Journal of Mechanical Science and Technology最新文献

This article addresses butt joining of 1.2 mm thick Ni-Cr superalloy 80 A sheets using a 3.5 kW CO₂ laser beam welding (LBW) process, which is critical for gas turbine components, nuclear tube supports and automotive valves in high temperature applications. LBW, chosen for its low heat input and minimal heat-affected zones, was optimized using Taguchi-based gray relationship analysis. The L9 orthogonal array experiment identified the optimal parameter sets for welding speed, laser power, focal length, and shielding gas flow rate that affect tensile strength, microhardness, penetration depth and weld bead. The results were validated using ANOVA analysis, fractography, hardness testing, micrographs and tensile tests. Microstructural variations in fusion and heat affected zones. The optimized parameters resulted in a tensile strength of 817 MPa and a microhardness of 292 HV, demonstrating improved weld quality.

Redesigning existing parts is challenging without keen insights into capabilities afforded by AM and with the absence of systematic design methods for the early product design stage. Therefore, this study proposes a systematic design method to leverage fully AM design benefits for reconceptualizing product architecture, which leads to consolidate parts and simplify product architecture. The proposed design method consists of four steps to achieve the goal. First, a baseline product architecture should be identified as a starting point of reconceptualization. Second, candidates for reconceptualization are identified by design principles for AM. Third, AM design benefits are recommended by the developed knowledgebase and query language, which can replace functions of the candidates. AM design benefits are applied to reconceptualize existing part design and lead to more efficient product architectures. Lastly, the reconceptualized product architecture is evaluated. To demonstrate usefulness of the proposed method, a case study is performed on an electric motorcycle.

The operation of wind turbines involves a complex interaction between aerodynamics, structural mechanics, and control systems. However, the control system is frequently overlooked. To investigate the impact of the control system on the aerodynamic characteristics and elastic deformations of wind turbines, this paper initially integrates the control system into the blade element momentum theory (BEMT) for calculating aerodynamic forces. Subsequently, the control system is incorporated into fluid-structure interaction (FSI) calculations to assess its influence on the overall performance of the turbine. The control system employs variable speed and pitch control, while the structural dynamics are modeled using the Euler-Bernoulli beam theory. When the control system is integrated with blade element momentum theory to calculate the aerodynamic forces of the wind rotor, it is observed that, below the rated wind speed, a portion of the torque error is transferred to the rotor speed. In contrast, above the rated wind speed, the entire torque error is transferred to the blade pitch angle (BPA). Crucially, when the control system is integrated, the rotor speed and BPA are no longer treated as known parameters. This approach enables the prediction of aerodynamic characteristics of the wind rotor, particularly under complex wind speed profiles. The control system exerts a significant influence on the FSI results, particularly in the range of wind speeds that correspond to larger blade deformations. This work can provide a reference for the calculation of aerodynamic characteristics and FSI of wind turbines under complex wind conditions.

This study utilizes the improved fast and adaptive bidimensional empirical modal decomposition (IFABEMD) method to study the impact of relative vibration between the tool and the workpiece on surface topography. The sieving stop condition is determined based on the difference in standard deviation, effectively addressing the issue of modal mixing in the traditional bidimensional empirical modal (BEMD) method. Additionally, a novel boundary extension method is proposed using the Gerchberg algorithm. The decomposition example demonstrates that the IFABEMD method can effectively handle modal mixing and boundary effects. Finally, the relative vibration frequencies between the tool and the workpiece are identified by analyzing the spatial spectrum of the feature surface with vibration information. Simulation and experimental surface morphology decomposition results validate the effectiveness of the IFABEMD method in identifying relative vibration between the tool and the workpiece.

A lightweight large-scale stainless-steel plate heat pipe (SPHP) fin has been developed for cooling high-power electronics, employing very thin 0.15 mm plates and a 3D rib-lattice structure to optimize heat pipe operation. Eco-friendly deionized (DI) water serves as a working fluid of the heat pipe, aligning with environmental concerns and regulations. To assess the heat dissipation effectiveness of the SPHP fin, a constant temperature water bath experiment was conducted to compare its heat transfer characteristics with those of a conventional solid aluminum fin. Furthermore, a test rig was constructed to simulate actual heat dissipation from high-power chips and to investigate the impact of different fin types, locations, and heat fluxes of the chips on cooling performance. Additionally, a simplified method was devised for analyzing the heat transfer of the cooling fin, replacing complex evaporation, condensation, and two-phase flow processes in heat pipes with a virtual solid of high thermal conductivity. Validated by experiments, this method aids in selecting the optimal SPHP fin placement based on chip location and heat flux. The study demonstrates that the SPHP fin significantly outperforms traditional aluminum fins, reducing the temperature of a 25 W heat source by 10 °C to 20 °C depending on the chip’s location.

Current fault detection methods for rolling bearings suffer from insufficient data, which limits the generalizability of the models. Typically, conventional approaches train models with a significant amount of labeled data to improve reliability. However, centralized training poses potential risks of data privacy leakage. To address this issue, we propose a federated learning-based fault diagnosis model. In this method, fault diagnosis models for different clients are collaboratively trained by multiple entities with distinct fault characteristics, eliminating the need for third-party aggregation and thereby reducing the risk of data leakage. Specifically, we design a multiscale residual neural network with the ability to perform direct feature extraction from fault data. This proposed network integrates attention units for various scales, emphasizing key features of bearing faults and enhancing the fault recognition capability of local models. Moreover, to address the inherent problem in traditional federated learning frameworks—disparities in client contributions, leading to suboptimal model quality and prolonged training times—this research introduces an innovative weighted strategy based on multiclass F1 scores. This strategy assigns higher weight to high-quality local clients, thereby enhancing both model quality and training speed. Experiments were conducted on two authentic bearing datasets, and the results demonstrate that the proposed method can achieve an average reduction of approximately 15 % in training iterations compared to the federated averaging algorithm, coupled with an average enhancement of approximately 5 % in fault diagnosis accuracy. The experimental results indicate that the proposed method exhibits outstanding accuracy and robustness.

We attempted to optimize the process-based parameters during friction stir welding of AZ80A - AZ31B Mg alloys with the objective of enhancing the mechanical properties of the fabricated joints. A response surface method based grey relational analysis was employed using three factors and three distinctive levels. A central composite design based multi-objective numerical model using the technique of grey relational analysis was formulated for optimizing the tool dependent parameters, namely tool’s rotational speed, its speed of traverse and geometry of the pin. Grey relational grade was determined for all the responses: tensile strength and elongation percentage. Analysis of variance was employed for attaining grey relational grade to determine the most influential parameter of the FSW process. It was observed that the geometry of the tool pin had a greater impact in ascertaining the quality of the fabricated Mg alloy joints, and the tool possessing tapered cylindrical pin geometry exhibited larger values of grey relational grade. Optimized process parameter settings based on the attained GRG values were recorded to be 1100 rpm rotational speed, speed of traverse of 1.5 mm/sec and a tool with taper cylindrical pin geometry. The anticipated values were validated through confirmation investigational runs performed during employment of optimized parameter combinations, which exhibited a perfect agreement with the investigational run values and the confirmatory joint exhibited a tensile strength of 260.42 MPa and elongation percentage of 6.53.

In this study, the atmospheric plasma (APP) surface modification technology was used to modify the surface of additive materials and mix them into the commercial thermal interface material (base-TIM). This technology allows for the preparation of a hybrid additives thermal interface material (HA-TIM) to improve the heat conduction performance of the base-TIM. The additives selected for HA-TIM include aluminum nitride (AlN), multi-walled carbon nanotubes (MWCNTs), and graphene flakes (GNFs) with different proportions. Additives with different sizes, shapes, and high thermal conductivity were expected to achieve a synergistic effect to produce an HA-TIM with high heat dissipation performance. After the preparation of the HA-TIM with different configuration ratios was completed, heat dissipation performance experiments would be carried out under different heating power and ambient temperature to determine the optimal configuration ratio of the HA-TIM. The results show that the HA-TIM prepared by adding 1 wt% GNFs and 1 wt% MWCNTs to the base-TIM has the best heat conduction performance. In the optimum configuration of the HA-TIM at the heating power of 50 W, 100 W, and 150 W, the heater surface temperature under the ambient temperature of 25 °C is 1.0 °C, 3.0 °C, and 4.2 °C lower than those of base-TIM, and the heater surface temperature under the ambient temperature of 30 °C is 1.1 °C, 3.2 °C, and 6.3 °C lower than those of base-TIM, respectively. Furthermore, the results show that HA-TIM has a better heat dissipation performance under high ambient temperature and heating power.

The synchronization control of its dual hydraulic cylinder moving beam in fatigue testing machine requires robustness and high synchronization accuracy. The synchronization control accuracy is affected by the time-varying nonlinearity of parameters, mutual coupling and load disturbance of the multi-cylinder system affect. A novel control strategy with single-cylinder self-immunity and double-cylinder fuzzy single neuron is proposed to overcome the slow compensation response and poor robustness of dynamic adjustability. First the working principle of the moving beam synchronization system of the fatigue testing machine is described, then the mathematical simulation model of the asymmetric single and double cylinders and their synchronization control system are established to carry out the simulation analysis of the tracking performance and synchronization performance. Finally the experimental is setup to verify the simulation analysis. The results show that the designed synchronization controller can meet the application requirements of the fatigue testing machine moving beam extremely well.

The actuation of traditional anthropomorphic hands is relatively complex, and there is little research on humanoid skin. In view of the above problems, this study proposed an anthropomorphic hand with biological and kinematic characteristics of the human hand, including structural skeletal parts and humanoid skin. Firstly, the overall structure and control system of the anthropomorphic hand were designed. Then, the humanoid skin was fabricated and its physical and mechanical properties were tested. Based on the overall structure of the rope-driven anthropomorphic hand, its motion characteristics were simulated and analyzed using multi-body dynamics software ADAMS/Cable. Finally, control experiment verifies the performance of the anthropomorphic hand described in this paper. The results are that the proposed anthropomorphic hand can reproduce the movement characteristics of the human hand well, and can grip objects of different shapes, different sizes and different weights stably, with a maximum grip force of 11.91 N measured.