Aeroengine Fault Correlation Analysis Based on Extension FP-Growth

Aeroengine configuration is complex, and there are many failure reasons and failure mechanisms of components, which bring difficulty to aeroengine fault diagnosis. Using machine learning method to analyze historical operation and maintenance data and mining fault association rules can provide effective support for aeroengine fault diagnosis. In this paper, an aeroengine fault correlation analysis method based on extension FP-Growth algorithm is proposed by combining extension superiority evaluation method with FP-Growth algorithm. Aiming at the problems of low value and weak relevance of operation and maintenance data, this paper uses FP-Growth to mine frequent items and search for feature items closely related to engine failure; aiming at the problems of slow search and low efficiency of transaction set in traditional FP-Growth, this paper removes the redundant items of transaction set, Improve the process of FP-Growth algorithm; for the problems of single evaluation index and over reliance on confidence in association rule mining, this paper introduces extension superiority evaluation method to comprehensively analyze the frequency characteristics, security, economy and maintenance, so as to improve the actual utility of association rules; and through data analysis and example verification, it proves the effectiveness and effectiveness of the algorithm proposed in this paper Advantages.