Current industrial image anomaly detection methods generally face challenges such as high dependency on anomalous samples, insufficient realism of synthetic samples, and limited perception capability for complex defects. To An industrial image anomaly detection method based on diffusion synthesis and feature mining was proposed to address these issues. Accordingly, a category-sensitive selective diffusion anomaly synthesis module was designed to generate pseudo-anomaly samples with adjustable intensity and category adaptability through controllable perturbation and category-sensitive loss, which could effectively alleviate data scarcity. Meanwhile, a multi-stage feature mining framework was constructed, including contrast-driven feature selection, multi-dimensional perception attention reconstruction, and residual refinement selection modules, enabling dynamic screening of anomaly-sensitive features and enhancement of structural details. Experimental results demonstrated that the proposed method achieved outstanding performance on the MVTec AD and MPDD datasets, with image-level AUROC scores of 99.7% and 98.4%, and pixel-level AUROC scores of 99.0% and 98.7%, respectively, validating its effectiveness and robustness.

Discrete element method (DEM) and Tavares breakage model were employed to investigate the rock fragmentation mechanism and the statistical characteristics of fragments under impact loading. Drop hammer impact crushing process of limestone specimens with different aspect ratios (1.0—2.0) and sizes (diameter 30—50 mm) was systematically simulated. Image J image processing technology was used to quantitatively analyze the fragment count, particle size distribution, and shape indices. The results showed that increasing the aspect ratio and the size significantly raised the crushing energy threshold, which led to a delay in the peak normal force, a reduction in its amplitude, and a slowdown of the overall fragmentation process. The equivalent particle size of fragments showed poor alignment with the sieve analysis curve, which indicated that the fragments commonly exhibited irregular geometric configurations such as flaky and rod-like shapes. Fragments from specimens with lower aspect ratios (1.0) and smaller sizes (diameter 30 mm) had lower roundness, more pronounced angularity, and greater surface roughness. A quantitative relationship model among geometric parameters–crushing response–fragment morphology can provide theoretical basis and regulatory pathways for the design of impact-resistant structures, the optimization of blasting parameters, and the control of particle morphology in rockfill grading.

To address the issue of inconsistent thermal stress between the upper and lower devices of the same bridge arm in converters under conditions of uneven aging or layout deviations, a novel thermal balancing method for upper and lower bridge arms based on dead-time shifting adjustment was proposed. By monitoring the on-state voltage of insulated gate bipolar transistor (IGBT) online, the real-time junction temperature of IGBTs was deduced based on the mapping relationship between the on-state voltage and junction temperature of IGBTs. Then, the temperature difference between the upper and lower devices of the same bridge arm was determined. When the thermal stress difference between the upper and lower bridge arms exceeded the threshold, the effective conduction time of the upper and lower bridge arms was adjusted by fine-tuning the position of the dead time, thereby actively regulating the power loss distribution of the upper and lower bridge arms. This achieved an increase in the loss of the bridge arm with lower temperature and a decrease in the loss of the bridge arm with higher temperature without affecting the total system loss, thus improving the thermal stress consistency of power devices in the converter and enhancing system reliability. Theoretical analysis and experimental results showed that the proposed method did not affect system efficiency and had no significant negative impact on output performance. Under typical operating conditions, the average temperature difference between the upper and lower devices was reduced by approximately 14% after adjustment, effectively improving the thermal stress consistency of the upper and lower bridge arms.

Significant progress in multimodal large language models (MLLMs) has driven advances in visual question answering, visual understanding, and reasoning tasks, and their potential for deployment on resource-constrained edge devices is increasingly recognized. However, large model sizes and the substantial costs of deployment and inference remain major barriers to practical adoption. Optimizing MLLMs for edge devices has become a critical research direction in this field. A comprehensive survey of recent advances in optimizing MLLMs for edge deployment was presented, along with the associated challenges and development trends. The research evolution of MLLMs on edge devices was reviewed, with particular emphasis on model architecture optimization and inference scheduling strategies. In model architecture optimization, techniques including visual information compression, sparse attention, and mixture-of-experts models were specifically analyzed. System-level optimizations involving computation scheduling, hardware adaptation, compilation optimization, and cloud-edge collaboration were investigated to enhance inference efficiency and energy efficiency. Furthermore, the key challenges of these models in practical applications were discussed, and a variety of task scenarios ranging from assistive to collaborative and autonomous types were covered, categorized by the perspective of autonomy levels. Finally, current limitations were summarized and future research directions regarding standardized deployment, efficient computing and storage, and multi-modal fusion optimization were outlined.

A lightweight micro-expression recognition method based on optical flow and convolutional vision Transformer was proposed to solve the problems of short duration, low motion intensity and insufficient sample size of micro-expressions. The optical flow and optical strain of human faces between the onset frame and the apex frame were extracted to highlight the movement of facial muscles, thereby effectively reducing the texture interference and lowering the feature dimension. The adversarial domain adaptation method based on identity domain was adopted to further remove the irrelevant components in the micro-expression features by making full use of the subjects’ labels. A lightweight multi-stage CNN-Transformer hybrid model named MiER-CvT, including the convolutional embedding layer, the convolutional Transformer block and the SeqSoftmax layer, was constructed to enhance the model’s capabilities of local representation and information integration for micro-expressions. The experimental results showed that the proposed method achieved a UF1 score of 0.9171 and a UAR score of 0.9192 on the MEGC 2019 dataset, and the parameter number and computational complexity of MiER-CvT were 7.5 M and 0.1 G, respectively. Compared with the existing methods, such as MiMaNet, the proposed method has the advantages of high precision and light weight.

A novel detection framework integrating automatic fluid target annotation and a lightweight YOLOv8n model was proposed, aiming at addressing the challenges of low annotation efficiency and insufficient model adaptability in real-time debris flow monitoring. An open-source debris flow dataset of 8 064 images was built. This dataset was achieved using multi-scale dynamic capture frames, which leverage fluid motion continuity, along with adaptive frame extraction and an improved binary classification model. To enhance the extraction of dynamic debris flow features and improve the irregular boundary localization, the LGSA module and C2f_GhostNetV2 structure were developed. The Shape-IoU loss function was also introduced into the model architecture. Experimental results demonstrated that the proposed YOLOv8-Mudslide model achieved an mAP@0.5 of 86.7%, which was 4.7% higher than that of the baseline model, and the detection speed reached 230.89 frame/s. This case provides reliable technical support for the real-time monitoring of debris flow disasters, and its framework can be further extended to other intelligent detection fields of fluid targets.

A collaborative control approach integrating multi-agent proximal policy optimization (MAPPO) and maximum pressure control (MPC) was developed to address traffic control challenges at mixed traffic intersections involving connected autonomous vehicles (CAVs), human-driven vehicles (HDVs), and pedestrians. A hierarchical state space and an action space were designed through the formulation of a decentralized partially observable Markov decision process (Dec-POMDP). Reward functions considering the balance among safety, traffic efficiency, phase switching frequency, and regulatory compliance were introduced. The proposed model was validated under low, medium, and high traffic flow conditions using a centralized training with decentralized execution framework on the SUMO simulation platform. Experimental results demonstrated that the proposed MAPPO-MPC method significantly improved the throughput (by up to 44.3%) while reducing the queue length (by up to 49.3%), average vehicle delay (by up to 43.2%), and pedestrian waiting time (by up to 43.53%). Moreover, the model exhibited more substantial performance advantages when the CAV penetration rate exceeded 50%, outperforming the traditional Webster-based methods and the baseline models.

To address the challenge of difficult identification of concrete cracks in extremely dark environments, a concrete crack recognition method based on biomimetic tactile technology was proposed. By simulating the tactile perception mechanism of organisms, a tactile sensing system and a deep learning model were constructed to achieve crack detection in dark environments. A high-sensitivity silicone tactile probe was designed and combined with a multi-light source imaging system and the Lambertian reflection model, and the three-dimensional texture of cracks was reconstructed through elastomer deformation and light-shadow gradient changes, overcoming the reliance on visual information in dark environments. On this basis, a tactile segmentation network for concrete cracks (TSCC-Net) was constructed. Through the collaboration of a semantic perception extractor, mutual fusion module, and auxiliary supervision module, precise segmentation of cracks in tactile images was achieved. Experimental results showed that TSCC-Net achieved an IOU of 86.3% in tactile image crack recognition, with a model parameter size of 12.86 MB and an inference speed of 114.5 fps, significantly outperforming traditional visual models. In particular, TSCC-Net demonstrated stronger robustness in narrow cracks and areas with complex textures compared to models such as U-Net, Segformer, and DeepLabV3+.

The Electric_Data inspection dataset was constructed, and a PDD-SPCIE algorithm based on context information enhancement and multi-target semantic perception was proposed within the RT-DETR framework, in order to improve the detection accuracy of multi-category power defects in complex substation inspection scenarios. The LDDM-ResNet module was designed to extract high- and low-frequency detailed features through multi-kernel parallel convolution. The DA-Encoder module was constructed to enhance the representation of multi-scale and irregularly shaped targets by incorporating deformable attention and dynamic sampling. The LGCA module was introduced to fuse global and local features using atrous convolution and spatial depth transformation convolution. The original framework was improved from the aspects of feature representation, context modeling, and feature fusion, thus enhancing the unified detection ability for multiple defect categories. Experimental results showed that the proposed model achieved a precision of 0.819, a recall of 0.681, and an mAP₅₀ of 0.704 on the self-built dataset, and a precision of 0.719, a recall of 0.642, and an mAP₅₀ of 0.712 on the VOC dataset.

A knowledge graph-guided 3D visual question answering method was proposed to capture the implicit common-sense semantic relationships between objects in the scene. By introducing external structured knowledge, the model was effectively enhanced in both semantic understanding and reasoning ability. Specifically, key semantic entities were extracted from the question text, and a knowledge graph-guided feature enhancement module was designed to obtain knowledge features using these key semantic entities. The knowledge features were fused with visual features extracted both from the question representation and from the 3D object detection network for answer prediction. Experimental results on the ScanQA dataset showed that the proposed method outperforms existing baseline models on metrics such as EM@1 and BLEU-4.

A reparameterization-based RUL prediction method that integrated channel decomposition and feature fusion was proposed to address the challenges in remaining useful life (RUL) prediction for aero-engines. Specifically, a channel decomposition and reconstruction module based on wavelet transform was designed to capture dependencies between time and channels while reducing redundancy. By fusing the raw input with the extracted feature information, the model enhanced the preservation of original information and mitigated the loss of critical information during deep network training. Furthermore, a frequency-enhanced channel attention mechanism was introduced to further suppress channel redundancy. A neural network was then employed to estimate the mean and variance of the features, and reparameterization was applied to map them into a Gaussian distribution for sampling-based prediction. In addition, the Whale optimization algorithm (WOA) was used to perform global optimization of key hyperparameters. Experimental validation on the C-MAPSS dataset from NASA demonstrated that the proposed method achieved an RMSE range of [11.58,13.72] and a Score range of [224.78,1025.29], with average decreases of 22.67% and 45.56%, respectively. Moreover, the constructed three-dimensional latent space supported the visualization of engine degradation processes.

To achieve fault type judgment, fault line localization, and fault distance judgment in the grid, a novel hybrid network model integrating fast Fourier transform (FFT), convolutional neural networks (CNN), and graph convolutional networks (GCN) was proposed for grid fault diagnosis. Voltage and current signals were decomposed in time and frequency domains by FFT to extract fundamental waveform amplitude and phase. CNN extracted the temporal features of the decomposed data, and layer normalization was introduced to enhance the model stability. The spatial topology of the grid was processed with GCN to extract and integrate spatial features. The model’s effectiveness was verified through modeling and simulation of the IEEE 39-bus power grid system. Experimental results show that the proposed model possesses strong generalization capabilities, and the fault diagnosis accuracy under various tasks, sampling intervals, and noise conditions outperforms existing models.

An improved lightweight small object detection method based on YOLOv8s was proposed to address the challenges of complex background interference, small object sizes, and limited device resources in unmanned aerial vehicle (UAV) image object detection. A high-resolution detection head was reconstructed to improve the sensitivity to tiny objects. A dynamic multi-scale aggregation network was designed to adaptively adjust the receptive field for scale-variant targets, while an efficient multi-scale feature deep fusion module was proposed to integrate features from different semantic levels. An efficient channel-spatial attention module was introduced to further refine fused features. Experiments on the VisDrone2021 dataset demonstrated that the proposed method achieved improvements of over 5.0% in average precision, precision at different intersection over union thresholds, and detection precision for small and medium-sized objects, with a 16.25% reduction in parameters. Tests on the DOTA dataset validated the generalization ability of the proposed model. Results demonstrate that the proposed method improves the detection performance while achieving the model light weighting, indicating its practical value for object detection in UAV aerial images.

A railway foreign object detection algorithm for complex weather conditions based on an improved RT-DETR, named FRP-DETR, was proposed. A feature complementary mapping module (FCM) and Pzconv units were introduced, which built complementary paths between shallow spatial details and deep semantic information to compensate for the limitations of single-scale feature representation, with the two components working collaboratively to enhance the perception capabilities of small and edge targets. A railway perception modulation fusion module (RMFM) was introduced, which was based on adaptive channel attention and spatial modulation mechanisms to enhance the model’s response to key semantic information in railway scenes. The original downsampling module was replaced with pinwheel-shaped convolution (PSConv), which enhanced edge texture extraction through multi-directional asymmetric padding and separable convolution. A railway foreign object detection dataset containing four weather conditions was constructed based on the SaMam style transfer method, by transferring sunny railway foreign object images to rainy, foggy, and snowy scenes. Experimental results showed that compared to the original RT-DETR-R18 model, this method achieved improvements of 1.65 percentage points and 3.4 percentage points in mAP@0.5 and mAP@0.5:0.95, respectively, with a 63.5% reduction in parameter count and an inference speed of 88 frames per second. The results verified that FRP-DETR achieved high accuracy, lightweight design, and real-time performance in railway foreign object detection under complex weather conditions.

A multi-agent graph convolutional reinforcement learning based on hybrid dilated attention was proposed to address the challenges of capturing diverse dynamic information and computational complexity in complex high-dynamic multi-agent reinforcement learning. The multi-agent graph convolutional reinforcement learning with ensemble-based hybrid dilated attention was proposed to balance computational complexity and receptive field size in order to enhance learning efficiency. The dynamics matrix of the underlying graph was obtained to capture the dynamic interactions of multiple agents. The hybrid dilated attention mechanism used the Bagging method to combine the advantages of the multi-head attention and the multi-scale dilated attention mechanisms in order to enable the model to effectively learn different types of features. A regularization was added into the loss function to enhance the stability of the model and address the overfitting problem. The improved temporally extended greedy method was utilized to choose the multi-agent’s actions in order to achieve the exploration and exploitation trade-off of actions. The simulation results demonstrate that the proposed method achieves higher accuracy and greater stability in policy optimization than the existing approaches in complex dynamic environments.

A hypersonic vehicle, where the injection surface was reasonably partitioned, was analyzed in order to enhance the cost-effectiveness of wall mass injection for high-speed vehicles. Numerical simulations were conducted to get sample data in order to construct surrogate models. A multi-objective optimization of the injection flow rates of different surface partitions was conducted based on surrogate models by using non-dominated sorting genetic algorithm II (NSGA-II), targeting the reduction of total heat flux and drag. The uniform mass injection scheme reduced total drag by 14.05%, friction drag by 38%, total heat flux by 35.92%, and peak heat flux by 1.38% compared with the no mass injection case under the conditions of 50 km altitude, Mach 15 freestream, 5° angle of attack, and a total air injection flow rate of 50 g/s. The optimized mass injection case reduced total drag by 22.56%, friction drag by 53.96%, total heat flux by 53.40%, and peak heat flux by 30.77%. Results indicate that the optimized mass injection case injects more cooling medium on regions with high aerodynamic and thermal loads, such as the nose and leading edges of the wings compared with the uniform mass injection case. This optimized case leverages the attachment and cumulative effects of the injected mass flow along the high-speed main flow, improves the distribution of surface film thickness and significantly enhances both the drag and heat reduction effects and efficiency of the overall and the critical local location, while maintaining a constant total injection flow rate.

In response to the challenges of high difficulty in modeling and the low control efficiency in multivariable nonlinear complex systems, fuzzy control has been identified as a promising research approach. Significant progress has been made over the years. Advancements in fuzzy control theory for these systems have been reviewed. Firstly, relevant knowledge of fuzzy theory and the basic components of fuzzy controllers were presented. Secondly, various fuzzy models and control methods for multivariable nonlinear complex systems were discussed, including fuzzy control systems based on semantic rules, model-based fuzzy control systems, and others. Special emphasis was placed on model-based fuzzy control systems, such as the T-S fuzzy model, fuzzy cell-to-cell mapping model, fuzzy neural network model, and fuzzy models based on the variation relationship between process input and output data. Detailed summaries and analyses of both model structure identification and parameter identification were provided. Finally, challenges related to the universality, interpretability, stability, and applicability of these systems were explored, along with potential directions for future research and development.

The development trajectory of coupled heat and moisture transfer (HAMT) models was elucidated, and the applicability and limitation of different formulation were compared in order to address the complexity of HAMT mechanism in building envelope, as well as the diversity of modeling approaches and experimental methods with markedly different levels of engineering applicability. Typical experimental methods and property measurement techniques were summarized, and their differences in reproducibility and controllability were analyzed. Engineering applications of HAMT were synthesized, with particular attention given to the specific challenges of the regions characterized by high humidity, abundant rainfall, and intermittent operation patterns. Existing research gaps were identified and future research directions were proposed in light of the ongoing enhancement of building envelope performance and recent advance in artificial intelligence methods, with the aim of providing systematic reference and methodological insight for subsequent study and the transition towards net-zero energy building.

An improved ship target detection algorithm DD-YOLO based on YOLOv8 was proposed in order to address the challenges of small target size, significant scale variations and complex background noise in synthetic aperture radar (SAR) images for ship detection. An enhanced C2f module was incorporated in the backbone network to strengthen multi-scale feature extraction and fusion capabilities, combined with a newly designed SPA module to optimize gradient flow information propagation, significantly improving multi-scale target detection performance. A more lightweight dynamic upsampling approach was adopted in the neck network, which reduced computational overhead and model complexity while enhancing the recognition of small ships in complex backgrounds. A multi-dimensional attention mechanism was integrated in the detection head and lightweight processing was conducted to improve the model’s sensitivity to key features in complex backgrounds, thereby increasing detection accuracy. Experiments conducted on two public datasets, HRSID and SSDD, demonstrate that DD-YOLO achieves mAP₅₀ scores of 92.2% and 98.5%, respectively, representing improvement of 2% and 2.2% over the baseline model. The model complexity is significantly lower than that of mainstream algorithms, achieving an optimal balance between accuracy and efficiency.

An aerial small target detection algorithm that balanced performance and resource consumption was proposed to address the issues of low detection accuracy and large model parameter size in small target detection of aerial images. On the basis of YOLOv8s, an adaptive detail-enhanced module (ADEM) was proposed by reducing the channel dimension and enhancing the focus on the high-frequency features to capture the fine-grained features of small targets while discarding the redundant information. A feature fusion network was optimized based on the PAN-FPN architecture to enhance the attention on shallow features. Multi-scale convolutional kernels were introduced to enhance the focus on the target contextual information, thereby adapting to the small object detection scenario. A parameter-adjustable Nin-IoU was constructed to overcome the limitations of traditional IoU in flexibility and generalization, and this adjustment achieved by introducing adjustable parameters allowed the Nin-IoU to be tailored to different detection tasks. A lightweight detection head was proposed to enhance the integration of multi-scale feature information while reducing redundant information transmission. Experimental results on the VisDrone2019 dataset indicated that the proposed algorithm achieved an mAP_0.5 of 50.3% with only 8.08×10⁶ parameters, representing a 27.4% reduction in parameters and an improvement of 11.5 percentage points in accuracy compared to the YOLOv8s benchmark algorithm. Experimental results on the DOTA and DIOR datasets further demonstrated the strong generalization capabilities of the proposed algorithm.