Hybrid Analytics for Sim2real Models
This talk will cover methods to create fast and precise predictive models for industrial assets with limited data, useful for control design, XiL models, and digital twins.
-
-
학생용 무료 소프트웨어에 액세스하기
차세대 엔지니어에게 힘을 실어주는 Ansys
학생들은 세계적 수준의 시뮬레이션 소프트웨어를 무료로 이용할 수 있습니다.
-
지금 바로 Ansys에 연결하십시오!
미래를 설계하기
시뮬레이션이 다음 혁신을 어떻게 지원할 수 있는지 알아보려면 Ansys와 연결하십시오.
국가
무료 트라이얼
제품 및 서비스
학습하기
회사 정보
Back
제품 및 서비스
Date/Time:
February 17, 2026
11 AM EST
Venue:
Virtual
Register Now
Overview
In this talk, we will describe methods to develop predictive and accurate models for industrial assets in the regime of small measurement data. These models are fast, precise, and valuable for plant models for control design, XiL models, and digital twins.
Models are built by using machine learning techniques that augment measurements with simulations to learn system behavior from limited data. We demonstrate that such hybrid models exhibit superior extrapolation and learning efficiency, as illustrated by relevant case studies.
What Attendees Will Learn
- Efficient ways to build reduced order models from simulations and measurement
- Combine simulations and measurement to build data efficient models
- See successful use cases from various industries
Who Should Attend
- Computational engineers, mathematical modelers, control engineer, and system modelers.
Speakers
- Srinivasa Mohan - Distinguished Engineer, Ansys
