CN-121978217-A - Acoustic emission detection method and system for pitting damage of gear tooth surface

CN121978217ACN 121978217 ACN121978217 ACN 121978217ACN-121978217-A

Abstract

The invention discloses a gear tooth surface pitting damage acoustic emission detection method and a system, which relate to the technical field of mechanical fault diagnosis and comprise the following steps of obtaining an attaching pressure sequence and an acoustic waveform sequence of a target object; and judging the morphological consistency of the attached pressure sequence according to a sliding window to generate a trusted window set, and establishing hyperbolic mapping for the sound wave form sequence in the trusted window set from the head end and the tail end and the attached pressure sequence of the corresponding trusted window respectively to generate a hysteresis mapping table. According to the scheme, the pressure representative value is biased and expanded according to the hysteresis mapping table, a bias expansion result which changes along with the hysteresis of the pressing force is formed, so that threshold mismatch caused by coupling drift is restrained, and a boundary judgment mark is generated by executing boundary judgment on the bias expansion result and a preset baseline proportion, so that the moment that the threshold needs to be adjusted is rapidly identified when the proportion is out of range, and the strong impact traction threshold is prevented from lifting.

Inventors

Qiu Shuicai
ZHANG LINGYAN

Assignees

常州大学怀德学院

Dates

Publication Date: 20260505
Application Date: 20260401

Claims (10)

1. The method for detecting the pitting damage acoustic emission of the gear tooth surface is applied to detecting the pitting damage acoustic emission of the gear tooth surface of an external gear box of a high-temperature kiln and is characterized by comprising the following steps of: acquiring an attaching pressure sequence and an acoustic waveform sequence of a target object; Judging the morphological consistency of the attached pressure sequence according to a sliding window to generate a trusted window set, and establishing hyperbolic mapping for the sound wave form sequence in the trusted window set from the head end and the tail end and the attached pressure sequence of the corresponding trusted window respectively to generate a hysteresis mapping table; performing bias expansion on the attached pressure sequence according to the hysteresis mapping table, and performing boundary discrimination and threshold updating on a bias expansion result and a preset baseline proportion to generate a threshold updating set; performing threshold comparison on the sound emission waveform sequence according to the threshold updating set, performing grouping count on a threshold comparison result, taking the number of poles to generate a cycle Zhou Houxuan, and performing equidistant extrapolation on the threshold comparison result according to the cycle Zhou Houxuan to generate a cycle reference sequence; Performing windowing and offset averaging on the turnover reference sequence according to the trusted window set to generate a position centroid sequence, and performing segment fitting on the position centroid sequence and the attached pressure sequence to generate a phase offset function; And carrying out compensation processing on the rotating reference sequence according to the phase offset function to generate a pitting corrosion detection result.
2. The method for detecting pitting damage to the tooth surface of the gear according to claim 1, wherein the method further comprises the step of generating a pressure direction mark after generating the trusted window set, and specifically comprises the following steps: Performing difference calculation on every two attached pressure sequences in the trusted window set to generate a pressure difference sequence, and performing symbol counting on the pressure difference sequence; The sign counting comprises the steps of respectively counting the number of elements larger than zero in the pressure difference value sequence to obtain a positive sign count, and counting the number of elements smaller than zero to obtain a negative sign count; And respectively calculating the duty ratio results of the positive sign count and the negative sign count in the non-zero elements in the pressure difference value sequence, and judging the size relation of the duty ratio results to generate a pressure direction mark.
3. The method for detecting pitting damage to a gear tooth surface according to claim 2, wherein creating a hysteresis map table by creating hyperbolic maps for the acoustic waveform sequence from head and tail ends and attached pressure sequences of corresponding trusted windows, respectively, in the set of trusted windows comprises: segmenting the sound emission waveform sequence in the trusted window set to generate a candidate waveform segment set, and sequencing the candidate waveform segment set according to the amplitude statistics and descending order of the segments to generate an energy sequencing result; calculating the ratio of the head end and the tail end of the energy sequencing result to the candidate waveform segment set respectively, and correspondingly generating a high energy segment proportion and a low energy segment proportion; respectively performing hyperbolic curve fitting on the pressure representative value and the high energy segment proportion, and the pressure representative value and the low energy segment proportion to generate a hysteresis mapping table; the pressure representative value is obtained by carrying out statistical analysis on the attached pressure sequence in the trusted window set.
4. The gear tooth surface pitting damage acoustic emission detection method according to claim 3, characterized by: performing bias expansion on the attached pressure sequence according to the hysteresis mapping table, and performing boundary discrimination and threshold updating on a bias expansion result and a preset baseline proportion, wherein the generation of the threshold updating set specifically comprises the following steps: performing bias expansion on the pressure representative value corresponding to the attached pressure sequence according to the hysteresis mapping table; Performing boundary discrimination on the bias expansion result and a preset baseline proportion to generate a boundary discrimination mark; the preset baseline proportion is obtained by extracting a reference waveform segment from the sound emission waveform sequence in the trusted window set and performing proportion mapping on the energy sequencing result from the head end according to the reference waveform segment; and carrying out updating processing on the current threshold data according to the boundary distinguishing mark to generate a threshold updating set, wherein the current threshold data is obtained by carrying out combination operation on a reference waveform segment corresponding to the last trusted window and a preset baseline proportion.
5. The method for detecting pitting damage to a gear tooth surface according to claim 1, wherein the step of comparing the acoustic waveform sequence with a threshold according to the threshold update set, and counting the threshold comparison result in groups and taking the number of poles, and the step of generating a cycle candidate period specifically comprises the steps of: Performing threshold comparison on the sound emission waveform sequence according to the threshold updating set, and performing difference on two adjacent execution time sequences of the threshold comparison result to generate an interval sequence; And performing time sequence interval comparison and screening processing on the interval sequence according to the threshold updating set to generate a purifying interval sequence, counting the purifying interval sequence according to numerical groups, and taking out the purifying interval value with the largest current times as a cycle candidate period.
6. The method for detecting pitting damage to a gear tooth surface according to claim 3, wherein the step of performing windowing and offset averaging on the rotating reference sequence according to the trusted window set to generate a position centroid sequence, and performing segment fitting on the position centroid sequence and the attached pressure sequence to generate a phase offset function specifically comprises: carrying out window division and difference solving processing on the circumference-shifting reference sequence according to the trusted window set to generate a circumference-shifting set, and carrying out mean value solving on the circumference-shifting set to generate a position centroid sequence; Merging the pressure representative values within the range of adjacent trusted windows according to the numerical value proximity degree to generate a pressure segmentation set; And carrying out segmentation statistics and mean value solving on the position centroid sequence according to the pressure segments according to the pressure segmentation set to obtain the average offset of the pressure segments, correspondingly combining the pressure representative value and the average offset to generate mapping pairs, and further carrying out curve fitting on all the mapping pairs to generate a phase offset function.
7. The method for detecting pitting damage of a gear tooth surface according to claim 6, wherein the step of performing compensation processing on the rotating reference sequence according to the phase shift function to generate a pitting detection result specifically comprises the following steps: Performing compensation clustering on the turnover reference sequence according to the phase offset function to generate a phase belt index set, and performing consistency analysis on the sound emission waveform sequence in combination with the pressure direction mark under the constraint of the phase belt index set to generate a pitting evidence score; Performing weighted synthesis and differential operation on the threshold updating set, the pitting evidence score and the phase belt index set to generate a score gradient sequence; And performing gradient matching on the grading gradient sequence to generate a fastening suggestion sequence, and performing rechecking treatment after the fastening suggestion sequence is executed to generate a pitting corrosion detection result.
8. The method for detecting pitting damage to a gear tooth surface according to claim 7, wherein the step of performing compensation clustering on the circumferential reference sequence according to the phase offset function to generate a phase band index set, and performing consistency analysis on the acoustic waveform sequence in combination with the pressure direction mark under the constraint of the phase band index set, and the step of generating a pitting evidence score specifically comprises the following steps: Performing difference processing on a threshold comparison result and the circumference-shifting reference sequence to generate an event circumference-shifting quantity sequence, and performing compensation and cluster analysis on the event circumference-shifting quantity sequence according to the phase shifting function to generate a phase band index set; normalizing the attached pressure sequence according to the hysteresis mapping table to generate a proportional weight, calculating in-band and out-of-band energy comparison of the sound wave sequence under the constraint of the phase band index set and a threshold comparison result, weighting according to the proportional weight, and further summarizing by combining the pressure direction marks to generate an evidence consistency matrix; And carrying out consistency calculation on the evidence consistency matrix to generate a pitting evidence score.
9. The method for detecting pitting damage of a gear tooth surface according to claim 8, wherein the step of performing gradient matching on the scoring gradient sequence to generate a fastening advice sequence, and performing rechecking processing after the fastening advice sequence is executed, the step of generating a pitting detection result specifically comprises the steps of: performing discrete action matching on the target pressure interval and the scoring gradient sequence to generate a fastening suggestion sequence; The target pressure interval is generated by screening, sampling and extremum taking the attached pressure sequence according to the trusted window set; After the fastening suggestion sequence is executed, rechecking and recalculating the attached pressure sequence and the sound emission waveform sequence to generate a rechecking evidence grading and rechecking evidence consistency matrix, and carrying out differential synthesis judgment on the rechecking evidence grading and the rechecking evidence consistency matrix corresponding to the pitting evidence grading and the evidence consistency matrix respectively to generate a pitting detection result.
10. Gear tooth surface pitting damage acoustic emission detection system, which is characterized by comprising: the data acquisition module is used for acquiring an attaching pressure sequence and an acoustic waveform sequence of the target object; The sound signal processing module is used for judging the morphological consistency of the attached pressure sequence according to a sliding window to generate a trusted window set, and establishing hyperbolic mapping for the sound waveform sequence in the trusted window set from the head end and the tail end and the attached pressure sequence of the corresponding trusted window respectively to generate a hysteresis mapping table; The threshold updating module is used for carrying out bias expansion on the attached pressure sequence according to the hysteresis mapping table, and carrying out boundary judgment and threshold updating on a bias expansion result and a preset baseline proportion to generate a threshold updating set; the reference generation module is used for carrying out threshold comparison on the sound emission waveform sequence according to the threshold updating set, carrying out grouping counting on a threshold comparison result and taking the number of poles to generate a cycle of Zhou Houxuan, and further carrying out equidistant extrapolation on the threshold comparison result according to the cycle of Zhou Houxuan to generate a cycle of revolution reference sequence; The function fitting module is used for performing window division and offset averaging on the turnover reference sequence according to the trusted window set to generate a position centroid sequence, and performing segment fitting on the position centroid sequence and the attached pressure sequence to generate a phase offset function; and the detection result generation module is used for carrying out compensation processing on the turnover reference sequence according to the phase shift function to generate a pitting corrosion detection result.

Description

Acoustic emission detection method and system for pitting damage of gear tooth surface Technical Field The invention relates to the technical field of mechanical fault diagnosis, in particular to a gear tooth surface pitting damage acoustic emission detection method and system. Background The gear transmission system is widely applied to operation equipment in industries such as metallurgy, cement, chemical industry and the like, particularly in an external gear box matched with a high-temperature kiln, a gear runs in a high-load, heat radiation and temperature circulation environment for a long time, the tooth surface is easy to contact with fatigue damage, and in order to realize early identification and state evaluation of the pitting of the tooth surface of the gear, an acoustic emission detection technology is often used for monitoring the damage of the gear because the acoustic emission detection technology is sensitive to transient high-frequency elastic wave responses such as microcrack initiation, partial spalling of materials and the like. In the prior art, an acoustic emission detection scheme aiming at gear pitting generally carries out event triggering by a fixed threshold or a self-adaptive threshold based on waveform statistics, counts triggering events, and forms a revolution reference by assisting a sensor when phase positioning is required, so that the abnormal event density of a specific meshing phase section is judged; however, in the external gearbox scene of the high-temperature kiln, the gearbox shell and the clamp are in continuous heat radiation and temperature circulation environments, the attachment state of the sensor changes along with time, and the coupling transmission of the acoustic emission signals shows overall scaling and local distortion, so that the triggering result of a fixed threshold or a conventional self-adaptive threshold fluctuates along with the coupling state, and further leakage triggering and false triggering coexist. Disclosure of Invention The invention aims to provide a gear tooth surface pitting damage acoustic emission detection method and system, which are used for solving the problems in the background technology. In order to achieve the above object, the technical scheme of the present invention is as follows: The invention discloses a method for detecting pitting damage acoustic emission of a gear tooth surface, which is applied to detecting pitting damage acoustic emission of the gear tooth surface of an external gear box of a high-temperature kiln, and comprises the following steps: acquiring an attaching pressure sequence and an acoustic waveform sequence of a target object; Judging the morphological consistency of the attached pressure sequence according to a sliding window to generate a trusted window set, and establishing hyperbolic mapping for the sound wave form sequence in the trusted window set from the head end and the tail end and the attached pressure sequence of the corresponding trusted window respectively to generate a hysteresis mapping table; performing bias expansion on the attached pressure sequence according to the hysteresis mapping table, and performing boundary discrimination and threshold updating on a bias expansion result and a preset baseline proportion to generate a threshold updating set; performing threshold comparison on the sound emission waveform sequence according to the threshold updating set, performing grouping count on a threshold comparison result, taking the number of poles to generate a cycle Zhou Houxuan, and performing equidistant extrapolation on the threshold comparison result according to the cycle Zhou Houxuan to generate a cycle reference sequence; Performing windowing and offset averaging on the turnover reference sequence according to the trusted window set to generate a position centroid sequence, and performing segment fitting on the position centroid sequence and the attached pressure sequence to generate a phase offset function; And carrying out compensation processing on the rotating reference sequence according to the phase offset function to generate a pitting corrosion detection result. In a second aspect, the invention discloses a gear tooth surface pitting damage acoustic emission detection system, comprising: the data acquisition module is used for acquiring an attaching pressure sequence and an acoustic waveform sequence of the target object; The sound signal processing module is used for judging the morphological consistency of the attached pressure sequence according to a sliding window to generate a trusted window set, and establishing hyperbolic mapping for the sound waveform sequence in the trusted window set from the head end and the tail end and the attached pressure sequence of the corresponding trusted window respectively to generate a hysteresis mapping table; The threshold updating module is used for carrying out bias expansion on the attached pressure sequence according to th