Experimental evaluation of a steel U-shaped embedded track for vibration and noise mitigation in metro line retrofit

Abstract

This study evaluates a newly developed steel U-shaped embedded track system designed for vibration and noise mitigation in existing metro lines. The system enables rapid modular installation and provides a compliant polymer support layer for improved vibration isolation. Comprehensive before–after field measurements were conducted on rail and tunnel vibration, in-train noise, and building structure-borne noise. The peak rail acceleration decreased from 219 to 85 m/s², and the peak tunnel-wall acceleration was reduced from 0.194 to 0.025 m/s². and lowered indoor structure-borne noise by 4.9 - 6.1 dB(A). In-train A-weighted noise decreased by 4.1 - 5.1 dB(A). The results confirm that the proposed embedded track provides an effective and practical retrofit solution that meets relevant vibration and noise limits and significantly improves the environmental performance of existing metro lines.

Keywords

embedded track metro retrofit vibration reduction noise control structure-borne noise field measurement

By 2025, China’s urban metro network has surpassed 9,000 kilometers in operation. While metro systems greatly facilitate urban mobility, the vibrations and noise they generate negatively impact nearby residents.¹ Controlling these vibrations and noise is essential to improve living quality and support harmonious urban development.

To address railway-induced environmental vibration, a variety of vibration-mitigation track systems have been developed and widely applied worldwide. These systems can be broadly categorized into track-type and fastener-type solutions. Track-type systems introduce elastic or isolation layers beneath the track slab to reduce the effective support stiffness and limit the transmission of wheel–rail vibrations into the substructure; representative examples include steel-spring floating slab tracks,² rubber floating slab tracks,³ and trapezoidal-sleeper tracks.⁴ Although such systems offer high insertion loss. In contrast, fastener-type systems use highly compliant elastic components to substantially reduce vertical dynamic stiffness. Representative systems include the Pandrol Vanguard⁵ fastener, Cologne Egg fastener,⁶ and double-layer resilient fasteners⁷ etc. The Vanguard system, for example, supports the rail through rubber components beneath the rail head and along the rail web, achieving stiffness about 6 kN/mm and significantly reducing the transmission of wheel–rail vibration energy into the track structure. Despite their widespread application, these mitigation measures still face challenges such as increased vibration energy within the wheel–rail system, accelerated rail corrugation,⁶ insufficient noise control, and high lifecycle costs. This highlights the need for new retrofittable track technologies that can simultaneously achieve vibration reduction and noise suppression for existing metro lines.

Embedded rail track systems have attracted considerable attention since their inception owing to their continuous rail support, high damping capacity, and ability to enhance wheel–rail contact, reduce environmental vibration, and lower radiated noise. The world’s first embedded rail track was installed on a heavy-duty line in Deurne, the Netherlands, in 1976, achieving substantial vibration and noise reduction during operation. Markine⁸ developed the SA42 embedded rail based on the UIC54 rail profile, which reduced the volume of infill material within the groove and decreased wheel–rail radiated noise by approximately 5 dB(A). Van Lier⁹ employed the TWINS model and finite element analysis to study the acoustic radiation of UIC54 embedded rails and found that the wheel–rail noise of embedded tracks was 4–6 dB(A) and 5.5–9 dB(A) lower than those of ballasted and slab track systems, respectively. Nilsson¹⁰ used boundary element and wavenumber finite element methods to compare the sound radiation of fastened and embedded track systems, revealing that embedded tracks radiate significantly less noise than fastened tracks in the 500–2000 Hz mid–high frequency range, whereas fastened tracks perform better in the low-frequency region around 500 Hz. Real et al.¹¹ established a quarter-vehicle–embedded rail–ground coupled model based on wave theory, pioneering studies of environmental vibrations induced by vehicles operating on embedded rails.

In China, research on embedded track systems has advanced rapidly in recent years. Yang Gang et al.¹² conducted frequency response function and longitudinal attenuation tests on embedded and fastened rails along Guangzhou Metro Line 14, showing that embedded tracks provide superior vibration reduction within the 550–1200 Hz range, particularly above 750 Hz. Shi Xiaodi¹³ applied explicit dynamic finite element analysis and a three-dimensional high-speed wheel–rail rolling model to investigate the impact forces generated by trains running at 400 km/h over embedded tracks with varying irregularities. He Aonan et al.¹⁴ applied hyperelastic constitutive theory to explore the influence of polymer material properties on wheel–rail dynamic interaction, concluding that polymers have limited suppression capability for low-frequency vibrations below 400 Hz. Han Jian et al.^15–19 developed a coupled metro train–embedded track dynamic model incorporating the inertia, elasticity, and damping properties of polymer materials. Their studies systematically evaluated how vertical and lateral polymer stiffness, slab support stiffness, and slab dimensions affect the dynamic response of embedded tracks and vehicle ride comfort. He Yuanpeng, Zhao Yue et al.^20–22 established a coupled wheel–embedded track model based on wheel–rail rolling sound radiation theory and finite element–boundary element methods, revealing the effects of vehicle elasticity, wheel–rail friction coefficient, and groove polymer modulus on rail vibration and acoustic radiation. Lei Xiaoyan et al.²³ conducted hammer impact tests on a 5 m full-scale embedded track model to analyze acceleration admittance characteristics and found that polymer infill materials effectively restrain rail vibration within the 100–300 Hz frequency band.

However, most of the aforementioned studies primarily focused on newly constructed lines or laboratory-scale simulations, while research on the retrofit application of embedded tracks to existing metro lines remains limited. In particular, few studies have provided full-scale field measurements to quantitatively evaluate the vibration and noise control performance of embedded track systems under real operating conditions. Moreover, the complex coupling among rail, tunnel structure, and surrounding buildings has not been fully validated through in-situ testing. To bridge this gap, the present study conducts an integrated experimental investigation on a metro line retrofitted with a newly developed steel U-shaped embedded track, systematically assessing its effectiveness in mitigating track and tunnel vibration, in-train noise, and building structure-borne noise. Section 1 introduces the design concept, structural characteristics, and construction method of the steel U-shaped embedded track. Section 2 describes the field test methodology and measurement point layout. Section 3 presents the test results and corresponding analysis of vibration and noise performance before and after the track retrofit.

1. Introduction of the steel U-shaped embedded track

The U-shaped embedded track is a novel continuously supported track structure designed for upgrading existing lines. Its design concept is to be compatible with conventional fastened tracks on existing lines, taking into account factors such as alignment, sleeper specifications, and fastener spacing. The system employs prefabricated steel U-shaped rail troughs combined with polymer damping materials to achieve continuous rail support and continuous locking, thereby providing active control of vibration and noise at both the source and along the transmission path. This type of track is suitable for mainline sections of urban rail transit with operating speeds of up to 120 km/h and can be applied to standard slab track.

The U-shaped embedded track structure consists of the rail, steel support trough, wedge blocks, polymer damping material, rail-adjusting components, fastening bolts, leveling plates, elastic pads, steel cover plates, eccentric cylinders, steel trough pads, sleepers, and track bed (as shown in Figure 1). The steel support trough is assembled on-site from a T-shaped plate (2a) and an L-shaped plate (2b). The standard steel trough has a length of 6.25 m and accommodates sleeper spacing of 625 ± 15 mm. The rail-adjusting components are symmetrically arranged on both sides of the rail web at a fixed 625 mm interval. The elastic pads are laid continuously along the bottom of the steel trough, and the polymer damping material wraps more than two-thirds of the rail height from the rail base.

Figure 1.

Structure diagram of steel groove embedded continuous support track.

The U-shaped embedded track structure tightly secures the rail using the steel support trough and polymer damping material within the trough, transforming the traditional fastened, discretely supported track system into a continuously supported system. Dynamic properties of the polymer material were characterized through DMA testing at frequencies of 1 and 50 Hz over a temperature range of −50 °C to 80 °C. The storage modulus decreases markedly with increasing temperature, from approximately 36–254 MPa at −50 °C to 4–7 MPa at 75 °C, indicating strong temperature-dependent stiffness. The loss modulus shows a similar decreasing trend, and the loss factor remains within 0.05–0.20 in the operating temperature range (0–40 °C), suggesting moderate damping capacity. These results confirm that the polymer maintains sufficient stiffness and energy-dissipation capability for vibration-isolation applications.

The continuous support structure completely eliminates the pinned–pinned resonance modes of the rail,²⁴ effectively suppresses corrugation formation. Moreover, since more than two-thirds of the rail surface is wrapped with polymer damping material, the rail web and base radiated noise are substantially suppressed, resulting in a significant reduction of interior noise of trains. The U-shaped embedded track renovation process consists of five main steps: rail and steel trough interface treatment (to enhance bonding between the polymer material, rail, and steel trough), fastener removal, steel trough installation, polymer material casting, and curing/fixation (approximately 1.5 hours). Laboratory tests show that the rapid-curing polymer continues to gain stiffness after the initial 1.5 h curing period and reaches a stable modulus within approximately 24 hours. All field measurements in this study were performed after this stabilization period, ensuring consistent material properties throughout the experiments. Based on on-site measurements, each maintenance window point can accomplish about 60 meters of single-track renovation, with a construction efficiency of approximately 1.2–1.5 meters per person.

2. Experimental setup and overview

The retrofitted section of the line is located near a station with a burial depth of approximately 15 m. Residential communities are situated to the south and north of the station, with the nearest buildings located about 25 m from the line (Figure 2). The line operates with eight-car Type A trains at a maximum speed of 80 km/h. Because the measurement section is located close to a station, the train speed varies as trains decelerate into and accelerate out of the platform. During all tests, the operating speed at the measurement point ranged approximately from 40 to 55 km/h. All trains followed the scheduled timetable and exhibited nearly identical speed profiles at this location. To address noise complaints from the nearby communities, embedded track systems were adopted to retrofit both the up and down tracks. The retrofit area includes the station section and an adjacent portion of the line. Photographs of the line before and after the retrofit are shown in Figure 3. The test section is on a straight track with stable wheel–rail conditions. Rail roughness and friction levels were unchanged before and after retrofit, ensuring that the measured vibration differences result from the modified rail support system rather than variations in contact conditions.

Figure 2.

Overview of the measurements.

Figure 3.

Changes in the track structure before and after the retrofit.

To comprehensively assess how retrofitting the line with embedded track affects vibrations of the track, vehicles, and surrounding buildings, vibration and acoustic sensors were deployed trackside, on board, and within nearby buildings, and a comprehensive measurement campaign was conducted before and after the retrofit. The detailed spatial arrangement of all vibration sensors, including their measurement directions and relative locations on the rail, track slab and tunnel wall, is shown in Figure 4. Rail vibration accelerometers were installed on the rail foot to measure vertical rail vibration. For the slab track, accelerometers were mounted at the center of the track slab. Tunnel wall vibration measurements were conducted at a height of 1.25 m above the rail head level, with sensors rigidly fixed using angle steel brackets to ensure reliable measurement of vertical vibration components. Rail accelerometers were installed using magnetic mounts, while accelerometers on the track slab and tunnel wall were rigidly bolted to steel plates or angle steel brackets, which were then adhesively bonded to the structural surfaces.

Figure 4.

Layout of trackside measurement points.

Indoor vibration testing in the building was conducted in an unrenovated, unfurnished room. The measurements included floor vibrations in different rooms (bedroom and living room) and in-room structural noise. Noise measurements were conducted using a GRAS 46AE 1/2-inch free-field microphone, and vibration measurements used a low-frequency, high-sensitivity 941B sensor (Figure 5).

Figure 5.

Building vibration and noise measurements.

Car interior noise measurements were conducted in accordance with GB 14892-2006: Urban Rail Transit—Train Noise Limits and Measurement Methods. The microphone was placed in the driver’s cab at the front in the direction of travel, positioned 1.2 m above the floor (Shown in Figure 6).

Figure 6.

Layout of in-train noise measurement points.

All parameters used in the tests are listed in Table 1.

Table 1.

Information of sensors used in the tests.

Model	Type	Sensitivity	Measurement range	Test location
PCB352C04	Vibration Accelerometer	10mV/g	±500g	Rail
PCB393B04	Vibration Accelerometer	1V/g	±5g	Track, Tunnel
941B	Vibration Accelerometer	3V/g	±2g	Indoor vibration
GRAS 46AE	Microphone	50mv/Pa	17-138 dB	Indoor noise, in-train noise

3. Experimental results and discussion

To verify the vibration and noise reduction performance of this track system during mainline operation, in-situ comparative measurements were conducted on sections before retrofitting (conventional fastened track on monolithic slab) and after retrofitting (steel-groove embedded track). The measurements included track and tunnel vibration, in-vehicle noise, and interior secondary structure noise within adjacent buildings (Figure 7).

Figure 7.

Vibration time history and frequency spectrum of rail before and after retrofitting.

3.1. Track and tunnel vibration measurement and analysis

To visually demonstrate the source-level vibration control provided by the steel-groove embedded track, comparisons were made between the rail and tunnel wall acceleration time histories, acceleration level spectra, and maximum Z vibration levels before and after retrofitting, as shown in Figure 8. The acceleration level (in dB) was calculated using an unweighted (Z-weighted) formulation. The reference acceleration was taken as $a_{0} = 1 \times 10^{- 6} m / s^{2}$ , and the acceleration level was defined as

L_{a} = 20 \log_{10} (\frac{a_{rms}}{a_{0}})

where

a_{rms}

is the root-mean-square value of the measured acceleration.

Figure 8.

Vibration time history and frequency spectrum of tunnel before and after retrofitting.

The peak rail acceleration decreased from 219 m/s² to 85 m/s² after retrofitting (Table 2), while the peak tunnel wall acceleration dropped from 0.32 m/s² to 0.02 m/s². The rail acceleration level increased by approximately 10 dB around 6.3 Hz and decreased by about 9 dB near 80 Hz. This behavior is consistent with the mechanical characteristics of embedded track systems and can be explained by the change in global stiffness–mass distribution of the rail support. The tunnel wall acceleration level was reduced by 16–24 dB across the 6.3–200 Hz frequency range. The maximum Z vibration level of the rail increased from 108.7 dB to 111.1 dB, whereas that of the tunnel wall decreased from 77.3 dB to 58.8 dB.

Table 2.

Peak and effective value of vibration acceleration of rail and tunnel before and after retrofitting.

Condition		Vertical vibration acceleration (m/s²)
		Rail		Tunnel
		Peak	RMS value	Peak	RMS value
Before retrofitting	Average	219.069	15.759	0.321	0.039
	Standard deviation	17.931	1.462	0.078	0.011
	Confidence Interval	(201.138, 237.000)	(14.297, 17.221)	(0.243, 0.399)	(0.028, 0.05)
After retrofitting	Average	84.657	9.135	0.025	0.004
	Standard deviation	10.436	0.700	0.004	0.001
	Confidence Interval	(74.221, 95.093)	(8.435, 9.835)	(0.021, 0.029)	(0.003, 0.005)

Figure 9(a) and (b) show the time–frequency maps of rail vibration before and after the retrofit, analyzed using the short-time Fourier transform (STFT). A Hanning window with a window length of fs/30 (approximately 33 ms) was adopted, with a 30% overlap and an FFT length of 1024. The corresponding frequency resolution is fs/1024. A pronounced difference is observed in the 700–1400 Hz band: the conventional track exhibits a broad affected range, indicating a relatively low vibration attenuation rate, whereas the embedded track effectively suppresses vibration propagation within this band, which is highly favorable for noise control.

Figure 9.

Time–frequency spectrograms of rail vibration before and after the retrofit. (a) Before retrofitting. (b) After retrofitting.

3.2. In-vehicle noise measurement and analysis

To provide a clear demonstration of the noise control effectiveness of the steel-groove embedded track, the single-value sound pressure levels and corresponding spectra inside the carriage before and after the retrofit are compared, as shown in Figure 10. The results indicate that, following the retrofit, the interior-train equivalent sound pressure level and the maximum sound pressure level are reduced by 5.1 dB(A) and 4.1 dB(A), respectively. Moreover, a pronounced reduction in sound pressure is observed across the 20–2000 Hz frequency range. These findings confirm that the U-shape embedded track can significantly suppress interior noise levels.

Figure 10.

Comparison of A-weighted sound pressure levels inside the train before and after the track retrofit.

The highlighted area represents the embedded track zone, where the acoustic characteristics change markedly. Before the retrofit, strong noise components were observed around 700 Hz and 1200 Hz (Figure 11). After entering the embedded track section, these components were largely eliminated, and the dominant energy shifted to 300–400 Hz, consistent with the rail vibration analysis. Figure 12 compares the noise spectra of the conventional and embedded tracks. The results indicate effective suppression of noise in the 80–125 Hz and 800–1250 Hz bands, while the 300–400 Hz component becomes more pronounced. Overall, the A-weighted sound pressure level decreased by about 2.5 dB(A) after the retrofit.

Figure 11.

Time–frequency distribution of in-train noise in the retrofitted section with the embedded track.

Figure 12.

Comparison of sound pressure levels between the conventional and embedded tracks.

3.3. Building structure-borne noise

To evaluate the effectiveness of the embedded track in controlling structure-borne noise in nearby buildings, the indoor noise levels in the living room and bedroom were compared before and after the track retrofit. Figure 13 presents the comparison of A-weighted sound pressure levels in the 20–400 Hz range, which is typically used to characterize structural noise. It can be observed that after the retrofit, the equivalent A-weighted sound pressure levels of secondary noise in the living room and bedroom decreased by 4.9 dB(A) and 6.1 dB(A), respectively. The A-weighted sound pressure level was calculated following the Chinese standard JGJ/T 170-2009, which specifies A-weighting for evaluating structure-borne noise in buildings induced by metro operation. Since the assessment is limited to 16–200 Hz.

Figure 13.

Comparison of structural noise in the living room and bedroom before and after the track retrofit.

To further analyze the frequency characteristics of the secondary structural noise, one-third-octave-band sound pressure levels within the 31.5–250 Hz range were evaluated according to GB 50355-2018. The Figure 14 results show that the band-limited A-weighted levels decreased by 2.7–10.3 dB(A) in the living room and 7.1–16.3 dB(A) in the bedroom, both below the Class I limit specified in the standard. Therefore, the steel-channel embedded track effectively reduces secondary structural noise in buildings. Figure 15 shows that the indoor vibration levels decreased slightly in the 40–63 Hz range after the retrofit, indicating that the embedded track effectively reduces building vibration (Table 3).

Figure 14.

Comparison of octave band sound pressure levels in the living room and bedroom before and after the track retrofit.

Figure 15.

Comparison of 1/3 octave band vibration levels in the living bedroom before and after the track retrofit.

Table 3.

A-weighted SPL (20-400 Hz) before and after retrofitting.

Condition		A-weighted SPL (20-400 Hz) (dB(A))
Condition		Living room	Bedroom
Before retrofitting	Average	38.8	38.1
	Standard deviation	0.93	1.15
	Confidence Interval	(37.87, 39.73)	(36.95, 39.25)
After retrofitting	Average	33.9	32
	Standard deviation	0.99	1.63
	Confidence Interval	(32.91, 34.89)	(30.37, 33.63)

4. Results and discussion

This study evaluated a newly developed U-shaped steel embedded track system designed for retrofitting existing metro lines. Comprehensive before-and-after field measurements demonstrate that the system provides substantial improvements in vibration and noise control.

(1) rail and tunnel vibration: Peak rail acceleration decreased from 219 m/s² to 85 m/s², and peak tunnel wall acceleration from 0.20 m/s² to 0.02 m/s². The retrofit reduced rail vibration by up to 9 dB near 80 Hz and tunnel vibration by 16–24 dB within 6.3–200 Hz.

(2) In-vehicle noise: The In-vehicle equivalent and maximum sound pressure levels were reduced by 5.1 dB(A) and 4.1 dB(A), respectively, following the retrofit. Noise components in the 80–125 Hz and 800–1250 Hz frequency bands were effectively suppressed, though energy in the 300–400 Hz band became more prominent.

(3) Building structure-borne noise: Indoor secondary structural noise in nearby residences decreased by 4.9–6.1 dB(A), and one-third-octave A-weighted levels in the 31.5–250 Hz bands were reduced by 2.7–16.3 dB(A), meeting the Class I limits of GB 50355-2018.

Overall, the newly developed steel U-shaped embedded track demonstrates clear effectiveness in reducing rail–tunnel vibration transmission and building structure-borne noise, offering a practical and quantifiable retrofit solution for existing metro and tram lines.

The embedded track system introduces a continuous polymer layer that lowers the rail support stiffness and increases the effective vibration participation mass. These changes shift the dominant resonance of the track toward lower frequencies and redistribute vibration energy across the spectrum. As a result, rail vibration may increase slightly at very low frequencies, while mid-frequency components (e.g., 20–200 Hz) are significantly reduced and less energy is transmitted to the tunnel and surrounding buildings. Furthermore, the polymer layer provides substantial damping, which attenuates high-frequency vibration propagation. These mechanisms collectively explain the observed vibration reduction in the tunnel wall and buildings, despite localized changes in the rail response.

In the future, a long-term monitoring program (at 6 and 12 months after installation) has been planned to evaluate potential aging or stiffness changes of the polymer layer, and the results will be reported in future work.

Footnotes

ORCID iDs

Qingjie Liu

Jiapeng Yuan

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

References

Yan

Liu

Zhang

, et al. Current status and prospects of vibration and noise control of buildings under the influence of rail transit. Railway Standard Design 2025; 69: 1–14.

Liu

Wang

Chen

, et al. Review on vibration transmission mitigation measures for rail transit systems. Railway Standard Design 2025; 30: 1–15.

Zou

Pang

, et al. Environmental noise and vibration characteristics of rubber-spring floating slab track. Environmental Science and Pollution Research 2021; 28(11): 13671–13689.

Zheng

. Vibro-acoustic characteristics of a rail transit viaduct paved with trapezoidal sleeper track. Journal of Vibration and Control 2024; 30(19-20): 4637–4649.

Faure

Bongini

. Mitigation measures installed on commercial track (Deliverable D3.5). RIVAS Project, WP3, European Commission FP7, 2013.

Cui

Peng

, et al. Fracture mechanism and control method of elastic strip of Cologne-egg fastener in the high-prevalence section of rail corrugation. Engineering Failure Analysis 2023; 152: 107463.

Xia

Wei

Jia-yu

, et al. Influence of frequency-dependent characteristics of double-layer nonlinear fastener system on wheel-rail dynamics characteristics. Journal of Traffic and Transportation Engineering 2024; 24(6): 159–171.

Markine

de Man

Jovanovic

, et al. Modelling and optimisation of an embedded rail structure. Rail International, 2000.

Van Lier

. The vibro-acoustic modelling of slab track with embedded rails. Journal of Sound and Vibration 2000; 231(3): 805–817.

10.

Nilsson

Jones

CIC

Thompson

, et al. A waveguide finite element and boundary element approach to calculating the sound radiated by railway and tram rails. Journal of Sound and Vibration 2009; 321(3): 813–836.

11.

Real

Martinez

Montalbán

, et al. Modelling vibrations caused by tram movement on slab track line. Mathematical and Computer Modelling 2011; 54(1): 280–291.

12.

Yang

Han

, et al. Study on vibration transmission of embedded track systems. Urban Mass Transit 2024; 27(7): 197–203.

13.

Shi

Han

, et al. Impact characteristics of embedded track based on a three-dimensional rolling contact model. Noise and Vibration Control 2024; 44(3): 165–169.

14.

Wang

, et al. Influence of hyperelastic mechanical properties of embedded track on the dynamic interaction between wheel and rail. Journal of Vibration and Shock 2023; 42(11): 48–57.

15.

Han

. Study on dynamic behaviour and vibro-acoustic characteristics of metro train and embedded track system. Doctoral dissertation, Southwest Jiaotong University, 2018.

16.

Han

Xiao

, et al. Dynamic behaviour of metro train and embedded track system I: Theoretical modelling, experimental analysis, and verification. Journal of Mechanical Engineering 2020; 56(22): 148–158.

17.

Han

Xiao

Yang

, et al. Dynamic behaviour of metro train and embedded track system II: Effect of track parameters on dynamic behaviour. Journal of Mechanical Engineering 2020; 56(24): 173–180.

18.

Han

Xiao

Wen

, et al. Dynamic behaviour of metro train and embedded track system III: Comparative analysis of dynamic responses of embedded and conventional tracks. Journal of Mechanical Engineering 2022; 58(6): 169–176.

19.

Han

Wang

, et al. Simulation and experimental study on vibration and acoustic characteristics of a continuous supported embedded track. Applied Acoustics 2021; 180: 108103.

20.

Jiao

Zhao

, et al. Influence of elastic material characteristics on vibration and acoustic radiation of embedded tracks. Noise and Vibration Control 2015; 35(3): 51–55.

21.

Han

Chen

, et al. Experimental analysis of curve squeal noise in embedded and fastener track systems. Journal of Mechanical Engineering 2023; 59(6): 194–203.

22.

Zhao

Wang

, et al. Time-domain modelling and analysis of curve squeal noise caused by tram and embedded rail track. Journal of Mechanical Engineering 2021; 57(2): 158–168.

23.

Lei

Wei

Liu

, et al. Experimental study on vibration reduction performance of embedded track based on hammering method. Noise and Vibration Control 2021; 41(6): 37–41.

24.

Yang

, et al. Performance analysis of continuously supported embedded rail track system in urban rail transit. Urban Mass Transit 2020; 23(7): 102–107.