Search results

Abnormal vibrations often occur in the liquid oxygen kerosene transmission pipelines of rocket engines, which seriously threaten their safety. Improper handling can result in failed rocket launches and significant economic losses. Therefore, this paper aims to examine vibrations in transmission pipelines.

In this study, a three-dimensional high-pressure pipeline model composed of corrugated pipes, multi-section bent pipes, and other auxiliary structures was established. The fluid–solid coupling method was used to analyse vibration characteristics of the pipeline under various external excitations. The simulation results were visualised using MATLAB, and their validity was verified via a thermal test.

In this study, the vibration mechanism of a complex high-pressure pipeline was examined via a visualisation method. The results showed that the low-frequency vibration of the pipe was caused by fluid self-excited pressure pulsation, whereas the vibration of the engine system caused a high-frequency vibration of the pipeline. The excitation of external pressure pulses did not significantly affect the vibrations of the pipelines. The visualisation results indicated that the severe vibration position of the pipeline thermal test is mainly concentrated between the inlet and outlet and between the two bellows.

The results of this study aid in understanding the causes of abnormal vibrations in rocket engine pipelines.

The causes of different vibration frequencies in the complex pipelines of rocket engines and the propagation characteristics of external vibration excitation were obtained.

Limited by the types of sensors, the state information available for musculoskeletal robots with highly redundant, nonlinear muscles is often incomplete, which makes the control face a bottleneck problem. The aim of this paper is to design a method to improve the motion performance of musculoskeletal robots in partially observable scenarios, and to leverage the ontology knowledge to enhance the algorithm’s adaptability to musculoskeletal robots that have undergone changes.

A memory and attention-based reinforcement learning method is proposed for musculoskeletal robots with prior knowledge of muscle synergies. First, to deal with partially observed states available to musculoskeletal robots, a memory and attention-based network architecture is proposed for inferring more sufficient and intrinsic states. Second, inspired by muscle synergy hypothesis in neuroscience, prior knowledge of a musculoskeletal robot’s muscle synergies is embedded in network structure and reward shaping.

Based on systematic validation, it is found that the proposed method demonstrates superiority over the traditional twin delayed deep deterministic policy gradients (TD3) algorithm. A musculoskeletal robot with highly redundant, nonlinear muscles is adopted to implement goal-directed tasks. In the case of 21-dimensional states, the learning efficiency and accuracy are significantly improved compared with the traditional TD3 algorithm; in the case of 13-dimensional states without velocities and information from the end effector, the traditional TD3 is unable to complete the reaching tasks, while the proposed method breaks through this bottleneck problem.

In this paper, a novel memory and attention-based reinforcement learning method with prior knowledge of muscle synergies is proposed for musculoskeletal robots to deal with partially observable scenarios. Compared with the existing methods, the proposed method effectively improves the performance. Furthermore, this paper promotes the fusion of neuroscience and robotics.

Polymer laser powder bed fusion (PBF-LB/P) is an additive manufacturing technology that is sustainable due to the possibility of recycling the powder multiple times and allowing the fabrication of gears without the aid of support structures and subsequent assembly. However, there are constraints in the process that negatively affect its adoption compared to other additive technologies such as material extrusion to produce gears. This study aims to demonstrate that it is possible to overcome the problems due to the physics of the process to produce accurate mechanism.

Technological aspects such as orientation, wheel-shaft thicknesses and degree of powder recycling were examined. Furthermore, the evolving tooth profile was considered as a design parameter to provide a manufacturability map of gear-based mechanisms.

Results show that there are some differences in the functioning of the gear depending on the type of powder used, 100% virgin or 50% virgin and 50% recycled for five cycles. The application of a groove on a gear produced with 100% virgin powder allows the mechanism to be easily unlocked regardless of the orientation and wheel-shaft thicknesses. The application of a specific evolutionary profile independent of the diameter of the reference circle on vertically oriented gears guarantees rotation continuity while preserving the functionality of the assembled mechanism.

In the literature, there are various studies on material aging and reuse in the PBF-LB/P process, mainly focused on the powder deterioration mechanism, powder fluidity, microstructure and mechanical properties of the parts and process parameters. This study, instead, was focused on the functioning of gears, which represent one of the applications in which this technology can have great success, by analyzing the two main effects that can compromise it: recycled powder and vertical orientation during construction.

Based on the weak seismic performance and low ductility of coupled shear walls, engineered cementitious composites (ECC) is utilized to strengthen it to solve the deformation problem in tall buildings more effectively and study its mechanical properties more deeply.

The properties of reinforced concrete coupled shear wall (RCCSW) and reinforced ECC coupled shear wall (RECSW) have been studied by numerical simulation, which is in good agreement with the experimental results. The reliability of the finite element model is verified. On this basis, a detailed parameter study is carried out, including the strength and reinforcement ratio of longitudinal rebar, the placement height of ECC in the wall limb and the position of ECC connecting beams. The study indexes include failure mode and the skeleton curve.

The results suggest that the bearing capacity of RECSW is significantly affected by the ratio of longitudinal rebar. When the ratio of longitudinal rebar increases from 0.47% to 3.35%, the bearing capacity of RECSW increases from 250 kN to 303 kN, an increase of 21%. The strength of longitudinal rebar has little influence on the bearing capacity of RECSW. When the strength of the longitudinal rebar increases, the bearing capacity of RECSW increases little. The failure mode of RECSW can be improved by lowering the casting height of the ECC beam in a certain range.

In this paper, ECC is used to strengthen the coupled shear wall, and the accuracy of the finite element model is verified from the failure mode and skeleton curve. On this basis, the casting height of the ECC casting wall limb, the strength and reinforcement ratio of longitudinal rebar and the position of the ECC beam are studied in detail.

There has been a lot of research on how to set marketing budgets, but the overlooked aspect was how allocating funds influences business performance in a multi-goal context. This study aims to examine the relationship between business performance, the process of allocating funds to multiple goals and the interaction among the goals.

Ratio data were generated through “a constant sum scale” from a sample of 362 managers from the B2C sector, besides data on after-tax revenue for two years. The data file was created. Then, a factor analysis was performed on the data. Furthermore, an econometric model with interaction terms was fitted to the data.

The results show that allocating funds to multiple marketing goals – demand generation, customer experience, brand image, marketing competency and purchase intention – influences business performance. Furthermore, a goal’s impact on business performance is higher when coupled with other goals than in isolation.

The findings of the study should assist managers in increasing revenue while spending less on marketing and shifting funds from less efficient goals and pairs of goals to highly efficient ones.

By extending the relevant theory on the relationship between the process of marketing fund allocation, multiple goals and business performance, this study contributes to the literature on marketing.

The purpose of this paper is to improve autonomous flight performance of a fixed-wing unmanned aerial vehicle (UAV) via simultaneous morphing wingtip and control system design conducted with optimization, computational fluid dynamics (CFD) and machine learning approaches.

The main wing of the UAV is redesigned with morphing wingtips capable of dihedral angle alteration by means of folding. Aircraft dynamic model is derived as equations depending only on wingtip dihedral angle via Nonlinear Least Squares regression machine learning algorithm. Data for the regression analyses are obtained by numerical (i.e. CFD) and analytical approaches. Simultaneous perturbation stochastic approximation (SPSA) is incorporated into the design process to determine the optimal wingtip dihedral angle and proportional-integral-derivative (PID) coefficients of the control system that maximizes autonomous flight performance. The performance is defined in terms of trajectory tracking quality parameters of rise time, settling time and overshoot. Obtained optimal design parameters are applied in flight simulations to test both longitudinal and lateral reference trajectory tracking.

Longitudinal and lateral autonomous flight performances of the UAV are improved by redesigning the main wing with morphing wingtips and simultaneous estimation of PID coefficients and wingtip dihedral angle with SPSA optimization.

This paper originally discusses the simultaneous design of innovative morphing wingtip and UAV flight control system for autonomous flight performance improvement. The proposed simultaneous design idea is conducted with the SPSA optimization and a machine learning algorithm as a novel approach.

The paper aims to tackle the classification of Remote Sensing Images (RSIs), which presents a significant challenge for computer algorithms due to the inherent characteristics of clustered ground objects and noisy backgrounds. Recent research typically leverages larger volume models to achieve advanced performance. However, the operating environments of remote sensing commonly cannot provide unconstrained computational and storage resources. It requires lightweight algorithms with exceptional generalization capabilities.

This study introduces an efficient knowledge distillation (KD) method to build a lightweight yet precise convolutional neural network (CNN) classifier. This method also aims to substantially decrease the training time expenses commonly linked with traditional KD techniques. This approach entails extensive alterations to both the model training framework and the distillation process, each tailored to the unique characteristics of RSIs. In particular, this study establishes a robust ensemble teacher by independently training two CNN models using a customized, efficient training algorithm. Following this, this study modifies a KD loss function to mitigate the suppression of non-target category predictions, which are essential for capturing the inter- and intra-similarity of RSIs.

This study validated the student model, termed KD-enhanced network (KDE-Net), obtained through the KD process on three benchmark RSI data sets. The KDE-Net surpasses 42 other state-of-the-art methods in the literature published from 2020 to 2023. Compared to the top-ranked method’s performance on the challenging NWPU45 data set, KDE-Net demonstrated a noticeable 0.4% increase in overall accuracy with a significant 88% reduction in parameters. Meanwhile, this study’s reformed KD framework significantly enhances the knowledge transfer speed by at least three times.

This study illustrates that the logit-based KD technique can effectively develop lightweight CNN classifiers for RSI classification without substantial sacrifices in computation and storage costs. Compared to neural architecture search or other methods aiming to provide lightweight solutions, this study’s KDE-Net, based on the inherent characteristics of RSIs, is currently more efficient in constructing accurate yet lightweight classifiers for RSI classification.

A novel type of heat transfer fluid known as hybrid nanofluids is used to improve the efficiency of heat exchangers. It is observed from literature evidence that hybrid nanofluids outperform single nanofluids in terms of thermal performance. This study aims to address the stagnation point flow induced by Williamson hybrid nanofluids across a vertical plate. This fluid is drenched under the influence of mixed convection in a Darcy–Forchheimer porous medium with heat source/sink and entropy generation.

By applying the proper similarity transformation, the partial differential equations that represent the leading model of the flow problem are reduced to ordinary differential equations. For the boundary value problem of the fourth-order code (bvp4c), a built-in MATLAB finite difference code is used to tackle the flow problem and carry out the dual numerical solutions.

The shear stress decreases, but the rate of heat transfer increases because of their greater influence on the permeability parameter and Weissenberg number for both solutions. The ability of hybrid nanofluids to strengthen heat transfer with the incorporation of a porous medium is demonstrated in this study.

The findings may be highly beneficial in raising the energy efficiency of thermal systems.

The originality of the research lies in the investigation of the Darcy–Forchheimer stagnation point flow of a Williamson hybrid nanofluid across a vertical plate, considering buoyancy forces, which introduces another layer of complexity to the flow problem. This aspect has not been extensively studied before. The results are verified and offer a very favorable balance with the acknowledged papers.

The purpose of this paper is to construct an event-triggered finite-time fault-tolerant formation tracking controller, which can achieve a time-varying formation control for multiple unmanned aerial vehicles (UAVs) during actuator failures and external perturbations.

First, this study developed the formation tracking protocol for each follower using UAV formation members, defining the tracking inaccuracy of the UAV followers’ location. Subsequently, this study designed the multilayer event-triggered controller based on the backstepping method framework within finite time. Then, considering the actuator failures, and added self-adaptive thought for fault-tolerant control within finite time, the event-triggered closed-loop system is subsequently shown to be a finite-time stable system. Furthermore, the Zeno behavior is analyzed to prevent infinite triggering instances within a finite time. Finally, simulations are conducted with external disturbances and actuator failure conditions to demonstrate formation tracking controller performance.

It achieves improved performance in the presence of external disturbances and system failures. Combining limited-time adaptive control and event triggering improves system stability, increase robustness to disturbances and calculation efficiency. In addition, the designed formation tracking controller can effectively control the time-varying formation of the leader and followers to complete the task, and by adding a fixed-time observer, it can effectively compensate for external disturbances and improve formation control accuracy.

A formation-following controller is designed, which can handle both external disturbances and internal actuator failures during formation flight, and the proposed method can be applied to a variety of formation control scenarios and does not rely on a specific type of UAV or communication network.

The purpose of this study is to develop a magnetic resonance imaging (MRI)-safe iron-free electrical actuator for MR-guided surgical interventions.

The paper deals with the design of an MRI compatible electrical actuator. Three-dimensional electromagnetic and thermal analytical models have been developed to design the actuator. These models have been validated through 3D finite element (FE) computations. The analytical models have been inserted in an optimization procedure that uses genetic algorithms to find the optimal parameters of the actuator.

The analytical models are very fast and precise compared to the FE models. The computation time is 0.1 s for the electromagnetic analytical model and 3 min for the FE one. The optimized actuator does not perturb imaging sequence even if supplied with a current 10 times higher than its rated one. Indeed, the actuator’s magnetic field generated in the imaging area does not exceed 1 ppm of the B₀ field generated by the MRI scanner. The actuator can perform up to 25 biopsy cycles without any risk to the actuator or the patient since he maximum temperature rise of the actuator is about 20°C. The actuator is compact and lightweight compared to its pneumatic counterpart.

The MRI compatible actuator uses the B₀ field generated by scanner as inductor. The design procedure uses magneto-thermal coupled models that can be adapted to the design of a variety actuation systems working in MRI environment.