Method for monitoring impact state of cold heading forming machine die

Abstract

The invention provides a method for monitoring an impact state of a cold heading forming machine die. The method comprises the steps of acquiring a vibration signal of a cold heading forming machine die working in a current impact state; building a variational modal decomposition model, and introducing a vibration signal to calculate an optimal solution, thereby obtaining K intrinsic mode components; performing M entropy calculation on each intrinsic mode component, and serially combining the same entropies calculated by the intrinsic mode components into corresponding eigenvectors, thereby forming a K*M eigenvector matrix; performing feature fusion on the formed K*M eigenvector matrix by using a kernel principal component analysis model, thereby obtaining an eigenvector matrix after feature fusion; and building a hidden unit logic model, and introducing the eigenvector matrix after feature fusion into the hidden unit logic model for solving, thereby obtaining a probability that the cold heading forming machine die works in the current impact state. By implementing the method, high-noise signals can be accurately extracted; good noise robustness is achieved; and the fault diagnosisprecision is improved.

Description

technical field [0001] The invention relates to the technical field of monitoring the machining process of mechanical equipment, in particular to a method for monitoring the impact state of a mold of a cold heading forming machine. Background technique [0002] For a long time, my country's fastener industry has learned from a large number of new tool and mold technologies from abroad, continuously developed and improved, and adopted some new molds, and achieved good results. However, with the development of industry, the market has higher and higher requirements for the strength and precision of fasteners. If fastener products want to reduce costs and improve product competitiveness in the market, it is necessary to improve the performance of cold heading forming machines. Therefore, it is very important to monitor the impact state of the mold of the cold heading forming machine. [0003] Due to the obvious characteristics of the vibration signal, the monitoring method is ...

AI Technical Summary

Problems solved by technology

[0003] Due to the obvious characteristics of the vibration signal, the monitoring method is mature and easy to implement, the condition monitoring method based on the vibration signal is also the most widely used diagnostic method at present. Therefore, it is of great research significance to extract the effective impact state

Images