A Research on Multi-Layer Perceptron Diagnosis Model Based on D Matrix

Autor: Xianjie Zhu, Nanpo Niu, Junyou Shi
Rok vydání: 2018
Předmět:
Zdroj: 2018 Prognostics and System Health Management Conference (PHM-Chongqing).
DOI: 10.1109/phm-chongqing.2018.00137
Popis: With the intelligent development of modern industry, the system integration structure is becoming more and more complex, so the research of intelligent fault diagnosis is becoming more and more significant, and the method of fault diagnosis using neural network is becoming the hotspot of research. First, this paper introduces the dependency matrix, based on which a Multi-layer Perceptron diagnosis model matrix is proposed and its workflow is illustrated. This method has the advantages of good robustness and high precision. In the future, it can further optimize its self-learning ability to be appropriate for appliaction in more scenes.
Databáze: OpenAIRE