the Budapest metro carriages. This assumption was corrected in 2010 when a train of the new type was run on line M2 for test purposes, and some comparative measurements and corrections were possible. RAILWAY DYNAMICS AND GROUND VIBRATIONS 3399 Impact hammer (stat. load = 0) rail and fastening drop sack vibration of invert tunnel structure vibration of tunnel wall soil vibration of founding building noise vibration trains (stat. load 0) Figure 1: Schematic representation of the vibration propagation and noise generation from rails to the rooms of buildings above metro tunnels We have also measured the vibration transmission between the rail, invert and tunnel wall in existing tunnels and used to verify an analytical model of the track, consisting of various flexible rail pads and elastically floated invert, also called mass-spring system. The vibration transmission between the tunnel and the building foundations was measured on existing metro lines under normal operating conditions, and also for artificial excitation. The obtained measurement results were used to validate a numerical propagation model, which was applied to calculate the vibration transmission from the planned, new line in a second phase of investigations. The soil parameters of this propagation model were derived by means of the in situ seismic method Spectral Analysis of Surface Waves (SASW), whenever appropriate conditions were there to perform the test. The vibration propagation within the affected buildings and the radiation of noise into the rooms from surrounding building structures was substituted by a hypothetical, average Budapest building, realized virtually by means of numerical simulation. Based on all these partial elements we have set up a full calculation scheme of the train-to-noise transmission system for three other test sites on the existing metro lines. The calculated results were verified by measuring and comparing maximum passby noise levels. Applying the developed calculation scheme a parametric study was performed for various relative distances between the tunnel (station) and the building-model, and for two types of soil. The obtained results were eventually used to derive a set of maximum allowable PSD-velocity curves on the tunnel walls. The limited coverage of this paper does not allow to discuss all the details of the procedure as outlined above. Therefore, only some of the steps are described below in some details in paragraphs 2.2 to Vibration transmission measurements on existing lines An essential step in the process of requirement specifications was to perform parallel tunnel wall vibration and noise measurements at three sites of the two existing metro line M2 and M3. The main parameters of the measurement sites are summarized in Table 1. The measuring system, shown in Figure 2, consisted of accelerometers, placed both in the tunnel (Figure 3) and in the building above. The acceleration signals were amplified, digitized and transferred to a computer both underground and on the surface. The applied devices were sound recording studio equipment: broadcast quality A/D converters and soundracks (Behringer AD8000, MOTU 828 MkII and RME Fireface 800). The soundracks were connected by a single, long coaxial cable to ensure full synchronism, led through nearby escalator or ventilation shafts. The use of standard studio equipment enabled us to get stability and good quality signals at reasonable price, while the necessary signal processing and post-processing was made by means of in-house software packages written in MATLAB. As depicted also in Figure 1, the vibration transmission of the investigated subsystems (from rail to invert and tunnel wall and from invert to building foundations) was also measured by artificial excitations, using an impact hammer and a leather sack filled with 80 kg lead shot (see Figure 4.) 3400 PROCEEDINGS OF ISMA2014 INCLUDING USD2014 A/D converter Accelerometers in the building MOTU soundrack Accelerometers in the tunnel RME soundrack A/D converter Figure 2: Measuring system used for parallel measurements in the tunnel and in buildings on surface Site of measurement Soil type around the tunnel Type of tunnel Type of rail fixture (see Figure 1.) Characteristics of the building above the tunnel Point of noise measurement #1 (on line M2 in Rákóczi street) #2 (on line M2, at Kossuth square) #3 (on line M3, at Kálvin square) sandy gravel sandy gravel sandy gravel cast iron tubbing & monolithic concrete Ortec, Delta lager (recently renewed) 1+3-storeyed traditional brickwork apartment block, approx. 100 years old in cellar on soil probe (vertical) and on wall (horizontal) monolithic reinforced concrete Ortec, Delta lager (recently renewed) 7-storeyed frame office building of reinforced concrete, steel and glass, built in 1972 in cellar on the floor pre-cast concrete with shotcrete surface Metro storeyed frame office building, reinforced concrete and glass, construction in progress in 2006 in the parking lot at level -4 and on the ground floor Table 1: Main parameters of the measurement sites selected on the existing metro lines, applied to reveal the vibro-acoustic characteristics of the transmission from tunnel to building Some typical examples of the gained transfer functions are depicted in Figure 5. While the vibration attenuation from the rail head to the tunnel wall varies in between 40 and 55 db in the audible frequency range, the transmission between the tunnel and the cellar wall starts as from 0 db and goes to values between -25 and -50 db. Significant differences can be observed between measurements at site #1 and #3 on two different lines, but also between two tunnel wall constructions at site #1. RAILWAY DYNAMICS AND GROUND VIBRATIONS 3401 Figure 3: Accelerometers in the tunnel at site #2 on the rail head and invert (left) and on the tunnel wall (right) Figure 4: Application of a 80 kg drop sack to exert artificial excitation in the tunnel (site #2, on the left) and for SASW test at the place of a would-be station of the new line (on the right). 3402 PROCEEDINGS OF ISMA2014 INCLUDING USD2014 Figure 5: Comparison of various frequency response functions. Left diagram: transmission from rail head (vertical) to tunnel wall (vertical), right diagram: transmission from tunnel wall (horizontal) to cellar walls of buildings (horizontal). Red lines: site #3, green lines: site #1 on concrete tubbing, blue curves: site #1 on cast iron tubbing, black line: site # Development of the numerical propagation model from tunnel to building The prediction model used to compute vibration propagation characteristics between the tunnel and the building is a coupled FE-BE model consisting of the following components. The tunnel is modelled by a 2.5D longitudinally invariant finite element model [6], where the model characteristics are the tunnel s material and geometrical parameters. The soil is modelled by a 2.5D longitudinally invariant BE model [5] capable of handle horizontally layered soils, where each homogeneous layer is characterised by dynamic material properties. The structure is modelled using a structural FE model. Finally, the re-radiated noise is computed using an acoustic spectral finite element model [5] capable of computing the acoustic behaviour of shoe-box shaped enclosures. Rather than going into details regarding the models which are described in [6], the following two sections demonstrate how the coupled model was validated with measurements along existing metro tunnels, and how the prediction tool was used to relate re-radiated sound pressure levels to tunnel vibrations in a parametric study Model validation The validation was performed by comparing measured and computed structural vibration levels in an existing building due to a drop weight excitation in the tunnel of existing metro line M3. The selected building is the Kálvin Center, located along the intersection of the existing metro line M3 and the planned M4. The tunnel centre of metro line M3 is located at a depth of 28.2 m below the building. The inner radius of the tunnel is 2.6 m, the outer radius is 3.1 m. The tunnel's material is described by a mass density of 2300 kg/m 3, a Young's modulus of 2x10 10 N/m 2, a Poisson's ratio of 0.16 and a material damping ratio or 2%. The soil is characterized by the material parameters determined during the SASW measurement in the garden of the nearby Hungarian National Museum [6]. The finite element mesh of the Kálvin Center is displayed in Figure 6. The model contains the main structural elements of the building. These are the box foundation, the columns and beams of the portal frame structure, the slabs of the floors, the central cores that surround the elevators and the staircases and finally the concrete wall of a neighbouring building. All these structural elements are made of reinforced concrete. The numerical model was used to compute the transfer function between the displacements of the tunnel's wall and the displacements in the building when an impulsive load is acting on the tunnel invert. During the computations, it has been assumed that only three low frequency modes of the tunnel's cross section are excited. These are (1) the rigid body vertical displacement of the tunnel, the (1) first vertical compres- RAILWAY DYNAMICS AND GROUND VIBRATIONS 3403 sional mode of the tunnel and (3) the first horizontal compressional mode of the cross section. These modes are displayed in Figure 7. Figure 6: The structural finite element model of the Kálvin Center As the impact force of the falling weight was not measured during the drop sack measurements and only the three displacement components were known from the testing, the frequency content of the loading force in the numerical model has been defined such that the computed tunnel accelerations matches the measured tunnel response. In this phase, the frequency content of the x and z components of the loading force were considered as unknown parameters, and the difference between the measured and computed tunnel wall accelerations were minimized by means of a least mean squares algorithm. Figure 7: Three tunnel modes taken into account in the validation Figure 8 shows the frequency content of the resulting accelerations in the tunnel and in the building's basement. The computed structural accelerations in the two points of the basement are also displayed in the figure. Comparing these values with the measured structural displacements, it can be seen that the model computes the structural vibrations with a reasonable accuracy. The difference between the computed and measured results is about 5 db. 3404 PROCEEDINGS OF ISMA2014 INCLUDING USD2014 Figure 8: Frequency content of the computed acceleration of the tunnel and the building due to the falling weight excitation. Blue - invert vertical, green - tunnel side wall radial, red - tunnel side wall tangential, yellow - building base mat vertical, cyan - building wall horizontal, magenta - building wall vertical. The gray curves display the measured structural vibrations Parametric study After the model has been validated, a parametric study has been carried out in order to determine the vibration specifications for the new line M4. In the parametric study, a typical portal frame office building located above the metro tunnel has been modelled, and the re-radiated noise level in the rooms of the building due to the passage of a metro train has been computed. These computations have been carried out with two types of soil properties, two type of underground structures (tunnel or station) and different tunnel-building distances. These parametric studies resulted in transfer functions between the vibration on the tunnel's base plate and the re-radiated noise in the building. With these transfer functions, the maximum allowable tunnel vibration levels could be determined so that the specified maximal sound pressure in the building's rooms is not exceeded. The average Budapest building model used in the parametric study is displayed in Figure 9. The dimensions of the building are 30 m x 20 m x 13.5 m. The structure is embedded in the soil to a depth of 4.5 m, and rests on a box foundation (not displayed in the figure). The box foundation is built of reinforced concrete of width 0.5 m. The building's floors are resting on r.c. columns with square cross section and width of 0.3 m. The reinforced concrete floors are modelled by m plate elements. The in-fill walls of 15 rooms are taken into account in the model. The masonry in-fill walls of the rooms are modelled with 0.1 m wide plate elements. In each floor, there are two rooms of dimensions 5 m x 4 m x 3 m, one smaller room of dimensions 2.5 m x 4 m x 3 m and two very small rooms of dimensions 2.5 m x 2 m x 3 m. In the parametric study, two different soil types have been considered. Both soil types are modelled as a homogeneous half space. The first, heavy soil is characterized
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x