Mobile heat supply unit fault diagnosis method

Abstract

The invention discloses a mobile heat supply unit fault diagnosis method, and belongs to the technical field of equipment fault diagnosis. The method comprises the following steps: S1, sound data collection: collecting information of equipment based on a non-contact method of an acoustic wave sensor; and S2, fault weak signal detection: using a wavelet packet energy spectrum method with lighter operand, and compared with mathematical methods such as machine learning, a deep neural network and minimum entropy convolution, the time complexity and the space complexity being lower. According to the invention, a non-contact method is introduced to detect the unit. The defect that a conventional method cannot go deep into a unit for signal reconstruction is overcome, more continuous separation and decomposition signals can be obtained, fault features are enhanced through wavelet packets and wavelet packet energy spectrum methods, a transfer learning method is applied to popularize differenttypes of heat exchange and supply equipment, and the method has high application value and popularization capacity.

Description

technical field [0001] The invention relates to the technical field of equipment fault diagnosis, in particular to a method for fault diagnosis of a mobile heating unit. Background technique [0002] 1"The method disclosed in the literature is applied to tool life detection: [0003] [0] Zhu Guokui, Zhang Minliang, Zhu He. Experimental Research on Tool State Based on Milling Sound Signal [J]. Light Industry Machinery, 2017,35(01):54-58+63. [0004] [1] Huo Pengcheng. Research on Tool Condition Monitoring System Based on Audio Signal Analysis in Cutting Processing [D]. Shanghai University of Engineering Technology, 2016. [0005] [2] Yang Minglun. Research on milling tool damage monitoring method based on noise-assisted empirical mode decomposition [D]. Shanghai Jiaotong University, 2013. [0006] [3] Chen Quntao. Research on Milling Cutter Condition Monitoring Technology Based on Acoustic Vibration Signal EMD and ICA [D]. Shanghai Jiaotong University, 2013. [0007] [4] X...

AI Technical Summary

Problems solved by technology

However, there are also calculation and pre-training problems in Document 1.

[0018] 4. The rest of the literature did not propose specific feasible algorithms and data analysis mechanisms

Images