基于密度峰值聚类的高斯混合模型核电运行工况划分

核电厂运行数据记录了核电厂的运行状态，对核电数据进行处理分析从而完成准确的工况划分是实现核电厂运行状态监测的重要基础。为提高核电厂的运行工况划分准确性，本文提出了基于密度峰值聚类的高斯混合模型对核电厂的运行工况进行划分。首先，采用PCA（主成分分析）算法进行数据降维，然后利用密度峰值聚类算法中的决策图确定工况个数，最后利用高斯混合模型完成工况划分。基于真实的核电厂运行数据开展工况划分实验。实验结果表明，本文提出的方法能合理有效地划分出核电运行工况，其三类工况的划分准确率分别达到了99.29%、100%、97.57%，且错误率仅为1.25%。

Nuclear power operating condition division based on density peak clustering using Gaussian mixture model

The operation data of nuclear power plants record the status of nuclear power plants. The data processing and analysis of nuclear power plants can realize the operation status inspection and monitoring of nuclear power plants, and the division of working conditions is the basis for the operation status monitoring of nuclear power plants. In this paper, the Gaussian mixture model based on density peak clustering is proposed to divide the operating conditions of nuclear power plants. Firstly, the PCA (principal component analysis) algorithm is used to reduce the dimensionality of the data, and then the decision graph in the density peak clustering algorithm is used to determine the number of working conditions, and finally the Gaussian mixed model is used to complete the classification of working conditions. Experiments are conducted based on real nuclear power plant operation data. The experimental results show that the proposed method can reasonably and effectively classify the operating conditions of nuclear power plants. The classification accuracy of the three types of operating conditions reaches 99.29%, 100% and 97.57% respectively, and the error rate is only 1.25%.