Deep learning methods have demonstrated impressive effectiveness in diagnosing bearing faults. Nevertheless, their performance tends to decline significantly in the presence of long-tailed distributions, where certain fault types are underrepresented, which is common in industrial environments where fault occurrences are rare. To address the challenges posed by long-tailed data in fault diagnosis, this paper proposes the Probabilistic Contrastive Quadratic Network (PCQNet), a novel framework designed to enhance the diagnostic capabilities of neural networks under class imbalance conditions. PCQNet consists of three core components: a quadratic convolutional residual network that strengthens feature representations through higher-order interactions; a Probabilistic Contrastive Loss (PCL) that models class features using the von Mises-Fisher (vMF) distribution, combining class expectations and prior probabilities to adaptively balance the contributions of each class, effectively mitigating long-tail bias; and a logit-adjusted cross-entropy loss that improves the discrimination ability for minority classes while enhancing classification fairness. The parabolic decay formula dynamically adjusts the dominance of these two loss functions during training, ensuring the model achieves both effective feature learning and accurate classification.The effectiveness of PCQNet is validated on one public dataset and two self-constructed datasets, demonstrating its superior performance in handling long-tailed data. Analysis of the attention mechanism derived from quadratic neurons further explains the excellent feature extraction capability of the quadratic network. Building on this analysis, ablation studies further verify the independent contributions and synergistic effects of each module, confirming their critical roles in improving diagnostic performance.

To address the issues of insufficient rigor in existing methods for quantifying apple watercore severity and the complexity and low accuracy of traditional classification models, this study proposes a method for watercore quantification and a classification model based on a deep convolutional neural network. Initially, visible/near-infrared transmission spectral data of apple samples were collected. The apples were then sliced into 4.5 mm thick sections using a specialized tool, and image data of each slice were captured. Using BiSeNet and RIFE algorithms, a three-dimensional model of the watercore regions was constructed from the apple slices to calculate the watercore severity, which was subsequently categorized into five distinct levels. Next, methods such as the Gramian Angular Summation Field (GASF), Gram Angular Difference Field (GADF), and Markov Transition Field (MTF) were applied to transform the one-dimensional spectral data into two-dimensional images. These images served as input for training and prediction using the ConvNeXt deep convolutional neural network. The results indicated that the GADF method yielded the best performance, achieving a test set accuracy of 98.73%. Furthermore, the study contrasted the classification and prediction of watercore apples using traditional methods with the existing quantification approaches for watercore levels. The comparative results demonstrated that the proposed GADF-ConvNeXt model is more straightforward and efficient, achieving superior performance in classifying watercore grades. Furthermore, the newly proposed quantification method for watercore levels proved to be more effective.

To address the challenge of nondestructively identifying watercore disease in apples during growth and maturation, a portable device was developed for real-time grading of apple watercore using visible/near-infrared (Vis/NIR) spectroscopy combined with a one-dimensional quadratic convolutional neural network (1DQCNN). The instrument enables rapid, nondestructive, and accurate detection of apple watercore grades. The AI-OX2000-13 micro-spectrometer is used as the core data acquisition unit, and an ARM processing system is built with the STM32F103VET6 as the main control chip. A 4G wireless communication module enables efficient and stable data transmission between the processor and computer, meeting the real-time detection needs of apple watercore content in orchard environments. To improve the scientific and accurate classification of watercore grades, this paper combines the BiSeNet and RIFE algorithms to construct a 3D model of apple watercore, allowing quantification of the degree of watercore and classification into four levels. Based on this, quadratic convolution operations are incorporated into a one-dimensional convolutional neural network (1DCNN), leading to the development of the 1D quadratic convolutional neural network (1DQCNN) model for watercore grade classification. Experimental results indicate that the model achieves a classification accuracy of 98.05%, outperforming traditional methods and conventional CNN models. The designed portable instrument demonstrates excellent accuracy and practicality in real-world applications.

Long-tailed data are very common in industrial scenarios because equipment failures occur with a low probability, resulting in far fewer faulty samples than normal ones. However, when facing long-tailed data distributions, existing deep learning methods suffer from a significant degradation in performance and exhibit high bias. To overcome this limitation, this paper proposes a network that combines balanced adaptive logit-compensated cross-entropy loss with quadratic convolution (BALQNet) to improve diagnostic performance under long-tailed data conditions. The proposed method mainly consists of a balanced adaptive logit-compensated cross-entropy loss (BAL) and a quadratic convolution backbone. By jointly incorporating logit compensation, label smoothing, and class reweighting, BAL enhances the optimization of minority-class samples, thereby improving the classifier’s ability to distinguish different categories without introducing additional architectural complexity. Meanwhile, quadratic convolution further improves the effectiveness of feature representation learning. Finally, experiments are conducted on self-built bearing, gear, and motor datasets. The results show that BALQNet maintains strong diagnostic performance when handling long-tailed data. In addition, the ablation results provide further evidence for the effectiveness of the proposed approach.

The content of the watercore in apples plays a decisive role in their taste and selling price, but there is a lack of methods to accurately assess it. Therefore, this paper proposes an OCRNet-based method for apple watercore content evaluation. A total of 720 watercores of apples from Mengzi, Lijiang, and Zhaotong City in Yunnan Province were used as experimental samples. An appropriate watercore extraction model was selected based on different evaluation indicators. The watercore feature images extracted using the optimal model were stacked, and the watercore content of apples in different regions was evaluated by calculating the fitted area of the stacked watercore region. The results show that the OCRNet model is optimal in all evaluation metrics when facing different datasets. The error of OCRNet is also minimized when extracting overexposed as well as underexposed images with 0.15% and 0.38%, respectively, and it can be used to extract the characteristics of the apple watercore. The evaluation result of the watercore content of apples in different regions is that Lijiang apples have the highest watercore content, followed by Mengzi apples, and Zhaotong apples have the least watercore content.

Late spring frost poses a serious threat to orchard production, especially in mountainous orchards where timely frost monitoring and adaptive protection are difficult to implement. To address this problem, this study developed an intelligent smoke-based anti-frost machine integrating a LoRa-based wireless temperature monitoring system, a smoke actuation unit, and a closed-loop control terminal. To overcome the slow response and large overshoot of conventional PID control under nonlinear field conditions, a fuzzy PID control strategy optimized by the Hippopotamus Optimization Algorithm (HOA) was proposed to regulate smoke release in real time. Comparative simulations were conducted using conventional PID, fuzzy PID, and HOA-fuzzy PID controllers, and field experiments were performed in an apple orchard. The results showed that the HOA-fuzzy PID controller achieved the best dynamic performance. Compared with conventional PID, the overshoot, rise time, and settling time were reduced by 60.12%, 33.21%, and 61.94%, respectively; compared with fuzzy PID, they were reduced by 22.85%, 50.45%, and 60.11%, respectively. Disturbance simulation further indicated improved control robustness. Field experiments showed that the prototype increased the orchard canopy temperature by 0.9–3.0 K, and the PM_2.5 distribution in the operational area indicated improved smoke coverage. The adaptive regulation strategy also avoided continuous fixed-output operation, suggesting its potential to improve energy-use efficiency. Overall, the proposed system provides a feasible field-operable approach for improving canopy thermal conditions and reducing frost-risk exposure in mountainous orchards, although further biological validation is still required.