A5038268580- Advanced Battery Technologies Research
- Fault Detection and Control Systems
- Advancements in Battery Materials
- Spectroscopy and Chemometric Analyses
- Electric Vehicles and Infrastructure
- Mineral Processing and Grinding
- Control Systems and Identification
- Advanced Battery Materials and Technologies
- Advanced Statistical Process Monitoring
- Ethics and Social Impacts of AI
- Reliability and Maintenance Optimization
- Hydraulic and Pneumatic Systems
- Machine Learning in Materials Science
- Machine Fault Diagnosis Techniques
- Electron and X-Ray Spectroscopy Techniques
- Blind Source Separation Techniques
- Advanced Electron Microscopy Techniques and Applications
- Privacy-Preserving Technologies in Data
- Advanced Multi-Objective Optimization Algorithms
- Machine Learning and Algorithms
- Optimal Experimental Design Methods
- Optimization and Search Problems
- Gaussian Processes and Bayesian Inference
- Metaheuristic Optimization Algorithms Research
- Process Optimization and Integration
Tsinghua University
2012-2025
Massachusetts Institute of Technology
2015-2023
Beijing University of Chemical Technology
2016-2018
Chinese Academy of Sciences
2018
Real-time and personalized lithium-ion battery health management is conducive to safety improvement for end-users. However, prognostic of the status still challenging due diverse usage interests, dynamic operational patterns limited historical data. We generate a comprehensive dataset consisting 77 commercial cells (77 discharge protocols) with over 140 000 charge–discharge cycles—the largest our knowledge its kind, develop transfer learning framework realize real-time prediction unseen...
Accurate evaluation of Li-ion battery (LiB) safety conditions can reduce unexpected cell failures, facilitate deployment, and promote low-carbon economies. Despite the recent progress in artificial intelligence, anomaly detection methods are not customized for or validated realistic settings due to complex failure mechanisms lack real-world testing frameworks with large-scale datasets. Here, we develop a deep-learning framework electric vehicle (EV) LiB detection. It features dynamical...
Abstract Flexibility has become increasingly important considering the intermittency of variable renewable energy in low-carbon systems. Electrified transportation exhibits great potential to provide flexibility. This article analyzed and compared flexibility values battery electric vehicles fuel cell for planning operating interdependent electricity hydrogen supply chains while degradation costs. A cross-scale framework involving both macro-level micro-level models was proposed compute...
Abstract Reaction rates at spatially heterogeneous, unstable interfaces are notoriously difficult to quantify, yet essential in engineering many chemical systems, such as batteries 1 and electrocatalysts 2 . Experimental characterizations of materials by operando microscopy produce rich image datasets 3–6 , but data-driven methods learn physics from these images still lacking because the complex coupling reaction kinetics, surface chemistry phase separation 7 Here we show that heterogeneous...
Porous electrode theory (PET) is widely used to model battery cycling behavior by describing electrochemical kinetics and transport in solid particles electrolyte, modeling thermodynamics fitting an open-circuit potential. The PET consists of tightly coupled nonlinear partial differential-algebraic equations which effective kinetic parameters are fit data, then the analyze effects variations design or operating conditions such as charging protocols. In a detailed identifiability analysis, we...
The fast charging problem of lithium-ion batteries with minimum time while limiting battery degradation is receiving increasing attention and a critical challenge to community. Difficulties in this optimization lie that: 1) the parameter space strategies high dimensional, budget experimental cost often limited; 2) evaluation strategies' performance expensive; 3) process strongly nonlinear, multiple mechanisms occur simultaneously leading difficulties for establishing accurate first-principle...
As batteries become widespread applications across various domains, the prediction of battery cycle life has attracted increasing attention. However, intricate internal mechanisms pose challenges to achieving accurate lifetime prediction, and inherent patterns within temporal data from experiments are often elusive. Meanwhile, commonality missing in real-world usage further complicates prediction. To address these issues, this article develops a self-attention-based neural network (NN)...
Fault diagnosis plays a key role in the safe and efficient operation of industrial processes. With emerging big data era, analytic methods based on probabilistic representations have attracted growing research interest. In this brief, dynamic minimax probability machine (DMPM) approach framework is proposed for diagnosing process faults, without imposing any assumptions distributions. addition, an information criterion put forward to determine optimal dimensionality reduction order time lags...
This paper proposes a stochastic representation of maximized mutual information analysis (MIA) method for quality monitoring in which manner imposing prior probability distributions over projection parameters is employed and subsequently, Bayesian estimation algorithm put forward learning. The proposed MIA (SMIA) based approach allows the enhanced performance fault detection due to following advantages classic methods. First, SMIA utilizes mechanism hierarchical priors an individual each...
Conventional battery simulation tools offer current, voltage, and power operating modes. This article presents General Operating Modes (GOMs), which move beyond these standard modes allow models of any scale to simulate novel such as constant temperature, lithium plating overpotential, concentration. The governing equations the model are solved alongside a single algebraic constraint that determines current. simulated efficiently deterministically inside differential-algebraic equation (DAE)...
Fault diagnosis gains increasing attention for its ability to enhance process safety and efficiency. This brief proposes a maximized ratio divergence analysis (MRDA) approach fault diagnosis, which maximizes the pairwise Kullback-Leibler (KL) between each pair of classes during dimensionality reduction step. In addition, an iterative algorithm based on deflation techniques is put forward learning loading vectors MRDA. The proposed MRDA-based allows improved power because following advantages...