This study develops an impact-acoustic framework for classifying predefined surface-damage states on steam-turbine blades, identifying one of three damage regions, and estimating class-dependent severity parameters. Impact signals were recorded for 5 s at 22,050 Hz under controlled impact and microphone angles. After event screening, each signal was converted into a 13-coefficient W-MFCC map that combines continuous-wavelet time-frequency localization, Mel-scale subband mapping, and cepstral compression. A hybrid ResNet-50-Transformer-PANet-CMTL network was designed to extract hierarchical local features, model long-range dependencies, fuse multi-scale information, and produce conditional multi-task outputs. The dataset contained 1,200 training samples and 300 validation samples covering crack, crack-erosion, and crack-abrasion conditions in Areas 1-3. On the reported validation split, W-MFCC correctly classified 296 of 300 samples (98.67 %), compared with 89.33 % for MFCC and 87.33 % for wavelet input, and its loss curves showed more stable convergence. The full multi-scale configuration also exhibited smoother training behavior than the ResNet-50 + Transformer baseline. These findings demonstrate the feasibility of combining W-MFCC representation learning with multi-scale conditional prediction for controlled turbine-blade condition assessment.
In order to improve ride comfort, mounting system was required to effectively isolate the vibrations generated by the engine, preventing them from being transmitted to the vehicle body and the cabin, while ensuring that the displacement of the powertrain remains within controllable limits under various operating conditions. This study investigated the vibration isolation performance of single bell plate hydraulic mount in a quarter‑car model. Key parameters including force transmissibility and relative displacement transmissibility were analyzed under varying damping of the main rubber spring, different mass combinations (engine, body, wheel) and different road excitations. The results showed that with optimized parameters ( K u = 3.39×10 4 N/m, B u = 2500 N·s/m, K t = 2.5×10 5 N/m, B r = 100 N·s/m), minimum force transmissibility in the low‑frequency region and low relative‑displacement transmissibility in the high‑frequency region can be achieved. Appropriate matching of M b , M w , and M e further reduced both force and relative‑displacement transmissibility. Under random road excitation, force transmissibility, mount stroke, and suspension stroke increased with road roughness, whereas relative displacement transmissibility between engine and frame decreased significantly (by 67.0 % on Grade B and 85.5 % on Grade C compared to Grade A). With impact excitation, force transmissibility changed variably with grade, while mount and suspension displacements were markedly amplified; nevertheless, the mount maintained effective control across all roughness levels.
Prefabricated bridge structures have gained increasing attention in modern civil engineering due to their advantages in construction efficiency and quality control. However, their application in seismic regions remains limited, primarily due to insufficient understanding of their seismic performance and damage mechanisms. This study investigates the seismic behavior of prefabricated double-column piers with grouted sleeve connections using a high-fidelity finite element model developed in ABAQUS. A quasi-static loading protocol is employed to evaluate key performance indicators, including hysteretic response, ductility, energy dissipation capacity, and damage evolution. The seismic performance of prefabricated piers is systematically compared with that of cast-in-place counterparts. Furthermore, the influence of pier configuration (single-, double-, and three-column systems) on seismic performance is examined. The results indicate that, although prefabricated double-column piers exhibit slightly reduced ductility, they demonstrate superior peak load capacity and enhanced energy dissipation capacity compared to cast-in-place piers. Among different configurations, double-column piers show the most balanced force distribution and minimal damage concentration. In contrast, single- and three-column systems exhibit significantly lower load capacity and energy dissipation, with reductions of 66.4 % and 23.3 %, respectively. These findings provide valuable insights into the seismic design of prefabricated bridge piers and support their application in seismic regions.
Drawbacks of continuous flexible manipulators, such as insufficient trajectory tracking accuracy and obvious flexible vibration, were targeted, and an integrated hierarchical strategy of trajectory replanning and disturbance rejection control was proposed for cable-driven flexible manipulators. A STO-MPC (Stochastic Trajectory Optimization Model Predictive Control) framework was constructed. Probabilistic obstacle avoidance constraints and gradient-independent solving mechanisms were introduced to tackle the high computational delay and poor dynamic adaptability of traditional planning approaches, while stiffness constraints and residual vibration suppression requirements of flexible structures were fully satisfied. A cooperative control framework of ADO-RITSMC (Adaptive Disturbance Observer-Rapid Integral Terminal Sliding Mode Control) was established. Model-free real-time compensation for multi-source disturbances was realized, finite-time convergence of tracking errors was achieved by the improved terminal sliding mode, and residual vibration of flexible links was effectively suppressed accordingly. Comprehensive comparative tests were conducted on standard O-shaped trajectories and high-curvature V-shaped trajectories, with GO-MPC (Gaussian Observer-based Model Predictive Control), APF-MPC (Artificial Potential Field-based Model Predictive Control) sliding mode algorithms. Experimental results demonstrated that STO-MPC can achieve the lowest peak computation time, completes the convergence of obstacle state estimation within 0.5 s, and yields a steady-state velocity estimation error of 0.0015 m/s. Its trajectory tracking RMSE (Root Mean Square Error) is 18.7 % and 36.8 % lower than that of GO-MPC and APF-MPC respectively, delivering superior real-time performance, estimation stability and tracking accuracy. The proposed ADO-RITSMC reduces vibration amplitude by 16.7 % with a peak vibration acceleration of –1.0 g, and exhibits faster vibration attenuation and slighter trajectory oscillation during dynamic obstacle avoidance, which fully verifies the hierarchical collaborative advantages of STO-MPC trajectory replanning and ADO-RITSMC vibration suppression.
To solve the engineering problems of severe vibration in the dehydration stage, insufficient reliability of load-bearing components, and the difficulty in balancing lightweight design and dynamic performance of industrial washing machines, mechanical performance analysis of key load-bearing components and optimization of vibration isolation systems were conducted. Dynamic modeling, finite element simulation and multi-objective optimization methods were integrated under extreme working load conditions. The dynamic model of the vibration isolation system was established by the Lagrange method. The rigid-flexible coupling simulation model of the cabinet and the refined finite element model of the inner cylinder were constructed. The Kriging surrogate model was adopted to fit nonlinear correlations between structural design variables and mechanical responses. Optimization results quantitatively demonstrate that the cabinet achieves an 11.0 % weight reduction and the inner cylinder realizes a 9.8 % mass cut without exceeding the original maximum stress and fundamental frequency constraints. After optimizing the matching of suspension spring stiffness and damping coefficient, the resonance peak amplitude of the inner cylinder center of mass vibration decreased by 58 %, effectively suppressing the equipment resonance amplification phenomenon and improving the dehydration stability and efficiency. This study provides a complete simulation-surrogate multi-objective optimization workflow for heavy laundry equipment bearing structures and vibration isolation systems under extreme service loads.
