论文部分内容阅读
间歇过程测量数据的高维、非线性、非高斯分布特征直接影响过程测量数据异常检测的准确性,为了融合多源数据异常检测信息,提升间歇过程测量数据异常检测精度,提出了一种基于多证据融合决策的间歇过程测量数据异常检测方法,该方法通过引入证据理论(Dempster-Shafer,D-S),采用主焦元判别伪证据和重新计算证据权重改进冲突证据处理方法,减小了冲突证据对多证据融合决策结果的影响,提高了间歇过程测量数据异常检测的准确率。构建了基于多证据融合的测量数据异常检测模型并将其应用到间歇过程测量数据异常检测决策判决中。实验结果表明,该方法能够融合多证据信息,有效地处理冲突证据,实现了间歇过程测量数据异常检测,降低了误检和漏检率。
The high dimensional, nonlinear and non-Gaussian distributions of the intermittent process measurement data directly affect the accuracy of the anomaly detection of the process measurement data. In order to fuse the information of the multi-source data anomaly detection and improve the accuracy of the anomaly detection of the intermittent process measurement data, This method improves the conflict evidence processing method by introducing Dempster-Shafer (DS), using the main focal element to discriminate the false evidence and recalculating the weight of the evidence, and reduces the conflict evidence. The effect of multi-evidence fusion decision-making results improves the accuracy of anomaly detection of intermittent process measurement data. An abnormality detection model based on multi-evidence fusion is constructed and applied to decision-making of anomaly detection of intermittent process measurement data. The experimental results show that the proposed method can fuse multiple evidences, process conflict evidences effectively, and achieve anomaly detection of intermittent process measurement data, reducing false positives and false negatives.