Foundation models (FMs) have shown great promise in single-cell genomics, yet current approaches, such as scGPT, Geneformer, and scFoundation, rely on centralized training and language modeling objectives that overlook the tabular nature of single-cell data and raise significant privacy concerns. We present TABULA, a foundation model designed for single-cell transcriptomics, which integrates a novel tabular modeling objective and federated learning framework to enable privacy-preserving pretraining across decentralized datasets. TABULA directly models the cell-by-gene expression matrix through column-wise gene reconstruction and row-wise cell contrastive learning, capturing both gene-level relationships and cell-level heterogeneity without imposing artificial gene sequence order. Extensive experiments demonstrate the effectiveness of TABULA: despite using only half the pretraining data, TABULA achieves state-of-the-art performance across key tasks, including gene imputation, perturbation prediction, cell type annotation, and multi-omics integration. It is important to note that as public single-cell datasets continue to grow, TABULA provides a scalable and privacy-aware foundation that not only validates the feasibility of federated tabular modeling but also establishes a generalizable framework for training future models under similar privacy-preserving settings.

The dynamics simulation of complex railway vehicles requires a dedicated vehicle model, such as multi-body dynamics model. However, the multi-body model is time-consuming in long-distance simulation due to its computational complexity. This issue can be alleviated by using a data-driven vehicle dynamics model due to its effective generalization and computational speed. Firstly, the construction of the physical model of the vehicle system is carried out to obtain the coupling relationship between the components. Secondly, the coupling relationship between the components is embedded into the loss function of the deep neural network as physics-informed constraints. Further, the network parameters satisfying certain physical laws are obtained by minimizing the loss function. Finally, the proposed lightweight 3D convolutional neural network is used to predict the vibration state of the vehicle system. The dynamic response resulting from both the data-driven simulation model and the multi-body simulation model are investigated and compared. The simulation results show that the data-driven dynamics simulation model can accurately predict the vibration state of the vehicle system. The data-driven simulation model has smaller size and faster operation speed, which can be applied to long-distance prediction research of vehicle systems.

Since the cyclostationarity in vibration signals is the key to judge the rotating machine health state, spectral harmonics-to-interference ratio (SHIR) has been used to construct single node feature learning network (SHIR-based blind deconvolution, BD-SHIR) to realize condition monitoring of machine. However, BD-SHIR still has several obvious limitations, including the need for prior fault information and the tendency to fall into local optimal solutions which will affect its feature learning and state monitoring performance. Therefore, this paper proposes a multi-node feature learning network based on BDSHIR, which does not rely on prior information, multi-node adaptive BDSHIR (MABD-SHIR). It uses each filter in the 1/3 binary tree filter bank to design each initial node (filter) in the learning network correspondingly, and introduces an adaptive fault characteristic frequency (FCF) detection process. Since this initialization can give some filters close to the fault resonant band, iterations starting from these nodes are more likely to detect the precise FCF and converge to the optimal solution. Finally, MABD-SHIR can obtain multiple local optimal solutions, among which the one with the largest SHIR is generally closer to the global optimal solution than the one obtained by BD-SHIR. Simulation and experimental data of defective rotating machinery verify the effectiveness of MABD-SHIR in machinery condition monitoring and fault feature learning.Note to Practitioners—This paper is motivated by the problems that the rotating machine condition monitoring based on BDSHIR is easy to prematurely fall into local optimal solution and requires prior fault information. These problems can be effectively solved by extending BDSHIR from single node structure to multi-node structure and introducing adaptive period detection technique (A-DPT). The core idea is to start iterations from multiple initial filters and to detect the FCF adaptively during each iteration. Multiple features conforming to filtered signals’ SHIR maximization can be finally learned. Therefore, in this paper, MABD-SHIR is proposed, which is equivalent to a multi-node feature learning network. Since the objective function of BD is usually non-convex, MABD-SHIR can obtain multiple local optimal solutions without requiring prior FCFs. This is more likely to achieve a global optimal solution than BDSHIR, which starts iteration from only one initial filter and ends up with one local optimal solution. In addition, the objective function of MABD-SHIR has lower mathematical complexity, which reduces the difficulty of optimization.

Harmonics-to-noise ratio (HNR) is an important health index of rotating machine, which has been applied in blind deconvolution (BD) method to realize periodic impulse detection. However, most fault impulses are not strictly periodic, but pseudo-cyclostationary, which will affect the performance of HNR in fault characterization to some extent. This limits its applications. Therefore, in this paper, a novel BD method, maximum squared envelope spectrum harmonic-to-interference ratio deconvolution (MSESHIRD), is proposed to more effectively achieve fault identification. The proposed method seeks a target filter by maximizing squared envelope spectrum harmonic-to-interference ratio (SESHIR). Since harmonic components corresponding to repetitive fault impulses in SES are less sensitive to random fluctuations, SESHIR can more accurately distinguish repetitive fault impulses from irrelevant interference in vibration signals. Therefore, BD based on SESHIR has better performance than BD based on HNR in measuring fault features in signals. Through simulation and experimental case analysis, the proposed method is compared with several public methods Results show that the proposed method has better performance in fault characteristic extraction. In addition, it is implemented on bearing run-to-failure data for condition monitoring to show that the proposed method has excellent ability of early fault detection. Note to Practitioners—This paper is motivated by the problems of automatic operating condition monitoring and early defect diagnosis of rotating machines. These problems can be effectively solved by designing a BD method based on reliable and efficient health indices. HNR defined on autocorrelation function (AF) is an excellent health index to characterize the signal-to-noise ratio (SNR) of repetitive fault impulse in signals. However, this paper uses mathematical models of HNR to show that it has very strict requirements on the period and SNR of fault impulse signals. A fluctuating fault period or a low SNR might make HNR unable to accurately estimate the energy of fault components in signals, thus weakening its performance in fault characterization. Compared with HNR, SESHIR has better fault characterization ability due to that SES can more accurately obtain the periodicity (frequency) and energy of fault components in signals. Therefore, this paper proposes a novel BD method based on SESHIR maximization for repetitive impulse monitoring. Its effectiveness and robustness are verified by both theoretical justification and experimental results.

Second-order IIR adaptive notch filters (ANF) andbanks (ANFB) have been extensively utilized in many applications ranging from biomedical engineering to control systems. In this paper, a new efficient ANFB is proposed that is composed of two filter banks, namely a prefilter bank with cascaded IIR notch filters and a parallel structure IIR ANFB. The second-order IIR ANFs or filter bank cells in the proposed ANFB are updated by utilizing a modified normalized gradient (ModNG) algorithm. The proposed ANFB requires the same number of cells as does a conventional parallel form ANFB. However, the filter weight initialization drawback that the conventional parallel form ANFB inherently suffers from is significantly alleviated in the proposed ANFB. Equipped with the ModNG algorithm, the proposed ANFB yields considerably improved frequency estimation performance as compared to the conventional parallel form ANFB, for a wide range of frequencies and signal-to-noise ratios (SNR). Furthermore, preliminary steady-state analysis is provided for both the proposed and conventional ANFBs, which significantly enhances our understanding of their capabilities and features. Extensive case study simulations are conducted to demonstrate the superiority of the proposed ANFB over the conventional parallel form ANFB. The proposed ANFB is also applied to real noise and vibration signals to reveal its promising analytical and practical capabilities.

In the above article[1],Table I,III, andIVshould show “N/m” instead of “kN/m” and they should also show “Ns/m” instead of “kNs/m.” The revised tables are shown below.

A novel data-driven framework to monitor the health status of high-speed rail suspension system by measuring train vibrations is proposed herein. Unlike existing methods, this framework does not rely on sophisticated dynamic models or high-fidelity simulations; it combines the power of data and domain knowledge to generate a model that can be trained quickly and adapted easily to different rail systems. In addition, the framework includes a module to generate a training dataset, tackling a typical challenge in real-world system monitoring, namely, the lack of labeled data due to practical limits. Based on the multi-output support vector regression (MSVR), the proposed framework can monitor the stiffness and damping coefficients of the suspension system using vibration signals measured on trains in real time. The framework comprises three modules. First, a simple suspension system dynamics model is built to generate a training dataset. Furthermore, key features are extracted from frequency response curves to reflect the impact of spring and damper degradation. Subsequently, a supervised learning model based on the MSVR is built to predict the stiffness and damping coefficients of suspension systems from features extracted in the second module. Once the model is built, real-time monitoring can be achieved by feeding the vibration signals as they are collected during operations. The proposed framework was evaluated on simulation data for its accuracy and tested on real-world operational data for its practicability.

Approximations of a concept by a variable precision rough-set model (VPRS) usually vary under a dynamic information system environment. It is thus effective to carry out incremental updating approximations by utilizing previous data structures. This paper focuses on a new incremental method for updating approximations of VPRS while objects in the information system dynamically alter. It discusses properties of information granulation and approximations under the dynamic environment while objects in the universe evolve over time. The variation of an attribute's domain is also considered to perform incremental updating for approximations under VPRS. Finally, an extensive experimental evaluation validates the efficiency of the proposed method for dynamic maintenance of VPRS approximations.