ASSESSMENT OF THE PERFORMANCE OF SONIC CRYSTAL NOISE BARRIERS FOR THE MITIGATION OF CONSTRUCTION NOISE

Transcription

1 ASSESSMENT OF THE PERFORMANCE OF SONIC CRYSTAL NOISE BARRIERS FOR THE MITIGATION OF CONSTRUCTION NOISE Heow Pueh LEE; Long Bin TAN; Kian Meng LIM Department of Mechanical Engineering, National University of Singapore, Singapore ABSTRACT Construction sites are common noise sources in major cities. Excessive noise can cause stress and health impacts. Sonic crystals have been reported as noise barriers for the reduction of traffic and environmental noise. This study aims to explore the innovative use of sonic crystal structures as noise barriers that will encompass the noisy equipment parts such as an excavator, a DC cutter, or a piling machine. The proposed sonic crystal noise barrier is in the form of multiple arrays of tubular columns with suitable spacing to target noise reduction for the frequency range of interest. In addition, other features such as noise absorbing materials and Helmholtz resonators can be incorporated into the tubular structures. The space and openings through the periodic tubular structures ensure a balance of natural ventilation for the heat dissipating from the machine as well as some visuals of the enclosed component for safety inspection. The noise profiles of various typical construction machines are first presented, followed by the finite element modelling and analysis of some proposed designs and the assessment of their acoustic performance for noise attenuation, including the benchmarking of the performance with some existing solutions. Keywords: sonic crystals; acoustic attenuation; construction noise I-INCE Classification of Subjects Number(s): INTRODUCTION Many big cities are facing the problem of noise from construction sites or activities due to the rapid urban development as well as redevelopment and renewal of matured cities to relieve the traffic congestion. The subway systems are being built or expanded to cater to the large numbers of people moving to the cities, and the roads cannot carry them all. Although these subway projects can relieve transportation pressure in cities when they are completed, they raise many environmental problems such as noise and dust pollution during the construction periods besides aggravating the nearby traffic congestion during the road closure or diversion. The problem is also aggravated by the relatively complex construction processes that involve the construction of underground stations. The construction duration is therefore relatively longer when the surrounding environment and the ground traffic will be affected greatly [1-4]. As the subway stations under construction are often at locations of high population density as what they are planned and intended for, the construction sites could be just located next to the existing residential or commercial buildings. Figure 1 shows a typical subway construction site in Singapore surrounding by many buildings. Figure 1 A typical subway construction site in Singapore 4220

2 Due to the proximity of the surrounding buildings to the construction sites, noise is a common complaint from residents of the neighborhoods. As the construction site is near the surrounding buildings, typical noise barriers located at the peripheries of the construction sites may not be effective for the high floors of the surrounding buildings. The lower floors could be shielded from the construction noise but would be less effective if the noise is of lower frequencies. An alternative solution is to shield the noise source closer to the source, for example, the use of noise barriers surrounding a BC Trench cuter or at the impact location of a vibratory pile driver (Figure 2). For effective design of the noise shields, a better understanding of the noise profile from the machines would be required. Figure 2 Typical noise barriers for construction equipment In this study, instead of using the typical type 1 Sound Level Meters for the measurement of noise profiles emitted from construction equipment and processes as commonly done in construction industry, we attempt to characterize the noise profiles of common construction equipment at their sources using an Acoustic Array or Acoustic Camera. This type of information is usually not available from equipment vendors. Moreover, the noise generated may be dependent on actual site condition such as ground hardness and other properties. Noise measurement to obtain such information would not be possible on site due to the surrounding on-site conditions. The method using the acoustic camera will enable the characterization of the noise profile at the respective noise source instead of the use of Sound Level Meter which will measure the noise from all processes and machines from the construction site that reach the measurement location. The information collected for the noise source profile would then be used to design the sonic crystal barriers for the mitigation of noise from construction equipment and processes. Sonic crystals or periodic structures have been reported for applications as noise barriers. For example, a sculpture consisting of hollow stainless steel cylinders located in Madrid was reported to have a sound attenuation peak at 1670Hz [5]. Kushwaha [6] reported that multiple periodic system in tandem could create sound attenuation in the human audible range of frequency. Martinez -Sala et al. [7] reported the possible use of periodic arrays of trees as green acoustic screens. Although there are several reports and even patents (for example, patent WO ) on sonic crystal noise barriers, there is no reported use of such structures for the mitigation of construction noise and in particular for construction machineries. 2. METHODOLOGY 2.1 Noise measurement As a pilot study, the use of acoustic camera for the measurement of the noise profile is demonstrated for a vibratory pile drive. The measurements were carried out using the Norsonic Acoustic Camera Array (Nor 848A) as shown in Figure 3. It is a 1-meter diameter array holding 256 microphones. The measurements were also benchmarked with measurement carried out using Larson Davis Type 1 Sound Level Meter Model 831. The noise profile as well as the frequency for peak noise level can be used as the inputs for the design of sonic crystal noise barriers

3 Figure 3 A typical vibratory pile drive and noise measurement using the acoustic camera 2.2 Finite Element Analysis The finite element analysis was carried out using ABAQUS. Three rows of sonic crystals are proposed for the cage design for construction equipment. The proposed sonic cage encloses the sound source only so as to minimally affect the mobility of the construction equipment and also not to hinder the visibility of the driver or operator. The existing methods of mounting the noise barrier to the different machines are used for demonstration. As shown in Figure 4, the crane s engine compartment is enclosed by panels which are supported from the rotatable platform of the machine. This allows the barrier to rotate together with the cab of the crane, and being suspended, provides also a lot of flexibility and mobility together with the equipment. A potential sonic cage design riding on the same mounting technique is appended next to the same figure. Figure 4 A typical barrier of a crane and how a sonic cage can be designed to replace the original barrier 3. RESULTS AND DISCUSSION 3.1 Sample noise measurement results Table 1 shows the results obtained from the various equipment. The numbers stated are the average of three 1-minute measurements. The db values presented are Leq for a one-minute duration. Table 1 noise measurements from various equipment Larson Davis Norsonic db (A) db (C) The acoustic camera was used to identify the characteristics of the vibratory pile driver as shown in Figures 5 and 6. Figure 5 shows the db(a) results from the acoustic camera whereas Figure 6 shows the db(c) results for the same measurement. A-weighting or db(a) is the most commonly used of a

4 family of curves defined in the International standard IEC 61672:2003 and various national standards relating to the measurement of sound pressure level. Typical regulatory measures are also prescribed in terms of db(a). A-weighting is applied to instrument-measured sound levels in an effort to account for the relative loudness perceived by the human ear, as the ear is less sensitive to low audio frequencies. On the other hand, Frequency less than 20 Hz is not accounted for in db(a). In fact, for A weighting, the contribution from frequencies below 1kHz is gradually reduced. On the other hand, The C weighting is practically linear over several octaves and is suitable for subjective measurements at higher sound pressure levels. In this study, db(c) is also presented to highlight the contribution from the lower frequencies. The contribution from the lower frequency is reflected in the difference in db levels measured by both Larson Davis and the acoustic camera. Figure 5 The noise profile in db(a) for the vibratory pile driver Figure 6 The noise profile in db(c) for the vibratory pile driver 3.2 Finite Element Analysis In the finite element simulations, only a slice model (2.5D) is adopted due to the huge problem size. Each slice of x 9 x m and comprises of approximately 8.5M linear elements, that results in an acoustic accuracy of up to approximately 1800 Hz; and a height models up to 4 ½ levels. The density and bulk modulus of the air properties used are kg/m 3 and N/m 2 (or Pa) respectively. The two side surfaces are assigned with radiative boundary condition. The ground surface is assigned with ground impedance values, while a sound source of 1 Pa is assigned at the engine compartment height of approximately 0.73m. The low magnitude of the sound source is inconsequential since the problem is linear and we are only interested in the parametric study on the insertion loss characteristics. A typical computational domain is shown in Figure 7. The Steady State Dynamics, Direct Analysis procedure is used to perform the frequency sweep analysis from 200 Hz to 2000 Hz, at a linear interval of 100 Hz

5 Figure 7 A typical computational domain Besides looking at contours, acoustic pressure (Pa) at 6 regions are extracted & converted to sound pressure level, SPL (db) to quantitatively assess which barrier designs are better (Figure 8). For each frequency, the sound pressure of the nodes within the regions is averaged before being converted to SPL (db) values. Sampling locations: 0.5 x 0.5 x 0.5 m cube region R1/R4 at 1.2 m above ground examines sound at level 1 R2/R5 at 6.0 m above ground examines noise at 2nd floor (level 3) R3/R6 at 12.0 m above ground examines noise at 4th floor (level 5) Figure 8 The six sample positions for the comparison of db Typical results for 800 and 900 Hz for the sonic cage and comparison with other existing barriers are shown in Figure 9. For the corrugated plate with holes (CPH) barrier design, the specification is shown in Figure 10 while those for the first sonic cage model is shown in Figure 11. Figure 9 Comparison of the performance of various noise barriers

6 Figure 10 Specification of the CPH barrier Figure 11 Specification of the first sonic cage model Comparison of the insertion losses for the different designs is shown in Figure 12. The vertical panel and corrugated plate with holes (CPH) designs have poor performance for R2, R3, R5, R6 but better performance at lower floor due to fully covered plate. The 3-layer sonic cage has better performance at Hz especially for R2 and R6 locations due to curved top edge. The simulation results also show that the barrier top-edge is important for sound attenuation at higher floors and noise leakage at barrier can occur if gaps are too wide. An improved sonic cage design was implemented with twin beams used instead of separate sonic crystal beams for reducing the leakage at lower elevation as shown in Figure 13. The larger overhang and top-edge deflection angle would also reduce the upward noise projection. The gap depth between the rows and slit sizes are tuned to filter frequencies of construction noise ( Hz) resulting in better insertion loss characteristics. The comparison of the insertion loss characteristics of the two sonic cage designs are shown in Figure 14. Figure 12 Comparison of the performance of various noise barriers

7 Figure 13 The second or improved sonic cage design Figure 14 Comparison of the two sonic cage designs 4. CONCLUSIONS The first sonic cage design was found to perform worse than the existing barrier at R1, R4, but slightly better at R2, R5, R6. The CPH design is found to perform worse than the existing barrier at R1 (much worse), R3, R4, R6, but slightly better at R5. The second sonic crystal design has marked improvement for all locations against the first sonic cage design as well as the CPH design. It has better performance than the existing barrier at all locations except R1. The model is currently being improved with reduction in weight for test bedding at construction sites. ACKNOWLEDGEMENTS The authors would like to acknowledge the assistance from the intern students, researchers and laboratory technologists at the Applied Mechanics Laboratory, Department of Mechanical Engineering, National University of Singapore for assisting in the site measurements. This material is based on research/work supported by the Land Transport Authority of Singapore under the LTA Innovation grant (WBS no: R ). REFERENCES 1. Koushki PA, Kartam N, AL-Mutairi N. Urban Development and Construction Noise: A study in Metropolitan Kuwait. Kuwait Journal of Science and Engineering 2003;30(2): Wu WX. Noise Assessments for Construction Noise Impacts. Proc Internoise 2008, October, Shanghai, China, Lee SC, Jeon JY. Assessment of Annoyance Cause by Different Types of Construction Noises. Transactions of the KSNVE 2013;23(12): 1111~ Golmohammadi R, Mohammadi H, Bayat H, Mohraz H, Soltanian AR. Noise Annoyance Due to Construction Worksites. JRHS 2013;13(2): Martinez-Sala R, Rubio C, Garcia-Raffi M, Sanchez-Perez JV, Sanchez-Perez EA, Llinares L. Control of noise by trees arranged like sonic crystals. Journal of Sound and Vibration. 2006;291: Kushwasa MS. Stop-bands for periodic metallic rods: Sculpture that can filter the noise. Applied Physics Letters. 1997;70: Martinez-Sala R, Rubio C, Garcia-Raffi M, Sanchez-Perez JV, Sanchez-Perez EA, Llinares L. Control of noise by trees arranged like sonic crystals. Journal of Sound and Vibration. 2006;291: