CN-121973023-A - On-machine intelligent detection method and system for geometric parameters and grinding damage states of cutter

CN121973023ACN 121973023 ACN121973023 ACN 121973023ACN-121973023-A

Abstract

The invention relates to the technical field of precision machining and discloses an on-machine intelligent detection method and system for geometric parameters and grinding damage states of a cutter, wherein the detection method comprises the following steps of irradiating the cutter by a light source and collecting cutter images; the method comprises the steps of acquiring a cutter image acquired under irradiation, processing the cutter image acquired under irradiation, acquiring a grinding damage measured value based on the processed cutter image, extracting an edge image, identifying the cutter type and extracting cutter geometric parameters based on the edge image, determining a grinding damage threshold value in cutter wear detection based on the cutter type and the cutter geometric parameters, and comparing the grinding damage measured value with the grinding damage threshold value to judge whether the cutter parameter exceeds the tolerance or whether the grinding damage exceeds the standard. The invention solves the problems of false matching of cutter types, difficult timely identification of grinding damage, manual omission of detection, potential safety hazard, low reliability and the like in the prior detection technology in the processing of high-strength high-hardness materials.

Inventors

LIU HONGGUANG
YANG JIANXIN
FANG YUHENG
YANG ZHE
Jiao Yitong
Xue Yugun
ZHANG JUN
ZHAO WANHUA

Assignees

西安交通大学

Dates

Publication Date: 20260505
Application Date: 20260323

Claims (10)

1. An on-machine intelligent detection method for geometric parameters and grinding damage states of a cutter is characterized by comprising the following steps: illuminating the cutter by adopting a light source, and collecting a cutter image; Processing the cutter image collected under irradiation; Based on the processed cutter image, acquiring a grinding damage measured value, extracting an edge image, and identifying the cutter type and extracting cutter geometric parameters based on the edge image; And comparing the measured value of the grinding damage with the threshold value of the grinding damage to judge whether the parameters of the cutter are out of tolerance or whether the grinding damage is out of standard.
2. The on-machine intelligent detection method for geometric parameters and wear state of a cutter according to claim 1, wherein the cutter is irradiated by a light source and an image of the cutter is acquired, comprising the steps of: and carrying out combined irradiation on the cutter by using a Dome light source and a point light source at the other side of the back parallel light source, and collecting the cutter image at the other side of the back parallel light source.
3. The on-machine intelligent detection method for tool geometric parameters and wear state according to claim 1, wherein the processing of the tool image collected under irradiation comprises the following steps: and performing one or more of five processing steps of image preprocessing, gaussian filtering, bilateral filtering, otsu binarization and morphological operation on the cutter image.
4. The on-machine intelligent detection method for tool geometry and wear and tear state according to claim 1, wherein the step of obtaining a wear and tear measurement value based on the machined tool image comprises the steps of: And introducing a Markov random field into a break area of the processed cutter image for modeling, dividing the break area of the cutter by a pixel second-order neighborhood system and a Gibbs energy function, obtaining the division of a perfect area, the break area and the background by solving global energy minimization, and calculating the proportion of the break area.
5. The on-machine intelligent detection method for tool geometry and wear state according to claim 1, wherein the edge image is extracted based on the processed tool image, comprising the steps of: and obtaining a subpixel level edge by adopting a Canny operator and a Zernike moment method on the processed cutter image.
6. An on-machine intelligent detection system for geometrical parameters and grinding damage states of a cutter is used for realizing the on-machine intelligent detection method for the geometrical parameters and grinding damage states of the cutter according to any one of claims 1-5, and is characterized by comprising a box body (1), a sliding assembly, a lifting adjusting bracket (6), side plates (10), a light source, an industrial camera (7) and an intelligent processor, wherein the sliding assembly comprises a sliding rail (3) and a sliding block (13), the light source comprises a main light source and a back parallel light source (12), the sliding rail (3) is arranged in the box body (1), a supporting block (4) is fixedly connected to the sliding rail (3), a fixing plate (5) is fixedly connected to the supporting block (4), the lifting adjusting bracket (6) is provided with the main light source and the industrial camera (7), the side plates (10) are fixedly connected to the sliding block (13), the side plates (10) are connected with the back parallel light source (12), and the intelligent camera (7) is electrically connected to the industrial camera.
7. The intelligent detection system for the geometric parameters and the grinding damage state of the cutter according to claim 6 is characterized by further comprising a crank-rocker assembly and a miniature motor, wherein the crank-rocker assembly and the miniature motor are arranged in the box body (1), the light source further comprises a point light source (8), a rocker of the crank-rocker assembly is fixedly connected with the point light source (8), and a crank of the crank-rocker assembly is driven by the miniature motor.
8. The intelligent detection system for tool geometry and wear state according to claim 6, wherein, The intelligent processor includes: The raspberry pie is used for controlling and processing data of the intelligent detection system; the storage module is used for storing the grinding damage threshold value data; The GPIO expansion module is used for supporting signal transmission and interacting with external equipment; The computer module is used for collecting, analyzing and storing edge image data of the cutter and outputting display signals; the machine tool numerical control system network port is used for interacting with an external numerical control machine tool and reading on-machine tool geometric parameters; The raspberry pie comprises: the power supply interface module is used for being externally connected with the power supply module; The cooling fan module is used for radiating heat for raspberries; the GPIO pin module is used for being externally connected with a GPIO expansion module, a miniature motor and a control end of an industrial camera (7); The USB3.0 module is used for externally connecting a data end of the industrial camera (7); The gigabit Ethernet port is used for externally connecting with a network port of a numerical control system of the machine tool; And the memory module is used for storing data.
9. A computer readable storage medium having instructions stored therein, which when run on an electronic device, cause the electronic device to perform a method of on-machine intelligent detection of a tool geometry and wear state according to any one of claims 1-5.
10. A computer program product comprising instructions which, when run on an electronic device, cause said electronic device to perform a method of intelligent detection of a tool geometry and wear state according to any of claims 1-5.

Description

On-machine intelligent detection method and system for geometric parameters and grinding damage states of cutter Technical Field The invention relates to the technical field of precision machining, in particular to an on-machine intelligent detection method and system for geometric parameters and grinding damage states of a cutter. Background In the fields of high-end manufacturing of aerospace, precision dies and the like, high-strength and high-hardness materials such as 15-5PH and the like are increasingly widely applied, and the materials are high in hardness and cutting force and provide strict requirements for stability and machining consistency of cutters. However, in actual production, the mismatching of the cutter type (such as using a rough machining cutter for a finish machining scene) or the failure of grinding damage is not recognized in time (such as that the abrasion loss of a rear cutter surface exceeds 0.2mm and the abrasion loss still continues to be used), so that the surface quality of a workpiece is out of tolerance, the dimensional accuracy is out of control, even the linkage faults such as vibration of a machine tool spindle and bearing damage are caused, the continuous and stable operation of an unmanned production line is severely restricted, and the production efficiency cliff type drop and the maintenance cost are greatly increased. The existing tool detection technology has obvious short plates, wherein manual detection is dependent on experience and is easily influenced by factors such as fatigue and emotion, missing detection occurs, radioactive substances are required to be introduced in an isotope labeling method, potential safety hazards exist and the environment is polluted, spindle current monitoring is easily interfered by load fluctuation, vibration acoustic wave analysis is sensitive to environmental noise, reliability is obviously insufficient, contact detection is required to stop for dismounting tools, an off-line scheme is required to be provided with additional equipment, single detection cost is high and low in efficiency, part of machine vision schemes realize non-contact detection but are influenced by factors such as illumination, reflection and greasy dirt, stability and repeated positioning accuracy fluctuation are large, and the requirement of high-end manufacturing on micron-level accuracy is difficult to be met. Disclosure of Invention The invention aims to provide an on-machine intelligent detection method and system for geometric parameters and grinding damage states of a cutter, which are used for solving the problems that in the processing of high-strength and high-hardness materials, the cutter type is mismatched, grinding damage is difficult to identify in time, manual omission is caused by hidden danger, reliability is low, cost is high, precision is insufficient and the like in the existing detection technology, the cutter type is mismatched by adopting a 'first calibration and then monitoring' process design, quick response is realized through a full-process strong real-time design, high-precision analysis of grinding damage is realized by a rear-end computer depending on calculation force, an acquisition device is adapted to shooting requirements of different cutters, an adjustable support movable mechanism and a customized light supplementing module are carried, shooting requirements of cutters with different specifications can be adapted, light interference of a processing site is eliminated, and image acquisition quality and parameter extraction precision are ensured. In order to achieve the above object, the present invention provides the following technical solutions. The invention provides an on-machine intelligent detection method for the geometric parameters and the grinding damage state of a cutter, which comprises the following steps of irradiating the cutter by adopting a light source, collecting cutter images, processing the cutter images collected under irradiation, obtaining grinding damage measured values based on the cutter images after processing, extracting edge images, identifying the cutter type and extracting the geometric parameters of the cutter based on the edge images, determining a grinding damage threshold value in cutter abrasion detection based on the cutter type and the geometric parameters of the cutter, and comparing the grinding damage measured values with the grinding damage threshold value to judge whether the cutter parameters exceed the tolerance or whether the grinding damage exceeds the standard. The invention further improves the method by adopting a light source to irradiate the cutter and collect the cutter image, and the method comprises the following steps of adopting a back parallel light source to irradiate the cutter on one side and collect the cutter image on the other side of the back parallel light source, adopting a Dome light source and a point light source to irradiate the cutter on the other s