In the industrial sector, accurate fault identification is paramount for ensuring both safety and economic efficiency throughout the production process. However, due to constraints imposed by actual working conditions, the motor state features collected are often limited in number and singular in nature. Consequently, extending and extracting these features pose significant challenges in fault diagnosis. To address this issue and strike a balance between model complexity and diagnostic accuracy, this paper introduces a novel motor fault diagnostic model termed FSCL (Fourier Singular Value Decomposition combined with Long and Short-Term Memory networks). The FSCL model integrates traditional signal analysis algorithms with deep learning techniques to automate feature extraction. This hybrid approach innovatively enhances fault detection by describing, extracting, encoding, and mapping features during offline training. Empirical evaluations against various state-of-the-art techniques such as Bayesian Optimization and Extreme Gradient Boosting Tree (BOA-XGBoost), Whale Optimization Algorithm and Support Vector Machine (WOA-SVM), Short-Time Fourier Transform and Convolutional Neural Networks (STFT-CNNs), and Variational Modal Decomposition-Multi Scale Fuzzy Entropy-Probabilistic Neural Network (VMD-MFE-PNN) demonstrate the superior performance of the FSCL model. Validation using the Case Western Reserve University dataset (CWRU) confirms the efficacy of the proposed technique, achieving an impressive accuracy of 99.32%. Moreover, the model exhibits robustness against noise, maintaining an average precision of 98.88% and demonstrating recall and F1 scores ranging from 99.00% to 99.89%. Even under conditions of severe noise interference, the FSCL model consistently achieves high accuracy in recognizing the motor’s operational state. This study underscores the FSCL model as a promising approach for enhancing motor fault diagnosis in industrial settings, leveraging the synergistic benefits of traditional signal analysis and deep learning methodologies.

Attention to the natural environment is equivalent to observing the space in which we live. Plant roots, which are important organs of plants, require our close attention. The method of detecting root system without damaging plants has gradually become mainstream. At the same time, machine learning has been achieving good results in recent years; it has helped develop many tools to help us detect the underground environment of plants. Therefore, this article will introduce some existing content related to root detection technology and machine detection algorithms for root detection, proving that machine learning root detection technology has good recognition capabilities.

Water scarcity is a global problem. Deficit irrigation (DI) reduces evapotranspiration, improving water efficiency in agriculture. Reference evapotranspiration ${(ET}_0)$ is an important factor in determining DI. ${ET}_0$ forecasting predicts field water consumption and enables proactive irrigation decisions, offering guidance for water resource management. However, implementation of ${ET}_0$ forecasting faces challenges due to complex calculations and extensive meteorological data requirements. This project aims to develop a machine learning system for ${ET}_0$ forecasting. The project involves studying ${ET}_0$ methods and identifying required meteorological parameters. Historical meteorological data and weather forecasts were obtained from meteorological websites and analyzed for accuracy after preprocessing. A machine learning-based model was created to forecast reference crop evapotranspiration. The model’s input parameters were selected through path analysis before it was optimized using Bayesian optimization to reduce overfitting and improve accuracy. Three forecasting models were developed: one based on historical meteorological data, one based on weather forecasts, and one that corrects the weather forecasts. All three models achieved good accuracy, with root mean square errors ranging from 0.52 to 0.81 mm/day. Among them, the model based on weather forecast had the highest accuracy; the RMSE six days before the forecast period was between 0.52 and 0.75 mm/day, and the RMSE on the seventh day of the forecast period was 1.12 mm/day. In summary, this project has established a mathematical model of ${ET}_0$ prediction based on machine learning, which can achieve more accurate predictions for within a few days.

Root diameter is an important indicator of plant growth and development to a large extent. However, the field monitoring of these parameters is severely limited by the lack of appropriate methods, and some traditional methods may harm the plant and its growing environment. Ground-penetrating radar (GPR) is a new nondestructive detection method for underground root systems. A new method for the estimation of the diameter of coarse roots using GPR with 900 MHz frequency was proposed in this paper. First, a simulation model was established to simulate the root system under natural conditions, and the root diameter estimation model based on the scanning results of GPR was obtained. Secondly, by studying the influence of soil and root relative permittivity on the diameter estimation model, a method was found to devise a coarse root diameter estimation model under different soil and root conditions of relative permittivity. Thirdly, the applicability of the diameter estimation model to roots with different growth orientations was tested by simulating roots with different growth orientations. Finally, the practical applicability of the estimation method was verified by field experiments. The results suggest that the root diameter estimation model can be constructed by extracting the pixel distance (∆p) of waveform parameters from the 900 MHz scanning results. This method can be used to estimate the diameter of coarse roots with diameters of no less than 2 cm and a relative permittivity greater than 5, and to estimate the diameter of roots in any orientation and soil environment effectively. At the same time, the application in the field experiment also resulted in a good estimation effect. This method provides a new opportunity to achieve more reliable root diameter estimation in complex situations. The estimation of coarse root diameter provides an experimental basis and data support for the healthy growth of trees, and also provides some information for the study of coarse root ecology.

Ginkgo leaf disease poses a grave threat to Ginkgo biloba. The current management of Ginkgo leaf disease lacks precision guidance and intelligent technologies. To provide precision guidance for disease management and to evaluate the effectiveness of the implemented measures, the present study proposes a novel disease progression prediction (DPP) method for Ginkgo leaf blight with a multi-level feature translation architecture and enhanced spatiotemporal attention module (eSTA). The proposed DPP method is capable of capturing key spatiotemporal dependencies of disease symptoms at various feature levels. Experiments demonstrated that the DPP method achieves state-of-the-art prediction performance in disease progression prediction. Compared to the top-performing spatiotemporal predictive learning method (SimVP + TAU), our method significantly reduced the mean absolute error (MAE) by 19.95% and the mean square error (MSE) by 25.35%. Moreover, it achieved a higher structure similarity index measure (SSIM) of 0.970 and superior peak signal-to-noise ratio (PSNR) of 37.746 dB. The proposed method can accurately forecast the progression of Ginkgo leaf blight to a large extent, which is expected to provide valuable insights for precision and intelligent disease management. Additionally, this study presents a novel perspective for the extensive research on plant disease prediction.

Accurate forest monitoring and resource assessment are crucial for sustainable forest management, with tree diameter at breast height (DBH) serving as a key metric for tree growth assessment and carbon storage estimation. In this study, we developed a comprehensive mobile-LiDAR-based point cloud processing pipeline to segment individual trees and estimate the DBH of trees. We first conducted terrain extraction using a resolution-passing method combined with a cloth simulation filter. Then, by leveraging the vertical structural characteristics of trees and changes in point cloud density, we achieved high-performance tree trunk segmentation. On this basis, we deployed the Randomized Hough Transform algorithm to estimate the DBH of the trees. Finally, a large-scale experiment was conducted in a forest (Olympic Forest Park, Beijing, China) and we provided experimental results comparing our trunk segmentation and DBH estimation to ground-truth measurements recorded manually. Eventually, our results showed that 97.4% of the trees were accurately segmented, and the DBH estimation error was reduced to 3.2 cm, which shows that the proposed pipeline is able to achieve high-accuracy trunk segmentation and high-precision DBH estimation. Further, this research demonstrates that integrating MLS with SLAM technology can enhance the efficiency and accuracy of forest surveys, providing a valuable tool for future forest management strategies.