Genetic algorithms with niching for conceptual design studies

The paper describes the use of genetic algorithms with the concept of niching for the conceptual design of superconducting magnets for the Large Hadron Collider, LHC at CERN. The method provides the designer with a number of possible solutions which can then be further optimized for field quality and manufacturability. Two 6-block dipole coils were found and proved to have advantages compared to the standard 5-block version. Index terms - Design automation, genetic algorithms, magnetic fields, superconducting accelerator magnets, optimization methods