Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2408.03674 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910559095488512 |
|---|---|
| author | Most, Thomas Krenz, Peter Lampert, Ralf |
| author_facet | Most, Thomas Krenz, Peter Lampert, Ralf |
| contents | Antennas are more prevalent than ever enabling 5G connectivity for wide ranging applications like cellular communication, IoT, autonomous vehicles, etc. Optimizing an antenna design can be challenging and employing traditional optimization techniques oftentimes require evaluating ~100s of design variations, which is time prohibitive to tight design deadlines. Instead, the common optimization approach relies on engineering expertise and solving only a handful design variations to manually determine a design with good-enough performance. Here, we present a novel approach for antenna optimization, which utilizes analytical derivatives calculated by Ansys HFSS to dramatically reduce the number of necessary simulation runs. In the presented approach a local updating scheme, which converges quickly to the next local optimum, is extended to a global search strategy intended to detect several local and global optima. This global search procedure uses a consistent interpolation model to approximate the simulation model outputs and its derivatives depending on the design parameters values. The presented approach is implemented in an optimization workflow, which automatically interacts with the HFSS simulation model. Since this optimizer relies on the analytical derivatives calculated by HFSS, it is applicable to any design optimization where geometry or material properties are varied, and the device response is described by SYZ-parameters or far-field quantities. The usability of this new optimizer is therefore wide ranging from antenna design to signal integrity applications. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_03674 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | A global optimization approach for antenna design using analytical derivatives from high-frequency simulations Most, Thomas Krenz, Peter Lampert, Ralf Optimization and Control Antennas are more prevalent than ever enabling 5G connectivity for wide ranging applications like cellular communication, IoT, autonomous vehicles, etc. Optimizing an antenna design can be challenging and employing traditional optimization techniques oftentimes require evaluating ~100s of design variations, which is time prohibitive to tight design deadlines. Instead, the common optimization approach relies on engineering expertise and solving only a handful design variations to manually determine a design with good-enough performance. Here, we present a novel approach for antenna optimization, which utilizes analytical derivatives calculated by Ansys HFSS to dramatically reduce the number of necessary simulation runs. In the presented approach a local updating scheme, which converges quickly to the next local optimum, is extended to a global search strategy intended to detect several local and global optima. This global search procedure uses a consistent interpolation model to approximate the simulation model outputs and its derivatives depending on the design parameters values. The presented approach is implemented in an optimization workflow, which automatically interacts with the HFSS simulation model. Since this optimizer relies on the analytical derivatives calculated by HFSS, it is applicable to any design optimization where geometry or material properties are varied, and the device response is described by SYZ-parameters or far-field quantities. The usability of this new optimizer is therefore wide ranging from antenna design to signal integrity applications. |
| title | A global optimization approach for antenna design using analytical derivatives from high-frequency simulations |
| topic | Optimization and Control |
| url | https://arxiv.org/abs/2408.03674 |