Detecting bearing defects accurately and efficiently is critical for industrial safety and efficiency. This paper introduces Bearing-DETR, a deep learning model optimised using the Real-Time Detection Transformer (RT-DETR) architecture. Enhanced with Dysample Dynamic Upsampling, Efficient Model Optimization (EMO) with Meta-Mobile Blocks (MMB), and Deformable Large Kernel Attention (D-LKA), Bearing-DETR offers significant improvements in defect detection while maintaining a lightweight framework suitable for low-resource devices. Validated on a dataset from a chemical plant, Bearing-DETR outperformed the standard RT-DETR, achieving a mean average precision (mAP) of 94.3% at IoU = 0.5 and 57.5% at IoU = 0.5–0.95. It also reduced floating-point operations (FLOPs) to 8.2 G and parameters to 3.2 M, underscoring its enhanced efficiency and reduced computational demands. These results demonstrate the potential of Bearing-DETR to transform maintenance strategies and quality control across manufacturing environments, emphasising adaptability and impact on sustainability and operational costs.

In industrial manufacturing, bearings are crucial for machinery stability and safety. Undetected wear or cracks can lead to severe operational and financial setbacks. Thus, accurately identifying bearing defects is essential for maintaining production safety and equipment reliability. This research introduces an improved bearing defect detection model, YOLOv8-LMG, which is based on the YOLOv8n framework and incorporates four innovative technologies: the VanillaNet backbone network, the Lion optimizer, the CFP-EVC module, and the Shape-IoU loss function. These enhancements significantly increase detection efficiency and accuracy. YOLOv8-LMG achieves a mAP@0.5 of 86.5% and a mAP@0.5–0.95 of 57.0% on the test dataset, surpassing the original YOLOv8n model while maintaining low computational complexity. Experimental results reveal that the YOLOv8-LMG model boosts accuracy and efficiency in bearing defect detection, showcasing its significant potential and practical value in advancing industrial inspection technologies.

With the development of IoT and 5G technologies, edge computing has become a key driver for providing compute, network and storage services. The dramatic increase in data size and the complexity of AI computation models have put higher demands on the performance of edge computing. Rational and optimal scheduling of AI data-intensive computation tasks can greatly improve the overall performance of edge computing. To this end, a particle swarm algorithm based on objective ranking is proposed to optimize task execution time and scheduling cost by designing a task scheduling model to achieve task scheduling in an edge computing environment. It is necessary to fully understand the concept of symmetry of resource utilization and task execution cost indicators. The method utilizes nonlinear inertia weights and shrinkage factor update mechanisms to improve the optimization-seeking ability and convergence speed of the particle-to-task scheduling solution space. The task execution time and scheduling cost are greatly reduced. Simulation experiments are conducted using the Cloudsim toolkit to experimentally compare the proposed algorithm TS-MOPSO with three other particle swarm improvement algorithms, and the experimental results show that the task execution time, maximum completion time and total task scheduling cost are reduced by 31.6%, 23.1% and 16.6%, respectively. The method is suitable for handling large and complex AI data-intensive task scheduling optimization efforts.

In this paper, a class of predator-prey systems with fear effect is investigated, the integrability conditions of the origin and the positive equilibrium are obtained, and the fact that three limit cycles can be bifurcated from the positive equilibrium is proved, so bistable phenomenon can occur for this system.