About

Jing Sun is a Professor in the Department of Mechanical Engineering at the University of Michigan and holds the title of Michael G. Parsons Collegiate Professor of Naval Architecture and Marine Engineering. He serves as the Chair of Naval Architecture and Marine Engineering. His research interests include the theory, methodologies, and tools for control system integration and optimization, with applications to marine and automotive systems, connected and automated transportation systems, and renewable energy systems. Dr. Sun has been recognized as a Fellow by the National Academy of Inventors in 2018, reflecting his significant contributions to innovation and research in his field.

Research topics

• Artificial Intelligence
• Computer Science
• Control engineering
• Engineering
• Electrical engineering

Selected publications

• In Memoriam: Neng (Eva) Wu, 1956–2025 [Obituary]

  IEEE Control Systems · 2026-04-01

  article1st authorCorresponding
  PublisherDOI

• Analysis of the Transformation and Value of Contemporary Jingdezhen Incised and Scratched Flower Art from an Interdisciplinary Perspective

  2026-03-31

  articleSenior author

  本文以当代景德镇陶瓷刻划花艺术为研究对象，从跨学科视角研究其由“线”向“面”的转化机制与文化意义。研究结合材料工艺实证、艺术理论审美融合、中西美学对话等多元方法，通过陶瓷材料科学分析、工艺历史脉络梳理等，发现当代刻划花是以宋代青白瓷薄釉特质为线刻艺术奠基，结合“半刀泥” 技法、陶瓷材料物理属性及工艺革新为线面转化提供支撑，而现代设计与西方造型观念的融入，是实现“线—面—空间”三重转型的重要成因。这一转化是中西美学的对话融合，也是本土陶瓷艺术在全球化语境下的文化身份重塑。

  PublisherDOI

• Multivariate Load Interval Prediction Considering Dispatch Costs For Integrated Coal Mine Energy Systems

  2025-06-30

  article

  Given the high complexity and significant uncertainty of the multiple loads in integrated coal mine energy systems, interval prediction provides more comprehensive information for developing dispatch strategies, thus improving the reliability and robustness of the decision-making process. However, traditional interval prediction evaluation metrics fail to adequately reflect the impact of interval prediction results on the optimal dispatch cost. To address this, this paper proposes a day-ahead interval prediction method for the multivariate load of an integrated coal mine energy system, explicitly considering the dispatch cost. First, multi-task learning is combined with quantile regression to explore the coupling characteristics between multiple loads systematically. Next, an optimization model is constructed that simultaneously accounts for interval prediction performance and dispatch cost to determine the optimal upper and lower quantile combinations, adapting to dynamic load changes. Finally, the upper and lower quantile combinations are optimized using genetic algorithms, and the prediction intervals of multiple loads for the day ahead are obtained based on the optimized quantile combinations and the trained model. The proposed method is applied to an integrated coal mine energy system and compared with existing methods. Experimental results demonstrate that the proposed method ensures the prediction accuracy of the day-ahead multivariate load interval and significantly reduces the system’s dispatch cost.

  PublisherDOI

• PCM-integrated bionic cold plate inspired by banyan aerial roots for enhanced battery thermal management

  International Journal of Thermal Sciences · 2025-11-29 · 4 citations

  articleCorresponding
  PublisherDOI

• The integration of virtual reality and augmented reality technology in immersive media art

  Systems and Soft Computing · 2025-08-18 · 3 citations

  articleOpen access1st authorCorresponding

  With the rapid development of digital technology, immersive media art has gradually become a new field of integrating art and technology. This study focuses on integrating virtual reality (VR) and augmented reality (AR) technologies to explore the potential and practice of immersive media art creation. This paper proposes the Kalman filter algorithm to optimize the virtual-real fusion algorithm, which aims to improve the user's sense of immersion and realism when interacting with the virtual world. This study first analyzes the application status of augmented reality technology in media art and discusses the importance of natural features in the fusion of virtual reality. Then, this paper introduces in detail the virtual-real fusion technology based on the Kalman filter algorithm, which can effectively predict and filter noise and improve the accuracy and stability of scene fusion. Experiments are carried out in the system simulation by constructing the interaction model between the virtual environment and the natural scene. It verifies the effectiveness and feasibility of the algorithm. The simulation results show that the system using the Kalman filter algorithm improves the accuracy of virtual-real fusion by 20% compared with the traditional method, and the fluency of user interaction is also significantly increased. The technology proposed in this study can provide a richer and more authentic experience for immersive media art.

  PublisherDOI

• A Hybrid Wind-Wave Energy Converter with Structural and Functional Synergy: Concept Design and Simulation

  2025-09-29

  articleSenior author

  This paper presents the design and dynamic analysis of a novel hybrid wind-wave platform that co-locates a floating wind turbine and three flap-type wave energy converters (WECs) on a shared hexagonal semi-submersible structure. Rather than simply adding WECs to a floating wind system, our concept structurally integrates them to function synergistically. The WEC flaps provide buoyancy, enhance hydrostatic stability, and reduce platform motions while converting wave energy. Hydrodynamic coefficients were computed using Capytaine, and time-domain simulations were performed in WEC-Sim under irregular wave and turbulent wind conditions. A hydrostatic stability analysis confirmed that the platform maintains positive metacentric height within <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\pm 20^{\circ}$</tex> of pitch, ensuring robust performance under varying sea states. Dynamic simulations further showed that platform pitch and surge remained within safe operational limits. The wind turbine consistently generated power near its rated output at a mean wind speed of 15 m/s. Among the WECs, the flap oriented perpendicular to the wave direction achieved the highest performance, with a capture width ratio (CWR) of 36.67 %, while the total CWR across all three flaps was 23.6 %, and the total output from all three flaps reached 0.85 MW. These results demonstrate the effectiveness of the hybrid windwave configuration in enhancing total energy extraction while maintaining acceptable platform responses. The concept supports higher energy yield per ocean footprint and presents a viable path toward more compact, multi-functional offshore renewable energy systems.

  PublisherDOI

• Research on battery thermal management system of electric mining truck based on topology optimization and PCM coupling

  International Communications in Heat and Mass Transfer · 2025-11-27 · 4 citations

  articleCorresponding
  PublisherDOI

• Dynamic decision making simulation with limited data via causal inference

  Scientific Reports · 2025-12-06

  articleOpen access1st author

  The application of machine learning in decision-making has become widespread. However, investigating the impact of interventions in decision-making tasks poses significant challenges, particularly in scenarios with limited observable data. Specifically, when interventions are introduced into the decision-making system, their effects extend beyond specific variables, creating a cascade of causal relationships that influence the entire decision-making process. Consequently, traditional decision-making methods based on static data prove insufficient and fail to consider the comprehensive data post-intervention. In response to these challenges, this paper proposes a framework for simulating interventions in decision-making tasks. By uncovering causal relationships within the dataset, the framework infers post-intervention data. Two inference methods are also designed within the framework: direct computation of weights and model-fitted weights. We employ our proposed framework and algorithms to simulate the data changes under two scenarios: PCOS Prediction and The Law School Admissions. By integrating validated knowledge, the experimental results demonstrate that our framework more realistically simulates the intervention process, providing more reliable outcomes for machine learning tasks, compared with the static decision-making with interventions.

  PublisherOA PDFDOI

• Analysis and numerical approximation for space-dependent variable-order fractional diffusion equation in the flowing media

  IMA Journal of Numerical Analysis · 2025-06-20

  article

  Abstract We investigate a fully discrete scheme for a space-dependent variable-order fractional diffusion equation in the flowing media, which can be derived by introducing a velocity field to continuous time random walk model with waiting time obeying a spatially dependent power-law distribution. We provide regularity estimates for the solution under some regularity assumptions on the variable-order αx and the velocity field v. A temporal semidiscrete scheme generated by the backward Euler convolution quadrature method is proposed, and an Oτ convergence rate is obtained by some skillful error analyses. Then, the fully discrete scheme is built by using finite element method to approximate the spatial operator, and an optimal spatial error estimate is obtained by introducing some discrete operators, i.e., the convergence order can well match the order of optimal spatial regularity of the solution. Finally, various numerical examples are presented to validate our theoretical results.

  PublisherDOI

• An Improved Diffusion Model for Remote Sensing Image Spatiotemporal Fusion

  2025-08-03

  article

  Due to the trade-off between spatial and temporal resolutions in satellite sensors, it is a challenge to conduct long-term, dynamic, and high-precision studies of observed targets. Among existing solutions, spatiotemporal fusion is one of the most practical approaches. In this paper, an improved diffusion model is proposed for spatiotemporal fusion of remote sensing image. To address the issue of noise diffusion not adapting to the characteristics of different input data, a dynamic noise intensity injection module is introduced. This module better aligns with the diverse types of data, enhancing the model’s effectiveness. In addition, considering the spectral continuity between bands in remote sensing imagery, a novel loss function is introduced. This new loss function combines the traditional mean squared error (MSE) with spectral angle mapper (SAM), enhancing the model's ability to recover high-frequency details and maintain spectral consistency. An improved diffusion model for remote sensing image spatiotemporal fusion (IDiff-STF) outperforms state-of-the-art (SOTA) methods. Furthermore, it achieves superior performance in downstream classification tasks, delivering the highest accuracy and robustness.

  PublisherDOI

Recent grants

• Control Methodologies for Fuel Cell CHP Systems Integration and Optimization

  NSF · $280k · 2005–2009

Frequent coauthors

• Ilya Kolmanovsky

  University of Michigan–Ann Arbor

  95 shared

• João P. Hespanha

  75 shared

• Thomas Parisini

  74 shared

• Magnus Egerstedt

  65 shared

• Miroslav Krstić

  University of California, San Diego

  60 shared

• Jorge Cortés

  University of California, San Diego

  59 shared

• Andrew G. Alleyne

  University of Minnesota

  59 shared

• Anna G. Stefanopoulou

  58 shared

Awards & honors

• 2018 National Academy of Inventors Fellow