CN-121980482-A - Wear detection method, system, terminal and storage medium

CN121980482ACN 121980482 ACN121980482 ACN 121980482ACN-121980482-A

Abstract

The application relates to a wear detection method, a system, a terminal and a storage medium, and relates to the technical field of wear detection, comprising the steps of controlling a preprocessing device to preprocess a wear detection device, and then controlling the wear detection device to detect lubricating oil so as to generate a wear detection signal; the method comprises the steps of inputting a wear detection signal into a time sequence feature extraction submodule to conduct feature extraction so as to generate a signal time sequence feature vector, analyzing the signal time sequence feature vector based on a random forest algorithm so as to generate a copper-containing abrasive particle analysis result, judging whether the copper-containing abrasive particle analysis result is an abrasive particle copper-containing result or an abrasive particle copper-free result, continuously controlling a wear detection device to detect lubricating oil so as to generate a wear detection signal to conduct cycle judgment if the abrasive particle copper-free result is obtained, and analyzing the signal time sequence feature vector so as to generate copper abrasive particle parameters if the abrasive particle copper-containing result is obtained. The application has the effect of improving the timeliness of abrasion detection.

Inventors

QIN YAGUANG
LI ZIHENG
ZHANG DESHUAI
QI FEIXIANG
PAN QI
Guo Mengni

Assignees

浙江华东工程建设管理有限公司
中国电建集团华东勘测设计研究院有限公司

Dates

Publication Date: 20260505
Application Date: 20260408

Claims (10)

1. A wear detection method, comprising: Controlling a preset preprocessing device to preprocess a preset abrasion detection device, and controlling the preprocessed abrasion detection device to detect lubricating oil in a lubricating oil tank of a preset cone crusher so as to generate an abrasion detection signal; Inputting the abrasion detection signal into a preset time sequence feature extraction submodule to perform feature extraction so as to generate a signal time sequence feature vector; Analyzing the signal time sequence feature vector based on a preset random forest algorithm to generate a copper-containing abrasive particle analysis result; Judging whether the analysis result of the copper-containing abrasive particles is a preset abrasive particle copper-containing result or a preset abrasive particle copper-free result; if the abrasive particles do not contain copper, continuously controlling the abrasion detection device to detect the lubricating oil in the lubricating oil tank of the cone crusher so as to generate an abrasion detection signal for cycle judgment; If the result is the copper-containing result of the abrasive particles, analyzing the signal time sequence feature vector to generate copper abrasive particle parameters; Uploading the copper abrasive particle parameters to a preset upper computer for wear early warning.
2. A wear detection method according to claim 1, wherein the step of controlling the preprocessed wear detection device to detect the lubricating oil in the preset cone crusher lubricating oil tank to generate the wear detection signal comprises: controlling a wear detection device to detect lubricating oil in a lubricating oil tank of the cone crusher so as to generate a basic detection signal; controlling a preset LC resonance module to process the basic detection signal so as to generate an amplitude lifting signal; controlling a preset pre-amplification module to process the amplitude boost signal to generate an amplified detection signal; Controlling a preset low-pass filtering module to process the amplified detection signal so as to generate a noise reduction detection signal; And controlling a preset coherent demodulation module to process the noise reduction detection signal so as to generate a wear detection signal.
3. The method of claim 1, wherein the step of analyzing the signal timing feature vector to generate the copper abrasive parameters comprises: Analyzing the signal time sequence feature vector based on a preset depth residual error network to generate a confidence coefficient of the copper abrasive particle material; judging whether the confidence coefficient of the copper abrasive grain material meets the requirement of a preset material confidence coefficient threshold value or not; If yes, determining a signal amplitude according to the signal time sequence feature vector; Substituting the signal amplitude value into a preset cubic polynomial relation to calculate and count so as to generate the diameter of the copper abrasive particles and the number of the copper abrasive particles; If the number of the copper abrasive particles does not accord with the number of the copper abrasive particles, analyzing the signal time sequence feature vector according to a preset digital twin model to generate the diameter of the copper abrasive particles and the number of the copper abrasive particles; and analyzing the diameter of the copper abrasive particles and the number of the copper abrasive particles to generate copper abrasive particle parameters.
4. A wear detection method according to claim 3, wherein the third order polynomial relationship is: , Wherein, the For the signal amplitude value, The coefficients are fitted for the cubic term, The coefficients are fitted for the quadratic term, The coefficients are fitted for the first order term, The coefficients are fitted for the constant terms, Is the diameter of the copper abrasive particles.
5. A wear detection method according to claim 3, wherein the step of analyzing the signal timing feature vector according to a predetermined digital twin model to generate the copper abrasive grain diameter and the number of copper abrasive grains comprises: Controlling the digital twin model to simulate according to preset abrasive particle combination parameters so as to generate an aliasing signal feature vector; Analyzing the signal time sequence feature vector and the aliasing signal feature vector to generate signal feature similarity; judging whether the signal feature similarity meets the requirement of a preset signal similarity threshold value or not; If not, eliminating the abrasive particle combination parameters; If the parameters are met, determining the diameter of the copper abrasive particles and the number of the copper abrasive particles according to the abrasive particle combination parameters.
6. A method of wear detection according to claim 3, wherein the step of analyzing the diameter of the copper abrasive grains and the number of copper abrasive grains to generate the copper abrasive grain parameters comprises: calculating the sum of the diameters of the copper abrasive particles to generate the total diameter of the copper abrasive particles; Calculating the quotient of the total diameter of the copper abrasive particles and the preset reference abrasive particle diameter to generate the number of corrected copper abrasive particles; Calculating the quotient of the number of the corrected copper abrasive particles and the preset lubricating oil volume to generate copper abrasive particle concentration; the copper abrasive particle diameter, the number of copper abrasive particles and the copper abrasive particle concentration are correlated to generate copper abrasive particle parameters.
7. The method of claim 1, wherein the step of uploading the copper abrasive parameters to a predetermined host computer for wear pre-warning comprises: Determining copper abrasive particle concentration according to copper abrasive particle parameters; Collecting the reference concentration of copper abrasive particles; calculating the difference between the concentration of the copper abrasive particles and the reference concentration of the copper abrasive particles to generate a concentration variation; calculating the quotient of the concentration variation and the reference concentration of the copper abrasive particles to generate a concentration increase rate; judging whether the concentration increasing rate is larger than a preset concentration increasing rate threshold value or not; If yes, carrying out abrasion early warning; If not, continuously uploading the copper abrasive particle parameters to an upper computer for cycle judgment.
8. A wear detection system, comprising: The collection module is used for collecting the reference concentration of the copper abrasive particles; a memory for storing a program of a wear detection method according to any one of claims 1 to 7; a processor, a program in memory being capable of being loaded by the processor and implementing a wear detection method as claimed in any one of claims 1 to 7.
9. An intelligent terminal comprising a memory and a processor, wherein the memory has stored thereon a computer program that can be loaded by the processor and that performs a wear detection method according to any of claims 1 to 7.
10. A computer-readable storage medium, characterized in that a computer program capable of being loaded by a processor and executing a wear detection method according to any one of claims 1 to 7 is stored.

Description

Wear detection method, system, terminal and storage medium Technical Field The present application relates to the field of wear detection technologies, and in particular, to a wear detection method, a wear detection system, a wear detection terminal, and a storage medium. Background The abrasion detection means a technical method for capturing an abrasion detection signal generated by abrasion of a copper shaft sleeve of the cone crusher in operation through an abrasion detection device, judging whether equipment is in normal abrasion or abnormal abrasion, and ensuring that the abrasion condition of the shaft sleeve can be reflected timely and accurately. In the related art, the identification and analysis of ferromagnetic particles are focused mostly, through firstly collecting a lubricating oil sample used after a cone crusher runs for a certain period, then installing an inductive abrasive particle sensor in a lubricating system pipeline, detecting the concentration and the size distribution of ferromagnetic particles flowing through oil in real time, finally correlating and deducing the state of a copper shaft sleeve, and because the copper shaft sleeve is made of nonferromagnetic materials, copper abrasive particles generated by the copper shaft sleeve cannot be effectively detected by the inductive sensor, when iron parts such as a main shaft or an eccentric sleeve generate abnormal high wear signals, the copper shaft sleeve of the cone crusher is proved to be worn out and lose efficacy, so that the iron parts are directly rubbed. Aiming at the related technology, when the abrasion detection device detects and correlates the abnormal high abrasion signal of the iron part to infer the state of the copper shaft sleeve, the key signal of early abrasion of the shaft sleeve cannot be timely and accurately captured due to the fact that copper abrasive particles and ferromagnetic particles are not distinguished, so that the timeliness of abrasion detection is poor, and the improvement is still in existence. Disclosure of Invention In order to improve timeliness of wear detection, the application provides a wear detection method, a wear detection system, a wear detection terminal and a storage medium. In a first aspect, the present application provides a wear detection method, which adopts the following technical scheme: A wear detection method, comprising: Controlling a preset preprocessing device to preprocess a preset abrasion detection device, and controlling the preprocessed abrasion detection device to detect lubricating oil in a lubricating oil tank of a preset cone crusher so as to generate an abrasion detection signal; Inputting the abrasion detection signal into a preset time sequence feature extraction submodule to perform feature extraction so as to generate a signal time sequence feature vector; Analyzing the signal time sequence feature vector based on a preset random forest algorithm to generate a copper-containing abrasive particle analysis result; Judging whether the analysis result of the copper-containing abrasive particles is a preset abrasive particle copper-containing result or a preset abrasive particle copper-free result; if the abrasive particles do not contain copper, continuously controlling the abrasion detection device to detect the lubricating oil in the lubricating oil tank of the cone crusher so as to generate an abrasion detection signal for cycle judgment; If the result is the copper-containing result of the abrasive particles, analyzing the signal time sequence feature vector to generate copper abrasive particle parameters; Uploading the copper abrasive particle parameters to a preset upper computer for wear early warning. By adopting the technical scheme, the preprocessing device is controlled to preprocess the wear detection device, the preprocessed wear detection device is controlled to detect lubricating oil in the lubricating oil tank of the cone crusher to determine wear detection signals, the wear detection signals are input into the time sequence feature extraction submodule to perform feature extraction to determine signal time sequence feature vectors, the signal time sequence feature vectors are analyzed according to a random forest algorithm to determine copper-containing abrasive particle analysis results, when the copper-containing abrasive particle analysis results are determined to be abrasive particle copper-containing results, the signal time sequence feature vectors are analyzed to determine copper abrasive particle parameters, and finally the determined copper abrasive particle parameters are uploaded to an upper computer to perform wear early warning, so that key signals of early wear of a shaft sleeve are timely and accurately captured, and timeliness of wear detection is improved. Optionally, the step of controlling the preprocessed wear detection device to detect the lubricating oil in the preset cone crusher lubricating oil tank to generate the wear