In the study of brain-computer interfaces (BCIs) based on steady-state visual evoked potentials (SSVEPs), how to improve the classification accuracies of BCIs has always been the focus of researchers. Canonical correlation analysis (CCA) is widely used in BCI systems of SSVEPs because of its rapidity and scalability. However, the classical CCA algorithm always encounters the difficulty of low accuracy in a short time. For targetless stimuli, this paper proposes a fusion algorithm (CCA-CWT-SVM) that is combined with CCA, a continuous wavelet transform, and a support vector machine (SVM) to improve the low classification accuracies when a single feature extraction method is used. This fusion algorithm achieves high accuracies and information transfer rates (ITRs) in the SSVEP paradigm with few targets. Through the study of 400 groups of experimental data from 10 subjects, the results show that CCA-CWT-SVM has a classification accuracy of 91.76% within 2 s and an ITR of 48.92 bits/min, which are 10.88% and 13.18 bits/min higher than those of the standard CCA. Compared with a mainstream EEG decoding algorithm, filter bank canonical correlation analysis (FBCCA), the classification accuracy and ITR of the CCA-CWT-SVM algorithm also improved (4.45% and 5.69 bit/min, respectively). Using a dataset from Tsinghua University (THU), we also showed that the fusion algorithm is better than the classical algorithms. The CCA-CWT-SVM algorithm obtained an 89.1% accuracy and a 39.91 bit/min ITR in a time window of 2 s. The results were significantly improved compared with those of CCA and the FBCCA (CCA: 79.44% and 28.23 bits/min, FBCCA: 84.03% and 33.4 bits/min). Hence, this work provides an experimental basis for designing an SSVEP-based BCI system with a high task classification accuracy in some crucial biomedical applications.

The rolling element bearings used in rotating machinery generally include multiple coexisting defects. However, individual defect–induced signals of bearings simultaneously arising from multiple defects are difficult to extract from measured vibration signals because the impulse-like fault signals are very weak, and the vibration signal is commonly affected by the transmission path and various sources of interference. This issue is addressed in this study by proposing a new compound fault feature extraction scheme. Vibration signals are first preprocessed using resonance-based signal sparse decomposition to obtain the low-resonance component of the signal, which contains the information related to the transient fault–induced impulse signals, and reduce the interference of discrete harmonic signal components and noise. The objective used for adaptively selecting the optimal resonance-based signal sparse decomposition parameters adopts the ratio of permutation entropy to the frequency domain kurtosis, as a new comprehensive index, and the optimization is conducted using the cuckoo search algorithm. Subsequently, we apply multipoint sparsity to the low-resonance component to automatically determine the possible number of impulse signals and their periods according to the peak multipoint sparsity values. This enables the targeted extraction and isolation of fault-induced impulse signal features by multipoint optimal minimum entropy deconvolution adjustment. Finally, the envelope spectrum of the filtered signal is used to identify the individual faults. The effectiveness of the proposed scheme is verified by its application to both simulated and experimental compound bearing fault vibration signals with strong interference, and its advantages are confirmed by comparisons of the results with those of an existing state-of-the-art method.

With the advancement of complex system diagnosis, prediction, and health management technologies, digital twin technology has become a prominent research area in the fields of intelligent manufacturing and system operation and maintenance. However, due to the high complexity of practical systems, the difficulty of data acquisition, and the low accuracy of modeling techniques, current digital twin modeling suffers from low accuracy, and the generalization ability of models is poor when applied in model transfer. To address this issue, a novel fault diagnosis method is proposed, which integrates a digital twin model based on transfer learning. The framework introduces an innovative approach to construct multiple digital twin models using both mechanistic and data-driven models. The mechanism twin constructs a universal simulation model based on physical equipment and updates it with system response measurement data. The data twin consists of a high-dimensional fully connected-generative adversarial network twin for extracting deep features from data and an long- and short-term memory twin for extracting time series features. Subsequently, transfer learning is introduced to achieve deep fusion in the multiple digital twins system. The mechanism twin is used to obtain source domain samples to construct a diagnostic network, and the data twin is used to extract target domain features to correct the diagnostic network, thereby improving the accuracy and reliability of fault diagnosis. Finally, the proposed framework is applied to the fault diagnosis of triplex pump equipment. The accuracy of diagnosis continuously improves as the system is updated and ultimately reaches 89.28%, demonstrating the effectiveness of the algorithm and providing a novel solution for the generalization limitations of current digital twin models.

The energy loss and substantial costs associated with faults in photovoltaic (PV) systems impose significant limitations on their efficiency and reliability. Addressing current issues in PV fault diagnosis such as the lack of typical fault data, imbalanced data distribution, and poor diagnostic performance, this paper proposes an intelligent fault diagnosis method for PV systems, deep digital twins (DDT) with information gain stacking sparse autoencoders (IGSSAEs). Initially, the method designs a novel DDT modeling framework tailored to actual PV system specifications. This framework utilizes a mechanism simulation model to generate typical data under various states. Simultaneously, a deep data model is constructed to learn the distribution characteristics of the mechanism model and complete data diversification, achieving the fusion and complementation of data from both models. Subsequently, a diagnostic network using IGSSAE is introduced. This network utilizes information gain ratio to assess feature classification contributions, enabling automatic feature selection. Based on the input features, a stacked sparse autoencoder fault classification network is designed, incorporating multi-level feature compression to enhance the model’s stability and diagnostic accuracy. Finally, a case study is conducted using a 250 kW grid-connected PV system, thoroughly validating the method’s effectiveness with a diagnostic accuracy of 98.4%.

The extreme unbalance of training samples among different working conditions caused by complex and variable external environment makes the fault diagnosis of chiller based on domain adaptation (DA) poor performance. Although recently emerging fault diagnosis methods based on domain generalization (DG) can learn domain-invariant knowledge from multiple source domains and can generalize to unseen target domains, these methods still rely on multiple similar source domain data and rarely consider how to enhance the ability to distinguish the joint distribution of features extracted from the source domain samples. To address these problems, a generative domain-generalized framework for chillers fault diagnosis, namely, vision transformer generative adversarial domain generalization (VIT-GADG), is proposed. In VIT-GADG, a novel VIT domain generation network (VIT-DGN) is firstly designed to reduce DG’s dependence on multi-source domain data by improving the diversity of the distribution of the source domain samples. Then, a new adversarial domain generalization network called VIT conditional adversarial domain generalization network (VIT-CADGN) is designed to extract domain-invariant knowledge from source domain and latent domains that can be generalized to unseen target domains. Specifically, VIT module can effectively extract the global statistical feature of input samples, which is conducive to the identification of joint distribution. Simultaneously, the collaborative conditional domain discrimination strategy is introduced to improve the distribution discrimination ability of the extracted global statistical features while simultaneously aligning its conditional distribution. In addition, a personalized adaptive weight strategy is proposed to improve the performance of VIT-CADGN. Finally, the comprehensive case study shows that VIT-GADG has a satisfactory ability to extract invariant features, which improves the diagnosis accuracy in the unseen target domain.