Forschung
Artificial Intelligence

Using Artificial Intelligence to Realistically Model Complex Dynamic Systems in Real-Time

Simulation driven engineering is an essential part in the development process and a key component in the digitalization of industry. Based on virtual replicas of real dynamic systems engineers can develop optimal patient specific implants, sustainable automotive technologies or light weighted long-life batteries.

However for a reliable prediction using virtual replicas accurate models and solution schemes are a prerequisite. Many mathematical models have been formulated in recent years in order to reproduce the real behavior as accurately as possible. Additionally, new solution schemes were developed in order to reduce the computational effort and to increase the range of applications. Nevertheless even with these improved approaches a strong mismatch between simulation output and real physical behavior is observed in many cases.

Artificial neural networks have the potential to close this gap.These algorithms from Machine Learning are well suited to improve the computational models by directly including experimental or onboard measured data. Hence the simulation output of the virtual replica automatically assimilates to the real physical behavior. Additionally, artificial neural networks can enable real-time simulations by solving differential equations directly without using time consuming solution schemes.

Providing these enhanced modeling concepts the digital twin concept can become reality. For each dynamic system a virtual replica is build which automatically improves by comparison with real physical data. It can be used in the design process to develop improved dynamic systems or for life cycle predictions. Furthermore the digital twin can give machines a mind allowing them to make decisions self-dependly.

Artificial Intelligence

  • Physics-Informed Data-Driven Simulation
    This project investigates to what extent simulation with neural networks on the one hand and data-based empirical modeling on the other hand can be combined in a symbiotic manner. The ultimate goal is the generation of reliable models for complex dynamical systems known as digital twins.
    Leitung: H. Wessels, P. Wriggers
    Team: H. Wessels
    Jahr: 2020
  • Using Machine Learning to Improve the Modelling of Machining and Cutting Processes
    Metal cutting is a fundamental process in industrial production. The fast and accurate on-line prediction of metal cutting processes is crucial for the Intelligent Manufacturing (IM). With the advent of high-speed computing, robust numerical algorithms and machine learning technology, computational modelling serves as a tool for not only accurate but also fast predicting the complex machining processes and understanding the complex physics. In this work, the machine learning based numerical model is developed for simulation of metal cutting processes.
    Leitung: C. Weißenfels, P. Wriggers
    Team: M.Sc. Dengpeng Huang
    Jahr: 2018
    Förderung: China Scholarship Council (CSC)

PROJECT COORDINATORS

Prof. Dr.-Ing. habil. Dr. h.c. mult. Dr.-Ing. E. h Peter Wriggers
Geschäftsführende Leitung
Adresse
An der Universität 1
30823 Garbsen
Prof. Dr.-Ing. habil. Dr. h.c. mult. Dr.-Ing. E. h Peter Wriggers
Geschäftsführende Leitung
Adresse
An der Universität 1
30823 Garbsen