CN111251187A - Method and device for fusing information and extracting characteristics of blade grinding burn - Google Patents
Method and device for fusing information and extracting characteristics of blade grinding burn Download PDFInfo
- Publication number
- CN111251187A CN111251187A CN202010044861.5A CN202010044861A CN111251187A CN 111251187 A CN111251187 A CN 111251187A CN 202010044861 A CN202010044861 A CN 202010044861A CN 111251187 A CN111251187 A CN 111251187A
- Authority
- CN
- China
- Prior art keywords
- grinding
- information
- grinding burn
- burn
- blade
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B24—GRINDING; POLISHING
- B24B—MACHINES, DEVICES, OR PROCESSES FOR GRINDING OR POLISHING; DRESSING OR CONDITIONING OF ABRADING SURFACES; FEEDING OF GRINDING, POLISHING, OR LAPPING AGENTS
- B24B49/00—Measuring or gauging equipment for controlling the feed movement of the grinding tool or work; Arrangements of indicating or measuring equipment, e.g. for indicating the start of the grinding operation
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B24—GRINDING; POLISHING
- B24B—MACHINES, DEVICES, OR PROCESSES FOR GRINDING OR POLISHING; DRESSING OR CONDITIONING OF ABRADING SURFACES; FEEDING OF GRINDING, POLISHING, OR LAPPING AGENTS
- B24B49/00—Measuring or gauging equipment for controlling the feed movement of the grinding tool or work; Arrangements of indicating or measuring equipment, e.g. for indicating the start of the grinding operation
- B24B49/003—Measuring or gauging equipment for controlling the feed movement of the grinding tool or work; Arrangements of indicating or measuring equipment, e.g. for indicating the start of the grinding operation involving acoustic means
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B24—GRINDING; POLISHING
- B24B—MACHINES, DEVICES, OR PROCESSES FOR GRINDING OR POLISHING; DRESSING OR CONDITIONING OF ABRADING SURFACES; FEEDING OF GRINDING, POLISHING, OR LAPPING AGENTS
- B24B49/00—Measuring or gauging equipment for controlling the feed movement of the grinding tool or work; Arrangements of indicating or measuring equipment, e.g. for indicating the start of the grinding operation
- B24B49/006—Measuring or gauging equipment for controlling the feed movement of the grinding tool or work; Arrangements of indicating or measuring equipment, e.g. for indicating the start of the grinding operation taking regard of the speed
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J11/00—Manipulators not otherwise provided for
- B25J11/005—Manipulators for mechanical processing tasks
- B25J11/0065—Polishing or grinding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Robotics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Acoustics & Sound (AREA)
- Finish Polishing, Edge Sharpening, And Grinding By Specific Grinding Devices (AREA)
Abstract
The invention discloses a method for fusing and extracting characteristics of blade grinding burn information, which specifically comprises the following steps: establishing a grinding monitoring system by utilizing multiple sensors; performing robot grinding and polishing processing on the blade according to different preset grinding and processing parameters; collecting grinding burn multi-sensing information through the monitoring system to obtain grinding burn multi-sensing information under different processing parameters; carrying out information fusion and feature extraction on the grinding burn multi-sensing information; and determining a characteristic value which is in accordance with a preset standard as a grinding burn invariance characteristic value by comparing the grinding burn multi-sensing information under different processing parameters. Through the scheme, comprehensive monitoring on the machining of the robot with the complex curved surface can be realized, so that the machining quality of the robot with the blades is effectively guaranteed. And the abrasive belt grinding burn invariance characteristic of the blade robot can be effectively extracted, the grinding burn misjudgment caused by the change of the processing parameters is avoided, and the processing efficiency and the processing quality of the blade are further ensured.
Description
Technical Field
The invention belongs to the technical field of monitoring and control of a machining process, and particularly relates to a method and a device for fusing and extracting blade grinding burn information.
Background
The blade is one of the key parts of the aircraft engine, and the quality of the processing quality of the blade is directly related to the overall mechanical performance of the aircraft engine. Due to the complexity of the profile of the blade, the control of the robot machining process is extremely difficult, so that the surface quality of the blade cannot be well guaranteed, particularly the phenomenon of grinding burn is always a difficult point in the field of blade robot machining, and the popularization and application of the robot machining technology in the field of two machines are greatly limited.
Grinding burn is the phenomenon that the surface layer tissue of a workpiece changes under the action of transient high temperature in the grinding process, and oxidation and discoloration appear on part of the processed surface layer, which can cause the surface material of the workpiece to be re-hardened, so that the surface of the workpiece generates residual stress, and the fatigue life and the stress corrosion performance of the workpiece are seriously reduced. At present, most of existing grinding burn feature extraction methods are based on single-sensor information extraction, such as: the method mainly utilizes a single sensor to collect information in the processing process and extracts information characteristics related to grinding burn from the information so as to judge whether the grinding burn occurs.
However, the above-mentioned extraction of single-sensor information has many disadvantages: firstly, because the extracted grinding burn characteristics (such as signal amplitude, effective value and the like) are extremely sensitive to the change of grinding processing parameters, the misjudgment of the grinding burn is easily caused; secondly, the information detected by a single sensor is difficult to comprehensively reflect the processing process of the robot and cannot be used as a reliable characteristic for judging whether grinding burn occurs or not.
Therefore, there is a need for a leaf grinding burn information fusion and feature extraction method based on multiple sensors to overcome the defects of the prior art.
Disclosure of Invention
The invention aims to provide a method for fusing and extracting characteristics of blade grinding burn information, which can realize comprehensive monitoring on the machining of a complex robot, thereby effectively ensuring the machining quality of a blade robot.
In order to solve the above technical problems, the present invention provides the following technical solutions, including:
establishing a grinding monitoring system by utilizing multiple sensors;
performing robot grinding and polishing processing on the blade according to different preset grinding and processing parameters;
collecting grinding burn multi-sensing information through the monitoring system to obtain grinding burn multi-sensing information under different processing parameters;
carrying out information fusion and feature extraction on the grinding burn multi-sensing information;
and determining a characteristic value which is in accordance with a preset standard as a grinding burn invariance characteristic value by comparing the grinding burn multi-sensing information under different processing parameters.
Preferably, the system for establishing grinding monitoring by using multiple sensors comprises:
mounting a first sensor on a flange at the tail end of an industrial robot, and calibrating the first sensor;
mounting a blade clamp at the first sensor end;
calibrating a second sensor, and installing the second sensor at a first preset position;
calibrating the thermal imager, and installing the thermal imager at a second preset position.
Preferably, the first sensor is a six-dimensional force sensor;
correspondingly, calibrating the second sensor and installing the second sensor in a first predetermined location comprises:
calibrating an acoustic emission sensor, smearing a coupling agent on the tail end of the blade clamp, and installing the acoustic emission sensor on the tail end of the blade clamp;
calibrating an acceleration sensor, and mounting the acceleration sensor on a contact wheel mounting beam of the abrasive belt grinding and polishing machine;
and calibrating a current sensor and a voltage sensor, and installing the current sensor and the voltage sensor at the input end of the abrasive belt grinding and polishing machine motor.
Preferably, the performing robot grinding and polishing processing on the blade according to different preset grinding and polishing parameters comprises:
presetting different grinding processing parameters;
accordingly, the presetting of different grinding parameters includes:
selecting processing parameters;
fixing any two of the processing parameters, and gradually changing the rest parameters;
carrying out grinding and polishing processing on the blades by using the processing parameters, and detecting whether the blades are burnt or not;
wherein the step of detecting whether the leaves are burnt is as follows:
and repeating the test for 3 times, recording the set of processing parameters if the burn and the non-burn appear in the 3 tests, and otherwise, continuously changing the rest parameters until the burn of the leaves appears.
Preferably, the information fusion and feature extraction of the grinding burn multi-sensing information comprises:
fusing the collected multi-sensing information based on a specific analysis method;
performing wavelet transformation on the fused information according to a preselected wavelet basis function to obtain an original wavelet coefficient;
carrying out denoising processing on the original wavelet coefficient by using a preselected threshold denoising function to obtain a reconstructed wavelet coefficient;
reconstructing a signal according to the reconstructed wavelet coefficient to obtain denoised fusion information;
performing Hilbert-Huang transformation on the de-noised fusion information, wherein the Hilbert-Huang transformation comprises an empirical mode decomposition part and a Hilbert transformation part, and obtaining an intrinsic mode component of the de-noised fusion information after the empirical mode decomposition;
performing a hilbert transform and a fourier transform on the eigenmode components;
extracting the statistical features of each of the eigenmode components includes, but is not limited to: root mean square value, average energy, kurtosis, average amplitude, power spectrum characteristic value.
Preferably, the performing wavelet transform on the fused information according to the preselected wavelet basis function comprises:
performing wavelet transformation on the fused information according to the preselected wavelet basis function and the decomposition layer number;
the fusion of the collected multi-sensing information based on a specific analysis method comprises the following steps of;
and fusing the collected multi-sensing information based on a kernel principal component analysis method.
Preferably, the determining a characteristic value conforming to a preset standard as a grinding burn invariance characteristic value by comparing the grinding burn multi-sensing information under the different processing parameters includes:
and determining a characteristic value which is high in association degree with the grinding burn and small in conversion along with the processing parameters as a grinding burn invariance characteristic value by comparing the grinding burn multi-sensing information under different processing parameters.
In addition, in order to achieve the above object, the present invention further provides a blade grinding burn information fusion and feature extraction device, including:
the monitoring unit is used for establishing a grinding monitoring system by utilizing multiple sensors;
the grinding and polishing unit is used for carrying out robot grinding and polishing on the blade according to different preset grinding and polishing parameters;
the information acquisition unit is used for acquiring grinding burn multi-sensing information through the monitoring system so as to obtain the grinding burn multi-sensing information under different processing parameters;
the information processing unit is used for carrying out information fusion and feature extraction on the grinding burn multi-sensing information;
and the characteristic determining unit is used for determining a characteristic value which is in accordance with a preset standard as a grinding burn invariance characteristic value by comparing the grinding burn multi-sensing information under different processing parameters.
Preferably, the information collecting unit includes:
the information fusion unit is used for fusing the collected multi-sensing information based on a specific analysis method;
the wavelet transformation unit is used for performing wavelet transformation on the fused information according to a preselected wavelet basis function to obtain an original wavelet coefficient;
a denoising unit; the system comprises a pre-selection threshold denoising function, a reconstruction wavelet coefficient and a wavelet coefficient acquisition unit, wherein the pre-selection threshold denoising function is used for denoising the original wavelet coefficient to obtain the reconstruction wavelet coefficient;
a reconstructed signal unit; the fusion information processing unit is used for reconstructing a signal according to the reconstructed wavelet coefficient to obtain denoised fusion information;
the first information transformation unit is used for performing Hilbert-Huang transformation on the de-noised fusion information, wherein the Hilbert-Huang transformation comprises an empirical mode decomposition part and a Hilbert transformation part, and an intrinsic mode component of the de-noised fusion information is obtained after the empirical mode decomposition;
a second information transformation unit, configured to perform hilbert transform and fourier transform on the eigenmode component;
a feature extraction unit, configured to extract statistical features of each of the eigenmode components, including but not limited to: root mean square value, average energy, kurtosis, average amplitude, power spectrum characteristic value.
Preferably, the feature determination unit includes:
and the characteristic value which is high in association degree with the grinding burn and small in conversion along with the processing parameters is determined as the grinding burn invariance characteristic value by comparing the grinding burn multi-sensing information under different processing parameters.
The invention provides a method for fusing and extracting characteristics of blade grinding burn information, which comprises the following steps: establishing a grinding monitoring system by utilizing multiple sensors; performing robot grinding and polishing processing on the blade according to different preset grinding and processing parameters; collecting grinding burn multi-sensing information through the monitoring system to obtain grinding burn multi-sensing information under different processing parameters; carrying out information fusion and feature extraction on the grinding burn multi-sensing information; and determining a characteristic value which is in accordance with a preset standard as a grinding burn invariance characteristic value by comparing the grinding burn multi-sensing information under different processing parameters. Through the scheme, comprehensive monitoring on the machining of the robot with the complex curved surface can be realized, so that the machining quality of the robot with the blades is effectively guaranteed. And the abrasive belt grinding burn invariance characteristic of the blade robot can be effectively extracted, the grinding burn misjudgment caused by the change of the processing parameters is avoided, and the processing efficiency and the processing quality of the blade are further ensured.
Drawings
FIG. 1 is a flow chart of a method for fusing information and extracting characteristics of blade grinding burn according to an embodiment of the present invention;
FIG. 2 is a diagram of a multi-sensor monitoring system architecture in a method for information fusion and feature extraction for blade grinding burn according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a multi-sensor monitoring system in a method for information fusion and feature extraction of blade grinding burn according to an embodiment of the present invention;
FIG. 4 is a schematic flow chart illustrating a method for obtaining grinding burn processing parameters in a blade grinding burn information fusion and feature extraction method according to an embodiment of the present invention;
FIG. 5 is a schematic flow chart of a combined feature extraction method based on wavelet transform and Hilbert-Huang transform in the leaf grinding burn information fusion and feature extraction method according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of a device for information fusion and feature extraction of blade grinding burn according to an embodiment of the present invention;
in all the figures, the same reference numerals denote the same features, in particular: 1-six-dimensional force sensor, 2-acoustic emission sensor, 3-acceleration sensor, 4-current voltage sensor, 5-infrared thermal imager, 6-low speed acquisition card, 7-high speed data acquisition card, 8-preamplifier, 9-preamplifier power supply separation signal, 10-charge amplifier, 11-contact wheel installation beam of abrasive belt polishing machine, 12-robot control cabinet, 13-PC end and 14-robot.
Detailed Description
The principles and spirit of the present invention will be described with reference to a number of exemplary embodiments. It is understood that these embodiments are given solely for the purpose of enabling those skilled in the art to better understand and to practice the invention, and are not intended to limit the scope of the invention in any way. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
Referring to fig. 1, the invention provides a flow chart of a method for information fusion and feature extraction of blade grinding burn, which specifically comprises the following steps:
s100: establishing a grinding monitoring system by utilizing multiple sensors;
s200: performing robot grinding and polishing processing on the blade according to different preset grinding and processing parameters;
s300: collecting grinding burn multi-sensing information through the monitoring system to obtain grinding burn multi-sensing information under different processing parameters;
s400: carrying out information fusion and feature extraction on the grinding burn multi-sensing information;
s500: and determining a characteristic value which is in accordance with a preset standard as a grinding burn invariance characteristic value by comparing the grinding burn multi-sensing information under different processing parameters.
The embodiment of the invention provides a blade grinding burn information fusion and feature extraction method, which is described below by taking the abrasive belt grinding of an aircraft engine blade robot as an example, but the application scene of the invention is not limited to the abrasive belt grinding of the aircraft engine blade robot.
S100: establishing a grinding monitoring system by utilizing multiple sensors;
a grinding monitoring system is established by utilizing multiple sensors, which is the premise of blade grinding and burning information fusion and characteristic extraction, and grinding and burning multiple sensing information under different grinding and processing parameters is acquired on line through the system. A specific system architecture diagram is shown in fig. 2, which is an architecture diagram of a multi-sensor monitoring system in a blade grinding burn information fusion and feature extraction method according to an embodiment of the present invention, and a specific system schematic diagram is shown in fig. 3, which is a schematic diagram of a multi-sensor monitoring system in a blade grinding burn information fusion and feature extraction method according to an embodiment of the present invention. In particular, in fig. 3, like reference numerals denote like features, in particular: 1-six-dimensional force sensor, 2-acoustic emission sensor, 3-acceleration sensor, 4-current voltage sensor, 5-infrared thermal imager, 6-low speed acquisition card, 7-high speed data acquisition card, 8-preamplifier, 9-preamplifier power supply separation signal, 10-charge amplifier, 11-contact wheel installation beam of abrasive belt polishing machine, 12-robot control cabinet, 13-PC end and 14-robot.
Wherein, the six-dimensional force sensor 1 is arranged at the tail end of a flange plate of a robot 14 and is connected with a robot control cabinet 12. The acoustic emission sensor 2 is bonded at the end of the clamp through a coupling agent and is connected with the high-speed data acquisition card 7 through a preamplifier 8 and a preamplifier power supply separation signal device 9 in sequence. The acceleration sensor 3 is arranged on a contact wheel mounting beam 11 of the abrasive belt grinding and polishing machine and is connected with the high-speed data acquisition card 7 through a charge amplifier 10. The voltage and current sensor 4 is arranged at the input end of the motor of the abrasive belt grinding and polishing machine and is connected with the low-speed acquisition card 6. The position where the thermal infrared imager 5 is mounted is not limited to the mounting position shown in fig. 3. The low-speed acquisition card 6, the high-speed acquisition card 7, the robot control cabinet 12 and the infrared thermal imager 5 are all connected with a PC end 13.
Further, the system for establishing grinding monitoring by utilizing the multiple sensors mainly comprises the following steps:
s101: a first sensor is arranged on a flange at the tail end of the industrial robot 14 and is calibrated, and the first sensor is connected with the robot control cabinet 12 through a data line;
s102: mounting a blade clamp at the first sensor end;
s103: calibrating a second sensor, and installing the second sensor at a first preset position;
s104: calibrating the thermal imager 5, and installing the thermal imager 5 at a second preset position.
Wherein, S101: the first sensor is mounted on the end flange of the industrial robot 14 and calibrated.
Specifically, a first sensor is mounted on the end flange of the industrial robot and calibrated, wherein the first sensor is connected with the robot control cabinet 12 through a data line. The first sensor is preferably a six-dimensional force sensor 1. The six-dimensional force sensor 1 is one of multi-dimensional force sensors, and the multi-dimensional force sensor refers to a force sensor capable of measuring force and moment components in more than two directions simultaneously, and the force and the moment can be respectively decomposed into three components in a Cartesian coordinate system, so that the most complete form of the multi-dimensional force is a six-dimensional force/moment sensor, namely, a sensor capable of measuring three force components and three moment components simultaneously, and the widely used multi-dimensional force sensor is the sensor. The six-dimensional force sensor 1 is an important sensor of the intelligent robot, and can simultaneously detect full force information of a three-dimensional space (a cartesian coordinate system), namely three force components and three moment components. Therefore, the six-dimensional force sensor 1 is preferably mounted as a first sensor on the end flange of the industrial robot 14 and calibrated in the embodiment of the invention.
S102: mounting a blade clamp at the first sensor end;
specifically, an air vane clamp is arranged at the tail end of the six-dimensional force sensor 1 and mainly plays a role in fixing.
S103: and calibrating a second sensor, and installing the second sensor at a first preset position.
Specifically, a second sensor is calibrated and mounted at a first preset position. The second sensor here preferably comprises 4 sensors, in particular: acoustic emission sensor 2, acceleration sensor 3, current sensor and voltage sensor 4.
Illustratively, calibrating the second sensor and mounting the second sensor at the first predetermined location comprises:
S103A: calibrating an acoustic emission sensor 2, smearing a coupling agent on the tail end of the blade clamp, and installing the acoustic emission sensor 2 at the tail end of the blade clamp;
S103B: calibrating an acceleration sensor 3, and mounting the acceleration sensor 3 on a contact wheel mounting beam 11 of the abrasive belt grinding and polishing machine;
S103C: calibrating a current sensor and a voltage sensor 4, and installing the current sensor and the voltage sensor 4 at the input end of the abrasive belt grinding and polishing machine motor;
wherein, S103A: calibrating the acoustic emission sensor 2, smearing a coupling agent on the tail end of the blade clamp, and installing the acoustic emission sensor 2 at the tail end of the blade clamp. In addition, the couplant is fully filled between the two ends to ensure the signal quality, and the acoustic emission sensor 2 is connected with the high-speed data acquisition card 7 sequentially through a preamplifier 8 and a preamplifier power supply separation signal device 9.
Wherein, S103B: and calibrating the acceleration sensor 3, and mounting the acceleration sensor 3 on a contact wheel mounting beam 11 of the abrasive belt grinding and polishing machine. The acceleration sensor 3 is connected with the high-speed data acquisition 7 card through a charge amplifier 10;
wherein, S103C: calibrating a current sensor and a voltage sensor 4, and installing the current sensor and the voltage sensor 4 at the input end of the motor of the abrasive belt grinding and polishing machine. The current sensor and the voltage sensor 4 are connected with a low-speed acquisition card 6
Wherein, S104: calibrating the thermal imaging camera 5, and installing the thermal imaging camera 5 at a second preset position, wherein the second preset position is a position convenient for measuring the grinding temperature, and is not limited to the position shown in fig. 3. The robot control cabinet 12, the low-speed acquisition card 6/the high-speed acquisition card 7 and the thermal imager 5 are all connected with a PC end 13. The thermal imager 5 according to the embodiment of the present invention is preferably an infrared thermal imager.
Further, S200: performing robot grinding and polishing processing on the blade according to different preset grinding and processing parameters;
in practical application, the blade is subjected to robot grinding and polishing according to different preset grinding and polishing parameters, different grinding and polishing parameters are mainly set, the aircraft engine blade is subjected to robot grinding and polishing, and force, temperature, vibration, power and acoustic emission signals in the machining process when grinding and burning occur are acquired on line.
At S200: the robot grinding and polishing processing of the blade according to different preset grinding and processing parameters comprises the following steps:
s200': presetting different grinding processing parameters;
specifically, referring to fig. 4, a schematic flow chart of a method for acquiring grinding burn processing parameters in a blade grinding burn information fusion and feature extraction method according to an embodiment of the present invention is shown, where the presetting of different grinding processing parameters includes:
s201: selecting processing parameters;
s202: fixing any two of the processing parameters, and gradually changing the rest parameters;
s203: carrying out grinding and polishing processing on the blade by using the processing parameters;
s204: detecting whether the leaves are burnt or not;
s205: if yes, the burn frequency i is i + 1;
s206: the test times n are n + 1;
s207: if not, the test times n are n + 1;
s208: judging whether the test times n are more than 3;
if not, returning to the step S203;
if yes, go to step S209: judging whether i is equal to 0;
if yes, returning to the step S202;
if not, recording the group of parameters and reselecting the parameters.
For example, the processing parameters are first selected, any two of the processing parameters are fixed again, and the remaining parameters are gradually changed, wherein any two of the processing parameters may be two of the rotation speed of the abrasive belt, the feeding speed of the robot and the grinding force, and the remaining parameters are the other parameters. Recording the test result: the number of tests n is 0 and the number of burns i is 0. Using the processing parameters to carry out grinding and polishing processing on the blades by a robot, and detecting whether the blades are burnt or not; if the burn is i ═ i +1, further, n ═ n +1, and if there is no burn, n ═ n + 1. And judging whether n is more than 3 times, if not, returning to the step S203, if so, further judging whether i is equal to 0, if so, returning to the step S202, otherwise, recording the group of parameters, and reselecting the parameters for testing.
Further, S300: collecting grinding burn multi-sensing information through the monitoring system to obtain grinding burn multi-sensing information under different processing parameters;
here, S300: and acquiring grinding burn multi-sensing information through the monitoring system so as to obtain the grinding burn multi-sensing information under different processing parameters. The built multi-sensor monitoring system is used for collecting grinding burn multi-sensing information for subsequent information fusion and feature extraction.
Further, S400: carrying out information fusion and feature extraction on the grinding burn multi-sensing information;
in the present embodiment, S400: and performing information fusion and feature extraction on the grinding burn multi-sensing information, namely processing the information acquired in the step S300.
The method mainly comprises the following steps:
s401: fusing the collected multi-sensing information based on a specific analysis method;
s402: performing wavelet transformation on the fused information according to a preselected wavelet basis function to obtain an original wavelet coefficient;
s403: carrying out denoising processing on the original wavelet coefficient by using a preselected threshold denoising function to obtain a reconstructed wavelet coefficient;
s404: reconstructing a signal according to the reconstructed wavelet coefficient to obtain denoised fusion information;
s405: performing Hilbert-Huang transformation on the de-noised fusion information, wherein the Hilbert-Huang transformation comprises an empirical mode decomposition part and a Hilbert transformation part, and obtaining an intrinsic mode component of the de-noised fusion information after the empirical mode decomposition;
s406: performing a hilbert transform and a fourier transform on the eigenmode components;
s407: extracting the statistical features of each of the eigenmode components includes, but is not limited to: root mean square value, average energy, kurtosis, average amplitude, power spectrum characteristic value.
Wherein, S401: fusing the collected multi-sensing information based on a specific analysis method;
in this step, the collected multi-sensing information is fused based on a specific analysis method, that is, the collected multi-sensing information is fused by the analysis method. The particular analysis method used herein is preferably a kernel principal component analysis method. The kernel principal component analysis method specifically comprises the following steps: in the continuous production process, various faults inevitably occur in the production system during long-term operation and production load, the production quality is influenced, and even great economic loss is caused, while the chemical production system generally has the characteristics of accurate process, difficult modeling, numerous process variables, strong coupling among the process variables, influence of various random factors in practice and the like. This makes the use of diagnostic methods based on mechanistic models extremely inconvenient. The Kernel Principal Component Analysis (KPCA) is a modeling method independent of process mechanisms, which only needs to perform statistical modeling through information of process data and then realizes monitoring of a process based on the model. Therefore, the principal component analysis is a mature multivariate statistical monitoring method.
Wherein, S402: performing wavelet transformation on the fused information according to the preselected wavelet basis function and the decomposition layer number to obtain an original wavelet coefficient;
in this step, wavelet transform is performed on the fused information according to a preselected wavelet basis function to obtain an original wavelet coefficient.
Wherein, S403: carrying out denoising processing on the original wavelet coefficient by using a preselected threshold denoising function to obtain a reconstructed wavelet coefficient;
in the step, the original wavelet coefficient is denoised by using a preselected threshold denoising function to obtain a reconstructed wavelet coefficient. The embodiment of the invention preferably adopts a wavelet threshold denoising algorithm. The wavelet threshold denoising algorithm has the advantages of simple algorithm, high calculation efficiency and the like, and can ensure that the useful information and the noise information in the fusion information are separated on a wavelet transform domain to the maximum extent, so that the useful information is reserved and recovered to the maximum extent.
However, the specific threshold denoising function method is not limited as long as the noise can be effectively removed.
Wherein, S404: reconstructing a signal according to the reconstructed wavelet coefficient to obtain denoised fusion information;
in this step, a signal is reconstructed according to the reconstructed wavelet coefficient, and the de-noised fusion information is obtained.
Wherein, S405: performing Hilbert-Huang transformation on the de-noised fusion information, wherein the Hilbert-Huang transformation comprises an empirical mode decomposition part and a Hilbert transformation part, and obtaining an intrinsic mode component of the de-noised fusion information after the empirical mode decomposition;
in this step, hilbert-yellow transform is performed on the de-noised fusion information, where the hilbert-yellow transform includes two parts, namely empirical mode decomposition and hilbert transform, and an intrinsic mode component of the de-noised fusion information is obtained after the empirical mode decomposition. Hilbert-Huang Transform (HHT), the main content of HHT comprises two parts, the first part is Empirical Mode Decomposition (EMD), which is proposed by Huang; the second part is Hilbert Spectroscopy (HSA). Briefly, the basic process of HHT processing non-stationary signals is: firstly, decomposing a given signal into a plurality of Intrinsic Mode functions (expressed by Intrinsic Mode functions or IMFs, also called Intrinsic Mode functions) by using an EMD method, wherein the IMFs are components meeting certain conditions; then, Hilbert transformation is carried out on each IMF to obtain a corresponding Hilbert spectrum, namely each IMF is represented in a combined time-frequency domain; finally, summing the Hilbert spectra of all IMFs results in the Hilbert spectrum of the original signal.
Referring to fig. 5 in particular, a flow diagram of a combined feature extraction method based on wavelet transform and hilbert-yellow transform in the blade grinding burn information fusion and feature extraction method according to an embodiment of the present invention is shown.
The method mainly comprises the following steps:
s601: selecting a wavelet basis function;
s602: determining the number N of wavelet decomposition layers;
s603: performing wavelet decomposition on the fusion information of the sensor;
s604: determining a wavelet threshold function;
s605: judging whether the number I of the current wavelet layers is larger than N;
S605A: if not, determining the wavelet coefficient threshold of the current layer;
S605B: updating the wavelet coefficient of the current layer;
S605C: updating I to I + 1;
S615A: if yes, reconstructing multi-sensor fusion information by using the updated wavelet coefficients;
S615B: performing empirical mode decomposition on multi-sensor fusion information to obtain a plurality of layers of IMF components;
S615C: calculating the time domain statistic of each layer of IMF component;
S615D: performing HHT transformation and FFT transformation on each layer of IMF;
S615E: computing the frequency domain statistics of each layer of IMF, such as: root mean square value, kurtosis, etc.;
S615F: recording and storing all calculation results as characteristic values;
S615G: and judging all the characteristic values, and selecting the grinding burn invariance characteristic value.
Specifically, the device mainly comprises two parts: firstly, selecting a proper wavelet basis function and a proper decomposition layer number, and carrying out denoising processing on fused multi-sensing information by combining a proper wavelet threshold function; then, performing Hilbert-Huang transform on the denoised fusion information, wherein the process mainly comprises two steps: firstly, carrying out empirical mode decomposition on de-noised fusion information to obtain an intrinsic mode component; then, hilbert transform and fourier transform are performed on each eigenmode component, time domain statistics of the hilbert transform and fourier transform are calculated, and features such as root mean square value, average energy, kurtosis, average amplitude, power spectrum characteristic value and the like of each eigenmode component are extracted. And finally, analyzing and judging all the obtained characteristic values, and selecting grinding burn invariance characteristics as a basis for identifying the occurrence of the grinding burn.
Wherein, S406: performing a hilbert transform and a fourier transform on the eigenmode components;
in this step, hilbert transform and fourier transform are performed on the eigenmode component, which are already mentioned in step S405 and will not be described in detail here.
Wherein, S407: extracting statistical characteristics of each eigenmode component, including but not limited to root mean square value, average energy, kurtosis, average amplitude, power spectrum characteristic value;
in this step, the statistical features extracted from each of the eigenmode components include, but are not limited to, root mean square value, average energy, kurtosis, average amplitude, and power spectrum feature value, which have already been mentioned in step S405 and will not be described in detail here.
By the technical scheme, nonlinear and non-stationary signals in the processing process of the blade robot are subjected to linearization and stationary processing by empirical mode decomposition, the characteristics of data are kept in the decomposition process, and the problem that the nonlinear/non-stationary signal characteristics cannot be effectively extracted by a traditional signal processing method (such as Fourier transform, power spectrum estimation and the like) is solved.
In addition, the method for carrying out information fusion and feature extraction through the grinding burn multi-sensing information has the advantages of low application cost, high automation degree, simplicity in operation and the like.
Further, S500: and determining a characteristic value which is in accordance with a preset standard as a grinding burn invariance characteristic value by comparing the grinding burn multi-sensing information under different processing parameters.
Specifically, a characteristic value which is in accordance with a preset standard is determined as a grinding burn invariance characteristic value by comparing the grinding burn multi-sensing information under different processing parameters. The preset standard is a characteristic value which is high in grinding burn relevance and small in processing parameter transformation, and grinding burn invariance characteristics are selected through the preset standard and serve as a basis for identifying grinding burn occurrence.
The embodiment of the invention provides a method for fusing and extracting the information and the characteristics of blade grinding burn, which comprises the following steps: establishing a grinding monitoring system by utilizing multiple sensors; performing robot grinding and polishing processing on the blade according to different preset grinding and processing parameters; collecting grinding burn multi-sensing information through the monitoring system to obtain grinding burn multi-sensing information under different processing parameters; carrying out information fusion and feature extraction on the grinding burn multi-sensing information; and determining a characteristic value which is in accordance with a preset standard as a grinding burn invariance characteristic value by comparing the grinding burn multi-sensing information under different processing parameters. By the scheme, a blade grinding burn information fusion and feature extraction method is formed, a blade robot abrasive belt grinding process monitoring system is developed by taking an acoustic emission sensor 2 as a main part and taking a six-dimensional force sensor, an acceleration sensor 3, a current sensor, a voltage sensor 4 and a thermal imager 5 as auxiliary parts, signals such as acoustic emission, force, vibration and temperature can be acquired on line in real time, comprehensive monitoring on the machining of a robot with a complex curved surface can be realized, and the machining quality of the blade robot is effectively guaranteed. And the abrasive belt grinding burn invariance characteristic of the blade robot can be effectively extracted, the grinding burn misjudgment caused by the change of the processing parameters is avoided, and the processing efficiency and the processing quality of the blade are further ensured.
Further, referring to fig. 6, the present invention further provides a device for information fusion and feature extraction of blade grinding burn, which specifically includes:
a monitoring unit 10 for establishing a grinding monitoring system using a plurality of sensors;
the grinding and polishing unit 20 is used for performing robot grinding and polishing on the blade according to preset different grinding and polishing parameters;
the information acquisition unit 30 is used for acquiring grinding burn multi-sensing information through the monitoring system so as to obtain the grinding burn multi-sensing information under different processing parameters;
the information processing unit 40 is used for carrying out information fusion and feature extraction on the grinding burn multi-sensing information;
and the characteristic determining unit 50 is used for determining a characteristic value which is in accordance with a preset standard as a grinding burn invariance characteristic value by comparing the grinding burn multi-sensing information under different processing parameters.
Further, the information collecting unit 30 further includes:
the information fusion unit 31 is used for fusing the collected multi-sensing information based on a specific analysis method;
a wavelet transform unit 32 for performing wavelet transform on the fused information according to a preselected wavelet basis function to obtain an original wavelet coefficient;
a denoising unit 33; the method is used for carrying out denoising processing on the original wavelet coefficient by utilizing a preselected threshold denoising function to obtain a reconstructed wavelet coefficient
A reconstructed signal unit 34; the fusion information processing unit is used for reconstructing a signal according to the reconstructed wavelet coefficient to obtain denoised fusion information;
the first information transformation unit 35 is configured to perform hilbert-yellow transformation on the denoised fusion information, where the hilbert-yellow transformation includes an empirical mode decomposition part and a hilbert transformation part, and an intrinsic mode component of the denoised fusion information is obtained through the empirical mode decomposition;
a second information transformation unit 36 configured to perform hilbert transform and fourier transform on the eigenmode component;
a feature extraction unit 37, configured to extract statistical features of each of the eigenmode components, including but not limited to root mean square value, average energy, kurtosis, average amplitude, and power spectrum feature value;
further, the feature determination unit includes:
and the characteristic value which is high in association degree with the grinding burn and small in conversion along with the processing parameters is determined as the grinding burn invariance characteristic value by comparing the grinding burn multi-sensing information under different processing parameters.
The invention provides a device for fusing information and extracting characteristics of blade grinding burn, which comprises: a monitoring unit 10 for establishing a grinding monitoring system using a plurality of sensors; the grinding and polishing unit 20 is used for performing robot grinding and polishing on the blade according to preset different grinding and polishing parameters; the information acquisition unit 30 is used for acquiring grinding burn multi-sensing information through the monitoring system so as to obtain the grinding burn multi-sensing information under different processing parameters; the information processing unit 40 is used for carrying out information fusion and feature extraction on the grinding burn multi-sensing information; the characteristic determining unit 50 is configured to determine a characteristic value conforming to a preset standard as a grinding burn invariance characteristic value by comparing the grinding burn multi-sensing information under the different processing parameters. Through the scheme, comprehensive monitoring on the machining of the robot with the complex curved surface can be realized, so that the machining quality of the robot with the blades is effectively guaranteed. And the abrasive belt grinding burn invariance characteristic of the blade robot can be effectively extracted, the grinding burn misjudgment caused by the change of the processing parameters is avoided, and the processing efficiency and the processing quality of the blade are further ensured.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. A method for fusing and extracting information and characteristics of blade grinding burn is characterized by comprising the following steps:
establishing a grinding monitoring system by utilizing multiple sensors;
performing robot grinding and polishing processing on the blade according to different preset grinding and processing parameters;
collecting grinding burn multi-sensing information through the monitoring system to obtain grinding burn multi-sensing information under different processing parameters;
carrying out information fusion and feature extraction on the grinding burn multi-sensing information;
and determining a characteristic value which is in accordance with a preset standard as a grinding burn invariance characteristic value by comparing the grinding burn multi-sensing information under different processing parameters.
2. The method for information fusion and feature extraction of blade grinding burn according to claim 1, wherein the establishing of the grinding monitoring system by using multiple sensors comprises:
mounting a first sensor on a flange at the tail end of an industrial robot, and calibrating the first sensor;
mounting a blade clamp at the first sensor end;
calibrating a second sensor, and installing the second sensor at a first preset position;
calibrating the thermal imager, and installing the thermal imager at a second preset position.
3. The method for information fusion and feature extraction of blade grinding burn according to claim 2,
the first sensor is a six-dimensional force sensor;
correspondingly, calibrating the second sensor and installing the second sensor in a first predetermined location comprises:
calibrating an acoustic emission sensor, smearing a coupling agent on the tail end of the blade clamp, and installing the acoustic emission sensor on the tail end of the blade clamp;
calibrating an acceleration sensor, and mounting the acceleration sensor on a contact wheel mounting beam of the abrasive belt grinding and polishing machine;
and calibrating a current sensor and a voltage sensor, and installing the current sensor and the voltage sensor at the input end of the abrasive belt grinding and polishing machine motor.
4. The method for fusing and extracting the information and the characteristics of the blade grinding burn according to claim 1, wherein the robot grinding and polishing the blade according to different preset grinding parameters comprises the following steps:
presetting different grinding processing parameters;
accordingly, the presetting of different grinding parameters includes:
selecting processing parameters;
fixing any two of the processing parameters, and gradually changing the rest parameters;
carrying out grinding and polishing processing on the blade by using the processing parameters;
detecting whether the leaves are burnt or not;
wherein the step of detecting whether the leaves are burnt is as follows:
and repeating the test for 3 times, recording the set of processing parameters if the burn and the non-burn appear in the 3 tests, and otherwise, continuously changing the rest parameters until the burn of the leaves appears.
5. The blade grinding burn information fusion and feature extraction method according to claim 1, wherein the information fusion and feature extraction of the grinding burn multi-sensing information comprises:
fusing the collected multi-sensing information based on a specific analysis method;
performing wavelet transformation on the fused information according to a preselected wavelet basis function to obtain an original wavelet coefficient;
carrying out denoising processing on the original wavelet coefficient by using a preselected threshold denoising function to obtain a reconstructed wavelet coefficient;
reconstructing a signal according to the reconstructed wavelet coefficient to obtain denoised fusion information;
performing Hilbert-Huang transformation on the de-noised fusion information, wherein the Hilbert-Huang transformation comprises an empirical mode decomposition part and a Hilbert transformation part, and obtaining an intrinsic mode component of the de-noised fusion information after the empirical mode decomposition;
performing a hilbert transform and a fourier transform on the eigenmode components;
extracting the statistical features of each of the eigenmode components includes, but is not limited to: root mean square value, average energy, kurtosis, average amplitude, power spectrum characteristic value.
6. The method for fusing and extracting the information and the characteristics of the blade grinding burn according to claim 5, wherein the performing wavelet transform on the fused information according to the preselected wavelet basis function comprises:
performing wavelet transformation on the fused information according to the preselected wavelet basis function and the decomposition layer number;
the fusion of the collected multi-sensing information based on a specific analysis method comprises the following steps of;
and fusing the collected multi-sensing information based on a kernel principal component analysis method.
7. The blade grinding burn information fusion and feature extraction method according to any one of claims 1 to 6, wherein the determining a feature value in compliance with a preset standard as a grinding burn invariance feature value by comparing the grinding burn multisensory information under the different processing parameters comprises:
and determining a characteristic value which is high in association degree with the grinding burn and small in conversion along with the processing parameters as a grinding burn invariance characteristic value by comparing the grinding burn multi-sensing information under different processing parameters.
8. The utility model provides a leaf grinding burn information fusion and feature extraction device which characterized in that includes:
the monitoring unit is used for establishing a grinding monitoring system by utilizing multiple sensors;
the grinding and polishing unit is used for carrying out robot grinding and polishing on the blade according to different preset grinding and polishing parameters;
the information acquisition unit is used for acquiring grinding burn multi-sensing information through the monitoring system so as to obtain the grinding burn multi-sensing information under different processing parameters;
the information processing unit is used for carrying out information fusion and feature extraction on the grinding burn multi-sensing information;
and the characteristic determining unit is used for determining a characteristic value which is in accordance with a preset standard as a grinding burn invariance characteristic value by comparing the grinding burn multi-sensing information under different processing parameters.
9. The blade grinding burn information fusion and feature extraction device according to claim 8, wherein the information acquisition unit comprises:
the information fusion unit is used for fusing the collected multi-sensing information based on a specific analysis method;
the wavelet transformation unit is used for performing wavelet transformation on the fused information according to a preselected wavelet basis function to obtain an original wavelet coefficient;
a denoising unit; the system comprises a pre-selection threshold denoising function, a reconstruction wavelet coefficient and a wavelet coefficient acquisition unit, wherein the pre-selection threshold denoising function is used for denoising the original wavelet coefficient to obtain the reconstruction wavelet coefficient;
a reconstructed signal unit; the fusion information processing unit is used for reconstructing a signal according to the reconstructed wavelet coefficient to obtain denoised fusion information;
the first information transformation unit is used for performing Hilbert-Huang transformation on the de-noised fusion information, wherein the Hilbert-Huang transformation comprises an empirical mode decomposition part and a Hilbert transformation part, and an intrinsic mode component of the de-noised fusion information is obtained after the empirical mode decomposition;
a second information transformation unit, configured to perform hilbert transform and fourier transform on the eigenmode component;
a feature extraction unit, configured to extract statistical features of each of the eigenmode components, including but not limited to: root mean square value, average energy, kurtosis, average amplitude, power spectrum characteristic value.
10. The blade grinding burn information fusion and feature extraction device according to claim 8, wherein the feature determination unit includes:
and the characteristic value which is high in association degree with the grinding burn and small in conversion along with the processing parameters is determined as the grinding burn invariance characteristic value by comparing the grinding burn multi-sensing information under different processing parameters.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010044861.5A CN111251187A (en) | 2020-01-16 | 2020-01-16 | Method and device for fusing information and extracting characteristics of blade grinding burn |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010044861.5A CN111251187A (en) | 2020-01-16 | 2020-01-16 | Method and device for fusing information and extracting characteristics of blade grinding burn |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111251187A true CN111251187A (en) | 2020-06-09 |
Family
ID=70945235
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010044861.5A Pending CN111251187A (en) | 2020-01-16 | 2020-01-16 | Method and device for fusing information and extracting characteristics of blade grinding burn |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111251187A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115383514A (en) * | 2022-09-06 | 2022-11-25 | 陕西法士特齿轮有限责任公司 | Grinding burn monitoring method based on vibration signal time domain analysis |
| CN115647819A (en) * | 2022-09-20 | 2023-01-31 | 玉环仪表机床制造厂 | Turning and grinding integrated compound machine and control method thereof |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101879690A (en) * | 2010-01-21 | 2010-11-10 | 湘潭大学 | Batch drilling process quality monitoring method based on multiple sensor signals |
| CN103962888A (en) * | 2014-05-12 | 2014-08-06 | 西北工业大学 | Tool abrasion monitoring method based on wavelet denoising and Hilbert-Huang transformation |
| CN104776908A (en) * | 2015-04-17 | 2015-07-15 | 南京理工大学 | EMD generalized energy-based wheeltrack vibration signal fault feature extraction method |
| CN106826565A (en) * | 2017-03-16 | 2017-06-13 | 中国人民解放军装甲兵工程学院 | A kind of utilization grinding force monitoring abrasion of grinding wheel and the method for grinding burn |
| CN110059639A (en) * | 2019-04-22 | 2019-07-26 | 桂林电子科技大学 | A kind of Frequency Hopping Signal detection method based on fractional wavelet transform and Hilbert-Huang transform |
-
2020
- 2020-01-16 CN CN202010044861.5A patent/CN111251187A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101879690A (en) * | 2010-01-21 | 2010-11-10 | 湘潭大学 | Batch drilling process quality monitoring method based on multiple sensor signals |
| CN103962888A (en) * | 2014-05-12 | 2014-08-06 | 西北工业大学 | Tool abrasion monitoring method based on wavelet denoising and Hilbert-Huang transformation |
| CN104776908A (en) * | 2015-04-17 | 2015-07-15 | 南京理工大学 | EMD generalized energy-based wheeltrack vibration signal fault feature extraction method |
| CN106826565A (en) * | 2017-03-16 | 2017-06-13 | 中国人民解放军装甲兵工程学院 | A kind of utilization grinding force monitoring abrasion of grinding wheel and the method for grinding burn |
| CN110059639A (en) * | 2019-04-22 | 2019-07-26 | 桂林电子科技大学 | A kind of Frequency Hopping Signal detection method based on fractional wavelet transform and Hilbert-Huang transform |
Non-Patent Citations (4)
| Title |
|---|
| 李娟莉等: "基于三层信息融合的提升机制动***故障诊断", 《振动、测试与诊断》 * |
| 李怀俊: "《基于核主元模糊聚类的旋转机机械故障诊断技术研究》", 31 July 2016 * |
| 杨振生: "面向磨削烧伤问题的间接监测技术研究", 《中国博士学位论文全文数据库 工程科技I籍》 * |
| 魏海增等: "基于CMT-HHT结合的永磁同步电机失磁故障诊断方法及其可行性分析", 《电机与控制应用》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115383514A (en) * | 2022-09-06 | 2022-11-25 | 陕西法士特齿轮有限责任公司 | Grinding burn monitoring method based on vibration signal time domain analysis |
| CN115647819A (en) * | 2022-09-20 | 2023-01-31 | 玉环仪表机床制造厂 | Turning and grinding integrated compound machine and control method thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114619292B (en) | Milling cutter wear monitoring method based on fusion of wavelet denoising and attention mechanism with GRU network | |
| CN111251187A (en) | Method and device for fusing information and extracting characteristics of blade grinding burn | |
| CN112461934B (en) | Aero-engine blade crack source positioning method based on acoustic emission | |
| CN116861313B (en) | Kalman filtering working condition identification method and system based on vibration energy trend | |
| CN108037034B (en) | The multisensor on-line checking and data processing system of wheel grinding performance | |
| CN109605128B (en) | Milling chatter online detection method based on power spectrum entropy difference | |
| CN112487890B (en) | Bearing acoustic signal fault diagnosis method based on parallel sparse filtering | |
| CN115563553A (en) | Aviation bearing fault diagnosis method of deep migration map convolution network under fluctuation working condition | |
| CN110553789A (en) | state detection method and device of piezoresistive pressure sensor and brake system | |
| CN111337767A (en) | Resonant wave reducer fault analysis method | |
| CN115741235A (en) | Wear prediction and health management method based on five-axis machining center cutter | |
| CN111352365B (en) | Dustproof ventilation type electric power and electrical equipment cabinet and control method | |
| CN109062051A (en) | A method of improving Identification of Dynamic Parameters of Amanipulator precision | |
| CN113326774A (en) | Machine tool energy consumption state identification method and system based on AlexNet network | |
| CN110994802A (en) | Method and device for monitoring running state of power transmission conductor | |
| CN114486252B (en) | Rolling bearing fault diagnosis method of vector mode maximum envelope | |
| CN113984392B (en) | Online testing and evaluating method for vibration quality of electric drive assembly system | |
| CN111753876B (en) | Product quality detection method based on deep neural network | |
| CN113962318A (en) | Transformer state anomaly detection method based on internal and external data fusion and machine learning | |
| CN113807431A (en) | Intelligent spindle state evaluation method and system based on multi-source information fusion | |
| Li et al. | A fault diagnosis equipment of motor bearing based on sound signal and CNN | |
| Guo et al. | Rotating machinery fault detection using a new version of intrinsic time-scale decomposition | |
| CN115213735B (en) | System and method for monitoring cutter state in milling process | |
| CN112485018B (en) | Mechanical equipment energy consumption abnormity detection method based on mechanism data fusion | |
| CN117929418B (en) | Integrated circuit defect detection method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200609 |