一种芯片供电网络的电磁谐振风险预测方法、设备及介质
By combining the PDN grid-to-qubit mapping rule, the spatiotemporal compressed sensing measurement matrix, and the neural network model of physical constraints, the problem of accuracy in predicting electromagnetic resonance risk in chip power supply networks in existing technologies has been solved, achieving accurate identification of resonance-sensitive regions and propagation paths, and accurate prediction of resonance intensity.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- 广东全芯半导体有限公司
- Filing Date
- 2025-08-04
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies cannot accurately identify resonance-sensitive areas and propagation paths when predicting electromagnetic resonance risks in chip power supply networks, and they do not fully incorporate physical constraints, resulting in insufficient prediction accuracy.
A target model based on the PDN grid-to-qubit mapping rule is adopted, and combined with the running data and the first influencing factor, the resonance level of each grid cell is accurately predicted. A model based on the spatiotemporal compressed sensing measurement matrix is introduced to accurately predict the resonance propagation path. A neural network model based on physical constraints is used to predict the resonance intensity. Local and global resonances are predicted through synchronous iteration. The resonance risk value is determined by combining the resonance sensitive area and intensity.
It achieves accurate prediction of electromagnetic resonance risks in chip power supply networks, improves prediction accuracy, can accurately locate resonance-sensitive areas and propagation paths, and ensures that prediction results conform to actual physical laws.
Smart Images
Figure CN120995000B_ABST