With a significant number of mechanical products approaching the retirement phase, the batch recycling of discarded mechanical parts necessitates a preliminary assessment of their surface condition. ...However, the presence of surface rust poses a challenge to defect identification. Therefore, this paper proposes a method for detecting heavily rusted surface defects based on an improved YOLOv8n network. In the Backbone, the C2f-DBB module of re-parameterized deep feature extraction was introduced, and the attention module was designed to improve the accuracy of information extraction. In the Neck part, a Bi-Afpn multiscale feature fusion strategy is designed to facilitate information exchange between features at different scales. Finally, Focal-CIoU is employed as the bounding box loss function to enhance the network's localization performance and accuracy for defects. Experimentally, it is proved that the improved network in this paper improves the Recall, Precision, and mAP0.5 by 1.2%, 2.1%, and 1.9%, respectively, on the original basis, which is better than other network models.
Deep learning is an effective way to improve the classification accuracy of coal images for the machine vision-based coal sorting. However, the related research on deep learning-based mineral image ...classification has not systematically considered the models for multi-coal and multi-class sorting. Additionally, the universal CNNs model for multi-coal image classification has not been proposed. Given the above problems, combined with deep learning and transfer learning and based on VGG Net, Inception Net, and Res Net, this study builds four CNNs models with different depth and structure for multi-coal and multi-class image classification. Finally, we take anthracite, gas coal, coking coal as the research objects and propose a universal CNNs model suitable for multi-coal and multi-class sorting. Moreover, with the Channel Visualization map, Heatmap, Gard-CAM map, and Guided Backpropagation map, the operational processes of CNNs model in coal image recognition and classification are revealed, and the features that affect the classification weights are analyzed.
●Establishing multi-class image classification models of multi-coal by deep learning●Developing a universal CNNs classification model for multi-coal images●Exploring the operational process of CNNs model in coal image classification
Display omitted
•Present CNN model for various wet ore with different water content.•Analyze the classification performance for different wet ore.•Explore the operational process of CNN model for wet ...ore classification.
Moisture is one of the important influencing factors on machine vision-based mineral image classification, and it has different effects on various ore particles. At present, deep learning is an effective measure to improve classification accuracy, but the effects of moisture have not been systematically investigated. Therefore, this paper establishes deep learning-based RGB image classification models for the classification tasks of various coal particles with two density level (<1.8 g/cm3 & >1.8 g/cm3) in different water gradients, and analyzes their classification performance. Moreover, the model operational process and the change of classification weight and accuracy under different water gradients are investigated through Channel Visualization, Heatmap, Guided Backpropagation, Grad-CAM.
The failure of bearings can have a significant negative impact on the safe operation of equipment. Recently, deep learning has become one of the focuses of RUL prediction due to its potent ...scalability and nonlinear fitting ability. The supervised learning process in deep learning requires a significant quantity of labeled data, but data labeling can be expensive and time-consuming. Cotraining is a semisupervised learning method that reduces the quantity of required labeled data through exploiting available unlabeled data in supervised learning to boost accuracy. This paper innovatively proposes a cotraining-based approach for RUL prediction. A CNN and an LSTM were cotrained on large amounts of unlabeled data to obtain a health indicator (HI), then the monitoring data were entered into the HI and the RUL prediction was realized. The effectiveness of the proposed approach was compared and analyzed against individual CNN and LSTM and the stacking networks SAE+LSTM and CNN+LSTM in the existing literature using RMSE and MAPE values on a PHM 2012 dataset. The results demonstrate that the RMSE and MAPE value of the proposed approach are superior to individual CNN and LSTM, and the RMSE value of the proposed approach is 54.72, which is significantly lower than SAE+LSTM (137.12), and close to CNN+LSTM (49.36). The proposed approach has also been tested successfully on a real-world task and thus has strong application value.
Large-scale and high-efficiency disassembly is an important means to improve the efficiency and benefits of remanufacturing of waste products. This paper proposes a 3D-based multi-objective ...collaborative disassembly sequence planning method for remanufacturing products, which can effectively improve efficiency of the disassembly line. Firstly, combining entropy weight method and PROMETHEE II method, the priority disassembly level of the waste product parts is established. According to the disassembly level ranking of the parts, one or more target disassembly parts with higher priority are selected. Secondly, rapid-growing random tree algorithm is used to layer the three-dimensional product assembly structure, and the disassembly sequence of the single-objective part is calculated by the constraint relationship of different layer parts. Finally, the co-evolution algorithm is used to coordinate multiple single-objective part disassembly sequences to obtain the overall optimal multi-objective part disassembly sequence, which is the sequence plan with the minimum required number of disassembly parts. Through the disassembly case analysis of two-stage reducer, the correctness and effectiveness of the planning method for coordinated disassembly of multi-objective parts are verified.
Recycling and remanufacturing logistics network affects the scale and efficiency of sustainable development of the manufacturing industry. This paper designs a multi-level closed-loop supply chain ...network with supplier, manufacturer, recycling centers, preprocessing centers and processing centers. An improved nonlinear grey Bernoulli-Markov model is developed to predict the recycled quantity. The capacity of recycling center and preprocessing center, the demand of manufacturer and the inventory of preprocessing center are formulated as constraints. A dynamic multi-objective model, which is based on scalable logistics facilities, takes into account the minimization of system operating costs and minimization of time costs related to the out-of-stock and inventory in each operating cycle. This model realizes the dynamic selection of the scale of facilities. Objective weighted genetic algorithm is adopted to transform multi-objective optimization problem into a single-objective. A scrap automobile products calculations are analyzed to verify the effectiveness and practicability of this model.
Mineral image classification technology based on machine vision is an efficient system for ore sorting. With the development of artificial intelligence and computer technology, the deep ...learning-based mineral image classification system is gradually applied to ore sorting. However, there is a bottleneck in improving classification accuracy, and the feature extraction ability of the CNNs model is relatively limited for multi-category mineral image classification tasks. Therefore, four visual attention blocks are designed and embedded in the existing CNNs model, and new mineral image classification models based on the visual attention mechanism and CNNs are proposed. Then, referring to the building strategies of the different depth ResNet, we build various CNNs model embedding with attention blocks for mineral image classification and visualize the models by Grad-CAM to observe the change in classification weight distributions and classification weight values. Finally, by using the confusion matrices, this experiment systematically evaluates the classification performance of the proposed models and analyzes the misjudgment rate.
Deep neural networks (DNNs) require large amounts of labeled data for model training. However, label noise is a common problem in datasets due to the difficulty of classification and high cost of ...labeling processes. Introducing the concepts of curriculum learning and progressive learning, this paper presents a novel solution that is able to handle massive noisy labels and improve model generalization ability. It proposes a new network model training strategy that considers mislabeled samples directly in the network training process. The new learning curriculum is designed to measures the complexity of the data with their distribution density in a feature space. The sample data in each category are then divided into easy-to-classify (clean samples), relatively easy-to-classify, and hard-to-classify (noisy samples) subsets according to the smallest intra-class local density with each cluster. On this basis, DNNs are trained progressively in three stages, from easy to hard, i.e., from clean to noisy samples. The experimental results demonstrate that the accuracy of image classification can be improved through data augmentation, and the classification accuracy of the proposed method is clearly higher than that of standard Inception_v2 for the NEU dataset after data augmentation, when the proportion of noisy labels in the training set does not exceed 60%. With 50% noisy labels in the training set, the classification accuracy of the proposed method outperformed recent state-of-the-art label noise learning methods, CleanNet and MentorNet. The proposed method also performed well in practical applications, where the number of noisy labels was uncertain and unevenly distributed. In this case, the proposed method not only can alleviate the adverse effects of noisy labels, but it can also improve the generalization ability of standard deep networks and their overall capability.
Remanufacturing has become an important way to realize sustainable development strategy. For remanufacturing closed-loop supply chain under different circumstances, many factors are considered, such ...as consumers’ different preferences for the purchase and payment of remanufactured products and the quality of recycled products. In this study, three models are presented for supply chain system, including a manufacturer lead and a retailer recycle. While nongovernment participation and non-supply-chain coordination are considered in model I, model II has government participation but non-supply-chain coordination, and model III has both government participation and supply chain coordination. The equilibrium decisions are derived for each model using the game-theoretic approach. The study found that consumers’ preferences affect the pricing decisions of remanufactured products. When consumers’ willingness to buy and pay for remanufactured products is close to that of new products, the total profit of the supply chain will significantly increase; the quality of recycled products will also influence the pricing decisions of new/remanufactured products, which in turn will affect the market demand for products. Government participation is conducive to encouraging manufacturer, retailer, and consumers to actively participate in the recycling and remanufacturing activities of used products. Supply chain coordination contracts can further enhance the role of the consumer market and promote the implementation of government policies. Under certain conditions, revenue sharing contract coordination can significantly increase the total profit of the supply chain system. The research conclusions have certain reference value for the coordination among the nodes of the remanufacturing supply chain and the formulation of policies with government participation.
Nowadays, environmental problems have gradually become the focus of world attention. In recent years, heat waves in many parts of Europe have increased ozone concentrations, fuelling ozone pollution. ...Therefore, this paper investigates the spatial and temporal distribution of tropospheric column ozone concentrations in South-western Europe, future trend changes, influencing factors, and potential source regions based on remotely sensed monitoring data from the OMI (Ozone Monitoring Instrument) from 2011 to 2021. The results show that the areas of high tropospheric column ozone concentrations are mainly concentrated in the northwest, Poland, and southeast coastal areas. At the same time, the monthly variation curve of column ozone concentration is bimodal. Trend change analyses indicate an upward trend in future column ozone concentrations in the southeastern part of the study area. The potential for increases also exists in parts of Germany, France, and Poland, which will need to be monitored. Random forest model projections found a slight decrease in column ozone concentrations in 2022 and 2023 of about 1–4 DU compared to tropospheric column ozone concentrations in 2021. The health risk assessment found that the number of all-cause premature deaths due to exposure to ozone was the highest in Germany. During the summer, when ozone pollution is high, the potential source area in the southeastern part of the study area is located at the border of the three countries, and synergistic management is recommended. In exploring the correlation between the influencing factors and ozone, it was found that there is a significant difference between the long-time and short-time series. In addition, the pathway analysis shows that the population size, distribution density, and forested area in southwestern Europe may be more sensitive to the production of tropospheric ozone.