模糊粒计算在工业动态复杂事件识别中的应用

Application of Fuzzy Granular Computing in Industrial Dynamic Complex Event Recognition

工业动态复杂事件识别可以为决策者提供故障诊断、流程优化的参考依据。如何在不完备数据下进行可解释的推理和识别是一个非常有挑战性的问题。为了解决这个问题,本文提出了一个新的工业动态复杂事件识别模型。通过将该模型应用于合成数据集以及真实数据集,结果显示,在不完备数据的工业场景下,本文模型的准确率和召回率均达到90%以上,预测未知事件的准确率也达到95%以上。

Industrial dynamic complex event recognition(CER)can provide decision makers with reference for fault diagnosis and process optimization.It is a very challenging problem to perform interpretable reasoning and recognition under incomplete data.In order to solve this problem,this paper proposes a new identification model of industrial dynamic complex events.The model is applied to synthetic data sets and real data sets.The results show that under the industrial scenario of incomplete data,the accuracy and recall rate of the model in this paper are more than 90%,and the accuracy of predicting unknown events is more than 95%.