MERL - Mitsubishi Electric Research Laboratories

Multi-Physical Modeling

Optimal design & robust control through multi-physical modeling.

Our work involves the development of state-of-art modeling and simulation tools for complex, heterogeneous systems. We apply these models for the optimal design and robust control of a variety of systems including HVAC systems, zero-energy buildings, automobiles, and robotic systems.

  • Quick Links

  • Researchers

  • News & Events

    •  NEWS    Ankush Chakrabarty co-organized three sessions at the ACC2023, and was nominated for Best Energy Systems Paper.
      Date: June 30, 2023 - June 2, 2023
      Where: San Diego, CA
      MERL Contact: Ankush Chakrabarty
      Research Areas: Applied Physics, Artificial Intelligence, Control, Data Analytics, Dynamical Systems, Machine Learning, Multi-Physical Modeling, Optimization, Robotics
      Brief
      • Ankush Chakrabarty (researcher, Multiphysical Systems Team) co-organized and spoke at 3 sessions at the 2023 American Control Conference in San Diego, CA. These include: (1) A tutorial session (w/ Stefano Di Cairano) on "Physics Informed Machine Learning for Modeling and Control": an effort with contributions from multiple academic institutes and US research labs; (2) An invited session on "Energy Efficiency in Smart Buildings and Cities" in which his paper (w/ Chris Laughman) on "Local Search Region Constrained Bayesian Optimization for Performance Optimization of Vapor Compression Systems" was nominated for Best Energy Systems Paper Award; and, (3) A special session on Diversity, Equity, and Inclusion to improve recruitment and retention of underrepresented groups in STEM research.
    •  
    •  TALK    [MERL Seminar Series 2023] Prof. Zoltan Nagy presents talk titled Investigating Multi-Agent Reinforcement Learning for Grid-Interactive Smart Communities using CityLearn
      Date & Time: Wednesday, March 29, 2023; 1:00 PM
      Speaker: Zoltan Nagy, The University of Texas at Austin
      MERL Host: Ankush Chakrabarty
      Research Areas: Control, Machine Learning, Multi-Physical Modeling
      Abstract
      • The decarbonization of buildings presents new challenges for the reliability of the electrical grid because of the intermittency of renewable energy sources and increase in grid load brought about by end-use electrification. To restore reliability, grid-interactive efficient buildings can provide flexibility services to the grid through demand response. Residential demand response programs are hindered by the need for manual intervention by customers. To maximize the energy flexibility potential of residential buildings, an advanced control architecture is needed. Reinforcement learning is well-suited for the control of flexible resources as it can adapt to unique building characteristics compared to expert systems. Yet, factors hindering the adoption of RL in real-world applications include its large data requirements for training, control security and generalizability. This talk will cover some of our recent work addressing these challenges. We proposed the MERLIN framework and developed a digital twin of a real-world 17-building grid-interactive residential community in CityLearn. We show that 1) independent RL-controllers for batteries improve building and district level KPIs compared to a reference RBC by tailoring their policies to individual buildings, 2) despite unique occupant behaviors, transferring the RL policy of any one of the buildings to other buildings provides comparable performance while reducing the cost of training, 3) training RL-controllers on limited temporal data that does not capture full seasonality in occupant behavior has little effect on performance. Although, the zero-net-energy (ZNE) condition of the buildings could be maintained or worsened because of controlled batteries, KPIs that are typically improved by ZNE condition (electricity price and carbon emissions) are further improved when the batteries are managed by an advanced controller.
    •  

    See All News & Events for Multi-Physical Modeling

  • Internships

    • MS1851: Dynamic Modeling and Control for Grid-Interactive Buildings

      MERL is looking for a highly motivated and qualified candidate to work on modeling for smart sustainable buildings. The ideal candidate will have a strong understanding of modeling renewable energy sources, grid-interactive buildings, occupant behavior, and dynamical systems with expertise demonstrated via, e.g., peer-reviewed publications. Hands-on programming experience with Modelica is preferred. The minimum duration of the internship is 12 weeks; start time is flexible. This internship is preferred to be onsite at MERL, but may be done remotely where you live if the COVID pandemic makes it necessary.

    • MS1958: Simulation, Control, and Optimization of Large-Scale Systems

      MERL is seeking a motivated graduate student to research numerical methods pertaining to the simulation, control, and optimization of large-scale systems. Representative applications include large vapor-compression cycles and other multiphysical systems for energy conversion that couple thermodynamic, fluid, and electrical domains. The ideal candidate would have a solid background in numerical methods, control, and optimization; strong programming skills and experience with Julia/Python/Matlab are also expected. Knowledge of the fundamental physics of thermofluid flows (e.g., thermodynamics, heat transfer, and fluid mechanics), nonlinear dynamics, or equation-oriented languages (Modelica, gPROMS) is a plus. The expected duration of this internship is 3 months.

    • MS2012: Residual Model Learning for Building Energy Systems

      MERL is looking for a highly motivated and qualified candidate to work on learning residual dynamics to augment ODE/DAE-based models of building energy systems. The ideal candidate will have a strong understanding of system identification, optimization, machine learning and/or function approximation; additional understanding of energy systems is a plus. Hands-on programming experience with numerical optimization solvers and Python is preferred; experience with Modelica/FMUs is a plus. PhD students are strongly preferred, as an expected outcome of the internship is a publication in a high-tier venue. The minimum duration of the internship is 12 weeks; start time is flexible.


    See All Internships for Multi-Physical Modeling

  • Recent Publications

    •  Chinchilla, R., Deshpande, V.M., Chakrabarty, A., Laughman, C.R., "Learning Residual Dynamics via Physics-Augmented Neural Networks: Application to Vapor Compression Cycles", American Control Conference (ACC), May 2023.
      BibTeX TR2023-051 PDF
      • @inproceedings{Chinchilla2023may,
      • author = {Chinchilla, Raphael and Deshpande, Vedang M. and Chakrabarty, Ankush and Laughman, Christopher R.},
      • title = {Learning Residual Dynamics via Physics-Augmented Neural Networks: Application to Vapor Compression Cycles},
      • booktitle = {American Control Conference (ACC)},
      • year = 2023,
      • month = may,
      • url = {https://www.merl.com/publications/TR2023-051}
      • }
    •  Paulson, J.A., Sorouifar, F., Laughman, C.R., Chakrabarty, A., "LSR-BO: Local Search Region Constrained Bayesian Optimization for Performance Optimization of Vapor Compression Systems", American Control Conference (ACC), May 2023.
      BibTeX TR2023-057 PDF
      • @inproceedings{Paulson2023may,
      • author = {Paulson, Joel A. and Sorouifar, Farshud and Laughman, Christopher R. and Chakrabarty, Ankush},
      • title = {LSR-BO: Local Search Region Constrained Bayesian Optimization for Performance Optimization of Vapor Compression Systems},
      • booktitle = {American Control Conference (ACC)},
      • year = 2023,
      • month = may,
      • url = {https://www.merl.com/publications/TR2023-057}
      • }
    •  Sakamoto, Y., Xu, Y., Wang, B., Yamamoto, T., Nishimura, Y., "Electric Motor Surrogate Model Combining Subdomain Method and Neural Network", Conference on the Computation of Electromagnetic Fields (COMPUMAG), May 2023.
      BibTeX TR2023-041 PDF
      • @inproceedings{Sakamoto2023may2,
      • author = {Sakamoto, Yusuke and Xu, Yihao and Wang, Bingnan and Yamamoto, Tatsuya and Nishimura, Yuki},
      • title = {Electric Motor Surrogate Model Combining Subdomain Method and Neural Network},
      • booktitle = {Conference on the Computation of Electromagnetic Fields (COMPUMAG)},
      • year = 2023,
      • month = may,
      • url = {https://www.merl.com/publications/TR2023-041}
      • }
    •  Xu, Y., Wang, B., Sakamoto, Y., Yamamoto, T., Nishimura, Y., "Comparison of Learning-based Surrogate Models for Electric Motors", Conference on the Computation of Electromagnetic Fields (COMPUMAG), May 2023.
      BibTeX TR2023-042 PDF
      • @inproceedings{Xu2023may,
      • author = {Xu, Yihao and Wang, Bingnan and Sakamoto, Yusuke and Yamamoto, Tatsuya and Nishimura, Yuki},
      • title = {Comparison of Learning-based Surrogate Models for Electric Motors},
      • booktitle = {Conference on the Computation of Electromagnetic Fields (COMPUMAG)},
      • year = 2023,
      • month = may,
      • url = {https://www.merl.com/publications/TR2023-042}
      • }
    •  Anantaram, V., Goldsmith, A., Wang, Y., "Flying Start of Sensorless Synchronous Machines with Reactive Power Injection", IEEE International Electric Machines and Drives Conference (IEMDC), May 2023.
      BibTeX TR2023-037 PDF
      • @inproceedings{Anantaram2023may,
      • author = {Anantaram, Varatharajan and Goldsmith, Abraham and Wang, Yebin},
      • title = {Flying Start of Sensorless Synchronous Machines with Reactive Power Injection},
      • booktitle = {IEEE International Electric Machines and Drives Conference (IEMDC)},
      • year = 2023,
      • month = may,
      • url = {https://www.merl.com/publications/TR2023-037}
      • }
    •  Liu, D., Varatharajan, A., Goldsmith, A., Kong, C., Sigatapu, L., Wang, Y., "Broken-bar Fault Detection by Injecting a Frequency Modulated Continuous Wave Signal", IEEE International Electric Machines and Drives Conference (IEMDC), May 2023.
      BibTeX TR2023-039 PDF
      • @inproceedings{Liu2023may,
      • author = {Liu, Dehong and Varatharajan, Anantaram and Goldsmith, Abraham and Kong, Chuizheng and Sigatapu, Laxman and Wang, Yebin},
      • title = {Broken-bar Fault Detection by Injecting a Frequency Modulated Continuous Wave Signal},
      • booktitle = {IEEE International Electric Machines and Drives Conference (IEMDC)},
      • year = 2023,
      • month = may,
      • url = {https://www.merl.com/publications/TR2023-039}
      • }
    •  Sakamoto, Y., Xu, Y., Wang, B., Yamamoto, T., Nishimura, Y., "Multi-Objective Motor Design Optimization with Physics-Assisted Neural Network Model", IEEE International Electric Machines and Drives Conference (IEMDC), May 2023.
      BibTeX TR2023-038 PDF
      • @inproceedings{Sakamoto2023may,
      • author = {Sakamoto, Yusuke and Xu, Yihao and Wang, Bingnan and Yamamoto, Tatsuya and Nishimura, Yuki},
      • title = {Multi-Objective Motor Design Optimization with Physics-Assisted Neural Network Model},
      • booktitle = {IEEE International Electric Machines and Drives Conference (IEMDC)},
      • year = 2023,
      • month = may,
      • url = {https://www.merl.com/publications/TR2023-038}
      • }
    •  Xu, Y., Wang, B., Sakamoto, Y., Yamamoto, T., Nishimura, Y., Koike-Akino, T., Wang, Y., "Tandem Neural Networks for Electric Machine Inverse Design", IEEE International Electric Machines and Drives Conference (IEMDC), May 2023.
      BibTeX TR2023-040 PDF
      • @inproceedings{Xu2023may2,
      • author = {Xu, Yihao and Wang, Bingnan and Sakamoto, Yusuke and Yamamoto, Tatsuya and Nishimura, Yuki and Koike-Akino, Toshiaki and Wang, Ye},
      • title = {Tandem Neural Networks for Electric Machine Inverse Design},
      • booktitle = {IEEE International Electric Machines and Drives Conference (IEMDC)},
      • year = 2023,
      • month = may,
      • url = {https://www.merl.com/publications/TR2023-040}
      • }
    See All Publications for Multi-Physical Modeling

  • Videos