Search Results (48)

To address the uneven distribution of secondary-suspension loads, the insufficient prediction accuracy of conventional mechanistic models, and the limited comprehensive optimization capability of existing leveling algorithms in the weighing and leveling process of high-speed railway carbodies, this study proposes a secondary-suspension load distribution optimization method that integrates radial basis function (RBF) neural-network error compensation with a multi-objective improved whale migration optimization algorithm. First, a mechanistic model describing the relationship between secondary-suspension loads and shim thickness is established based on the four-point weighing mechanics, and an RBF neural network is employed to compensate for the model prediction error, thereby improving the characterization accuracy of the actual loading state of the carbody. Second, a multi-objective optimization model for carbody weighing and leveling is formulated by taking the load standard deviation, total shim thickness, and number of shim positions as optimization objectives. Furthermore, the whale migration algorithm is improved according to the requirements of secondary-suspension load optimization, enabling collaborative multi-objective optimization of load deviation and total shim thickness. Finally, simulated-carbody tests and field tests on actual carbodies are carried out using a weighing and leveling test bench. The results show that the proposed method can identify the stress-free state of the carbody more accurately, reduce the error between theoretical calculations and measured data, and effectively improve the nonuniform distribution of secondary-suspension loads, demonstrating good engineering applicability. Full article

Internet of Things (IoT), sensor-network, and cyber-physical system (CPS) software increasingly relies on large language models (LLMs) and autonomous agents for code generation, maintenance, and vulnerability repair. However, LLM-generated edge services, telemetry APIs, configuration handlers, and data-aggregation routines can introduce SQL injection, path traversal, command injection, hard-coded credentials, and unsafe device-control logic, which may compromise sensing data integrity and system safety. Existing approaches largely rely on static post hoc analysis and lack a unified modeling of the generation process, making it difficult to achieve a principled trade-off between functionality and security. To address this challenge, we propose SafeCodeRL, a framework that integrates multi-agent collaboration with constrained reinforcement learning for trustworthy LLM-generated IoT/CPS software. SafeCodeRL models code generation as a security-aware sequential decision process, where Planner, Code, Security, Test, and Critic agents jointly optimize task decomposition, code synthesis, vulnerability auditing, and sandbox-based validation. We design a constraint-aware policy based on Proximal Policy Optimization, augmented with a Lagrangian mechanism and a shielding strategy to explicitly enforce security constraints. Experiments on real-world engineering and security benchmarks, including SWE-bench, SecurityEval, and CyberSecEval, show that SafeCodeRL reduces high-risk vulnerabilities by over 60% while maintaining high functional correctness. A scenario-level IoT/CPS case study further demonstrates that SafeCodeRL substantially improves secure pass rates for sensor telemetry, edge gateway, configuration-management, and data-aggregation tasks, providing a practical path toward trustworthy AI-assisted software development for sensor-driven systems. Full article

Deep neural networks can fit nonlinear bearing vibration responses, but their learned parameters are difficult to relate to contact deformation, rolling element angular position, and other acceleration-generating mechanisms. To improve physical traceability in data-driven bearing dynamics identification, this study develops a physics-informed SINDy-NN with a mechanism-guided feature library. This paper presents a novel approach for constructing a physics-informed SINDy-NN (Sparse Identification of Nonlinear Dynamics-based Neural Network) and demonstrates its application in identifying bearing dynamics. A 5-DoF (five Degrees of Freedom) bearing dynamics model is built, and the primary components influencing the acceleration response are analyzed. This analysis forms the basis for defining a physics-explainable basis function library for the SINDy-NN. For comparison, widely used polynomial and Fourier libraries are also employed to evaluate modeling accuracy and convergence speed. Furthermore, to address the limited number of bearing data, virtual states are generated by applying multiple finite differences to the acceleration signal, expanding the dimensionality of the model and enabling the use of a Multi-Input–Multi-Output (MIMO) model in SINDy-NN. Finally, experimental data from the FEMTO bearing test bench are utilized for validation. The results demonstrate that the physics-informed SINDy-NN offers superior modeling efficiency, with sufficient accuracy and improved interpretability compared to general SINDy-NN. Full article

Thumb degree-of-freedom (DOF) allocation in anthropomorphic robot hands involves a trade-off between functional mobility and mechanical-control complexity. This study presents a controlled multi-metric framework for comparing recurring thumb DOF configurations under common palm geometry, non-thumb finger structure, reference frames, Denavit–Hartenberg kinematics, and sampling assumptions. Five literature-derived thumb configurations, namely 3-1-1, 2-2-1, 2-1-1, 2-0-1, and 1-1-1, were evaluated to determine which thumb DOFs should be preserved when kinematic complexity is reduced. The theoretical evaluation included Kapandji Opposition Test reachability, opposition alignment, workspace volume, workspace compactness, cylindrical grasp opportunity, and Jacobian-based dexterity. A targeted experimental validation of the 2-1-1 and 2-0-1 prototypes was then performed on a tendon-driven test bench. The results showed that qualitatively similar thumb configurations are quantitatively unequal: several designs achieved identical Kapandji scores but differed substantially in workspace, alignment, dexterity, and grasp feasibility. Overall, 3-1-1 achieved the strongest overall capability, while 2-2-1 emerged as the strongest reduced-complexity alternative and achieved the best mean dexterity. Retaining two active carpometacarpal DOFs preserved a large share of dexterous function, whereas metacarpophalangeal fixation maintained selected cylindrical grasps but narrowed the feasible task boundary. Full article

Aircraft emergency hydraulic pumps are safety-critical units whose intermittent operation complicates condition assessment and reduces the diagnostic value of conventional bulk physicochemical oil properties. This study evaluates the applicability of tribotechnical oil analysis for monitoring degradation of the NP-27T emergency hydraulic pump under controlled bench conditions. Four NP-27T units were tested on a dedicated hydraulic bench, and oil samples were collected at defined intervals during operation and after test completion. The diagnostic methodology combined elemental spectrometry (ICP-OES), particle counting interpreted with reference to ISO 4406, and analytical ferrography. The results showed that flow performance deterioration was accompanied by measurable changes in oil-borne wear indicators, although the sensitivity of the individual diagnostic channels varied among the tested units. In several cases, the coarse particle fraction (>15 μm) exhibited the clearest response to degradation, while Fe and Cu concentrations provided useful but not uniformly monotonic trends across all pumps. Using a pragmatic early warning criterion based on the first exceedance of 100 particles/mL in the >15 μm fraction, the coarse particle signal provided lead times of approximately 13–345 min before the flow-based rejection limit was reached in the four tested units. Ferrographic analysis identified cutting and fatigue debris, together with larger wear particles, in units approaching or reaching the flow-based rejection limit. Overall, the findings demonstrate that the combined use of elemental analysis, particle counting, and ferrography provides a practical multi-indicator framework for relating oil-diagnostic signals to functional degradation of the NP-27T hydrogenerator. Under the present bench conditions, flow proved to be a more sensitive degradation indicator than pressure. The proposed approach therefore represents a useful complementary tool for maintenance decision-making and for integration with vibration-based condition monitoring of aircraft hydraulic systems. Full article

Blade Tip Timing (BTT) is an essential non-contact technique for monitoring vibrations in rotating machinery, but its practical accuracy is often degraded by noise, undersampling, and spectral leakage. This paper proposes a multi-stage robust Bayesian high-resolution identification framework that systematically addresses these challenges. A recursive digital algorithm based on Kalman filtering estimates the rotational speed without requiring once-per-revolution probes, effectively suppressing sensor noise. An attention-enhanced dynamic convolutional autoencoder then generates channel-specific window functions to minimize spectral leakage. The core identification algorithm extracts phases via all-phase FFT and employs sub-bin interpolation to overcome the resolution limitation of conventional FFT. A Tukey-biweight-based robust aggregation strategy is used to suppress the influence of abnormal or unequal-quality sensor channels during multi-channel phase fusion. A Bayesian prior distribution over the vibration order guides the estimation toward physically plausible values under noisy conditions. Finally, a coarse-to-fine multi-stage search strategy drastically reduces computational burden while preserving accuracy. Experiments on a rotor-blade test bench at constant and variable speeds show that the method reduces the noise floor by about 60 dB, achieves a maximum frequency identification error of 7.84%, and accelerates the search by approximately 48.6% compared to exhaustive search. The proposed method provides a reliable and efficient solution for blade health monitoring. Full article

Accurate classification of spindle rotational errors is critical for ensuring machining precision and operational reliability of CNC machine tools. However, existing methods face challenges in extracting discriminative feature information from vibration signals due to small inter-class differences and complex electromechanical interference. This paper proposes a novel deep learning model, MFAFENet, based on multi-sensor collaboration and multi-scale feature information adaptive fusion. Vibration signals from three mounting positions are transformed into time-frequency information representations via Short-time Fourier Transform. The proposed network adaptively fuses multi-scale feature information from parallel branches with different kernel sizes through a branch attention mechanism. An efficient channel attention module is then incorporated to recalibrate channel-wise feature responses. The cross-entropy loss function is employed to optimize the network parameters during training. Experiments on a spindle reliability test bench demonstrate that MFAFENet achieves 93.37% average test accuracy, outperforming other comparative methods. Ablation and comparative studies confirm the effectiveness of each module and the clear advantage of adaptive fusion over fixed-weight multi-scale methods. Multi-sensor fusion further improves accuracy by 7.23% over the best single-sensor setup. The proposed method establishes an effective end-to-end mapping between vibration signals and rotational errors, providing a promising solution for high-precision spindle condition monitoring. Full article

The cleaning unit is a key functional component of grain combine harvesters, yet its operating parameters are still predominantly adjusted according to operator experience, resulting in limited adaptability to fluctuating working conditions. To enhance the intelligence and stability of the cleaning process, this study develops a fuzzy control approach supported by data-driven performance modeling. Based on multi-condition bench experiments, feeding rate, fan speed, cleaning sieve vibration frequency, and sieve opening were selected as input variables. Gaussian Process Regression (GPR) models were established to describe the nonlinear relationships between operating parameters and cleaning loss rate and impurity rate, and impurity rate was inferred online to compensate for the absence of a reliable sensor. Taking feeding rate variation as the primary disturbance, a dual-input fuzzy control strategy was designed using loss rate monitoring and model-predicted impurity rate as feedback signals. Simulation and bench test results show that, under small and moderate load disturbances (±20% and ±35%), the proposed method reduces either impurity rate or cleaning loss rate through coordinated parameter adjustment. Under large disturbances (±50%), performance deterioration cannot be fully eliminated, but its extent is alleviated compared with open-loop conditions. Full article

As a critical mechanical component, the operational integrity of rolling bearings is essential for equipment safety. However, under strong noise interference, the weak fault features in vibration signals are difficult to extract. To address this issue, a novel fault diagnosis method is proposed in this paper, which integrates an improved speed time-varying filtering empirical mode decomposition (ISTVF-EMD) with an adaptive sine–cosine optimization algorithm (A-SCA), enabling precise and efficient extraction of fault features. The core of the proposed method lies in improving the conventional time-varying filtering empirical mode decomposition (TVF-EMD) by setting a maximum decomposition layer limit, effectively addressing issues of excessive components and low computational efficiency during the decomposition of low signal-to-noise ratio (SNR) signals. Furthermore, a multi-characteristic frequency energy concentration centrality (MCFECC) index is employed as a fitness function to guide A-SCA in adaptively searching for the optimal bandwidth threshold and fitting order parameters of ISTVF-EMD, thereby extracting components with the most enriched fault information. Validated through simulation and multiple test bench cases, the results indicate that the proposed method can not only significantly enhance the fault characteristic frequencies and their harmonics in the envelope spectrum, successfully diagnosing outer race, inner race, and rolling element faults, but also, compared with the original method, ISTVF-EMD substantially reduces the computational time while ensuring or even improving the decomposition quality. The method presented in this paper provides an effective solution for achieving precise and adaptive fault diagnosis of rolling bearings under strong noise interference. Full article

The power-shift system employed in agricultural multi-range continuously variable transmissions (CVTs) features a complex structure and control logic, presenting significant challenges to the reliability of agricultural machinery. To enable timely detection of faults, constructing an intelligent fault diagnosis classifier to monitor the system’s health status is essential. Typically, fault samples utilized for classifier development originate from ideal bench tests, characterized by uniform patterns and limited diversity, thereby hindering the algorithm’s generalization capability. This study addresses this issue by proposing a generative adversarial network (GAN) model, integrated with a triple loss function and a novel generator architecture, to augment the fault dataset under laboratory conditions. The generator architecture comprises a variational autoencoder module and an oil pressure point attention mechanism, enabling the generation of diverse and fluctuating virtual samples. Building on this augmented dataset, a fault classifier based on one-dimensional ConvNeXt was developed. Experimental results indicate that the classifier achieves an accuracy of 99.73%. While classifier accuracy decreases with increasing noise levels, the GAN-generated dataset provides more comprehensive training, resulting in an accuracy approximately 3% higher than that achieved using the original dataset. Full article

To address the requirements for multi-channel decoupling and high-precision control in quadrotor UAV systems, this paper proposes a novel intelligent controller (TD3-ADRC) which integrates Active Disturbance Rejection Control (ADRC) with the Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithm. Firstly, the dynamic model of the quadrotor is established. Secondly, a parameterized tanh function is introduced and applied to design the tracking differentiator, extended state observer, and nonlinear feedback control law. Then, the TD3 learning mechanism is incorporated to automatically learn and optimize controller parameters, thereby significantly enhancing the system’s disturbance rejection capability. Finally, simulation studies comparing conventional PID, ADRC, DDPG and the proposed TD3-ADRC algorithms are conducted in Simulink. In addition, a bench test system is developed using the PX4 flight controller. Experimental results show that, under complex environmental conditions, the proposed TD3-ADRC controller outperforms both conventional PID and linear ADRC methods in terms of reliability and adaptability, validating the effectiveness of the proposed control approach. Full article

Wind turbine (WT) misalignment fault diagnosis is constrained by critical signal processing challenges: weak fault features, intense background noise, and poor generalization. This study proposes a lightweight method for high-precision fault diagnosis. A fixed-threshold wavelet denoising method with the scene-specific pre-optimized parameter a (0 < a ≤ 1.3) is proposed: the parameter a is determined via offline grid search using the feature retention rate (FRR) as the objective function for typical wind farm operating scenarios. A multi-scale depthwise separable CNN (MS-DSCNN) captures multi-scale spatial features via 3 × 1 and 5 × 1 kernels, reducing computational complexity by 73.4% versus standard CNNs. An attention-based minimal peephole LSTM (AttMPLSTM) enhances temporal feature measurement, using minimal peephole connections for long-term dependencies and channel attention to weight fault-relevant signals. Joint L1–L2 regularization mitigates overfitting and environmental interference, improving model robustness. Validated on a WT test bench, the Adams simulation dataset, and the CWRU benchmark, the model achieves a 90.2 ± 1.4% feature retention rate (FRR) in signal processing, an over 98% F1-score for fault classification, and over 99% accuracy. With 2.5 s single-epoch training and a 12.8 ± 0.5 ms single-sample inference time, the reduced parameters enable real-time deployment in embedded systems, advancing signal processing for rotating machinery fault diagnosis. Full article

This paper presents the design and experimental investigation of a multifunctional friction test bench, aiming to characterize the frictional and transmission efficiency of rope–drum systems in high-altitude wind power generation. The study addresses a critical gap in the experimental validation of key components for this demanding application. The test bench, comprising loading, power, test, and data acquisition modules, was designed to measure the equivalent friction coefficient (a comprehensive macro-parameter, not the traditional material friction coefficient) between an ultra-high-molecular-weight polyethylene (UHMWPE) fiber rope and a drum, as well as the transmission efficiency of pulleys. Key parameters, including contact angle, gasket material (steel vs. polyamide (PA)), groove type (U vs. V), and rotational speed, were systematically tested using tension and speed and torque sensors for data acquisition. Experimental results show that the equivalent friction coefficient initially increased and then decreased with the contact angle, reaching a maximum of approximately 0.15 at 100°. The coefficient was positively correlated with rotational speed, increasing by about 40% for steel and 10% for PA linings as speed rose from 25 to 100 rpm. Steel linings exhibited a significantly higher equivalent friction coefficient (0.14–0.17) than PA linings (0.10–0.13). Similarly, in transmission tests, steel pulleys demonstrated superior efficiency compared to PA pulleys, while V-grooves slightly reduced efficiency compared to U-grooves. Furthermore, pulley misalignment was found to decrease transmission efficiency. This work provides essential experimental data and a robust testing platform, laying a foundation for optimizing the efficiency and reliability of high-altitude wind energy systems. Full article

A collaborative control strategy combining the hyperbolic sine-cosine optimization (SCHO) algorithm with fuzzy adaptive linear active disturbance rejection control is proposed to address the nonlinearity and uncertainties in the hydraulic position servo system of shock absorber test benches. First, based on the dynamic characteristics of the shock absorber fatigue test bench and the tested shock absorber, a linearized model of the valve-controlled hydraulic cylinder and its load was established. The coupling mechanism of system parameter perturbation and disturbance was also analyzed. A third-order LADRC (Linear Active Disturbance Rejection Control) was designed considering the linear model characteristics of the test bench hydraulic servo system model to quickly estimate internal system disturbances and perform real-time compensation. Secondly, a multi-objective optimization function was constructed by integrating system performance indicators and incorporating controller and observer bandwidths into the optimization objectives. The SCHO algorithm was used for the global search and optimization of key LADRC parameters. To enhance the controller’s adaptive capability of modeling uncertainties and external disturbances, a fuzzy adaptive module was introduced to adjust control gains online according to errors and their rates of change, further improving system robustness and dynamic performance. The results show that compared with traditional PID, under different working conditions, the proposed method reduced the maximum tracking error, overshoot, and system response time by an average of 45%, from 15% to 5%, and by approximately 30%, respectively. Meanwhile, the parameter combination obtained via SCHO effectively avoids the limitations of manual parameter tuning, significantly improving control accuracy and energy utilization. The simulation results indicate that this method can significantly enhance position-tracking accuracy compared with traditional LADRC, providing an effective solution for position-tracking control in hydraulic servo testing systems. Full article

A wearable closed-loop transcutaneous electrical nerve stimulation (TENS) platform has been developed to address the limitations of conventional open-loop neuromodulation systems. Unlike existing systems such as CLoSES—which targets intracranial stimulation—and electromyography-triggered functional electrical stimulation (EMG-FES) platforms primarily used for motor rehabilitation, the proposed device uniquely integrates low-latency surface electromyography (sEMG)-driven control with six-channel current stimulation in a fully wearable, non-invasive format aimed at ambulatory pain modulation. The system combines real-time sEMG acquisition, adaptive signal processing, a programmable multi-channel stimulation engine, and a high-voltage, boost-regulated power supply within a compact, battery-powered architecture. Bench-top evaluations demonstrate rapid response to EMG events and stable biphasic output (±22 mA) across all channels with high electrical isolation. A human-subject protocol using the Cold Pressor Test (CPT), heart rate variability (HRV), and galvanic skin response (GSR) has been designed to evaluate analgesic efficacy. While institutional review board (IRB) approval is pending, the system establishes a robust foundation for future personalized, mobile neuromodulation therapies. Full article