Electrocardiogram (ECG) signals are often used to diagnose cardiac status. However, most of the existing ECG diagnostic methods only use the time-domain information, resulting in some obviously lesion information in frequency-domain of ECG signals are not being fully utilized. Therefore, we propose a method to fuse the time and frequency domain information in ECG signals by convolutional neural network (CNN). First, we adapt multi-scale wavelet decomposition to filter the ECG signal; Then, R-wave localization is used to segment each individual heartbeat cycle; And then, the frequency domain information of this heartbeat cycle is extracted via fast Fourier transform. Finally, the temporal information is spliced with the frequency domain information and input to the neural network for classification. The experimental results show that the proposed method has the highest recognition accuracy (99.43%) of ECG singles compared with state-of-the-art methods.

The prediction of chaotic time series can well describe the physical properties and influencing factors of a sequence of numerical data points in successive order and has played an increasingly crucial role in various scientific and engineering communities. These time series data require the prediction model to have strong dynamic feature representation ability, which often leads to excessive calculation burden. As an emerging and effective learning method, a broad learning system (BLS) can be trained quickly by incremental learning, and the system can be reconstructed without requiring a lengthy retraining process, which has been shown to facilitate superior time-saving performance. However, the historical information of time series data is often ignored, resulting in the lack of dynamic feature representation ability of the original BLS algorithm. In this article, a novel intergroup cascade BLS (ICBLS) method and its variant with optimized parameters, termed MOICBLS, are proposed, which can combine the previous knowledge with current information to determine the output results by constructing a novel intergroup cascade structure for the enhancement nodes. The new algorithms try to retain the time-saving advantage of BLS and greatly enhance the extraction ability of dynamic features of chaotic time series data. In addition, it cannot be ignored that the network structure parameters have an unquestionable impact on the performance of a BLS. In response, two conflicting goals of prediction accuracy and diversity are formulated as multiobjective optimization functions in the training phase to generate the optimized network structure parameters needed by multiple candidate models. The experimental results on two benchmark chaotic time series dataset, the Lorenz system and Rossler system, and a geomagnetic disturbance storm time chaotic time series dataset demonstrate that our proposed algorithms are effective for prediction of chaotic time series, and the prediction accuracy is better than some existing approaches.

The surface defects of a hot-rolled strip will adversely affect the appearance and quality of industrial products. Therefore, the timely identification of hot-rolled strip surface defects is of great significance. In order to improve the efficiency and accuracy of surface defect detection, a lightweight network based on coordinate attention and self-interaction (CASI-Net), which integrates channel domain, spatial information, and a self-interaction module, is proposed to automatically identify six kinds of hot-rolled steel strip surface defects. In this paper, we use coordinate attention to embed location information into channel attention, which enables the CASI-Net to locate the region of defects more accurately, thus contributing to better recognition and classification. In addition, features are converted into aggregation features from the horizontal and vertical direction attention. Furthermore, a self-interaction module is proposed to interactively fuse the extracted feature information to improve the classification accuracy. The experimental results show that CASI-Net can achieve accurate defect classification with reduced parameters and computation.

Continuous stirred tank reactor (CSTR)is essential equipment found in chemical processing industries. Control of the CSTR process has long been a challenging issue due to the high complexity and strong nonlinearity of the chemical process. In this article, a novel event-triggered (ET) adaptive dynamic programming (ADP) optimal control algorithm is developed for the CSTR system. To reduce the computational load and the communication data between the controller and the actuator, a novel event-triggering condition based on the Taylor series expansion is designed for the ET controller. With the novel event-triggering condition, the control policy does not update unless the event-triggering condition is not satisfied. Compared with the time-triggering mechanism, the update frequency of the controller can be obviously reduced. Then, an identifer–actor–critic structure is developed to implement the ET ADP controller. In particular, to overcome the challenge of establishing an exact dynamic for the CSTR system, an identifier neural network (NN) is employed to reconstruct the unknown system dynamic based on offline data. Furthermore, the actor–critic structure is developed to obtain the ET control law and the value function. In actor and critic NNs, weights are turned just at the triggered instant and remained constant during the interevent times. Finally, the developed ET ADP controller is applied to the CSTR system. Experimental results show that the developed ET approach can cut down the update frequency to 66%, which is very significant for the real CSTR system.

Fault diagnosis in modern industrial and information systems is critical for ensuring equipment reliability and operational safety, but traditional methods have difficulty in effectively capturing spatiotemporal dependencies and fault-sensitive features in multi-sensor data, especially rarely considering dynamic features between multi-sensor data. To address these challenges, this study proposes DyGAT-FTNet, a novel graph neural network model tailored to multi-sensor fault detection. The model dynamically constructs association graphs through a learnable dynamic graph construction mechanism, enabling automatic adjacency matrix generation based on time–frequency features derived from the short-time Fourier transform (STFT). Additionally, the dynamic graph attention network (DyGAT) enhances the extraction of spatiotemporal dependencies by dynamically assigning node weights. The time–frequency graph pooling layer further aggregates time–frequency information and optimizes feature representation.Experimental evaluations on two benchmark multi-sensor fault detection datasets, the XJTUSuprgear dataset and SEU dataset, show that DyGAT-FTNet significantly outperformed existing methods in classification accuracy, with accuracies of 1.0000 and 0.9995, respectively, highlighting its potential for practical applications.

Remote sensing (RS) images contain a wealth of information with expansive potential for applications in image segmentation. However, Convolutional Neural Networks (CNN) face challenges in fully harnessing the global contextual information. Leveraging the formidable capabilities of global information modeling with Swin-Transformer, a novel RS images segmentation model with CNN (GLE-Net) was introduced. This integration gives rise to a revamped encoder structure. The subbranch initiates the process by extracting features at varying scales within the RS images using the Multiscale Feature Fusion Module (MFM), acquiring rich semantic information, discerning localized finer features, and adeptly handling occlusions. Subsequently, Feature Compression Module (FCM) is introduced in main branch to downsize the feature map, effectively reducing information loss while preserving finer details, enhancing segmentation accuracy for smaller targets. Finally, we integrate local features and global features through Spatial Information Enhancement Module (SIEM) for comprehensive feature modeling, augmenting the segmentation capabilities of model. We performed experiments on public datasets provided by ISPRS, yielding notably remarkable experimental outcomes. This underscores the substantial potential of our model in the realm of RS image segmentation within the context of scientific research.