arXiv subjects

Zhe Bai

Publications and source records attributed to Zhe Bai.

10 recordsLinked to original sources

arXiv2026

StateFormer: A Multivariate Transformer for Learning History-Dependent Battery State Dynamics and Long-Horizon Health Forecasting

This paper introduces a novel multivariate Transformer \emph{StateFormer} that forecasts degradation dynamics of large-scale battery systems. The model learns across time scales, from short-term thermal fluctuations to long-term aging trajectories, enabling accurate prediction of battery states including state of charge (SOC), state of health (SOH), or battery temperature. By capturing long-range dependencies and identifying the operating conditions that drive future degradation, the model simultaneously represents fast electrochemical and thermal processes and slow aging mechanisms within a unified framework. \emph{StateFormer} achieves robust and accurate predictions of battery state estimation across both synthetic and real-world datasets. It maintains high predictive performance under additive current/voltage noise levels ranging from $1\%$ to $10\%$ and a wide range of ambient temperatures in a synthetic battery fleet accommodating different types of electrode chemistry and manufacturing, as well as five years of field data collected from residential utility battery systems. The resulting model bridges the laboratory-to-field gap and provides predictive intelligence for maintenance planning, operational optimization, and economic decision-making.

eess.SY↗

arXiv2026

Predicting Wave Reflection and Transmission in Heterogeneous Media via Fourier Operator-Based Transformer Modeling

We develop a machine learning (ML) surrogate model to approximate solutions to Maxwell's equations in one dimension, focusing on scenarios involving a material interface that reflects and transmits electro-magnetic waves. Derived from high-fidelity Finite Volume (FV) simulations, our training data includes variations of the initial conditions, as well as variations in one material's speed of light, allowing for the model to learn a range of wave-material interaction behaviors. The ML model autoregressively learns both the physical and frequency embeddings in a vision transformer-based framework. By incorporating Fourier transforms in the latent space, the wave number spectra of the solutions aligns closely with the simulation data. Prediction errors exhibit an approximately linear growth over time with a sharp increase at the material interface. Test results show that the ML solution has adequate relative errors below $10\%$ in over $75$ time step rollouts, despite the presence of the discontinuity and unknown material properties.

cs.LG↗

arXiv2026

Low-dimensional geometry learning for turbulence prediction in optimized stellarators

The optimized stellarator is an attractive concept for which the averaged particle radial drift is zero, and the single particle loss can be significantly reduced. But for the reactor design, global physics such as turbulent transport also need to be optimized besides the confined single particle orbit, or properties estimated using local estimations and heuristic formulations. The first-principle global transport code is too computationally expensive to integrate into the optimization process. The fast surrogate global transport model based on machine learning is a good alternative choice, but the amount of data required to train the surrogate model is numerous due to the high degree-of-freedom of the stellarator design. The work shows that the stellarator design with quasi-helically(QH) symmetric geometry is approximately distributed in a low dimensional latent space, which can be explicitly found by deep learning. This discovery makes it possible to generate global gyrokinetic simulation data for training surrogate models to directly optimize the stellarator geometry for turbulent transport, energetic particle instability, and MHD modes. Using the low dimensional latent space and data analysis methods, the relation between linear zonal residues and axis-excursion is found, providing a simple guide to optimize low turbulent transport QH stellarators.

physics.plasm-ph↗

arXiv2025

LayerScope: Predictive Cross-Layer Scheduling for Efficient Multi-Batch MoE Inference on Legacy Servers

Mixture-of-Experts (MoE) models face memory and PCIe latency bottlenecks when deployed on commodity hardware. Offloading expert weights to CPU memory results in PCIe transfer latency that exceeds GPU computation by several folds. We present PreScope, a prediction-driven expert scheduling system that addresses three key challenges: inaccurate activation prediction, PCIe bandwidth competition, and cross-device scheduling complexity. Our solution includes: 1) Learnable Layer-Aware Predictor (LLaPor) that captures layer-specific expert activation patterns; 2) Prefetch-Aware Cross-Layer Scheduling (PreSched) that generates globally optimal plans balancing prefetching costs and loading overhead; 3) Asynchronous I/O Optimizer (AsyncIO) that decouples I/O from computation, eliminating waiting bubbles. PreScope achieves 141% higher throughput and 74.6% lower latency than state-of-the-art solutions.

cs.LG↗

arXiv2025

StFT: Spatio-temporal Fourier Transformer for Long-term Dynamics Prediction

Simulating the long-term dynamics of multi-scale and multi-physics systems poses a significant challenge in understanding complex phenomena across science and engineering. The complexity arises from the intricate interactions between scales and the interplay of diverse physical processes, which manifest in PDEs through coupled, nonlinear terms that govern the evolution of multiple physical fields across scales. Neural operators have shown potential in short-term prediction of such complex spatio-temporal dynamics; however, achieving stable high-fidelity predictions and providing robust uncertainty quantification over extended time horizons remains an open and unsolved area of research. These limitations often lead to stability degradation with rapid error accumulation, particularly in long-term forecasting of systems characterized by multi-scale behaviors involving dynamics of different orders. To address these challenges, we propose an autoregressive Spatio-temporal Fourier Transformer (StFT), in which each transformer block is designed to learn the system dynamics at a distinct scale through a dual-path architecture that integrates frequency-domain and spatio-temporal representations. By leveraging a structured hierarchy of \ours blocks, the resulting model explicitly captures the underlying dynamics across both macro- and micro- spatial scales. Furthermore, a generative residual correction mechanism is introduced to learn a probabilistic refinement temporally while simultaneously quantifying prediction uncertainties, enhancing both the accuracy and reliability of long-term probabilistic forecasting. Evaluations conducted on three benchmark datasets (plasma, fluid, and atmospheric dynamics) demonstrate the advantages of our approach over state-of-the-art ML methods.

cs.LG↗

arXiv2024

FTL: Transfer Learning Nonlinear Plasma Dynamic Transitions in Low Dimensional Embeddings via Deep Neural Networks

Deep learning algorithms provide a new paradigm to study high-dimensional dynamical behaviors, such as those in fusion plasma systems. Development of novel model reduction methods, coupled with detection of abnormal modes with plasma physics, opens a unique opportunity for building efficient models to identify plasma instabilities for real-time control. Our Fusion Transfer Learning (FTL) model demonstrates success in reconstructing nonlinear kink mode structures by learning from a limited amount of nonlinear simulation data. The knowledge transfer process leverages a pre-trained neural encoder-decoder network, initially trained on linear simulations, to effectively capture nonlinear dynamics. The low-dimensional embeddings extract the coherent structures of interest, while preserving the inherent dynamics of the complex system. Experimental results highlight FTL's capacity to capture transitional behaviors and dynamical features in plasma dynamics -- a task often challenging for conventional methods. The model developed in this study is generalizable and can be extended broadly through transfer learning to address various magnetohydrodynamics (MHD) modes.

physics.comp-ph↗

arXiv2023

AutoCT: Automated CT registration, segmentation, and quantification

The processing and analysis of computed tomography (CT) imaging is important for both basic scientific development and clinical applications. In AutoCT, we provide a comprehensive pipeline that integrates an end-to-end automatic preprocessing, registration, segmentation, and quantitative analysis of 3D CT scans. The engineered pipeline enables atlas-based CT segmentation and quantification leveraging diffeomorphic transformations through efficient forward and inverse mappings. The extracted localized features from the deformation field allow for downstream statistical learning that may facilitate medical diagnostics. On a lightweight and portable software platform, AutoCT provides a new toolkit for the CT imaging community to underpin the deployment of artificial intelligence-driven applications.

eess.IV↗

arXiv2021

Non-intrusive Nonlinear Model Reduction via Machine Learning Approximations to Low-dimensional Operators

Although projection-based reduced-order models (ROMs) for parameterized nonlinear dynamical systems have demonstrated exciting results across a range of applications, their broad adoption has been limited by their intrusivity: implementing such a reduced-order model typically requires significant modifications to the underlying simulation code. To address this, we propose a method that enables traditionally intrusive reduced-order models to be accurately approximated in a non-intrusive manner. Specifically, the approach approximates the low-dimensional operators associated with projection-based reduced-order models (ROMs) using modern machine-learning regression techniques. The only requirement of the simulation code is the ability to export the velocity given the state and parameters as this functionality is used to train the approximated low-dimensional operators. In addition to enabling nonintrusivity, we demonstrate that the approach also leads to very low computational complexity, achieving up to $1000\times$ reduction in run time. We demonstrate the effectiveness of the proposed technique on two types of PDEs.

cs.LG↗

arXiv2019

Randomized methods to characterize large-scale vortical flow network

We demonstrate the effective use of randomized methods for linear algebra to perform network-based analysis of complex vortical flows. Network theoretic approaches can reveal the connectivity structures among a set of vortical elements and analyze their collective dynamics. These approaches have recently been generalized to analyze high-dimensional turbulent flows, for which network computations can become prohibitively expensive. In this work, we propose efficient methods to approximate network quantities, such as the leading eigendecomposition of the adjacency matrix, using randomized methods. Specifically, we use the Nyström method to approximate the leading eigenvalues and eigenvectors, achieving significant computational savings and reduced memory requirements. The effectiveness of the proposed technique is demonstrated on two high-dimensional flow fields: two-dimensional flow past an airfoil and two-dimensional turbulence. We find that quasi-uniform column sampling outperforms uniform column sampling, while both feature the same computational complexity.

math.NA↗

arXiv2017

Dynamic mode decomposition for compressive system identification

Dynamic mode decomposition has emerged as a leading technique to identify spatiotemporal coherent structures from high-dimensional data, benefiting from a strong connection to nonlinear dynamical systems via the Koopman operator. In this work, we integrate and unify two recent innovations that extend DMD to systems with actuation [Proctor et al., 2016] and systems with heavily subsampled measurements [Brunton et al., 2015]. When combined, these methods yield a novel framework for compressive system identification [code is publicly available at: https://github.com/zhbai/cDMDc]. It is possible to identify a low-order model from limited input-output data and reconstruct the associated full-state dynamic modes with compressed sensing, adding interpretability to the state of the reduced-order model. Moreover, when full-state data is available, it is possible to dramatically accelerate downstream computations by first compressing the data. We demonstrate this unified framework on two model systems, investigating the effects of sensor noise, different types of measurements (e.g., point sensors, Gaussian random projections, etc.), compression ratios, and different choices of actuation (e.g., localized, broadband, etc.). In the first example, we explore this architecture on a test system with known low-rank dynamics and an artificially inflated state dimension. The second example consists of a real-world engineering application given by the fluid flow past a pitching airfoil at low Reynolds number. This example provides a challenging and realistic test-case for the proposed method, and results demonstrate that the dominant coherent structures are well characterized despite actuation and heavily subsampled data.

eess.SY↗