An automated gaseous tritium sampling measurement method
By using a dynamic memory entropy observer and a reinforcement learning decision model, the problems of sensor wall adsorption and environmental interference in the gaseous tritium monitoring system were solved, achieving high-precision, long-cycle tritium concentration measurement and extending sensor lifespan.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHAANXI QINZHOU NUCLEAR & RADIATION SAFETY TECHNONLOY CO LTD
- Filing Date
- 2026-02-26
- Publication Date
- 2026-05-26
AI Technical Summary
Existing gaseous tritium monitoring systems suffer from memory effects and environmental interference caused by adsorption on the sensor wall under high concentrations or long-term operation, which are difficult to effectively remove. This leads to measurement blind spots and deterioration of the signal-to-noise ratio. Furthermore, the lack of an adaptive cleaning and maintenance mechanism makes it difficult to achieve high-precision long-term operation.
By combining a dynamic memory entropy observer model with reinforcement learning decision-making, sampling and cleaning modes are dynamically switched using real-time ionization current data and environmental state data. The extended Kalman filter algorithm is used to calculate the degree of adsorption on the sensor wall, generate the optimal sampling control strategy, remove environmental interference, and extend the sensor life.
It achieves high-precision tritium concentration measurement under complex working conditions, eliminates environmental interference, prevents sensor blind spots, extends sensor lifespan, and ensures monitoring reliability and accuracy throughout the entire life cycle.
Smart Images

Figure CN121721677B_ABST