The base of machine tools is essential for ensuring performance, improving processing accuracy, and extending service life. Too few anchor bolts cannot provide effective support for the machine tool, causing structural deformation and directly affecting the processing accuracy of the machine tool. Too many anchor bolts increase the difficulty of leveling the machine tool and lead to additional assembly stresses. To coordinate the contradictions in the fixing scheme design of anchor bolts, this paper proposes an optimization method based on the dynamic modeling analysis of the base-bed joint connected by anchor bolts. Firstly, considering the complicated load conditions of the base, especially the negative effects on the bolted connection of tangential load and its change, a dynamic model of multi-bolted joints coupling normal and tangential vibrations involving static friction is established based on Hertz contact theory and fractal theory. Then, taking the T-shaped bed of a vertical precision machining center as an example, the 4th-order Runge-Kutta method with variable step length is used to analyze the modes and vibration responses via the mentioned dynamic model, investigate the influence of the number of anchor bolts, and determine the minimum number of anchor bolts required to meet the stability of the bed. On this basis, the influence weight of different anchor bolts on the modal performance of the bed is explored through the orthogonal experiment method, and a reasonable distribution of anchor bolts is obtained by categorizing the bolts into fixed and support ones. Finally, considering the differences between the simplified theoretical model and the real machine tool, the simulation signal of the theoretical model is compared with that of the modal experiment, and the consistency and accuracy of the proposed theoretical model are verified.
Ti-6Al-4V is widely used in aerospace and biomedical fields, but its low thermal conductivity and high chemical reactivity make it difficult to machine, leading to high cutting forces, severe tool wear and poor surface integrity. Conventional cooling and lubrication techniques often fail to simultaneously control heat generation and friction at the tool-chip interface, creating a critical bottleneck in improving machinability. To address this limitation, this study proposes a novel synergistic approach by employing surface-textured cutting tools, including linear grooves and bionic hexagonal textures combined with pulsated minimum quantity lubrication (PMQL), dry ice (i.e., solid carbon dioxide), and hybrid cooling (PMQL + dry ice). The objective is to enhance lubrication retention, improve heat dissipation, and modify tool-chip interaction under high-speed machining conditions. Machining performance was evaluated in terms of tool wear, machined surface roughness and morphology, subsurface microhardness, cutting forces, and crystallographic integrity. At cutting speeds of 100m/min and 250m/min, hybrid cooling showed significant improvements in cutting forces. For the bionic tool, cutting forces decreased by 40.8% to 40.1% compared to PMQL and 34.0% to 37.1% compared to dry ice. The linear tool showed reductions of 34.8% to 33.7% compared to PMQL and 27.6% to 30.3% compared to dry ice. In terms of surface roughness, hybrid cooling reduced it by 34.7% to 59.4% compared to PMQL and 15.9% to 26.8% compared to dry ice for the bionic tool, and 39.1% to 43.7% compared to PMQL and 15.5% to 26.1% compared to dry ice for the linear tool at both cutting speeds. At 250m/min, subsurface microhardness analysis showed that hybrid cooling reduced surface hardness by 8.0% compared to dry ice and 3.1% compared to PMQL for the bionic tool, and 9.2% and 3.8% for the linear-textured tool, respectively. Systematic microscopic characterization with SEM and confocal analysis confirmed reduced cutting tool adhesion, improved machined surface morphology, and enhanced crystallographic integrity for the hybrid cooling conditions.
Motion planning for autonomous vehicles has attracted significant attention; however, autonomous overtaking by lane borrowing on two-way, two-lane national highways remains underexplored. The dense presence of heavy trucks on these highways, combined with overtaking blind spots and narrow road boundaries, presents significant challenges for motion planning systems to generate safe and feasible trajectories. The Frenét frame is widely adopted in motion planning primarily due to its efficiency in decoupling lateral and longitudinal motion and its ability to transform nonconvex road boundaries into more manageable forms. However, it exhibits notable limitations when describing the geometric dimensions of both the host vehicle and obstacles. In contrast, the Cartesian frame offers an intuitive framework for vehicle dynamics modeling yet faces significant challenges in addressing nonconvex constraints associated with curved lanes. To fully leverage the complementary advantages of these two frames, this paper proposes a joint Cartesian-Frenét Constrained Iterative Linear Quadratic Regulator (CILQR) planning method and introduces an optimization objective aimed at extending the host vehicle's front field of view (FOV). The augmented Cartesian-Frenét state space must satisfy four core constraints: (1) a Cartesian-Frenét coupled vehicle model to ensure the practical trackability of the generated trajectory; (2) a set of Frenét road constraints to provide critical safeguards for driving safety; (3) a set of Cartesian obstacle geometry constraints to ensure collision avoidance; and (4) the Cartesian-Frenét transformation relationship that serves as the central link between the two frames. The motion planning problem is then formulated and solved using CILQR to handle the aforementioned constraints. Simulation results show the capacity of the CILQR algorithm to improve planning safety by extending the FOV of autonomous vehicles.
Reliable estimation of remaining useful life (RUL) is essential for enabling condition-based maintenance decisions. Nonetheless, most current RUL prediction techniques encounter considerable difficulties under varying operational conditions, mainly due to the complex and diverse degradation behaviors of harmonic drives and the limited diagnostic insight provided by single-sensor data. To tackle these challenges, this study proposes a novel transfer learning-based remaining useful life prediction framework driven by multisensor information fusion, specifically developed for harmonic drive applications. The proposed framework integrates a convolutional module, a temporal module, and a multi-head self-attention module, facilitating the adaptive extraction of spatiotemporal features from degradation signals. Additionally, to reduce discrepancies in data distributions arising from differing degradation scenarios, a joint maximum mean discrepancy loss function is employed. This mechanism aligns feature representations across multiple network layers, thereby boosting the model’s generalization capability in cross-domain prediction tasks. Comprehensive experiments confirm the robustness and enhanced performance of the proposed method across various RUL prediction scenarios.
Vehicle-to-Vehicle (V2V) cooperative localization mitigates the cumulative drift of standalone localization by exploiting position and ranging information from nearby vehicles. Compared with existing approaches that are based solely on inter-vehicle distance constraints, combining visual-inertial odometry (VIO) with ranging measurements improves accuracy while reducing the dependence on the number and density of cooperative vehicular anchors. Such methods are particularly promising for applications in intelligent transportation systems (ITS) and GNSS-degraded environments. Building on this, this paper proposes a robust V2V cooperative localization method that combines multi-epoch cooperative observations with Visual-Inertial Odometry (VIO) pose priors in a batch weighted least-squares framework. By exploiting multi-epoch inter-vehicle ranging constraints and onboard VIO pose estimates, the proposed approach improves localization accuracy while reducing the reliance on dense cooperative participation, and remains effective even in two-vehicle scenarios. A robust M-estimator with a Hybrid Ordinary-Cauchy (HOC) kernel is employed to adaptively weight the VIO results and cooperative measurements by jointly considering theoretical observation reliability and actual measurement quality, thereby reducing the influence of low-quality observations on cooperative localization. A systematic theoretical analysis is then conducted and validated through simulations, including error distribution modeling, Cramér–Rao lower bound (CRLB) derivation, and an evaluation of the accuracy gains from VIO fusion. Finally, real-world experiment results demonstrate that the proposed method achieves higher positioning accuracy and robustness than the standalone VIO and representative cooperative localization baseline methods.
In low-temperature environments, electric vehicles (EVs) face dual challenges of increased thermal management energy consumption and degraded fast-charging capability due to insufficient battery preheating. Existing rule-based and short-horizon predictive control methods lack the capability for global trip-level coordination of cabin and battery heating. To address this limitation, a hierarchical control framework integrating reinforcement learning (RL) with nonlinear model predictive control (NMPC) is proposed. The upper-layer RL agent exploits global trip information, including remaining range and navigation data, to achieve global optimization of thermal energy allocation and to generate adaptive reference trajectories for cabin and battery temperatures, enabling proactive battery preheating for fast-charging readiness. The lower-layer NMPC tracks these references while explicitly satisfying system constraints and providing reliable multi-objective control performance that complements the RL-based global optimization. Simulations on urban and highway scenarios at ambient temperatures between -10 and -20 °C demonstrate that the proposed framework reduces thermal management energy consumption by 5%-8% and shortens the time to reach 80% state-of-charge by 18.0%–25.7% compared with rule-based baselines, validating its effectiveness in achieving global thermal optimization and enhanced fast-charging efficiency under cold-weather conditions.
This study compares the surface grinding performance of four types of wheels (white fused alumina, microcrystalline alumina, vitrified CBN, and electroplated CBN) to meet the strict corrosion resistance and surface integrity requirements for K444 superalloy marine turbine blades. By analyzing grinding forces, temperatures, surface topography, and wheel wear, we aim to optimize wheel selection and parameters to suppress thermal damage and micro-defects. Results show K444 has poorer grindability than Inconel 718. Severe material adhesion and wheel wear are the primary machining challenges, frequently causing burns and surface defects. Grinding performance varies significantly with wheel type: alumina wheels achieve a maximum stable material removal rate (MRR) of only 0.33mm³/(mm·s), though microcrystalline alumina offers better mechanical stability. Electroplated CBN provides the highest efficiency (MRR up to 5mm³/(mm·s)) with low forces and temperatures, but leaves deep grooves susceptible to pitting corrosion. Vitrified CBN machines stably up to 2.5mm³/(mm·s), yielding a uniform surface beneficial for anti-corrosion fatigue, but causes higher forces, temperatures, and severe abrasive wear flattening. Consequently, we propose a combined process strategy for highly corrosion-resistant blades: electroplated CBN for rough grinding, followed by vitrified CBN or microcrystalline alumina for finish grinding. This provides a proven basis for the high-efficiency, low-damage manufacturing of this alloy.
To address the limitations of existing nursing robot joints in terms of load capacity, surface smoothness, and contact compliance, which hinder their ability to meet human-robot compatibility requirements during elderly transfer (e.g., between bed and wheelchair), this study proposes a novel robotic joint integrating rigid-flexible structure and motion-coupling characteristics, along with systematic research on its dynamic modeling and parameter identification methods. A differential mechanism-based dual-motor torque-coupled joint was designed, achieving a 100% improvement in load capacity. Rubber-based flexible elements were incorporated into the transmission chain to accommodate nonlinear mechanical interactions during human-robot contact. The fundamental dynamic model of the joint was established using the Lagrange method, while a hyper-viscoelastic constitutive model was developed for the flexible elements, ultimately forming a complete joint dynamic modeling framework. Hyper-viscoelastic constitutive parameters were identified through rubber tensile tests, and the remaining dynamic parameters were determined via excitation trajectory tracking experiments. Experimental results demonstrate that the proposed joint design and modeling approach significantly enhance transfer safety and comfort, providing theoretical and technical insights for the development of nursing robot joints.
Achieving high-speed and high-precision motion performance in parallel mechanisms has been a long-standing challenge, particularly in areas like dynamic modeling and identification. In this paper, the dynamic performance of 3-PRR parallel mechanism based on a permanent magnet linear motor is studied for high-precision applications. Firstly, based on the analysis of friction, thrust fluctuations, and branched coupling force which collectively exhibit complex coupling and pose challenges for isolated analysis compensation, the concept of comprehensive resistance fluctuation is introduced to holistically represent these combined disturbances, enabling unified identification and compensation. Secondly, several groups of uniform-motion experiments are designed, and the identification of friction and thrust fluctuations is achieved through spectral analysis. The Newton-Euler dynamics equation is employed to investigate the influence of the branched coupling force on the joint motor of the 3-PRR parallel mechanism. Finally, in the torque control mode, the comprehensive resistance fluctuations at multiple positions are identified using an iterative learning algorithm. Compared to the experimental results mentioned above, the identification of friction force, thrust fluctuations, and comprehensive resistance fluctuation is shown to be reasonable.
Additive manufacturing (AM) of AlSi10Mg lattice structures have high surface quality requirements in practical applications, which limits their application. This paper employs the electrochemical polishing (ECP) method to effectively enhance the surface quality of AlSi10Mg lattice structures while minimizing deformation and material removal. The influence mechanism of different process parameters on the surface roughness and material removal amounts of the samples during ECP was analyzed. Under the optimized ECP process conditions, the surface roughness of the samples was reduced by 82.6%, and the polishing effect was good. The ultrasonically-assisted electrochemical polishing (UA-ECP) method was adopted, utilizing the synergistic effects of cavitation and electrochemical reactions to address issues in traditional ECP, such as insufficient electrolyte permeability and concentration differences between the internal and external regions of the structures due to complex lattice structures, as well as the hindrance to continuous and uniform reaction progresses caused by Si-containing reactants covering the sample surfaces. This method further improves the surface quality of samples and effectively enhances the efficiency of ECP. Research results demonstrate that the mechanical-electrochemical synergy has significant potential in improving the surface quality of AM complex structures.
The growing construction of space solar power stations increases demand for truss unit assembly systems, but current systems using collaborative robots for both individual truss building and multi unit on orbit assembly face challenges. These include unclear component sources that hamper tracking and replication, robots moving back and forth that reduces efficiency, and inconsistent component robot interactions that raise control needs while lowering reliability. To address these issues, this paper proposes a new on-orbit assembly system of truss units. It separates individual truss construction from multi unit assembly, assigns each task to dedicated robots for a pipeline process, adopts plug in locking for rods and joints with standardized robot end effector interaction, and designs two new robots with preliminary qualitative function and performance evaluation. Meanwhile, to solve low efficiency in robot lightweight design caused by frequent topological optimization and poor manufacturability when building stiffness weight mapping models, a lightweight method based on boundary specification and topological approximation is proposed. It enables one time rapid generation of the same component’s topological structure under identical mass retention ratio and stress conditions, eliminating repeated iterations in traditional topological optimization and manufacturability post processing. Verified via MTUA (multiple truss unit assembly) robot case study, results show only three topological optimizations are needed per robot connecting rod type to build stiffness mass mapping models, the structural material distribution is more reasonable and free of burrs, and the robot mass is reduced by 17.4% while meeting stiffness requirements. This research provides references for truss unit on orbit assembly robot configuration design and a foundation for similar robots’ lightweight design.
GH2132 superalloy is specifically used in the manufacture of aero-engine stator blades. However, its superior performance leads to challenges such as poor machinability and difficult control of surface integrity, which are critical factors affecting blade fatigue life. Balancing machining efficiency improvement and surface integrity maintenance remains a significant challenge in the precision machining of such components. This study focuses on a comparative analysis of machining surface integrity between two machining methods, Rotary Ultrasonic Elliptical Milling (RUEM) and Conventional Milling (CM), for GH2132 stator blades. It analyzes the intrinsic mechanism of action underlying the phenomena related to surface integrity from five aspects, encompassing surface topography, surface roughness, microhardness, residual stress, and plastic deformation layer thickness. Meanwhile, it investigates the feasibility of efficiency improvement of RUEM in the machining of such blades. A finite element simulation was performed to investigate the surface residual stress in the machined surface layer and its distribution characteristics along the depth direction under the two machining methods. Experimental results show that the RUEM-processed surface exhibits a regularly distributed micro-wavy texture morphology with no obvious surface damage defects, while the CM-processed surface displays significant pit holes, scratches, and other defect characteristics. Compared with CM, the microhardness of the RUEM-processed surface is increased by 17.2%, and the maximum residual compressive stress is 7.16 times that of CM. Further analysis reveals that the mechanical loads generated during RUEM induce significant plastic deformation in GH2132, leading to surface hardening. Notably, RUEM maintains a favorable thickness of the plastic deformation layer even at elevated feed rates. Research data indicate that RUEM can increase machining efficiency by 25%–35% and can effectively control surface integrity during the machining process of GH2132 stator blades. This achievement provides a brand-new technological approach for the processing of such blades.
Kinematic redundancy emerges as a prominent subject in the research of parallel mechanisms on account of the exceptional flexibility and substantial workspace it provides. However, for kinematically redundant parallel mechanisms, due to the dimensional difference between the actuator’s input and the end effector’s output, traditional Jacobian becomes undesirable for their analyses. To address this issue, this paper introduces a novel Jacobian method, designed for kinematically redundant parallel mechanisms with configurable platforms. The paper first analyzes the limitations of the traditional Jacobian and delves into their underlying causes on mathematical and kinematic basis. Then, the proposed Jacobian method is elucidated, and the general expressions of a novel kinematic Jacobian is derived. Also, to better demonstrate the proposed method and to prove its feasibility as well as universality, numerical examples are provided, using two typical planar and spatial kinematically redundant parallel mechanisms with configurable platforms as case studies. For mechanisms in this category, the newly derived Jacobian can simultaneously take into account the influence of redundant degree of freedom configurations on the original basis, thus can provide a more comprehensive description of the mechanism’s overall state. The newly derived Jacobian can be better suitable for the kinematic and static analyses of kinematically redundant parallel mechanisms, and can derive indices that are more accurate in the mechanisms’ performance evaluation.
Industrial fault diagnosis is a key component of intelligent manufacturing systems. Diagnostic accuracy directly influences the safety of production systems and the efficiency of equipment operation and maintenance. However, traditional bearing fault diagnosis methods are limited by three critical constraints. First, single-modal sensor data are relied upon, which fails to comprehensively characterize the complex operational states of equipment. Second, raw fault samples are scarce, resulting in poor model adaptability in small-sample scenarios. Third, generalization performance is significantly degraded during cross-operating-condition and cross-equipment transfer, with multimodal information being insufficiently integrated. Moreover, in the application of large language models (LLMs) to fault diagnosis, challenges such as insufficient data volume, modal gaps, and high fine-tuning costs remain unresolved. To address these challenges, this study proposes FM-LLM, a bearing fault diagnosis method that integrates simulation-based data augmentation with lightweight multimodal fine-tuning of LLMs. The implementation of the method is described as follows. First, a large number of simulated bearing fault datasets are generated through simulation technology. These datasets are fused with original measured datasets, effectively mitigating the limitation of scarce fault samples in industrial scenarios. Second, the fused data are processed via envelope analysis to suppress noise interference and enhance fault feature signals. Third, time-frequency images are extracted from the processed data as the visual modality. Concurrently, text-modal data are constructed by integrating core bearing information, such as model specifications, operating parameters, and fault type descriptions. This process produces a multimodal fine-tuning dataset composed of time-frequency images and text. Finally, parameter-efficient fine-tuning techniques, namely Low-Rank Adaptation (LoRA) and Quantized Low-Rank Adaptation (QLoRA), are adopted. These techniques facilitate accurate learning and classification of bearing fault features by the LLM. Additionally, computational costs are significantly reduced, and the forgetting of pre-trained knowledge is prevented.
Rack railways are often used on steep-slope lines because longitudinal forces can be transmitted through gear–rack meshing. Under braking, the combined effects of braking forces, gear–rack meshing forces, and wheel–rail contact forces can produce strong longitudinal dynamics, which may threaten the safety of the vehicle–track system. In this work, a dynamic model of a rack vehicle–track system under braking is established, and the dynamic characteristics of the vehicle and track structure at different braking speeds on steep-slope lines are investigated. Longitudinal and vertical acceleration responses of key components, including the car body, gears, and wheelsets, are analyzed during braking. The results show that when the braking speed increases from 10km/h to 50km/h, the peak longitudinal acceleration of the car body increases from 0.273m/s² to 0.995m/s², while the increase in vertical acceleration is limited. This indicates that the braking-induced response is mainly longitudinal. Gear responses are much stronger: the peak longitudinal acceleration increases from 57m/s² to 127m/s², which shows that gear–rack meshing plays a major role in amplifying longitudinal vibration. The evolution of gear–rack meshing forces and wheel–rail contact forces during braking is also examined. The peak longitudinal gear–rack meshing force increases from 321 kN to 596 kN, and the longitudinal wheel–rail contact force increases from 20 kN to 49 kN. At the structural level, the longitudinal stress of the rack increases sharply under high-speed braking and reaches 179.1MPa, which is much higher than that of the rail. In addition, excessive longitudinal slip of the rail occurs in the braking zone and exceeds the allowable elastic displacement of small-damping fasteners (2mm), which indicates a risk of fastener failure. These results show the distinctive longitudinal dynamic behavior of rack railways under braking compared with conventional adhesion railways. The findings can support safety assessment, speed control, and track structure design for rack railways under braking.
Directed Energy Deposition (DED) technology has huge potential in key component repairing with high density energy input, contributing to maintain the mechanical properties of the repaired component. Compared to the conventional infrared laser used in DED, the novel blue laser is promising in the laser assisted manufacturing due to the higher metal absorption rate. However, there are few works reported on the hybrid blue-infrared laser in Al2024 component repairing, which is commonly used in key aerospace component. This work thus focuses on the development of the novel hybrid blue-infrared laser in the Al2024 component repairing. Specially, a series of laser material interaction experiment was conducted, including the single point melting, single track melting and deposition, and defects repairing. Optimal process parameters were further analyzed to fulfill the potential of the hybrid blue-infrared laser. The reported work demonstrates the feasibility of the hybrid blue-infrared laser in the Al2024 component repairing.
Wick structures are considered the core components of heat pipes, which operate under complex load conditions and provide superior capillary mass transfer performance in next-generation aerospace equipment. However, there is significant potential to improve both specific strength and mass transfer performance, which are determined by the manufacturing process and materials. In this study, body-centered cubic lattice wicks with four different strut radii were fabricated using AlMgScZr powder via laser powder bed fusion (LPBF). Submillimeter structures with high forming quality (relative density of 98.3%) and high forming precision (minimum dimensional deviation of only 2.2%) were achieved. The influence of strut radius on permeability (K), capillary performance, and flexural properties of the lattice wicks was investigated through experiments and finite element analysis. Results indicated that the lattice wick with a strut radius of 120 μm (R120) exhibited the highest K at 2.90 × 10−10 m2 and K increased as strut radius decreased. The predictive equation for K of high-strength aluminum lattice wicks fabricated by LPBF was optimized. Improved capillary performance was observed in wicks with increasing strut radius, showing a monotonic increase in capillary equilibrium height as a function of strut radius. For R210, the heights reached 44.64mm. The lattice wicks simultaneously exhibited high K (1.64–2.90 × 10−10 m2) and capillary pressure (0.78−1.82 kPa), demonstrating a more balanced coexistence of high permeability and capillary performance compared with conventional wicks. Flexural performance showed a positive dependence on strut radius, with R210 demonstrating the highest flexural strength (2.35MPa). There were mutual constraints and contradictions among these different multifunctional performances (formability, lightweight, mass transfer, and bending resistance), making synergistic control difficult. Therefore, a coupling evaluation method was established for the synergistic control of multifunctional performance to quantitatively assess wick performance. The coupling evaluation value of R150 reached 0.75, indicating excellent integrated performance. This paper provides a novel perspective on the design, manufacturing, performance evaluation, and control of wicks in aerospace applications.
To address the challenging issue of fault-tolerant control in steer-by-wire (SBW) systems under the conditions of actuator faults, external disturbances, and input saturation constraints, this paper proposes an adaptive fault-tolerant control strategy that integrates an extended state observer (ESO), radial basis function neural network (RBFNN), and command-filtered backstepping (CFB) method. Firstly, a system model is constructed with the motor drive voltage serving as the direct control variable, and an ESO and an RBFNN are designed to perform real-time estimation and linear approximation of system states and disturbances, thereby significantly reducing the dependency of the fault-tolerant control strategy on sensor performance. Furthermore, to address the computational explosion issue inherent in traditional backstepping control, command filtering techniques are employed along with an error compensation mechanism, which simplifies the controller design while ensuring the tracking accuracy of the fault-tolerant control strategy. Finally, a hyperbolic tangent function is utilized to smoothly handle the actuator input saturation characteristics, and an auxiliary signal is incorporated to modify and compensate for the fault-tolerant control law, thereby ensuring the stability of the SBW system under actuator physical constraints. The conducted simulations and bench tests demonstrate that the proposed method can swiftly identify torque failure faults in the steering motor, accurately estimate the states of the front wheels and the motor, effectively solve both internal motor disturbances and external vehicle disturbances, and significantly mitigate the impact of input saturation constraints on system performance. It enhances the tracking accuracy of the front wheel angle and improves the fault-tolerant ability of SBW systems, surpassing the performance of those that do not consider input constraints as well as PD fault-tolerant methods. Therefore, the designed adaptive backstepping fault-tolerant control strategy exhibits excellent and effective control performance, enabling the steering system to maintain superior tracking capability and robustness online even in the presence of actuator faults, external disturbances, and input saturation.
In-wheel motor drive systems enable independent, wheel-specific torque control, providing high flexibility to support chassis-dynamics performance enhancement, but are hindered by increased unsprung mass and limited power density. To overcome these drawbacks, an Electric Torque Vectoring Drive System (E-TVDS) is developed in this paper, integrating distributed and centralized drive principles via a multi-stage planetary gear set to achieve accurate left-right torque allocation. The system's dynamic characteristics are investigated through ADAMS simulations, focusing on structural configuration and operational behavior. A hardware-in-the-loop (HIL) platform is further employed to validate transmission performance under representative load conditions. Test results indicate that the E-TVDS achieves precise and rapid torque vectoring with low response latency, enhancing vehicle stability and maneuverability in real-world driving scenarios.
In this paper, a novel ferrofluid shock absorber is studied numerically and experimentally, which has a nonmagnetic inertial mass block levitated in the ferrofluid in the presence of magnetic field as a working element. The effects of the ferrofluid mass, the geometry and density of the inertial mass block and the arrangement of magnets on the shock absorber performances have been investigated. As ferrofluid mass increases, the damping effect becomes better and then worse, and there exists an optimum value. The cylindrical and annular inertial mass blocks present similar damping effects, indicating that the annular one can reduce the total mass of the shock absorber. The results show that the higher the density of the inertial mass block, the better the damping effect. The annular magnet helps the inertial mass block back to the center but hinders the movement. The damping effects of the shock absorber are measured by the percentage of vibration attenuation， which is defined as the difference of 1 and the ratio of the settling time with the ferrofluid shock absorber and the settling time with an additional weight having the same mass as the shock absorber. The percentage of vibration attenuation can reach 95.68%.
