基于Transformer的β液体多道能谱处理方法
By constructing a Transformer-based multichannel energy spectrum processing method, combined with a multi-stage MCFormer module and a cross-attention mechanism, the problems of low efficiency, insufficient accuracy, and weak anti-interference ability in β liquid scintillation multichannel energy spectrum data processing are solved, and efficient and accurate nuclide detection is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- XIAN CNNC NUCLEAR INSTRUMENT CO LTD
- Filing Date
- 2026-05-14
- Publication Date
- 2026-07-17
AI Technical Summary
Existing methods for processing β-liquid scintillation multichannel energy spectrum data suffer from low efficiency, insufficient accuracy, weak anti-interference capability, and poor temporal fusion.
A Transformer-based β-liquid multichannel energy spectrum processing method is adopted. By constructing a network structure including an input layer, batch embedding layer, encoding layer, feature fusion layer, decoding layer and output layer, and combining a multi-stage MCFormer module and cross-attention mechanism, global dependency capture and local feature extraction are achieved, thereby improving data processing efficiency and accuracy and enhancing anti-interference ability.
It significantly improves the processing efficiency and accuracy of multichannel energy spectrum data, reduces the nuclide misidentification rate, and meets the real-time and accuracy requirements of scenarios such as nuclear power plant leak monitoring.
Smart Images
Figure CN122194227B_ABST