About

Yongzhi Qu is an Assistant Professor at the University of Utah and the Principal Investigator of the Utah Lab of Artificial Intelligence Powered Systems. He earned his Ph.D. from the University of Illinois Chicago in 2014. His research group focuses on scientific machine learning, emphasizing the integration of applied mathematics and programming skills. Professor Qu is actively seeking motivated graduate students with strong backgrounds in these areas to join his research efforts.

Research topics

• Computer Science
• Artificial Intelligence
• Data Mining
• Engineering
• Machine Learning
• Automotive engineering
• Marine engineering
• Composite material
• Algorithm
• Environmental science
• Petroleum engineering
• Materials science
• Mathematics
• Mechanical engineering
• Process engineering

Selected publications

• Spindle performance monitoring via accelerometer measurements in data-driven models

  Procedia CIRP · 2026-01-01

  articleOpen accessSenior author

  The future of manufacturing depends on transitioning traditional machines into intelligent machine tools that can monitor and control themselves. As the spindle is an essential component of machine tools, the performance of machine tool spindles should be tracked for quality control. For example, smart spindles could be equipped with accelerometers for monitoring the spindle performance via models that relate the measured accelerations to the spindle error motions. Various data-driven models were created that estimate spindle-related displacements from on-machine accelerations. The estimated displacements were compared, revealing the advantages and disadvantages of each model to monitor the spindle performance.

  PublisherDOI

• Fredholm integral equations neural operator (FIE-NO) for data-driven boundary value problems

  Machine Learning Engineering · 2026-02-24

  articleOpen accessSenior author
  PublisherDOI

• Acceleration-based spindle monitoring based on geometric error motions

  Procedia CIRP · 2025-01-01 · 1 citations

  articleOpen accessSenior author

  The development of intelligent machine tools requires integrating sensors that allow the machine to monitor, diagnose, and control the machining process. However, existing sensors for monitoring tool vibrations are intrusive and difficult to integrate in a production machining system. This paper describes how spindle-mounted accelerometers can be used to model spindle-speed dependent error motions in situ . It extends on previous work by applying a geometric model for interpreting tool holder displacements and spindle error motions from the motion of a laser that is coaxially mounted in a tool holder, providing an accurate and interpretable basis for training models that output geometric error motions via acceleration inputs. Trained data-driven models show the potential for use of spindle-mounted accelerometers to accurately estimate geometric error motions, even during spindle rotation with radial loads.

  PublisherDOI

• State space neural network with nonlinear physics for mechanical system modeling

  Reliability Engineering & System Safety · 2025-02-20 · 11 citations

  articleSenior authorCorresponding
  PublisherDOI

• Novel 3D Sensing Framework for Safety Monitoring in Human-Robot Collaboration Work Cells

  Annual Conference of the PHM Society · 2025-10-26

  articleOpen access

  The demand for work safety protection in Human-Robot Interaction (HRI) work cells is rapidly increasing, driven by the projected 34.3% Compound Annual Growth Rate (CAGR) of the global Collaborative Robot (Cobot) market from 2020 to 2030 [1]. According to IRF-World Robotics 2023, it is reported that there are nearly 4 million industrial robots in operation worldwide, with approximately 10% of them being cobot [2]. A NIOSH report highlighted 61 robot-related fatalities between 1992 and 2015, with an expectation of further rising due to the increasing use of industrial robots and cobots in the US work environment [3]. A recent study in [4] delved into 355 robot accidents documented by KOSHA between 2009 and 2019, revealing that 95% occurred in manufacturing businesses. Pinch and crush incidents accounted for 52% of the accidents, while impacts and collisions accounted for 36%, and the remaining 12% involved falls, flying objects, trips/slips, cuts, burns, etc. These findings align with US data reported in [5].The rising integration of cobot units among major manufacturers emphasizes the critical need for enhancing cobot safety in manufacturing. Owing to safety considerations and regulatory requirements, existing cobots frequently operate at significantly reduced speeds and are restricted from undertaking complex interaction tasks in shared workspace. This limitation has curtailed the full potential utilization and productivity of cobots in manufacturing. This paper introduces a novel 3D sensing framework designed to address these limitations by enabling safety assurance in workspaces requiring close human-robot interaction. The framework generates 3D human pose information and relays it to the robot for real-time safety monitoring. Our methodology begins with data collection from a single RGB-D camera capturing human-robot interactions in a manufacturing environment. Human shape and pose are predicted using deep neural networks, which then incorporate depth information and undergo 3D geometric transformations to deduce size, shape, and translation. This process produces a reconstructed 3D avatar with pose, size, and location. Following 3D human posture estimation, this data is then integrated into a virtual environment with a real robot for real-time monitoring. Results demonstrate successful reconstruction of 3D human geometry within human-robot collaboration settings. By integrating both the reconstructed mesh and real-time robot state into a unified virtual environment, we achieved real-time, offline, continuous monitoring of the critical distance between robot and human throughout operation. These distance measurements provide crucial data for developing collision detection, prediction, and avoidance capabilities when incorporated into the robot control feedback loop.

  PublisherOA PDFDOI

• Perspective: revisiting surface roughness in electrochemical machining and the paradoxes

  Surface Science and Technology · 2024-06-06 · 4 citations

  articleOpen access

  Abstract Electrochemical machining (ECM) represents a prominent electrochemistry-driven technique for surface flattening, post-processing, and (ultra-)precision machining, attracting considerable research interests recently. The method exhibits advantages in the machining of hard-to-machine nickel (Ni) superalloys, particularly those created via additive manufacturing approaches such as laser powder bed fusion (LPBF), in which enhanced microstructural features and mechanical properties are achieved with compromised surface quality. This study explores the intricate relationship between Ni alloy-specific microstructures, such as carbide precipitates, and the principles of electrochemistry integral to ECM. It further emphasizes the emerging requirement to re-examine the surface quality outcomes of ECM. We present a concise overview of the inherent paradoxes in ECM, encompassing the prediction of surface roughness range, the quantification of charge transfer coefficients, the efficiency of material removal, and the temporal dependence of the ECM process. These paradoxes necessitate systematic experimental and theoretical research to advance our understanding, and we wish to welcome, stimulate, and urge more raised awareness and attention to this matter about ECM surface quality control and prediction.

  PublisherOA PDFDOI

• Transfer Operator Learning with Fusion Frame

  arXiv (Cornell University) · 2024-08-20

  preprintOpen accessSenior author

  The challenge of applying learned knowledge from one domain to solve problems in another related but distinct domain, known as transfer learning, is fundamental in operator learning models that solve Partial Differential Equations (PDEs). These current models often struggle with generalization across different tasks and datasets, limiting their applicability in diverse scientific and engineering disciplines. This work presents a novel framework that enhances the transfer learning capabilities of operator learning models for solving Partial Differential Equations (PDEs) through the integration of fusion frame theory with the Proper Orthogonal Decomposition (POD)-enhanced Deep Operator Network (DeepONet). We introduce an innovative architecture that combines fusion frames with POD-DeepONet, demonstrating superior performance across various PDEs in our experimental analysis. Our framework addresses the critical challenge of transfer learning in operator learning models, paving the way for adaptable and efficient solutions across a wide range of scientific and engineering applications.

  PublisherOA PDFDOI

• Physics-Informed Multi-Task Learning for Material Removal Rate Prediction in Semiconductor Chemical Mechanical Planarization

  2024-06-17 · 8 citations

  articleSenior author

  The chemical mechanical planarization (CMP) process is a complex and critical operation in the semiconductor manufacturing industry, involving a wide variety of process parameters. As a universally employed technique in semiconductor fabrication, it has received extensive research attention over the past several decades. Despite the development of various physics-based and data-driven models aimed at assessing the quality and productivity of CMP tools, particularly through the metric of material removal rate (MRR), a universally accepted prediction method has yet to emerge, primarily due to the complicated chemical/physical process. To help remedy this situation, this paper explores multi-task learning and the integration of physics-based knowledge constraints into convolutional neural networks (CNNs) to enhance the predictive accuracy of CNNs. The hybrid approach aims to leverage the strengths of both data-driven models and physics-based relationships, offering a robust framework for predicting MRRs with higher confidence. The validation results show that the auxiliary tasks in multi-task learning improve the accuracy of the primary task. Also, the physics-informed strategy can clearly increase the generalization capability of the baseline CNN model.

  PublisherDOI

• A novel fiber Bragg grating-based smart clamp with macro strain measurement: design, modeling, and application to incipient looseness detection

  Structural Health Monitoring · 2024-05-24 · 3 citations

  article

  Ensuring the safety and optimal performance of hydraulic pipelines, particularly in aircraft, is of paramount importance. The challenge of detecting incipient looseness in fasteners within a time-varying temperature environment has garnered widespread recognition, making it a focal point in the field of structural health monitoring. In this study, we propose an innovative solution—a smart clamp based on Fiber Bragg Grating (FBG). This clamp aims to detect incipient looseness in hydraulic pipe systems by measuring its macro strain. Notable advantages include its ability to avoid the chirping of the FBG, the ease of building a distributed sensing network, especially for the blind-hole connected structure, and its applicability for monitoring clamps connected with small-diameter bolt. The analytical model of the clamp under the preload of the connected bolt, along with the unique condition for the measurement, is provided. Demonstrating a force resolution <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"> <mml:mrow> <mml:mn>0</mml:mn> <mml:mo>.</mml:mo> <mml:mn>1</mml:mn> <mml:mi mathvariant="normal">kN</mml:mi> </mml:mrow> </mml:math> , the clamp exhibits a strong capability in detecting incipient looseness. Furthermore, environmental temperature interference is mitigated by configuring two FBGs. Conducting application experiments in a hydraulic system, the results indicate that the proposed clamp can effectively detect the loosening and tightening processes in real time, even under the time-varying temperature environment.

  PublisherDOI

• Fredholm Integral Equations Neural Operator (FIE-NO) for Data-Driven Boundary Value Problems

  arXiv (Cornell University) · 2024-08-20

  preprintOpen accessSenior author

  In this paper, we present a novel Fredholm Integral Equation Neural Operator (FIE-NO) method, an integration of Random Fourier Features and Fredholm Integral Equations (FIE) into the deep learning framework, tailored for solving data-driven Boundary Value Problems (BVPs) with irregular boundaries. Unlike traditional computational approaches that struggle with the computational intensity and complexity of such problems, our method offers a robust, efficient, and accurate solution mechanism, using a physics inspired design of the learning structure. We demonstrate that the proposed physics-guided operator learning method (FIE-NO) achieves superior performance in addressing BVPs. Notably, our approach can generalize across multiple scenarios, including those with unknown equation forms and intricate boundary shapes, after being trained only on one boundary condition. Experimental validation demonstrates that the FIE-NO method performs well in simulated examples, including Darcy flow equation and typical partial differential equations such as the Laplace and Helmholtz equations. The proposed method exhibits robust performance across different boundary conditions. Experimental results indicate that FIE-NO achieves higher accuracy and stability compared to other methods when addressing complex boundary value problems with varying numbers of interior points.

  PublisherOA PDFDOI

Frequent coauthors

• David He

  42 shared

• Zude Zhou

  Wuhan University of Technology

  24 shared

• Liu Hong

  Wuhan University of Technology

  17 shared

• Eric Bechhoefer

  14 shared

• Zechao Wang

  Chinese University of Hong Kong

  14 shared

• Yuegang Tan

  Wuhan University of Technology

  13 shared

• Junda Zhu

  11 shared

• Xueyi Li

  11 shared

Labs

• Utah Lab of Artificial Intelligence Powered SystemsPI

Education

• Ph.D, Department of Mechanical and Industrial Engineering

  University of Illinois at Chicago

  2014