Commenced in January 2007
Frequency: Monthly
Edition: International
Paper Count: 2

genetic optimization Related Abstracts

2 A Genetic-Neural-Network Modeling Approach for Self-Heating in GaN High Electron Mobility Transistors

Authors: Anwar Jarndal


In this paper, a genetic-neural-network (GNN) based large-signal model for GaN HEMTs is presented along with its parameters extraction procedure. The model is easy to construct and implement in CAD software and requires only DC and S-parameter measurements. An improved decomposition technique is used to model self-heating effect. Two GNN models are constructed to simulate isothermal drain current and power dissipation, respectively. The two model are then composed to simulate the drain current. The modeling procedure was applied to a packaged GaN-on-Si HEMT and the developed model is validated by comparing its large-signal simulation with measured data. A very good agreement between the simulation and measurement is obtained.

Keywords: Neural Networks, GaN HEMT, computer-aided design and modeling, genetic optimization

Procedia PDF Downloads 256
1 An Application of Meta-Modeling Methods for Surrogating Lateral Dynamics Simulation in Layout-Optimization for Electric Drivetrains

Authors: Christian Angerer, Markus Lienkamp


Electric vehicles offer a high variety of possible drivetrain topologies with up to 4 motors. Multi-motor-designs can have several advantages regarding traction, vehicle dynamics, safety and even efficiency. With a rising number of motors, the whole drivetrain becomes more complex. All permutations of gearings, drivetrain-layouts, motor-types and –sizes lead up in a very large solution space. Single elements of this solution space can be analyzed by simulation methods. In addition to longitudinal vehicle behavior, which most optimization-approaches are restricted to, also lateral dynamics are important for vehicle dynamics, stability and efficiency. In order to compete large solution spaces and to find an optimal result, genetic algorithm based optimization is state-of-the-art. As lateral dynamics simulation is way more CPU-intensive, optimization takes much more time than in case of longitudinal-only simulation. Therefore, this paper shows an approach how to create meta-models from a 14-degree of freedom vehicle model in order to enable a numerically efficient drivetrain-layout optimization process under consideration of lateral dynamics. Different meta-modelling approaches such as neural networks or DoE are implemented and comparatively discussed.

Keywords: meta-modeling, genetic optimization, driving dynamics, drivetrain layout, lateral dynamicx

Procedia PDF Downloads 261