A Long-term Dependent and Trustworthy Approach to Reactor Accident Prognosis based on Temporal Fusion Transformer

TL;DR

This paper introduces a novel application of the Temporal Fusion Transformer model for long-term prognosis of reactor accidents, improving accuracy and robustness in predicting key parameters after incidents like LOCAs.

Contribution

It is the first to apply the TFT model to reactor accident prognosis, integrating multi-covariate data and uncertainty quantification for enhanced prediction and robustness.

Findings

TFT outperforms existing deep learning methods in accuracy and confidence.

The model demonstrates robustness against noise and static covariate variations.

Application to HPR1000 reactor LOCAs shows practical effectiveness.

Abstract

Prognosis of the reactor accident is a crucial way to ensure appropriate strategies are adopted to avoid radioactive releases. However, there is very limited research in the field of nuclear industry. In this paper, we propose a method for accident prognosis based on the Temporal Fusion Transformer (TFT) model with multi-headed self-attention and gating mechanisms. The method utilizes multiple covariates to improve prediction accuracy on the one hand, and quantile regression methods for uncertainty assessment on the other. The method proposed in this paper is applied to the prognosis after loss of coolant accidents (LOCAs) in HPR1000 reactor. Extensive experimental results show that the method surpasses novel deep learning-based prediction methods in terms of prediction accuracy and confidence. Furthermore, the interference experiments with different signal-to-noise ratios and the ablation experiments for static covariates further illustrate that the robustness comes from the ability to extract the features of static and historical covariates. In summary, this work for the first time applies the novel composite deep learning model TFT to the prognosis of key parameters after a reactor accident, and makes a positive contribution to the establishment of a more intelligent and staff-light maintenance method for reactor systems.