7 Purdue Engineering faculty earn NSF CAREER awards

Seven Purdue Engineering faculty members have received National Science Foundation Faculty Early Career Development (CAREER) Program awards to fund projects beginning from September 2024 to August 2025.
This highly respected award provides a grant to early-career faculty who have the potential to serve as academic role models in research and education and to lead advances in the mission of their department or organization.
“Prestigious awards for faculty near the start of their academic appointments, such as the NSF CAREER awards, are an important way the broader engineering community recognizes the potential we see in our people,” said David Bahr, senior associate dean of faculty for Purdue’s College of Engineering. “We’re very excited that seven of our colleagues have received NSF CAREER recognition, which helps faculty reach even higher.”
Jie Cai
Associate Professor of Mechanical Engineering
Project: CAS- Climate: An altruistic game theoretic framework to characterize environmental responsiveness of residential electricity consumption
Total Intended Award: $507,978
Start Date: Dec. 1, 2024
Estimated End Date: June 30, 2028
Challenge/Opportunity: For most electricity markets in the U.S., the marginal cost and the emission intensity of electricity generation exhibit opposite diurnal trends: during peak demand hours, the electricity price is high while the emission rate is low due to the operation of costly but less polluting natural gas “peaker” plants. Energy consumers may respond to the conflicting price and emission signals very differently. Understanding the behavioral variety in energy use and its impact on sustainability engineering of the electrical infrastructure is of critical importance to accelerating global clean energy.
Solution: This project will establish an altruistic game theoretic framework to understand the interplay of an individual’s financial and environmental goals in shaping energy use behaviors and to evaluate the effect of behavioral variety on system-level performances of the electric grid. The game framework models each energy consumer as a partially altruistic entity whose perceived cost is a weighted sum of the direct electricity cost and the social cost of energy-related emission. The weighting factor characterizes an individual’s valuation of energy-related emission, which in turn influences energy use behaviors.
Customer behavioral models will be developed from data collected through online human subject tests with a custom-designed demand response game, as well as sociotechnical experiments in a multifamily apartment complex in Indiana.
Project goals include generating new knowledge of residential customers’ valuation of energy-related emission and its impact on energy use behaviors; developing a statistical behavioral model characterizing how energy-related altruism correlates with socio-demographic variables; synthesizing learning-based model predictive control strategies to enable automated demand response; establishing an altruistic game theoretic framework to ease analyzing effects of behavioral variety on financial and environmental performances of the electric grid; and designing distributed and privacy-preserving Nash equilibrium solution algorithms to accommodate distributed decision making for flexible electrical load control.
Kirsten Davis
Assistant Professor of Engineering Education
Project: Characterizing the Process of Global Formation for Engineers
Total Intended Award: $650,921
Start Date: Aug 1, 2025
Estimated End Date: July 31, 2030
Challenge/Opportunity: Addressing complex and global issues often requires the collaboration of teams of engineers from across cultural and national boundaries. However, while international and intercultural teamwork can enhance innovation, participants must have the skills necessary to support such collaboration.
Solution: This project will explore how engineering students develop the skills needed to function in a globalized engineering work environment. The project will characterize global engineering competency (GEC) development by studying how this learning occurs in many contexts, including programs involving travel abroad and experiences without travel. Understanding this learning process will support development of curricular and co-curricular experiences that help all engineering students acquire these essential skills.
A longitudinal video reflection method will be used to collect weekly video reflections from 50 engineering students participating in four categories of experiences: study abroad, global courses at home, internship abroad, and internship at home. These reflections will be analyzed to create a framework describing GEC development.
In addition, the project will produce two online educational modules that can enhance experiential learning which will be made publicly accessible online. All study participants will complete the two modules – Reflecting on Experiential Learning and Connecting Culture and Engineering – before their intercultural experiences. Two workshops and a webinar also will be presented to various faculty audiences to provide training in how to implement these modules in different educational contexts.
The research and educational activities will contribute theoretical understanding and practical tools that enhance our ability to educate engineers who can work on teams to address the global challenges facing the world today.
Woongkul Lee
Assistant Professor of Electrical and Computer Engineering
Project: Magnetically Integrated Electric Drive with Rare-Earth-Free Motors
Total Intended Award: $542,129
Start Date: Oct. 1, 2024
Estimated End Date: Feb. 28, 2029
Challenge/Opportunity: Electric motors and generators play a crucial role in various industries, facilitating the widespread adoption of electric vehicles, enhancing industrial productivity, and harnessing clean energy sources. Nonetheless, existing technologies depend heavily on costly rare-earth permanent magnets, which present significant obstacles to achieving sustainability engineering objectives and promoting electrification efforts.
The demand for rare-earth permanent magnets, such as neodymium iron boron (NdFeB), is expected to surge more than 20 times to fulfill sustainable energy and transportation goals by 2050. The projected demand for rare-earth magnets in offshore wind turbines and electric vehicles will reach 36.3% (273.7 kt) and 35.3% (266 kt), respectively, which is more than 70% of the total demand. Therefore, identifying motor drives that are free from rare-earth elements yet maintain high-performance and efficiency is a significant area of research.
Solution: This project aims to develop a magnetically integrated electric drive system with a topologically optimized magnet-free and brushless wound-field flux-switching (WFFS) motor, eliminating the need for expensive rare-earth permanent magnets. The research involves optimizing the motor design to avoid inefficient magnetic flux paths, co-designing the motor and the inverter for high torque density and thermal performance, and accurately quantifying motor and drive losses for advanced control technique development.
Research outcomes will have broader implications for many sectors, including electric vehicles, renewable energy systems, more electric aircraft, and industrial processes, where electric drives are vital components. By providing an alternative to rare-earth magnet-based drives, the project seeks to revolutionize industry and contribute to development of sustainable and high-power density electric drives.
Additionally, the research addresses environmental concerns and supply chain challenges associated with rare-earth materials, promoting sustainability engineering in the transportation and clean energy sectors. The research results will contribute to the body of knowledge in the field and have a lasting impact on the transition to cleaner energy sources and transportation systems.
Can Li
Assistant Professor of Chemical Engineering
Project: Novel Neural Network Architectures Inspired by Optimization Algorithms
Total Intended Award: $500,000
Start Date: May 1, 2025
Estimated End Date: April 30, 2030
Challenge/Opportunity: Machine learning, particularly through deep neural networks, has revolutionized society, transforming such fields as computer vision and natural language processing and profoundly impacting daily life. In chemical engineering and process systems engineering (PSE), neural networks have contributed significantly across various scales – from designing new molecules for drug discovery to simulating chemical plant operations. However, the “black box” nature of these models makes them difficult to interpret, often resulting in outputs that do not adhere to essential physical laws. This lack of reliability is especially concerning in safety-critical applications like process design and control, in which compliance with strict physical constraints is crucial.
To address this issue, physics-informed neural networks (PINNs) have emerged, embedding physical laws within neural networks to increase their accuracy and reliability. Still, PINNs have limitations, particularly in enforcing nonlinear and logical constraints that are common in PSE.
Solution: This project proposes developing optimization-inspired neural networks (OINNs), a new class of neural architectures that integrate optimization principles to rigorously enforce physical and logical constraints. This approach not only improves model reliability but also aims to broaden the utility of machine learning in engineering and beyond. In addition to scientific advancements, the project includes educational initiatives to equip the next generation of chemical engineers with foundational AI and optimization skills.
As principal investigator, Li will redesign an existing Purdue data science course, emphasizing AI's role in engineering and integrating machine learning with core chemical engineering principles. He will recruit undergraduate researchers through the Research Experiences for Undergraduates (REU) program. Li also will engage K-12 educators and students through outreach programs, including development of a video game, seeking to widen public understanding of engineering and inspire young learners to explore STEM fields.
OINNs developed will incorporate strict, physically meaningful constraints directly within their architecture, reducing limitations of PINNS. The OINN architecture integrates optimization theory, enabling stringent adherence to linear and nonlinear constraints, as well as embedding domain-specific knowledge and accommodating uncertainties in parameters. To achieve these goals, the research will leverage optimization techniques from linear programming, conic programming, and mixed-integer programming to build layers representing various physical and logical constraints. The OINN framework will be benchmarked against PINNs and other models in process design, process control, and chemical structure prediction – assessing improvements in prediction accuracy, interpretability, and constraint satisfaction. The project aims to establish a new approach for reliable and explainable machine learning models in PSE.
Joseph Makin
Assistant Professor of Electrical and Computer Engineering
Project: Brain-Machine Interfaces for Speech
Total Intended Award: $600,932
Start Date: Oct. 1, 2024
Estimated End Date: Sept. 30, 2029
Challenge/Opportunity: Decoding what a person intends to say, through analysis of electrical signals recorded directly from the brain, has transformative potential to restore speech communication ability to those who have lost it. Neural speech decoders have been deployed along these lines in academic studies, but they are not yet good enough to displace noninvasive, low-tech alternatives. Extremely high accuracy has been achieved only at the price of task complexity or generality.
Solution: The project aims to soften this trade-off considerably by collecting much more data and using it better. These improvements will require integration of recent ideas in machine learning, efficient experimental design, and data collection from a large volunteer pool. The ultimate goal is to enable implantable decoders for restoring speech to those who have lost it through ALS, stroke, or other traumatic brain injury.
Modern machine learning relies on techniques (artificial neural networks) that scale very favorably with the amount of available training data, but intracranial recording (iEEG) data are scarce. The project is organized around a set of methods for increasing the amount of effective training data. One major focus is pretraining, including self-supervised learning – for example, training models to solve "pretext" tasks involving iEEG data but no annotations, speech or otherwise; generative models for speech audio, trained entirely on audio waveforms, for iEEG-to-audio; and large language models (trained entirely on text) for iEEG-to-text. A second key focus is transfer learning across multiple subjects – training a single model on multiple subjects' data.
At least two dozen participants from several collaborating hospitals and research groups are expected to provide data. The project also includes an educational component intended to create an interdisciplinary workforce in neuroscience and machine learning, and to make the problem of algorithm design for brain-machine interfaces accessible to a wider audience.
Caitlin Proctor
Assistant Professor of Agricultural and Biological Engineering & Sustainability Engineering and Environmental Engineering
Project: Designing stable biofilms for drinking water systems
Total Intended Award: $500,000
Start Date: Aug. 1, 2025
Estimated End Date: July 31, 2030
Challenge/Opportunity: Drinking water is not sterile. Many microbes thrive in biofilms along water pipe walls. While most of these microbes are harmless, some are pathogens that can cause respiratory infections estimated to cost more than $2.4 billion in healthcare expenses each year. Water distributors and building owners try to kill biofilm bacteria with disinfectants and high temperatures, but biofilms are resilient.
Current strategies to control these pathogens in plumbing systems, if effective, are unsustainable, with high energy, maintenance, and material costs. Even with disinfectant, biofilm formation in plumbing is largely inevitable. New strategies are required to control these biofilms and reduce exposure to pathogens from drinking water. An approach embracing biology can be more sustainable and provide greater stability in plumbing systems.
Solution: The project goal is to engineer a biofilm that promotes good bacteria that can outcompete unwanted bacteria. Working with biofilms, instead of against them, will lead to more stable drinking water quality. The newly developed biofilm also will be more sustainable, with lower energy and material costs. Understanding more about biofilms in water pipes will enhance guidance on making drinking water safe. Educational and outreach activities will strengthen the nation’s STEM workforce and empower the public to take control of their drinking water quality.
This project will engineer drinking water biofilms to have long-term biological stability and resistance to invasion, especially by pathogens. The scalable approach will establish a competitive model microbial consortium specific to drinking water. At the bench scale, coupon experiments and a cell printing method will be used to control initial biofilm colonization. A unique plumbing testing facility will be used to assess long-term biofilm stability at real scale. Finally, the same facility and pure culture experiments will be used to explore and validate models of biofilm-water interactions.
This approach will consider the unique reality of plumbing, with high surface area to volume ratios and intermittent flow. Using a community ecology framework will broaden understanding of dispersal and selection processes. Educational activities will teach the next generation of public health, biological engineering and environmental engineering students. These efforts also will inform the public about drinking water and biofilm engineering, increasing scientific literacy.
Junjie Qin
Assistant Professor of Electrical and Computer Engineering
Project: Towards Grid-Responsive Electrified Transportation Systems: Modeling, Aggregation, and Market Integration
Total Intended Award: $500,000
Start Date: September 1, 2024
Estimated End Date: August 31, 2029
Challenge/Opportunity: The massive trend of transportation electrification presents new challenges and opportunities at the nexus of electric power systems and transportation systems. Making the electrified transportation system grid-responsive can unlock significant economic value by monetizing its spatiotemporal charging flexibility for grid services.
Solution: This project aims to establish the conceptual and algorithmic bedrock for grid-responsive electrified transportation systems. The research will facilitate transportation electrification, support renewable integration, and speed up the sustainable transition of transportation and electricity systems. These goals will be achieved by developing novel concepts and models for coupled power and transportation systems, as well as a comprehensive algorithmic toolkit to bring the spatiotemporal flexibility in electric vehicle (EV) charging loads to electricity markets.
The intellectual merits of the project include advancing knowledge at the interface of power systems and transportation engineering, by integrating expertise in foundational disciplines of control theory, systems science, optimization (convex and nonconvex), game theory, economics, and statistical learning. The broader impacts include contributing to better matching charging loads with renewable generation while creating new revenue streams for EV owners; spurring wider adoption of EVs without subsidy while defining new entrepreneurial opportunities in charging aggregation for diverse grid services; training a generation of engineers at the intersection of power engineering, transportation engineering and systems engineering; and generating multiple engagement opportunities for K-12 and adult education students.
Realizing the vision of grid-responsive electrified transportation systems requires deep integration of expertise in power and transportation systems. This project will make original contributions to this emerging interdisciplinary field by examining fundamental couplings between transportation and power systems introduced by transportation electrification, uncovering their implications, and developing unified models for the coupled systems; devising computationally efficient algorithms to aggregate and control spatiotemporal flexibility of a large EV fleet, and identifying standardized representations of the resulting aggregate flexibility; and exploring ways to integrate charging flexibility into transmission- and distribution-level electricity markets.