Abstract:As a crossover frontier of physics and mechanics, quantum computing is showing its great potential in computational mechanics. However, quantum hardware noise remains a critical barrier to achieving accurate simulation results due to the limitation of the current hardware. In this paper, we integrate error-mitigated quantum computing in data-driven computational homogenization, where the zero-noise extrapolation (ZNE) technique is employed to improve the reliability of quantum computing. Specifically, ZNE is utilized to mitigate the quantum hardware noise in two quantum algorithms for distance calculation, namely a Swap-based algorithm and an H-based algorithm, thereby improving the overall accuracy of data-driven computational homogenization. Numerical examples including a multiscale simulation of a composite L-shaped beam are conducted with the quantum computer simulator Qiskit, and the results validate the effectiveness of the proposed method. We believe this work presents a promising step towards using quantum computing in computational mechanics.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is how to reduce the errors caused by quantum - hardware noise when using quantum computing in data - driven computational homogenization, thereby improving the accuracy of quantum computing. Specifically, the paper focuses on how to mitigate the impact of quantum - hardware noise on distance calculation through the Zero - Noise Extrapolation (ZNE) technique when performing nearest - neighbor search in computational material databases, and further improve the overall computational accuracy. ### Background and Motivation - **Advantages of quantum computing**: Quantum computing utilizes the superposition and entanglement properties of quantum systems and can exponentially reduce computational complexity in certain tasks. Therefore, it has attracted the attention of an increasing number of researchers in the field of computational mechanics. - **Existing challenges**: Current quantum hardware has noise problems, which limit the practical application effects of quantum computing. Especially in data - driven computational homogenization, the complex structure - activity relationships of composite materials require higher computational accuracy, and quantum - hardware noise can seriously affect the accuracy of distance calculation. ### Solutions - **Zero - Noise Extrapolation (ZNE)**: The paper proposes using the ZNE technique to mitigate the impact of quantum - hardware noise. ZNE obtains noise - free results by intentionally increasing the noise level and then extrapolating on a classical computer. - **Specific methods**: - **Swap - based algorithm**: A quantum algorithm based on the Swap test, used to calculate the Euclidean distance between two vectors. - **H - based algorithm**: A quantum algorithm based on a Hadamard gate, also used to calculate the Euclidean distance, but requiring fewer qubits and gates and having certain advantages. - **k - d tree structure**: Combine the k - d tree data structure to reduce the number of distance calculations and further reduce computational complexity. ### Experimental Verification - **Numerical experiments**: Multiple numerical experiments were carried out using the quantum computer simulator Qiskit to verify the effectiveness of ZNE in improving the accuracy of distance calculation. - **Application examples**: The application value of the proposed method in practical engineering problems was demonstrated by multi - scale simulation of a composite L - shaped beam. ### Conclusion By introducing the ZNE technique, the paper successfully mitigated the impact of quantum - hardware noise on distance calculation in data - driven computational homogenization and improved the reliability of quantum computing. This work provides an important step for the practical application of quantum computing in the field of computational mechanics.

Quantum computing with error mitigation for data-driven computational homogenization

Quantum computing enhanced distance-minimizing data-driven computational mechanics

Error mitigation extends the computational reach of a noisy quantum processor

Hybrid quantum-classical algorithms and quantum error mitigation

Leveraging hardware-control imperfections for error mitigation via generalized quantum subspace

Mitigating algorithmic errors in quantum optimization through energy extrapolation

Quantum error mitigation in quantum annealing

Quantum-assisted Monte Carlo algorithms for fermions

Tackling Coherent Noise in Quantum Computing via Cross-Layer Compiler Optimization

Application of zero-noise extrapolation-based quantum error mitigation to a silicon spin qubit

Superposed Quantum Error Mitigation

Quantum Tunneling: From Theory to Error-Mitigated Quantum Simulation

Improving Zero-noise Extrapolation for Quantum-gate Error Mitigation using a Noise-aware Folding Method

Error mitigation in variational quantum eigensolvers using tailored probabilistic machine learning

Enhancing quantum models of stochastic processes with error mitigation

Quantum Zeno Monte Carlo for computing observables

Quantum Error Mitigation via Quantum-Noise-Effect Circuit Groups

Efficiently improving the performance of noisy quantum computers

Machine Learning for Practical Quantum Error Mitigation

Error mitigation with stabilized noise in superconducting quantum processors