Winding shape optimization for asymmetric confinement magnets

We discuss a numerical method for the optimization of the shape of noncircular coil windings for atomic physics applications. The method relies on two main features: first, a piecewise straight modeling of the curve which describes the winding; second, the definition of a step perturbation, based on pursuit curve theory, that changes the shape of the curve while conserving its length. The evolution resulting from a sequence of random steps is piloted towards the optimized shape by means of a Monte Carlo procedure. The method allows for significant improvement of the performance of magnetic configurations and is particularly suitable for dealing with cases in which geometrical constraints play a major role. Finally, as an example, the application of the method to the important case of a clover-leaf trap is discussed.