On-line detection method for diamond press top hammer cracks based on acoustic emission signals

Abstract

The invention proposes a new method for on-line detection of cracks in an anvil of a diamond press. The backtracking search optimization algorithm (BSA) method is used to optimize the parameters of the support vector machine model, which avoids the blindness of parameter selection; the application of the K-SPWVD method not only uses the template matching method and the K-S test, but also avoids The detection signals are subjected to complex time-frequency analysis, which improves the efficiency of the algorithm, and makes full use of the good time-frequency analysis performance of the smooth pseudo Wigner-Willi distribution. The method has high accuracy and fast calculation speed, which is beneficial to the on-line detection of cracks in the top hammer of the diamond press.

Description

technical field [0001] The invention relates to the field of hard alloy defect detection, in particular to an online detection method for cracks of a top hammer of a diamond press based on acoustic emission signals. Background technique [0002] Synthetic diamond has high hardness and good wear resistance, and can be widely used in machining industries such as cutting, grinding, drilling, and electronic industries such as semiconductors. my country's synthetic diamond production ranks first in the world, mainly using six-sided top presses for production. The anvil of the diamond press is made of hard alloy material, which is a brittle material, and is prone to cracks under high temperature and high pressure environment. If the crack is not detected in time, it will often cause a hammer collapse accident and bring huge economic losses to the enterprise. [0003] There are two main methods for detecting cracks in the top hammer of a traditional diamond press. One is to use a...

AI Technical Summary

Problems solved by technology

[0007] In the above two patents, the processing process of the acquired signal data does not establish an intelligent identification mode, there is no self-learning ability, and its scope of application is limited. For the diamond press anvil, the amount of data in the online detection process is large , the above two patented methods, in the case of a large amount of data, their reaction speed must have limitations, and they cannot judge dangerous situations in time, so it is difficult to directly use them for reference.

Images