Quantum Classical Hybrid Chemical Simulation for Large Scale Molecular Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current quantum computing-based chemical simulation methods are limited by hardware constraints, making it difficult to simulate large molecular systems and predict material properties at a practical scale.
Innovation Solution
A chemical simulation apparatus and method that combines quantum computing with machine learning, specifically using a quantum classical algorithm to generate training data for an artificial neural network, which predicts material energy and physical properties.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If quantum computing is used for chemical simulation, then calculation accuracy of molecular energy is improved, but the simulation scale is limited by hardware constraints
Solution Approach 1:
The patent divides the chemical simulation task into two segments: a quantum computing segment for calculating molecular energy with high accuracy, and a classical machine learning segment for handling large-scale simulations. The quantum computer processes small molecular systems to generate training data, while the classical neural network handles the broader simulation scope, thus resolving the contradiction between accuracy and scale.
Solution Approach 2:
The patent introduces a machine learning model as an intermediary between quantum computing and large-scale chemical simulation. The quantum computer trains the machine learning model on small molecular systems, and then the trained model serves as a mediator to predict properties of larger molecular systems, enabling both high accuracy and large simulation scale.
2Productivity
If quantum computer hardware is expanded to increase simulation scale, then computational capability is improved, but hardware development speed is slow
Solution Approach 1:
The patent creates a virtual copy of the quantum computing capability through a machine learning model. Instead of physically expanding quantum hardware, the model learns quantum calculation patterns from training data and replicates this capability on classical hardware, achieving quantum-level accuracy without requiring additional quantum hardware development time.
3Productivity
If machine learning is used to expand simulation scale, then computational efficiency is improved, but training data generation requires quantum computing resources
Solution Approach 1:
The patent applies partial quantum computing action by using quantum computers only to generate training data for a limited set of small molecular systems. Once the machine learning model is trained on this partial quantum data, it can efficiently handle much larger simulation scales using classical computation, thus reducing overall quantum computing resource consumption while maintaining high computational efficiency.
Data Source
AI summary
A chemical simulation apparatus may include a training data generation module configured to sample a molecular structure to be learned, set a wave function, and extract an energy value with respect to the molecular structure based on a quantum computing, a fingerprint conversion module configured to convert structure calculated in the quantum computing into a fingerprint, a learning module configured to perform a neural network learning by using the converted fingerprint as an input, and a prediction module configured to predict material energy or physical properties by using the learned neural network.