基于朴素贝叶斯和D-S证据理论的多时空数据融合

Multi-time-space data fusion based on naive Bayes and D-S evidence theory

针对泛在电力物联网中分布广泛的传感器以及各类设备采样周期不同的问题,本文提出一种基于朴素贝叶斯和D-S证据理论的多时空数据融合方法。该方法突出的优点是融合了多个时间段、多个不同地点传感器的数据。首先运用朴素贝叶斯分类器得到信度分配,克服了过去采用专家系统进行信度分配的缺点,然后运用D-S证据理论进行融合得到最终系统的状态评价,有效地将多时空数据进行融合。实验结果表明,本文提出的方法相比其他机器学习算法有了明显的改进,能够有效地评估系统的状态。

Aiming at solving the problem of different sampling periods between sensors and various devices in the electric internet of things,this paper proposes a multi-time-space data fusion method based on naive Bayes and D-S evidence theory.The outstanding advantage of this method is that it combines the data of sensors in multiple time periods and multiple different locations.Firstly,the naive Bayes classifier is used to obtain the reliability distribution,which overcomes the shortcomings of the original expert system for reliability distribution.Then D-S evidence theory is used to fuse,and finally get the state evaluation of the system,which effectively integrate multi-time-space data.The experimental results show that the proposed method has obvious improvement compared with other machine learning algorithms,and can effectively evaluate the state of the system.