锆合金广泛应用于核燃料的包壳材料，关于锆合金包壳的腐蚀过程的研 究一直都有进行，这是核工程中一个重要的问题，因为对于腐蚀过程的认知 会影响到核燃料运行的安全性和经济性。本文主要研究关于锆合金的核燃料 包壳腐蚀模型中所涉及的微分方程参数估计问题。

本文旨在通过利用贝叶斯方法来对该模型进行参数估计，来研究微分方 程的贝叶斯参数估计方法并考虑其在核工程中应用的有效性。本文基于刻画 腐蚀过程的原始数据，通过对数据的处理，模型的简化，以及贝叶斯方法迭 代更新参数的后验分布，最后得到对参数的估计结果。

对于该模型进行贝叶斯参数估计得到的结果，比之前基于线搜索的优化 算法给出的估计结果更好，误差更小，这说明贝叶斯的概率方法在核工程中 有关的参数估计问题上有着不错的效果。

Zirconium alloys are widely used as cladding materials for nuclear fuel, and the corrosion process of zirconium alloy cladding has been extensively studied. This is an important issue in nuclear engineering because understanding the corrosion process can affect the safety and economy of nuclear fuel operation. This paper focuses on the parameter estimation problem of the differential equations involved in the corrosion model for zirconium alloy nuclear fuel cladding.

The aim of this paper is to study the effectiveness of Bayesian methods in nuclear engineering by using them to estimate parameters in the corrosion model. Based on the raw data that characterizes the corrosion process, we simplify the model, process the data, and use the iterative updating of the posterior distribution of the parameters in Bayesian methods to obtain the estimation results.

The results obtained by Bayesian parameter estimation of the model are better and have smaller errors than those obtained by previous optimization algorithms based on line search. This indicates that the probabilistic Bayesian method has good performance in parameter estimation problems related to nuclear engineering.