Cross-condition bearing fault diagnosis suffers from severe performance degradation due to domain shift across different operating conditions, especially when only a few labeled target-domain samples are available. To address this challenge, this paper proposes a progressive attention-guided two-stage transfer learning framework for few-shot bearing fault diagnosis across different fixed operating points. First, the raw time-domain vibration signals are fused with frequency-domain representations extracted by short-time Fourier transform (STFT) to enhance fault feature representation. Then, a progressive attention-guided feature learning strategy is developed by integrating dual efficient channel attention (ECA) modules into a deep one-dimensional convolutional neural network (1D-CNN), enabling the network to adaptively emphasize fault-sensitive features while suppressing redundant information. Subsequently, a two-stage transfer learning strategy is designed, consisting of transferable feature learning from the source domain and few-shot adaptation to the target domain. During target-domain adaptation, key feature extraction layers are frozen, and a sample-balanced optimization mechanism is introduced to alleviate the dominance of source-domain samples during joint training. Experimental results on the Case Western Reserve University (CWRU) bearing dataset demonstrate that the proposed method achieves an average accuracy of 99.96% across three cross-condition transfer tasks. Furthermore, experiments conducted on a self-built shaft system dataset show that the proposed method achieves an average accuracy of 87.11% under three representative transfer scenarios. The results verify that the proposed framework effectively mitigates domain shift and enables accurate bearing fault diagnosis with limited labeled target-domain samples.
Optimizing transmission efficiency is critical for advancing electric vehicle (EV) technology. This review paper covers advanced driveline schemes for electric cars and electric heavy vehicles through a selection and analysis of articles published over the past decade, focusing on passenger and commercial electric vehicle drivetrains, transmission efficiency modeling, gear optimization, and tribological and NVH outcomes. The selected sources are grouped into analytical themes to map current academic and industrial developments in transmission topologies, specifically comparing single-speed reduction units with two-speed and multi-speed configurations. A comparative summary is discussed to consolidate the characteristics, complexity, efficiency impact, and typical applications of the various gearbox types. The review article is complemented by an examination of loss mechanisms through the lens of international tribological research. Key focal points include mechanical power losses, gear and bearing friction, and the trade-offs associated with low-viscosity fluids for electric vehicles to provide a comprehensive support resource for researchers and designers in the field of transmissions for EVs. Future development trajectories are delivered regarding the increase in efficiency and specific challenges of EV gearboxes.
Passive tuned mass dampers (TMDs) can reduce chatter, but designing and fabricating an accurately tuned absorber remains challenging due to manufacturing constraints. This study proposes a Design of Experiments and Finite Element Analysis (DOE-FEA) based constrained design optimization framework for a passive two-degree-of-freedom (TDOF) TMD to suppress regenerative chatter in boring operations by considering practical and manufacturing constraints on absorber position, mass ratio, moment of inertia and fixed inter-spring distance. The proposed, additively manufactured TMD housing, made from polylactic acid (PLA), includes a mass block supported by two spring-damper elements that enable coupled translational and rotational interactions with the boring bar. A finite-element forced vibration analysis of the boring bar TMD system is developed to obtain the real and imaginary parts of the frequency response function (FRF), which are then used to construct the stability lobes. The minimum limiting depth of cut over the spindle speed range is used as the optimization criterion, and response surface methodology (RSM) is used to obtain optimum absorber parameters within realistic design constraints. The dynamic behaviour of the TDOF TMD is experimentally and numerically evaluated and compared with that of a single-degree-of-freedom (SDOF) TMD, attributing the relative improvement in performance primarily to the combined effects of independent absorber architecture, mass, stiffness and damping distribution and dynamic tuning. The results showed that the optimal TDOF TMD achieved a DOC of 11.054 mm, while the SDOF TMD achieved 4.335 mm. The experimental investigation of additively manufactured TDOF and SDOF TMDs demonstrated qualitatively similar dynamic phenomena to those of the corresponding numerically optimized absorbers. Time-domain acceleration response, spectrogram and power spectrum were used to compare these dynamic phenomena demonstrated by the SDOF and TDOF TMDs. A reduction in corresponding first and second amplitude peaks from −2.8 dB (670 Hz) and −22.7 dB (1360 Hz) in the case of the SDOF TMD to −17.6 dB (600 Hz) and −23.8 dB (1150 Hz) for the TDOF TMD verified the vibration attenuation and frequency redistribution phenomenon as exhibited by the FE-model.
Class imbalance is common in vibration-based fault diagnosis because normal-condition data are generally more abundant than fault data. This study proposes an auxiliary-classifier Wasserstein generative adversarial network with gradient penalty and spectral normalization, termed ACWGAN-SG, for fault-sample generation and progressive dataset augmentation. The method combines class-conditioned generation, Wasserstein adversarial learning, gradient penalty, spectral normalization, and PCC-CS-based sample screening. Experiments were conducted on the public CWRU bearing dataset and a self-built 12-class harmonic-reducer dataset, with the downstream diagnostic experiments covering balance ratios from 1:100 to 1:1. The CWRU and harmonic-reducer experiments were independently repeated five and three times, respectively. Under the balanced condition, ACWGAN-SG achieved mean diagnostic accuracies of 98.40% and 97.627% on the two datasets. On the harmonic-reducer dataset at BR = 1:2, the method obtained a Macro-F1 of 94.298%, a balanced accuracy of 94.333%, and an MCC of 0.9383. Repeated-run statistical analyses showed significant overall differences among the evaluated methods across the tested balance ratios. These results indicate that the proposed generation and progressive-augmentation procedure improves downstream diagnostic performance under the reported experimental settings.
Ultra-wideband (UWB) positioning provides cost-effective localization for quadrotors in global navigation satellite system (GNSS)-denied environments, but geometry-dependent, heavy-tailed errors challenge state estimation and safety-critical control. This paper presents a risk-aware framework linking geometry-aware error quantification with belief-space collision avoidance under a fixed four-anchor UWB configuration. Specifically, horizontal dilution of precision (HDOP) and nearest-anchor distance are used as spatial features in a Student’s-t process regression (STPR) model to predict UWB positioning errors and quantify the associated uncertainty. The compensated UWB measurements are then fused with inertial data through a Kalman filter to obtain a Gaussian belief state. A belief control barrier function (BCBF) maps ellipsoidal collision regions to a unit sphere, approximates them using tangent half-spaces, and is embedded in nonlinear model predictive control (NMPC). In outdoor flight experiments, the positioning RMSE is reduced to 0.071 m by the proposed STPR-KF method, compared with 0.299 m for raw UWB and 0.292 m for conventional KF. Feasible risk-aware obstacle avoidance and adjustable safety clearance are further demonstrated through numerical simulations. The feasibility of linking geometry-aware UWB error characterization with belief-space safety constraints for UWB-localized quadrotor navigation is therefore indicated.
Hydraulic short-circuit (HSC) operation is an important approach to enhancing the operational flexibility of pumped-storage power plants (PSPPs). However, under this new operating mode, the flow characteristics in the bifurcated pipe deteriorate significantly, posing a threat to the efficiency of the piping system and potentially affecting the inflow conditions for the turbine. In this study, six improved bifurcated pipe models were designed, and their internal flows under pumping, generating, and HSC modes were numerically simulated. Entropy production theory and vortex identification method were employed for flow field analysis. The results show that local modifications confined to the bifurcation are insufficient to simultaneously improve energy characteristics across different modes. In contrast, the bypass pipe enables early flow diversion, weakening the original high-dissipation regions while introducing controllable additional losses. M6 achieves an average energy loss reduction of 47.85% in the mid-to-high flow split ratio range (FSR > 0.3). A strong correlation is observed between vortex suppression and energy loss reduction: the bypass pipe substantially shortens the main vortex length at the inlet section of the generating branch, while simultaneously inducing new shear vortices at the junction; adjustment of its installation position is expected to further shorten their extension, thereby ensuring the normal operation of the turbine. This study provides a new technical pathway for extending the operating range of HSC operation and contributes to enhancing the grid-regulation capability of PSPPs.
A collision-free path is not sufficient for visibility-dependent mobile robot tasks: a moving obstacle can block the camera–target line of sight and cause inspection or visual-servoing failure even when the robot remains physically safe. Maintaining visual contact with targets is therefore important in Industry 4.0 environments, yet visibility-preserving maneuvers can conflict with navigation progress and collision avoidance. This work presents the Visibility-Informed Safety and Target Awareness framework with control barrier function filtering and occlusion-evasive local replanning (VISTA-CBF+ELR). The architecture combines visibility-risk planning, target-bearing control, an ELR supervisor, and a CBF quadratic program that keeps collision constraints hard while relaxing field-of-view and occlusion requirements through slack. Counterproductive interventions are limited through persistence, benefit–cost and feasibility gates, progress protection, bounded dwell, recovery, and cooldown. In locked factory simulations, redesigned VISTA achieved 67% and 73% strict-goal success under clean and nominal sensing, whereas Visibility-CEM-2D achieved 87% and 86% but with lower clearance. In matched Gazebo trials, strict success was 19/30 for redesigned VISTA, 26/30 without ELR, and 16/30 for Nav2 Smac+MPPI; zero-clearance collisions were 8/30, 3/30, and 14/30, with no difference surviving multiplicity correction. A separate CEM stress test sustained 6.875 Hz optimization, missed 26.31% of 100 ms deadlines, and held commands on 31.35% of ticks. The results demonstrate repair of the ELR pathology and conditional visibility-risk reduction while exposing safety–visibility trade-offs, transfer limitations, and real-time constraints.
Machining-induced residual stress, cutting force, and temperature govern the fatigue life, dimensional stability, and surface integrity of milled components, yet predictive models for these responses are still routinely validated only in-sample, concealing overfitting on small, single-laboratory datasets. This study re-examines a published AISI 1045 end-milling dataset (N = 24, combining one-factor-at-a-time and Taguchi L9 trials) using six regression paradigms: Multiple Linear Regression (MLR), random forest, gradient boosting, Support Vector Regression (SVR), Gaussian process regression (GPR), and a shallow neural network (ANN)—under leave-one-out cross-validation (LOO-CV). The previously reported in-sample R2 of 0.84 (from a smaller n = 9 subset) was substantially higher than the LOO-CV R2 of 0.167 obtained here on the full dataset; although this gap cannot be attributed to cross-validation alone, it shows a substantial in-sample/out-of-sample performance gap. SVR gave the strongest, bootstrap- and nested-CV-confirmed cross-validated residual-stress prediction (R2 = 0.575); its apparent force advantage (R2 = 0.558) was statistically indistinguishable from GPR and did not survive nested tuning, so it is reported cautiously. GPR was narrowly best for temperature (R2 = 0.492); the ANN and SVR underperformed the linear baseline there, though nested tuning traced this largely to a fixed hyperparameter rather than the kernel method itself. Random forest permutation importance identified feed rate as the dominant residual-stress predictor, consistent with the original ANOVA. The contribution is a cross-validated, multi-paradigm reassessment with explicit uncertainty and sensitivity analysis, together with a candidate low-cost screening surrogate for AISI 1045 process planning—not a replacement for XRD or FE—and a broader caution to match model complexity to sample size.
To address the influence of multi-source probabilistic uncertain parameters on the dynamic transmission error (DTE) of harmonic drives, this paper proposes a robust DTE modeling and multi-objective optimization method. First, a Chebyshev surrogate model is constructed by integrating the measured static transmission error (STE) probability model, system dynamic equations, and identified nominal parameters. Prototype validations show a prediction mean absolute percentage error (MAPE) of 8.14% and a mean absolute error (MAE) of 7.761″. Meanwhile, compared to the original dynamic equations, the surrogate model reduces the single-evaluation time from 0.147 s to 0.000003 s (a 49,000-fold acceleration), effectively overcoming the efficiency bottleneck of numerical integration in dynamic response evaluation. Secondly, to achieve the collaborative optimization of system transmission accuracy and anti-disturbance robustness, a Chebyshev–AMP–MOPSO algorithm integrating a diversity entropy state-driven weight and a pyramid-hierarchical dual-track search strategy is proposed, which improves upon the issues of local convergence and uneven solution set distribution in the classical MOPSO and NSGA-II algorithms. On this basis, parameter optimization under three decision preferences was completed. The accuracy-first scheme reduces the DTE mean by 3.67%, the robustness-first scheme reduces the standard deviation by 9.36%, and the balanced scheme improves both. Finally, comparative tests on five prototypes show the actual dynamic parameters’ deviation (Di) relative to the theoretical optimal configuration exhibits a consistent corresponding trend with measured DTE means. Prototypes with the minimum (Di = 0.365) and maximum (Di = 0.474) deviations yield the lowest and highest measured means, respectively, matching theoretical optimization expectations.
Intelligent following control for electric-drive unmanned ground vehicles (UGVs) is important for human–robot collaboration and autonomous mobility. However, target occlusion, detection noise, and short-term detection failures can reduce following stability and reliability. To address these challenges, this paper proposes a vision-based following method that integrates adaptive Kalman filtering with target-state-aware control. The proposed method updates the observation noise covariance online using an innovation-sequence sliding window and evaluates the validity of visual measurements based on the intersection over union between adjacent frames, thereby enabling target position prediction and compensation under unstable observation conditions. Meanwhile, the longitudinal velocity and lateral angular velocity are dynamically adjusted according to the target detection state, image-center offset, and distance variation to achieve continuous and stable following control. Experimental results show that the proposed method reduces the target position prediction error by approximately 10% compared with conventional Kalman filtering and maintains the longitudinal following distance at about 3 m, improving the robustness, continuity, and motion smoothness of electric-drive UGV target following in complex environments.
Cabin heating significantly affects the driving range of electric vehicles (EVs) under cold weather conditions, making energy-efficient thermal management essential. Conventional control strategies typically regulate a fixed cabin temperature setpoint, which may lead to suboptimal energy consumption and does not directly account for passenger thermal comfort. This study proposes a real-time thermal comfort-oriented nonlinear model predictive control (NMPC) framework for EV heat pump systems. To enable real-time implementation, a control-oriented model is developed by combining a data-driven model for heat pump performance with a linearized Predicted Mean Vote (PMV) model. The proposed NMPC optimizes the trade-off between passenger thermal comfort and compressor energy consumption while satisfying actuator constraints. A high-fidelity physics-based virtual plant is employed to evaluate the proposed strategy. Simulation results show that the proposed NMPC maintains thermal comfort within the recommended PMV range while reducing total energy consumption by 8.3% compared with a conventional rule-based controller. Furthermore, a parametric study involving 1210 gain-tuning cases of a rule-based controller and 100 NMPC weighting scenarios demonstrates a consistently superior comfort–energy trade-off. The average computation time remains below 0.15 s on the simulation platform, indicating that the proposed NMPC formulation can be solved within the selected sampling interval. These results highlight the potential of thermal comfort-oriented NMPC for improving both passenger comfort and energy efficiency in EV heat pump systems.
Commercial laser-processing systems are typically designed for a specific manufacturing process, limiting their adaptability in research and prototyping environments. This study presents the staged evolution of a laboratory-scale prototype originally conceived for metal powder bed fusion into a reconfigurable laser-processing platform. Successive mechanical, optical, electrical, and control developments are consolidated, including the integration and calibration of a 200 W fibre laser, a galvanometric scanning system, motorised powder feed and build platforms, a recoater mechanism, and a combined LabVIEW and weldMARK control architecture. The capabilities of the resulting platform were assessed through optical commissioning by laser marking and two experimental case studies involving single-layer fusion of AISI 316L powder and laser transmission welding of dissimilar thermoplastics. The marking trials provided qualitative confirmation of beam delivery, focal adjustment, and programmed path reproduction. In contrast, the powder experiments produced continuous fused regions, demonstrating controlled laser–powder interaction without constituting full multilayer powder-bed-fusion validation. Thermoplastic welding generated mechanically resistant joints, with failure occurring cohesively within the foam substrate rather than at the welded interface. These results demonstrate the potential of the developed system as a reconfigurable research platform for different laser-processing operations. Nevertheless, fully automated multilayer powder bed fusion still requires improvements in platform levelling, machine-zero integration, process synchronisation, and atmosphere monitoring.
The coordinated control of networked integrated energy systems (IESs) is complicated by the different response speeds of the electrical, gas, and thermal subsystems and by information exchange over directed communication graphs. Existing studies mainly consider the economic scheduling of a single IES, dynamic modeling of individual systems, or cooperative control of systems with two time scales, and therefore, they do not provide a unified supervisory framework for coordinating three energy domains with an explicit performance bound. This paper investigates whether a common leader-following framework can coordinate the principal variables of the three energy domains while limiting both regulation errors and control effort. Each IES is treated as an agent, the three energy domains are represented by separate reduced dynamic blocks sharing a common directed communication topology, and a distributed state feedback controller is developed using relative information, Riccati-based gain design, and complete Lyapunov analysis. Simulations of a network of four IESs show that the longest settling times in the electrical and gas domains are approximately 0.05 s and 5.41 s, respectively, whereas the thermal disagreement decreases by 96.5% over 300 s; the accumulated cost remains below its calculated upper bound in the nominal case and in all three time-scale settings, while a separate numerical communication reconfiguration case illustrates bounded responses during communication link removal and reconnection. These results demonstrate that the proposed framework can simultaneously coordinate variables with substantially different response speeds while accounting for regulation accuracy and control effort under the stated reduced model and fixed graph assumptions. The communication reconfiguration case provides a numerical illustration and does not constitute a general stability guarantee for arbitrary topology switching.
Pavement cracks pose significant challenges to the driving stability and operational safety of automated guided vehicles (AGV) in industrial and logistics environments. The complex geometry of cracks and their nonlinear coupling with vehicle dynamics make conventional rule-based or single-modal approaches insufficient for reliable engineering decision-making. To address this issue, this study proposes an artificial intelligence–driven multi-modal perception and engineering analysis framework for AGV speed optimization under representative operating conditions. From the artificial intelligence perspective, an improved lightweight instance segmentation model based on YOLO11 is developed by integrating a dynamic upsampling strategy, a hybrid multi-scale feature representation module, and a large-kernel attention mechanism, enabling robust and fine-grained crack extraction in complex pavement scenes. In addition, a multi-modal learning strategy is adopted to fuse two-dimensional visual features with three-dimensional point cloud-derived geometric parameters, allowing accurate quantification of crack width, depth, and surface damage. From the engineering analysis perspective, the relationship between AI-extracted crack geometric characteristics and AGV dynamic responses is established to investigate the influence of pavement defects on vehicle vibration behavior. A triaxial vibration acquisition system is constructed under controlled experimental conditions, and the relationship between crack severity, vibration characteristics, and driving speed is quantitatively analyzed. Based on vibration constraints, a hierarchical speed optimization strategy is formulated for different crack levels. Experimental results demonstrate that the proposed method achieves accurate crack perception and effective geometric feature characterization, providing a quantitative basis for AGV speed adjustment under different pavement conditions.
Accurate fault location in low-observability active distribution networks is hindered by uncertain inter-node relationships, underutilized early transients, and insufficient global context. To address these challenges, this paper proposes an adaptive Mamba-driven spatiotemporal graph neural network (AM-STGNN). It provides a unified task-driven spatiotemporal representation framework that progressively integrates fault-propagation modeling, global dependency modeling, transient-state learning, and topology-aware discriminative enhancement. Specifically, a prior-guided adaptive implicit topology is first learned to characterize task-dependent electrical coupling relationships among sparse observation nodes. Based on the resulting topology, topology-conditioned multi-order feature propagation and a dual-axis linear-attention module based on the spatiotemporal graph transformer (STGformer) are employed to capture local and global spatiotemporal dependencies. The resulting global spatiotemporal representation is subsequently processed by a Mamba selective state-space encoder to model input-dependent temporal evolution and emphasize informative fault transients. Finally, element-wise gated fusion, topology-aware differential output, and a margin constraint are employed to integrate the STGformer and Mamba representations and enhance the separability of adjacent faulted line sections with similar response characteristics. Extensive experiments demonstrate the effectiveness of AM-STGNN, while additional evaluations confirm its applicability to larger-scale networks, strongly phase-unbalanced conditions, and field-measured operating backgrounds. Robustness tests under individual and multi-level joint disturbances further demonstrate the practical relevance of the proposed architecture. Compared with the baseline models, AM-STGNN achieves consistent improvements in the macro-averaged F1 score (Macro-F1), exact accuracy, and one-hop accuracy under the clean IEEE 123-node condition. More importantly, it maintains clear performance advantages under identical mild, moderate, and severe joint disturbances, demonstrating improved robustness and practical relevance under simulated non-ideal operating conditions.
When a robot performs a grinding operation, the steady-state force/position tracking accuracy of its end-effector is critical to achieving high grinding quality. To solve this problem, a fuzzy adaptive impedance method is incorporated into the robot’s compliant control framework. Firstly, the robot dynamics model is established based on the Newton–Euler method. To describe the robot dynamics more comprehensively, a linear friction compensation model is also introduced. Secondly, a dynamic feedforward trajectory-tracking controller is proposed based on the dynamic model, and its stability is verified using a Lyapunov function. The impedance parameters are adjusted in real time according to the feedback contact force and its rate of change, thereby enabling dynamic equilibrium between the end contact force and end position. This allows the robot end-effector to exhibit compliance during external environmental interactions. Finally, a control platform of a force/position compliance controller was constructed, and two grinding conditions of plane and arc were designed to validate the effectiveness of force/position compliance control based on impedance control. Compared with the fixed impedance approach, the proposed method reduces overshoot by 11.6% (plane) and 12.45% (arc), improves surface roughness from Ra 0.042 μm to Ra 0.021 μm, and achieves faster force tracking with fewer oscillations.
Reviews of artificial intelligence (AI) for mobile robots usually cover one competence—perception, SLAM, path planning, control, or reinforcement learning—and rarely show how these combine into a working system. We take the opposite view and treat autonomy as one pipeline: sensing, perception, localization and mapping, prediction, planning, visual servoing and control, high-level decision-making, and continual learning. We survey how AI has reshaped each stage for industrial and service robots across Industry 4.0, 5.0, and the emerging Industry 6.0. Using a structured, PRISMA-informed protocol with explicit search strings, inclusion criteria, and cross-embodiment transfer rules, we screen the literature, analyze a corpus drawn mainly from the last five years, and position it against prior surveys with a coverage matrix that exposes their single-block focus. Four findings stand out. Perception and localization approach engineering maturity through multimodal fusion and foundation vision models. Planning and control stay effective but computationally demanding. Decision-making, now driven by large language and vision–language–action models, is powerful yet unverifiable and fails under safety constraints. Lifelong learning is almost absent from deployed systems. The decisive weaknesses sit at the interfaces: at the perception–planning, planning–control, and control–decision handoffs the sim-to-real gap, limited on-robot compute, and scarce industrial data compound. We compare AI families by technology readiness, catalog datasets and benchmarks, examine the safety-certification barrier, and consolidate cross-cutting gaps. We close with a staged roadmap toward Industry 6.0 and a next-generation architecture coupling a foundation perception backbone, a world model and digital twin, a continual-learning memory, and a reasoning core wrapped by a safety monitor. The aim is to move from cataloging algorithms to engineering integrated autonomy.
Freeform machining of impellers involves extended cycle times, leading to high energy consumption and costs necessitating efficient process optimization. This study develops a CAD/CAM-integrated hybrid Taguchi-Artificial Neural Network (ANN) model to optimize machining parameters for a freeform impeller. Four controllable factors, namely cutting feed (Cf), feed Z (Fz), retract feed (Rf), and cutter diameter (CD), were investigated at five levels using an L25 orthogonal array, with machining time as the response. Taguchi analysis identified cutting feed as the most dominant factor, while retract feed was insignificant, and a first-order regression model yielded an R2 of 95.88%. A two-layer feedforward neural network with six hidden neurons achieved an R2 of 0.9999 and a mean absolute error of 0.0976 min. To rigorously validate generalization, leave-one-out cross-validation was employed, identifying three hidden neurons as optimal with a cross-validated R2 of 0.9823, RMSE of 0.5350 min, and MAE of 0.3429 min. The final model trained on all samples achieved an R2 of 0.9996. Comparison with a quadratic regression model on the same test set confirmed the superior predictive capability of the ANN (R2=0.9992 vs. 0.9983). Optimal parameters (Cf=12,000 mm/min, Fz=600 mm/min, Rf=4000 mm/min, CD=6 mm) were validated through simulation, yielding a machining time of 11.05 min, representing a 52.6% reduction from 23.32 min. The hybrid Taguchi–ANN framework effectively optimizes freeform impeller machining, significantly enhancing productivity while maintaining process reliability.
No-till precision seeding is an important component of conservation tillage, but stable operation remains constrained by compacted undisturbed soil, dense crop residues, and uneven field surfaces. This review examines the evolution of no-till precision seeding equipment from mechanical soil–residue interaction to cyber–physical closed-loop regulation. It first summarizes how soil resistance and residue interference affect furrow opening, seed placement, and seed–soil contact. It then examines the development of residue-management, furrow-opening, covering, and compaction mechanisms, highlighting the transition from passive structural optimization toward active and adaptive operation. Advances in multi-source sensing, electric-drive metering, downforce control, and vibration suppression are further reviewed as enabling technologies for improving seeding stability under variable and high-speed conditions. Despite these advances, persistent trade-offs remain among residue-removal capacity, soil disturbance, energy demand, component durability, system complexity, and operational stability. Emerging approaches based on digital twins, adaptive damping, and cooperative autonomous systems may support further improvements, but their practical implementation still requires robust field performance and effective system integration. Overall, no-till seeding equipment is progressing toward perception-assisted and closed-loop intelligent regulation while continuing to face important mechanical and implementation challenges.
Accurate remaining useful life (RUL) prediction is essential for condition-based maintenance and safe aero-engine operation. To address non-stationary monitoring data, limited informative degradation samples, and the difficulty of jointly modeling local degradation patterns and temporal dependencies, this study proposes a degradation-aware dynamic masking augmentation method combined with a multiscale CNN–Transformer network. The 21 sensor variables in the NASA C-MAPSS dataset are first grouped by physical meaning and reconstructed into four-dimensional state features. A degradation-state score integrating local variance and trend slope is then used to adapt temporal masking probabilities across degradation stages, while feature masking probabilities are assigned according to feature importance. Masked positions are filled with adjacent unmasked observations, and invalid augmented samples are removed through trend consistency verification. A multiscale CNN with channel attention extracts local degradation features, and a Transformer encoder captures temporal dependencies. Bayesian optimization is used to determine key hyperparameters. On FD001, the proposed method achieves an MSE of 348.3970, an MAE of 8.3908, and an R2 of 0.9227, reducing MSE and MAE by 4.27% and 28.42%, respectively, compared with ML-RFR. It also achieves the highest R2 of 0.9369 on FD003. Cross-dataset and multi-seed experiments further confirm its applicability and stability.
