AI Technology for Faster Computation and High Efficiency in Control Development

DOI

Bibliographic Information

Other Title
  • AI技術による計算高速化と制御開発の高効率化

Description

<p>In addition to the Model Based Development (MBD) process that Mazda has been working on for some time, we are also working on improving the efficiency of development through the use of machine learning. As a specific initiative, working on significantly reducing the computation time by replacing each sub-model configured within the 1D-CAE (Computer Aided Engineering) model, which can analyze the energy and thermal behavior of the entire vehicle, with AI models using Neural Network technology. In addition, to improve the efficiency of control development for increasingly complex thermal-management systems, we have developed a process to automatically construct control models that can be utilized in the conceptual design phase at high speed, utilizing reinforcement learning technology that can construct controls. Furthermore, this control development process was applied to a transient control scene of a thermal management system assuming actual BEV development, and a case study of control optimization in a short period of time will be presented.</p>

Journal

Keywords

Details 詳細情報について

  • CRID
    1390022178087830016
  • DOI
    10.34338/mazdagihou.41.0_3
  • ISSN
    21863490
    02880601
  • Text Lang
    ja
  • Data Source
    • JaLC
  • Abstract License Flag
    Disallowed

Report a problem

Back to top