Weihua Ma

Fatigue assessment of maglev train levitation frames lacks dedicated criteria, and deterministic methods cannot account for the dispersion of fatigue life. This study proposes a probabilistic fatigue strength assessment methodology that integrates P-S-N curves with Goodman-Smith (GS) fatigue limit diagram, and applies it to a maglev train levitation frame. Fatigue test data of ZL101A base metal and 6005A aluminium alloy butt and corner welded joints were fitted with P-S-N curves at survival probabilities of 50%, 90%, 95% and 99% by combining a three-parameter Weibull distribution, maximum likelihood estimation and the Stromeyer model. The octagonal GS diagram was modified with probability-dependent symmetrical cyclic fatigue strengths corrected for stress concentration and welding residual stress. A refined finite element model of the frame was verified by mesh convergence and full-scale bench tests, with a maximum stress error of about 14%, and provided the stress states under 17 operational load cases. Case analysis results show that all critical locations lie within the probabilistic safety envelopes, and that the safety margins of the butt joint, corner joint and base metal exceed those of the traditional deterministic method by 44.32%, 6.33% and 46.11%, respectively.

Aiming at the lack of dedicated fatigue strength assessment criteria for maglev train levitation frames and the inadequacy of traditional rail train assessment methods, this study proposes a novel assessment methodology integrating the P-S-N curve with a modified Goodman-Smith (GS) fatigue limit diagram. Initially, a finite element analysis (FEA) model is established based on the structural and loading characteristics of the levitation frame, with its validity confirmed through full-scale bench testing. Subsequently, utilizing material fatigue test data from critical locations of the levitation frame, P-S-N curves under varying failure probabilities are fitted by combining the two-parameter Weibull distribution with the maximum likelihood estimation method. Thereafter, based on the characteristics of the P-S-N curve and the GS fatigue limit diagram, corrections for the symmetrical cyclic fatigue strength of both base metal and welded structures are accomplished and incorporated into the GS diagram for levitation frame fatigue strength assessment. Case analysis demonstrates that assessment stringency progressively increases with survival probability. The methodology comprehensively addresses multiaxial stress states and fatigue life dispersion, significantly enhancing both assessment rigor and engineering safety. This approach provides valuable references for lightweight design, weld optimization, and service safety evaluation of maglev levitation frames.

Maglev vehicles operate through non-contact levitation via electromagnetic forces, with track irregularities serving as a critical factor affecting levitation stability and vehicle vibration. Currently, field measurements of maglev track irregularities remain scarce, often necessitating the use of alternative track spectrum as equivalents for dynamic analysis. This work proposes an automated track inspection methodology capable of real-time data acquisition, which was implemented on the Qingyuan Maglev Line in China. Through data processing, filtering, and signal reconstruction, the true track irregularities were recovered. Mode decomposition based on CEEMDAN revealed that the wavelength components of irregularities align closely with structural periodicities such as sleepers and track panels. Specifically, the primary wavelength range for height irregularity spans 1.23–30.62 m, while for alignment irregularity it is 1.31–25.66 m. Furthermore, power spectral density (PSD) analysis of the reconstructed spatial irregularities was performed, and percentile-based spectral characteristics were extracted and compared. The 50% percentile spectrum was selected for PSD fitting, with a coefficient of determination (R2) approached 1, indicating that the constructed PSD function effectively represents the characteristics of the actual maglev track irregularities.

The traction device serves as the mechanical structure responsible for transmitting longitudinal loads between the car-body and the running gear. However, during emergency braking or collision impacts in maglev vehicles, the structural integrity of this device is compromised, with bolts pull-out failure having been documented in engineering practice. To enhance the connection reliability of the traction device, this study analyzes its stress states under various longitudinal impact scenarios. A finite element model of the maglev vehicle was established to compute transient dynamic responses during collisions, with a specific focus on the failure mechanisms of the traction device, particularly the connection bolts. Furthermore, an innovative structure featuring bidirectional buffering capabilities is proposed to protect the bolts. The results indicate that the inclination angle of the traction rod generates a vertical component force under longitudinal impact, which exerts a tensile load on the connecting bolts; this mechanism is identified as the critical factor leading to pull-out failures. The bolt connection failure process comprises four distinct stages: stability, loading, front-end damage, and rear-end damage. Bolt pull-out was observed in all four types of traction devices during collision events, with findings suggesting that longer traction rods contribute to an improved stress distribution. Based on the long traction rod configuration, the incorporation of a bidirectional buffering structure effectively absorbs a portion of the longitudinal impact energy, thereby preventing bolt damage. Following optimization, the average maximum von mises in the bolts decreased by over 20%, and no further connection failures were observed.

As a core technology in 3D measurement, laser stripe center extraction is widely applied in industrial inspection, robot navigation, and biomedicine. However, traditional methods struggle to balance denoising effectiveness and positioning accuracy when handling complex noise and non-uniform width stripes. To address this bottleneck, this paper proposes a denoising-adaptive weighted average width stripe center extraction algorithm based on an improved Hessian Matrix, integrating deep learning with traditional image processing for high-precision extraction. A U-Net++ denoising network with a spatial attention module is designed to focus on stripe regions, supplemented by a distance-aware mechanism that dynamically adjusts denoising intensity based on pixel-stripe distance. For center extraction, an improved Hessian Matrix algorithm is proposed, incorporating a curvature-adaptive FIR filter and adaptive weighted average width calculation to adapt to stripe morphology changes. Experimental results show the algorithm outperforms comparative methods, achieving 35.26 dB (PSNR), 0.962 (SSIM), and 6.14 (RMSE) in denoising. Under 200 μs, 500 μs, 1000 μs, and 1500 μs exposure conditions, the absolute radius errors are reduced to 0.2052 mm, 0.1743 mm, 0.0268 mm, and 0.0281 mm, respectively, verifying its reliability and stability in practical applications.

EMS (Electro-magnetic Suspension) maglev vehicles rely on actively controlled electromagnetic attraction for levitation. In the event of a failure, the levitation frame falls and collides with the track, subjecting the bolts at the support skid positions to shear forces and creating safety risks. Finite element models of the levitation frame and track were established, and the contact relationships between structural components were considered in the model. The transient dynamic behavior of the fall collision was simulated using ANSYS/LS-DYNA software, with a focus on analyzing the collision strength of the bolts, and structural optimization was implemented to enhance bolt protection. Research results indicate that during collision events, bolts at the support skid positions bear the primary impact load. The peak equivalent stress on bolts at both bottom and top positions exceeds 650[ ]MPa, with over 90% of bolts experiencing maximum stress surpassing allowable limits, rendering them critically vulnerable. By adding key connections, the stress state of bolts was optimized, with the contact forces and equivalent stresses significantly decreased, reducing the number of bolts exceeding allowable stress by 60%, thereby effectively ensuring bolts safety.

The wheel wear of heavy-duty locomotives is more complicated than that of other vehicles when it affects tractive force and adhesion control. The purpose of this article is to calculate wheel wear on a curved track using a co-simulation dynamic model of a heavy-duty locomotive with adhesion control. Additionally, the effect of creep threshold, decline slope, curve radius, and passing speed on wheel wear is further investigated. The findings demonstrate that wheel wear can be successfully decreased by adjusting the creep threshold and decline slope of adhesion control parameters. The wheel-rail stick-slip vibration behavior on the curve track can be eliminated by increasing the curve’s radius and passing speed, but the vertical wear at the left wheel’s flange will rise sharply with the increase of passing speed on right hand curve.

Maglev vehicles apply the entire vehicle load uniformly onto bridges through levitation forces. In assessing the dynamic characteristics of the maglev train–bridge coupling system, it is reasonable to simplify the distributed levitation force as a concentrated force. This article theoretically derives the analytical response of bridge dynamics under the action of a single constant force and conducts numerical simulations for a moving single constant force and a series of equally spaced constant forces passing over simply supported beams and two-span continuous beams, respectively. The topic of discussion is the response of bridge dynamics when different degrees of force concentration are involved. High-precision displacement and acceleration sensors were utilized to conduct tests on the Shanghai maglev line to verify the accuracy of the simulation results. The results indicate that when simplifying the distributed levitation force into a concentrated force model, a frequency ratio can be used to analyze the conditions for resonance between the train and the bridge and to calculate the critical speed of the train; the levitation distribution force of a high-speed maglev vehicle can be simplified into four groups of concentrated forces based on the number of levitation frames to achieve sufficient accuracy, with the dynamic response of the bridge being close to that under distributed loads.

A comparative study was conducted on Levitation Frame with Mid-set Air Spring (LFMAS) and Levitation Frame with End-set Air Spring (LFEAS). First of all, the work done by air spring force and by track excitation on two levitation frames were analysed, respectively, and the mechanism of the low dynamic interaction of LFMAS with guideway was explained theoretically; then, dynamics simulation and coupled vibration test on LFEAS and LFMAS were carried out using coupled vibration test bench of single levitation frame. Theoretical equations show that both the work done by the air spring forces and by track excitation on LFMAS are less than those on LFEAS under the same vibration condition, demonstrating that LFMAS has better levitation stability than LFEAS. Simulation and test results show that the fluctuation of levitation gap as well as the vibration acceleration of track beam platform, levitation frame and car-body of LFMAS are all smaller than those of LFEAS; with the increase of sine excitation frequency or vehicle speed, the superiority of LFMAS in dynamics performance is more obvious.

In recent years, low and medium speed maglev (LMSM) transportation system has been greatly promoted in China to meet different social demands. Accordingly, a LMSM train using levitation frames with mid-set air springs has been developed, and it is necessary to investigate the dynamic characteristics of its train-track-bridge coupled system. First, environmental vibration tests were conducted on the bridge to study its self-vibration characteristics. Then, field tests were carried out on this coupled system within the 60 km/h speed range. Subsequently, a three-dimensional rigid-flexible coupled dynamics model of the system was developed to evaluate its dynamic behavior at higher speeds. Results show that the first-order vertical and lateral bending frequencies of the bridge are 7.32 Hz and 6.00 Hz, respectively; train speed has little effect on the vertical dynamic displacement of the track-bridge system, yet has a significant impact on the acceleration responses of the coupled system; key dynamic indicators of the bridge meet the relevant requirements; the vertical ride quality of car body is not so good as that in the lateral direction, but on the whole the train has “Good” ride quality. These results can provide a useful reference for the optimal design of LMSM transportation systems.