Lei Gao

Lei Gao

Research Assistant

Hi, Welcome to Lei’s (磊) Page!

I am Lei Gao, a PhD candidate at Mechanical Engineering Department of Univeristy of Maryland, College Park. I’m supervised by Dr. Radermacher and Dr. Hwang @ CEEE. My PhD thesis topic is Optimum design and operation of combined cooling heating and power system under uncertainty. With the goal of achieving higher Energy efficiency, lower Economic cost and fewer Environment impact without losing Robustness (E3R), I broaden my interest in various energy related areas, including but not limited to energy demand analysis, energy system optimum design, and automation via optimization & machine learning. I occasionally collaborate with researchers with interdisciplinary knowledge outside the center.

Interests

  • Energy demand analysis
    As the premise of design and operation for energy systems, a robust, effective and precise enough demand prediction tool is the critical base of following decision-making. My interest within energy demand analysis ranges from nationwide to single building scale and hourly to annually granularity. Considering building sector takes a stable 40% share of total energy consumption from the first figure, building energy consumption would be an appropriate cutting point. Specifically, on contrary to physical model, data driven method (like machine learning) implicitly extracts features from data. The techniques in ML area have advantages of robustness, flexibility, and rapidity when applying for prediction tasks. The second figure shows the research trend focuses on ML based building energy prediction. Right now, I’m trying to implement not only advanced deep learning model for prediction but also trying to interpret performance model by model.

  • Thermal system optimum design
    A bottom-up optimization design framework is developed for thermal system optimization. With this framework, we are trying to realize end2end design, leverage more advanced cycles in terms of E3R and enable multiple objective optimization. The best thermal system considers the constraints of heat source, sink, component constrains. Contrary to conventional design stream with classic or expert recommended configuration. This framework starts with a comprehensive Configuration Optimization Framework for Energy Efficiency & Economics (COFEEE) based on graph knowledge. Here is an example of COFEEE applied for supercritical CO2 power cycle. The COFEEE is under intensively developed for more complicated case studies.

  • System operation under uncertainty
    Uncertainty issue persists around energy systems, instead of getting more accurate prediction capability, the more realistic strategy is to operate system with uncertainty information…

  • Disruptive innovations in HVAC area
    In CEEE, various innovations are developed to improve how energy is served for human beings with minimum destruction of environment. I was lucky to join in these projects when I was a junior PhD student, such as:

    • Thermoelastic cooling system test (participated in summer 16’)
    • 3D printed miniaturized heat exchanger design & test (participated from fall 16’ to spring 17’)

Education

  • Ph.D. University of Maryland, College Park, United States (22’, expected)
  • M.S. Xi’an Jiaotong University, Xi’an, China (16’)
  • B.S. Xi’an Jiaotong University, Xi’an, China (13’)

Projects

  • Microgrid system design & optimization
    (in active, independent)

  • Machine learning for flow boiling
    (in active, collaborate with research associate from NIST Dr. Lingnan Lin)

  • Low GWP & flammable refrigerant system test
    (completed, 17’, independent)

  • Compact screw compressor design and test
    (completed, 15’, independent)

Skills

  • Programming language: Python, Matlab, EES, GAMS, Julia
  • Software: AutoCAD, SolidWorks, LabVIEW, TRNSYS, EnergyPlus, CoilDesigner, Ansys, BCVTB
  • Machine Learning: GNN, deepGPR, seq2seq, CNN, LSTM, XGBoost, SVM, GPR

Publications

Under review

  • Stochastic dynamic optimal operation for combined cooling, heating and power system under uncertainty. Lei Gao, Yunho Hwang*, Reinhard Radermacher, Energy Conversion and Management
  • A comprehensive overview of configuration optimization for thermal energy conversion systems. Lei Gao, Yunho Hwang*, Reinhard Radermacher, Renewable and Sustainable Energy Review
  • A general model for flow boiling heat transfer in microfin tubes based on a new neural network architecture. Lingnan Lin*, Lei Gao, Mark A. Kedzierski, Yunho Hwang, submitted to International Journal of Heat and Mass Transfer on 2021.03.02

TAs

Others

  • Membership: ASHRAE, Association of Energy Engineers student member, Future Faculty Program Fellow (19’)
  • Reviewer: Applied Energy, Energy Conversion and Management, Energy, Sustainable Cities and Society, International Journal of Energy Research, Energy and AI
  • Certification: EPA Certified Universal Technician per Section 608 of Clean Air Act
  • Posts: here is where I post my thoughts and ideas (under construction)
Update on Aug 20, 2021