Machines

26 pages, 2461 KB

Open AccessArticle

Multi-Objective Structural Parameter Optimization for Stewart Platform via NSGA-III and Kolmogorov–Arnold Network

by Jie Tao, Yafei Xu, Yongjun Chen, Pin Cheng, Haikun Zhang, Jianping Wang and Huicheng Zhou

Machines 2025, 13(10), 887; https://doi.org/10.3390/machines13100887 - 26 Sep 2025

Abstract

The structural parameters of Stewart platforms play a critical role in enhancing dynamic performance, improving motion accuracy, and enabling effective control strategies. However, practical applications face several key limitations, including the metric balancing for optimization, the limited singularity-free workspace, and low computational efficiency. [...] Read more.

The structural parameters of Stewart platforms play a critical role in enhancing dynamic performance, improving motion accuracy, and enabling effective control strategies. However, practical applications face several key limitations, including the metric balancing for optimization, the limited singularity-free workspace, and low computational efficiency. To overcome those shortcomings, this work proposes a multi-objective optimal design of the structural parameters for Stewart platform based on Non-dominated Sorting Genetic Algorithm III (NSGA-III) and Kolmogorov–Arnold Network (KAN). Firstly, under the stroke constraints of the Stewart platform, this work focuses on optimizing the platform’s key structural parameters. This approach enables both the optimization of existing equipment and the design of new devices. Secondly, this work employs KAN to establish a model that characterizes the relationship between the structural parameters and diverse postures within the maximum singularity-free workspace. This approach not only enhances computational efficiency but also ensures high precision. Finally, this study proposes six performance metrics and utilizes NSGA-III to optimize the structural parameters, thereby achieving a trade-off among these diverse objectives. Simulation and experimental results demonstrate that KAN significantly outperforms the Multi-Layer Perceptron (MLP) in predicting workspace postures. Compared with MLP, KAN achieves higher prediction accuracy and lower error rates across both training and test datasets. When comparing NSGA-III with NSGA-II, the proposed approach demonstrates modest improvements in most performance metrics while preserving acceptable trade-offs between the optimization objectives. Full article

(This article belongs to the Section Machine Design and Theory)

20 pages, 2322 KB

Open AccessArticle

Transient Stability-Oriented Nonlinear Power Control of PMSG-WT Using Power Transfer Matrix Modeling with DC Link Behavior

by Muhammad Ali Bijarani, Ghulam S. Kaloi, Mazhar Baloch, Rameez Akbar Talani, Muhammad I. Masud, Mohammed Aman and Touqeer Ahmed Jumani

Machines 2025, 13(10), 886; https://doi.org/10.3390/machines13100886 - 26 Sep 2025

Abstract

In this paper, a nonlinear power transfer matrix model is presented for power control of Permanent Magnet Synchronous Generator (PMSG) wind turbines, incorporating the DC link dynamics to account for transient stability, thereby clarifying the technical aspect and purpose. The rising penetration of [...] Read more.

In this paper, a nonlinear power transfer matrix model is presented for power control of Permanent Magnet Synchronous Generator (PMSG) wind turbines, incorporating the DC link dynamics to account for transient stability, thereby clarifying the technical aspect and purpose. The rising penetration of wind turbines (WTs) into the power grid necessitates that they remain connected during and after faults to ensure system reliability. During voltage dips, the stator and grid-side converter (GSC) of a permanent magnet synchronous generator (PMSG) system are directly impacted by the sudden voltage changes. These disturbances can induce large transient voltages and currents in the stator, which in turn may lead to uncontrolled current flow in the rotor circuit and stress the converter components. Moreover, Low Voltage Ride-Through (LVRT) is a critical requirement for grid connection to Wind Energy Conversion Systems (WECS). It ensures that WTs remain connected and operational during short periods of grid voltage dips (faults), instead of disconnecting immediately. This capability is essential for maintaining grid stability. However, in this paper, the authors propose an LVRT scheme for a grid-connected PMSG-based WECS. A sequence of attempts was performed to validate the effectiveness of the proposed control scheme under fault conditions and to improve its overall performance. Full article

(This article belongs to the Section Electrical Machines and Drives)

► Show Figures

Figure 1

23 pages, 868 KB

Open AccessArticle

FRMA: Four-Phase Rapid Motor Adaptation Framework

by Xiangbei Liu, Chang Lu, Hui Wu, Bo Hu, Xutong Li, Zongyuan Li and Xian Guo

Machines 2025, 13(10), 885; https://doi.org/10.3390/machines13100885 - 25 Sep 2025

Abstract

In many real-world control tasks, agents operate under partial observability, where access to complete state information is limited or corrupted by noise. This poses significant challenges for reinforcement learning algorithms, as methods relying on full states or long observation histories can be computationally [...] Read more.

In many real-world control tasks, agents operate under partial observability, where access to complete state information is limited or corrupted by noise. This poses significant challenges for reinforcement learning algorithms, as methods relying on full states or long observation histories can be computationally expensive and less robust. Four-Phase Rapid Motor Adaptation (FRMA) is a reinforcement learning framework designed to address these challenges in high-frequency control tasks under partial observability. FRMA proceeds through four sequential stages: (i) full-state pretraining to establish a strong initial policy, (ii) auxiliary hidden-state prediction for LSTM memory initialization, (iii) aligned latent representation learning to bridge partial observations with full-state dynamics, and (iv) latent-state policy fine-tuning for robust deployment. Notably, FRMA leverages full-state information (

s_{t}

) only during training to supervise latent representation learning, while at deployment it requires only short sequences of recent observations and actions. This allows agents to infer compact and informative latent states, achieving performance comparable to policies with full-state access. Extensive experiments on continuous control benchmarks show that FRMA attains near-optimal performance even with minimal observation–action histories, reducing reliance on long-term memory and computational resources. Moreover, FRMA demonstrates strong robustness to observation noise, maintaining high control accuracy under substantial sensory corruption. These results indicate that FRMA provides an effective and generalizable solution for partially observable control tasks, enabling efficient and reliable agent operation when full state information is unavailable or noisy. Full article

(This article belongs to the Section Robotics, Mechatronics and Intelligent Machines)

► Show Figures

Figure 1

22 pages, 6248 KB

Open AccessArticle

Optimization Strategy and Evaluation of the Flow Heat Characteristics of the Cooling Plates of Electromagnetic Separators

by Jingjuan Du, Ke Li, Xiaoyuan Wang, Haiying Lv and Hongge Ren

Machines 2025, 13(10), 884; https://doi.org/10.3390/machines13100884 - 25 Sep 2025

Abstract

Electromagnetic separators are widely used in new energy battery purification, resource recycling, and mineral processing. However, coil heating can cause a decline in separation performance and damage to coil insulation. To ensure the stable operation of electromagnetic separators, cooling plates are employed to [...] Read more.

Electromagnetic separators are widely used in new energy battery purification, resource recycling, and mineral processing. However, coil heating can cause a decline in separation performance and damage to coil insulation. To ensure the stable operation of electromagnetic separators, cooling plates are employed to effectively mitigate temperature rise. To explore a high-performance and economical cooling method, this paper employs CFD finite element analysis for the structural optimization of cooling plates. First, the paper investigates the flow heat characteristics of S-shaped cooling plates. Numerical simulations are performed to analyze the variation of fluid characteristics with different numbers of water channels. Regression equations linking structural parameters to performance indicators are derived, and the optimal channel number and hydraulic diameter are determined. Furthermore, to enhance heat transfer efficiency, an innovative semicircular groove structure is introduced on the cooling plate walls. An optimization strategy based on a genetic algorithm is developed to determine the optimal groove parameters. A simulation shows that the optimized cooling plate reduces coil temperature by 12.63 °C with a decrease of 15.31% compared with the original design. Finally, a prototype with optimized parameters is manufactured after the experimental results of the two test points and the simulation results reveal errors of 0.26% and 0.96%, respectively. The experimental results align well with the simulations, confirming the reliability of the experimental results and the feasibility of the optimization strategy, and providing a reference for future cooling plate designs. Full article

(This article belongs to the Section Machine Design and Theory)

► Show Figures

Figure 1

37 pages, 11818 KB

Open AccessReview

Research Progress and Application of Vibration Suppression Technologies for Damped Boring Tools

by Han Zhang, Jian Song, Jinfu Zhao, Xiaoping Ren, Aisheng Jiang and Bing Wang

Machines 2025, 13(10), 883; https://doi.org/10.3390/machines13100883 - 25 Sep 2025

Abstract

Deep hole structures are widely used in the fields of aerospace, engineering machinery, marine, etc. During the deep hole machining processes, especially for boring procedures, the vibration phenomenon caused by the large aspect ratio of boring tools seriously restricts the machining accuracy and [...] Read more.

Deep hole structures are widely used in the fields of aerospace, engineering machinery, marine, etc. During the deep hole machining processes, especially for boring procedures, the vibration phenomenon caused by the large aspect ratio of boring tools seriously restricts the machining accuracy and production efficiency. Therefore, extensive research has been devoted to the design and development of damped boring tools with different structures to suppress machining vibration. According to varied vibration reduction technologies, the damped boring tools can be divided into active and passive categories. This paper systematically reviews the advancements of vibration reduction principles, structure design, and practical applications of typical active and passive damped boring tools. Active damped boring tools rely on the synergistic action of sensors, actuators, and control systems, which can monitor vibration signals in real-time during the machining process and achieve dynamic vibration suppression through feedback adjustment. Their advantages include strong adaptability and wide adjustment capability for different machining conditions, including precision machining scenarios. Comparatively, vibration-absorbing units, such as mass dampers and viscoelastic materials, are integrated into the boring bars for passive damped tools, while an energy dissipation mechanism is utilized with the aid of boring tool structures to suppress vibration. Their advantages include simple structure, low manufacturing cost, and independence from an external energy supply. Furthermore, the potential development directions of vibration damped boring bars are discussed. With the development of intelligent manufacturing technologies, the multifunctional integration of damped boring tools has become a research hotspot. Future research will focus more on the development of an intelligent boring tool system to further improve the processing efficiency of deep hole structures with difficult-to-machine materials. Full article

(This article belongs to the Section Machine Design and Theory)

► Show Figures

Figure 1

15 pages, 5189 KB

Open AccessArticle

Assembly Complexity Index (ACI) for Modular Robotic Systems: Validation and Conceptual Framework for AR/VR-Assisted Assembly

by Kartikeya Walia and Philip Breedon

Machines 2025, 13(10), 882; https://doi.org/10.3390/machines13100882 - 24 Sep 2025

Abstract

The growing adoption of modular robotic systems presents new challenges in ensuring ease of assembly, deployment, and reconfiguration, especially for end-users with varying technical expertise. This study proposes and validates an Assembly Complexity Index (ACI) framework, combining subjective workload (NASA Task Load Index) [...] Read more.

The growing adoption of modular robotic systems presents new challenges in ensuring ease of assembly, deployment, and reconfiguration, especially for end-users with varying technical expertise. This study proposes and validates an Assembly Complexity Index (ACI) framework, combining subjective workload (NASA Task Load Index) and task complexity (Task Complexity Index) into a unified metric to quantify assembly difficulty. Twelve participants performed modular manipulator assembly tasks under supervised and unsupervised conditions, enabling evaluation of learning effects and assembly complexity dynamics. Statistical analyses, including Cronbach’s alpha, correlation studies, and paired t-tests, demonstrated the framework’s internal consistency, sensitivity to user learning, and ability to capture workload-performance trade-offs. Additionally, we propose an augmented reality (AR) and virtual reality (VR) integration workflow to further mitigate assembly complexity, offering real-time guidance and adaptive assistance. The proposed framework not only supports design iteration and operator training but also provides a human-centered evaluation methodology applicable to modular robotics deployment in Industry 4.0 environments. The AR/VR-assisted workflow presented here is proposed as a conceptual extension and will be validated in future work. Full article

(This article belongs to the Special Issue Selected Papers from the AHFE International Conference on Human Factors in Design, Engineering, and Computing)

► Show Figures

Figure 1

19 pages, 4231 KB

Open AccessArticle

Deep Feature Decoupling Network for Ball Mill Load Signals

by Xiaoyan Luo, Wei Huang, Saisai He, Wencong Xiao and Zhihong Jiang

Machines 2025, 13(10), 881; https://doi.org/10.3390/machines13100881 - 24 Sep 2025

Abstract

Accurately identifying the load status of a ball mill is critical for optimizing grinding efficiency and ensuring operational stability. However, the one-dimensional vibration signals collected from ball mills exhibit strong non-stationarity and a high degree of entanglement between multi-scale local transient features and [...] Read more.

Accurately identifying the load status of a ball mill is critical for optimizing grinding efficiency and ensuring operational stability. However, the one-dimensional vibration signals collected from ball mills exhibit strong non-stationarity and a high degree of entanglement between multi-scale local transient features and long-range temporal evolution patterns. To address this, rather than relying on a purely black-box approach, this paper introduces a novel Deep Multi-scale Spatial–Temporal Feature Decoupling Network (DMSTFD-Net) guided by a clear feature decoupling philosophy to enhance model interpretability. The core of DMSTFD-Net lies in its hierarchical collaborative feature refinement mechanism. It first utilizes a one-dimensional residual network (ResNet) to adaptively capture and preliminarily decouple multi-scale spatial characteristics from the raw signal. Subsequently, the extracted high-level feature sequences are fed into a bidirectional gated recurrent unit (Bi-GRU) to decouple high-order temporal dynamic patterns. Experiments on a multi-condition dataset demonstrate that the proposed network achieves a state-of-the-art accuracy of 97.65%. Furthermore, dedicated cross-condition experiments and t-SNE visualizations validate the framework’s effectiveness. The results confirm that DMSTFD-Net provides a powerful, robust, and more interpretable solution for ball mill load identification. Full article

(This article belongs to the Section Advanced Manufacturing)

► Show Figures

Figure 1

21 pages, 2989 KB

Open AccessArticle

Numerical Investigation of Hydrogen Substitution Ratio Effects on Spray Characteristics, Combustion Behavior, and Emissions in a Dual-Fuel Compression Ignition Engine

by Takwa Hamdi, Fathi Hamdi, Samuel Molima, Victor M. Domínguez, José Rodríguez-Fernández, Juan José Hernández and Mouldi Chrigui

Machines 2025, 13(10), 880; https://doi.org/10.3390/machines13100880 - 23 Sep 2025

Viewed by 166

Abstract

Hydrogen is a promising alternative fuel for internal combustion engines due to its high specific energy, fast flame speed, and carbon-free combustion. In dual-fuel operation, it offers a practical route to reducing greenhouse gas emissions while remaining compatible with existing engine hardware. This [...] Read more.

Hydrogen is a promising alternative fuel for internal combustion engines due to its high specific energy, fast flame speed, and carbon-free combustion. In dual-fuel operation, it offers a practical route to reducing greenhouse gas emissions while remaining compatible with existing engine hardware. This work evaluates how the hydrogen energy substitution ratio (HSR = 50, 70, and 90%) influences spray dynamics, combustion characteristics, and emissions in a heavy-duty compression ignition engine. Simulations are validated against experiments and use a URANS RNG k–ε framework with a hybrid combustion model: the Eddy Dissipation Concept (EDC) coupled with detailed kinetics (111 species, 768 reactions) for auto-ignition and diffusion burning of diesel, and a G-equation for propagation of a hydrogen-rich premixed flame. The results reveal clear spray–combustion linkages. At HSR 50, the higher Weber number induces stronger breakup, yielding a smaller Sauter mean diameter and higher number-averaged droplet velocity; at HSR 90, the spray is more stable and less atomized, with larger droplets and a shorter vapor penetration length. Increasing the HSR reduces unburned hydrocarbons (UHCs) by more than 50% from HSR 50 to HSR 90 while modestly altering combustion phasing (a later CA50 and a shorter burn duration due to faster hydrogen flame propagation). The validated model provides a practical tool for optimizing dual-fuel settings and HSR–EGR–SOI trade-offs to balance efficiency and emissions. Full article

(This article belongs to the Section Electromechanical Energy Conversion Systems)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Machines, Volume 13, Issue 10 (October 2025) – 8 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI