基于多源数据融合的突发水污染事故可靠预警方法

在突发水污染事故自动监测领域中,传感器节点监测数据的异常是影响自动监测系统预警可靠性的重要原因.考虑到多传感器信息之间的互补性和相关性,该文提出了一种基于多源数据融合的突发水污染事故可靠预警方法.基于改进的D-S证据理论,利用综合权重对节点证据进行加权修正,并用D-S融合规则对多源数据进行两两融合,最终根据融合结果对突发水污染事故进行预警决策.案例分析及实验结果表明:与传统方法相比,该方法能得到可靠度更高、聚焦性更好的预警结论.

The Reliable Warning Method for Sudden Water Pollution Based on Multi-Source Data Fusion

In the field of automatic monitoring of sudden water pollution accidents,the abnormal data caused by sensor nodes is an important reason to affect the warning reliability for automatic monitoring system.Considering the complementarity and correlation between multi-sensor information, a new reliable warning method for accidental water pollution based on multi-source data fusion is presented in this paper.The comprehensive weight is used to modify the original evidences based on the improved Dempster-Shafer(D-S)evidence theory, then the multi-source evidences are fused by utilizing the combination rule of D-S theory,and finally the early warning decision for sudden water pollution accidents can be made according to the fusion result.Compared with the traditional methods,the case analysis and experimental results show that the proposed method can make the warning decisions with higher credibility and better focus.