Machine learning based system performance prediction model for reactor control

A machine learning based system performance prediction model is currently created to support the development of autonomous control for small reactors, such as the Transportable Fluoride-salt-cooled High-temperature Reactor (TFHR), which is a 20 MWth compact core proposed by MIT for remote site applications. The prediction model consists of a reactor physics model and a thermal-hydraulic model. It is presently constructed using support vector regression (SVR) with training data generated by multiple cases of a one-dimensional reactor system model. A particle filtering framework is utilized to estimate and update model parameters with noisy instrument measurements. Verifications of the prediction and filtering models have been carried out using the TFHR reactivity insertion events. Satisfactory performance in predicting the core behavior and in recognizing transient parameters such as reactivity insertion timing and rate is concluded.