Automatic multi-parameter design optimization for superconducting quantum devices

We present an automated framework for the precise optimization towards user-specified Hamiltonian parameter targets in superconducting quantum circuits, crucial for both small and large-scale quantum implementations.

Our design optimization package extends Qiskit Metal’s popular design environment and the ANSYS HFSS solver, to enable an efficient multi-component circuit design of qubits, resonators, and their linear and non-linear couplings, also to readout and control lines.

Through a combination of eigenmode analysis, capacitance extraction, and participation ratio studies, we realize an iterative physics-guided multi-parameter optimization wrapping around high-accuracy simulations and accounting for potential parameter interdependencies. The framework can break large-scale circuit layouts into smaller studies for efficient simulations on desktop computers.

Additionally, we provide a practical example of a multi-qubit chip, complementing our online available design optimization package.

The package should offer an accessible, low-effort entry point for students and researchers designing superconducting quantum circuits.