当前位置: X-MOL 学术Nucl. Eng. Technol. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
The ensemble approach in comparison with the diverse feature selection techniques for estimating NPPs parameters using the different learning algorithms of the feed-forward neural network
Nuclear Engineering and Technology ( IF 2.6 ) Pub Date : 2021-06-20 , DOI: 10.1016/j.net.2021.06.030
Khalil Moshkbar-Bakhshayesh

Several reasons such as no free lunch theorem indicate that there is not a universal Feature selection (FS) technique that outperforms other ones. Moreover, some approaches such as using synthetic dataset, in presence of large number of FS techniques, are very tedious and time consuming task. In this study to tackle the issue of dependency of estimation accuracy on the selected FS technique, a methodology based on the heterogeneous ensemble is proposed. The performance of the major learning algorithms of neural network (i.e. the FFNN-BR, the FFNN-LM) in combination with the diverse FS techniques (i.e. the NCA, the F-test, the Kendall's tau, the Pearson, the Spearman, and the Relief) and different combination techniques of the heterogeneous ensemble (i.e. the Min, the Median, the Arithmetic mean, and the Geometric mean) are considered. The target parameters/transients of Bushehr nuclear power plant (BNPP) are examined as the case study. The results show that the Min combination technique gives the more accurate estimation. Therefore, if the number of FS techniques is m and the number of learning algorithms is n, by the heterogeneous ensemble, the search space for acceptable estimation of the target parameters may be reduced from n × m to n × 1. The proposed methodology gives a simple and practical approach for more reliable and more accurate estimation of the target parameters compared to the methods such as the use of synthetic dataset or trial and error methods.



中文翻译:

与使用前馈神经网络的不同学习算法估计 NPP 参数的多种特征选择技术相比,集成方法

没有免费午餐等几个原因定理表明,没有一种通用的特征选择 (FS) 技术优于其他技术。此外,一些方法,例如使用合成数据集,在存在大量 FS 技术的情况下,是非常繁琐和耗时的任务。在这项研究中,为了解决估计精度对所选 FS 技术的依赖性问题,提出了一种基于异构集成的方法。神经网络主要学习算法(即 FFNN-BR、FFNN-LM)与各种 FS 技术(即 NCA、F-test、Kendall's tau、Pearson、Spearman 和浮雕)和异质系综的不同组合技术(即最小值、中值、算术平均值和几何平均值)被考虑在内。布什尔核电站 (BNPP) 的目标参数/瞬态作为案例研究进行了检查。结果表明,最小组合技术给出了更准确的估计。因此,如果 FS 技术的数量是m并且学习算法的数量为n,通过异构集成,可接受的目标参数估计的搜索空间可以从n × m减少到n × 1。与使用合成数据集或试错法等方法相比,所提出的方法提供了一种简单实用的方法,可以更可靠、更准确地估计目标参数。

更新日期:2021-06-20
down
wechat
bug