Burden distribution is one of the most important operations, and also an important upper regulation in blast furnace ( BF ) iron-making process. Burden distribution output behaviors ( BDOB ) at the throat of BF is a 3-dimensional spatial distribution produced by burden distribution matrix ( BDM ), including burden surface output shape ( BSOS ) and material layer initial thickness distribution ( MLITD ). Due to the lack of effective model to describe the complex input-output relations, BDM optimization and adjustment is carried out by experienced foremen. Focusing on this practical challenge, this work studies complex burden distribution input-output relations, and gives a description of expected MLITD under specific integral constraint on the basis of engineering practice. Furthermore, according to the decision variables in different number fields, this work studies optimization of BDM with expected MLITD, and proposes a multi-mode based particle swarm optimization ( PSO ) procedure for optimization of decision variables. Finally, experiments using industrial data show that the proposed model is effective, and optimized BDM calculated by this multi-model based PSO method can be used for expected distribution tracking.

In the context of resource constrained embedded terminals, the current deep learning-based technology for detecting surface defects in strip steel faces challenges with low detection accuracy and efficiency. This study introduces a lightweight convolution neural network (CNN) model for detecting surface defects of strip steel, which is based on deep learning. Through the integration of non-parameter attention mechanism SimAM, inverse residual structure, and depth-wise convolution module, a lightweight attention mobile backbone network is developed to achieve optimal feature extraction. To address the challenge of detecting small surface defects the context augmentation module is introduced to provide more information for small defects detection by using multi-scale features. To improve the efficiency of feature fusion and reduce parameter redundancy, the GVFPN neck network is proposed. The network aims to represent and deal with multi-scale features objectively while minimizing costs. The experimental results indicate that on the NEU-DET dataset, the proposed network attains an mAP of 78.8% and FPS at 97 frame/s. Moreover, the model requires only 2.39 M parameters and 3.1 G FLOPs. Compared to YOLOv5s, the proposed network reduces parameters by 65.9% and FLOPs by 80.7%, while achieving a 1.3% higher mAP and a 30 frame/s increase in detection speed. These results effectively demonstrate the effectiveness and superiority of the proposed method.