WO2024076326A1 - A system for detecting machine tool chatter - Google Patents

A system for detecting machine tool chatter Download PDF

Info

Publication number
WO2024076326A1
WO2024076326A1 PCT/TR2023/051053 TR2023051053W WO2024076326A1 WO 2024076326 A1 WO2024076326 A1 WO 2024076326A1 TR 2023051053 W TR2023051053 W TR 2023051053W WO 2024076326 A1 WO2024076326 A1 WO 2024076326A1
Authority
WO
WIPO (PCT)
Prior art keywords
chatter
data
control unit
detection system
machine tool
Prior art date
Application number
PCT/TR2023/051053
Other languages
French (fr)
Inventor
Hakki Ozgur Unver
Ahmet Murat OZBAYOGLU
Batihan SENER
Original Assignee
Tobb Ekonomi Ve Teknoloji Universitesi
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Priority claimed from TR2022/015107 external-priority patent/TR2022015107A1/en
Application filed by Tobb Ekonomi Ve Teknoloji Universitesi filed Critical Tobb Ekonomi Ve Teknoloji Universitesi
Publication of WO2024076326A1 publication Critical patent/WO2024076326A1/en

Links

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Programme-control systems
    • G05B19/02—Programme-control systems electric
    • G05B19/18—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form
    • G05B19/406—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form characterised by monitoring or safety
    • G05B19/4065—Monitoring tool breakage, life or condition
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
    • B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
    • B23Q15/00—Automatic control or regulation of feed movement, cutting velocity or position of tool or work
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/37—Measurements
    • G05B2219/37258—Calculate wear from workpiece and tool material, machining operations
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/37—Measurements
    • G05B2219/37432—Detected by accelerometer, piezo electric
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/41—Servomotor, servo controller till figures
    • G05B2219/41256—Chattering control

Definitions

  • the present invention relates to a chatter detection system for detecting self-excited chatter which is generated by the dynamic interactions between the cutting tool and the workpiece during manufacturing, particularly in machining industry.
  • the shaping of the raw material to be used in the manufacturing process, in which the design to be manufactured is predetermined, on the machine tools suitable for the manufacturing process, with the operations of the specified cutting tool machines is generally called machining.
  • Machining is realized by creating tension on the workpiece upon movement of the cutting tool and the workpiece in relation to each other.
  • the process of shaping metal, plastic, wood and similar materials by removing material from the surface or inner parts thereof is called machining.
  • the said material removal process is applied to the workpiece by the cutting tool.
  • the whole process is called the machining process.
  • the waste materials that are removed from the workpiece are called chips.
  • Machining is frequently preferred due to its advantages such as the ability to process different materials, having a precise margin of error, ability to manufacture products with different dimensions and geometries, and having fewer corner and surface defects.
  • Chatter causes decrease of the workpiece surface quality and shortening of the cutting tool life. Chatter occurs when the material removal rate increases and the cutting force decreases. Variation of the material removal rate and variation of the shear angle cause chatter. Chatter causes negative effects on the workpiece in the manufacturing process such as decreased precision, surface defects, and damage to the cutting tool.
  • chatter The vibration occurring during separation of the materials, which are removed from the workpiece and called chips, independent of the machine tool and the external environment is called chatter.
  • Machining is used in almost every industry, especially in aviation, defense industry, automotive and biomedical products.
  • the machining method is frequently used in the production of high value-added products.
  • the surface quality of the materials affects the application performance and the service life of the materials. It is extremely important to prevent surface defects of the materials manufactured by the machining method especially in the aviation and automotive industries.
  • chatter detection In an application in the state of the art, there are methods used to reduce chatter.
  • the said application enables to relatively reduce the chatter by using low cutting speeds.
  • the machining method with low cutting speeds reduces the chatter but increases the manufacturing time. Especially in applications requiring mass production, the slow manufacturing time has a significant impact on production and procurement processes.
  • artificial intelligence and system analysis methods are used for chatter detection.
  • the said application only the chatter that occurs during the production of certain parts on specified machine tools is detected.
  • the biggest disadvantage of the said application is that it cannot detect the chatter that occurs during the production of different parts and when different machine tools are used. Frequently new parts are manufactured in production facilities where custom production is carried out and R&D production centers in which new inventions are produced.
  • the application in the state of the art is insufficient for detection of the chatter.
  • artificial intelligence is used to detect the chatter.
  • the artificial intelligence in the said application learns with multiple data inputs. Numerous different parts are manufactured using a large number of machine tools for the artificial intelligence to learn the chatter detection. This increases the time for the artificial intelligence to predict the chatter and requires more materials to be produced for optimal results.
  • a system for chatter detection which collects chatter data using an impact hammer and an accelerometer and has a low error margin upon processing the said data and decomposing it into its modes by EEMD and IMF methods during preprocessing, and which is applicable to existing machine tools.
  • a further object of the present invention is to provide a system for chatter detection, in which the data is used to train the artificial intelligence module provided therein by using Hilbert-Huang Transform.
  • Another object of the present invention is to provide a system for chatter detection, which enables to detect chatter by putting the data obtained during machining through the same processes.
  • Yet another object of the present invention is to provide a system for chatter detection, which performs mode decomposition by EEMD and IMF methods during preprocessing.
  • the machine tool chatter detection system of the present invention is used particularly in machine tools in the machining process to detect the chatter that occurs due to the dynamic interactions between the cutting tool and the workpiece during manufacturing.
  • machining The method of manufacturing by removing material from a workpiece using a cutting tool is called machining. Machining is frequently preferred in industries such as aviation, aerospace, automotive and biomedical industries.
  • a system for detecting machine tool chatter which is developed to achieve the objects of the present invention and is defined in the first claim and the other claims dependent thereon, comprises a control unit; an impact hammer adapted to measure the forces occurring between the cutting tool and the workpiece; an accelerometer adapted to measure the acceleration occurring between the cutting tool and the workpiece; a data acquisition system adapted to digitally collect the data obtained from the impact hammer and accelerometer.
  • the machine tool chatter detection system of the present application is used to detect the chatter that occurs during machining. Dynamic interactions between the cutting tool and the workpiece during machining cause chatter formation.
  • the chatter occurring between the cutting tool and the workpiece in the machine tools increases the surface defects of the manufactured material. In some cases, the chatter damages the cutting tool and thus the machine tool. This causes production to slow down and the cost to increase.
  • the machine tool chatter detection system of the present application enables to detect the chatter occurring in the machine tool by training an artificial intelligence model.
  • the data on the chatter in the machine tool is theoretically generated and processed using analytical and numerical models.
  • An impact hammer and an accelerometer collect chatter data during machining.
  • the chatter data obtained in the machine tool is used for artificial intelligence training and generation of artificial intelligence module by using Hilbert-Huang Transform.
  • the machine tool chatter detection system of the present application performs a preprocessing.
  • the said preprocessing is performed by decomposition into the intrinsic mode functions by EEMD.
  • the data obtained by processing is transferred to the machining system and the chatter state is determined.
  • Figure 1 is a schematic view of the chatter detection system of the present invention.
  • Figure 2 is a schematic view of the chatter detection system of the present invention.
  • Control unit 500 Control unit
  • a chatter detection system (100) used in machine tools in order to detect the chatter caused by the dynamic interactions occurring between the cutting tool and the workpiece especially during machining basically comprises the following: at least one impact hammer (200) adapted to measure the forces between the cutting tool and the workpiece on the machine tool, which cause the chatter, at least one accelerometer (300) adapted to measure the acceleration between the cutting tool and the workpiece on the machine tool, which causes the chatter, a data acquisition system (400) used to enable the data obtained by the impact hammer (200) and accelerometer (300) to be digitally collected, stored and transferred to the control unit (500), at least one control unit (500), positioned on or away from the machine tool, having an artificial intelligence or comprising models generated by an artificial intelligence, and adapted to perform the following technical elements;
  • the chatter detection system (100) of the present application is used in production carried out by machining, especially in the aviation, automotive, medical and defense industries.
  • the chatter detection system (100) of the present application is used in machine tools that perform machining.
  • the process in which a cutting tool removes material from a material to be manufactured, i.e. a workpiece, is called machining.
  • the said process is carried out on a machine tool.
  • the cutting tool removes material from the workpiece.
  • the said material removal causes a dynamic interaction between the cutting tool and the workpiece.
  • the dynamic interaction occurring between the cutting tool and the workpiece causes chatter. Chatter causes defects on the surface of the workpiece and damage to the workpiece. Chatter also causes damage to the cutting tool and therefore to the machine tool. It is extremely important to detect chatter in order to reduce manufacturing cost and facilitate production.
  • the chatter detection system (100) of the present application comprises at least one impact hammer (200), at least one accelerometer (300), at least one data acquisition system (400) and at least one control unit (500).
  • the impact hammer (200) provided in one embodiment of the invention is adapted to measure the force generated due to the dynamic interactions.
  • the said impact hammer (200) measures the direction and magnitude of the forces applied to the workpiece on the machine tool.
  • the data obtained by the impact hammer (200) is transmitted to the control unit (500).
  • the accelerometer (300) provided in one embodiment of the invention is adapted to measure the direction and magnitude of the acceleration that occurs due to the dynamic interactions.
  • the said accelerometer (300) measures the direction and magnitude of the acceleration applied to the workpiece on the machine tool.
  • the data obtained by the accelerometer (300) is transmitted to the control unit (500).
  • the data acquisition system (400) provided in one embodiment of the invention is used to collect data digitally.
  • the said data acquisition system (400) is called DAQ.
  • the data obtained by means of the impact hammer (200) and the accelerometer (300) are collected digitally by the data acquisition system (400).
  • the data obtained by the data acquisition system (400) is transmitted to the control unit (500) in order to train the artificial neural networks.
  • the control unit (500) provided in one embodiment of the invention is adapted to store and process the chatter data.
  • the control unit (500) is adapted to theoretically generate the chatter data using analytical and numerical models.
  • the said chatter data is processed by analytical and numerical models in the control unit (500).
  • Analytical and numerical dynamic models are solved.
  • Artificial neural networks are obtained by analytical and numerical model solutions.
  • stability lobe diagrams are obtained. Artificial neural networks and stability lobe diagrams are automatically labeled.
  • the control unit (500) provided in one embodiment of the invention is adapted to store and process the chatter data.
  • the control unit (500) uses the obtained chatter data in artificial neural network training using Hilbert-Huang Transform.
  • the said labeled data is processed and converted into two-dimensional images called “HHT images”.
  • the said data set is used to train a convolutional artificial neural network (CNN).
  • CNN convolutional artificial neural network
  • the control unit (500) provided in one embodiment of the invention is adapted to store and process the chatter data.
  • the control unit (500) enables to determine the chatter by putting the processed data through the same machining process.
  • the artificial neural networks trained using the Hilbert-Huang Transform are transferred to the machining environment. In the said actual machining environment, the actual chatter data is obtained.
  • the actual chatter data from the actual environment is labeled and converted into two-dimensional images called "HTT images".
  • the actual data is compared by the artificial neural networks trained by the data obtained from the analytical and numerical solutions.
  • the artificial neural network model distinguishes between the stable or chatter state of the said machining process.
  • the chatter detection system (100) of the present application is characterized such that its components are in communication with each other.
  • the impact hammer (200), the accelerometer (300) and the data acquisition system (400) are dynamically characterized.
  • the data obtained using the data acquisition system (400) is processed.
  • the frequency response function of the data obtained by the data acquisition system (400) and the system is modeled.
  • the said models generate vibration signals that appropriately simulate the material removal process performed during machining.
  • stability lobe diagrams are plotted for automatic labeling.
  • the said stability lobe diagram automatically generates a plurality of signal samples required for training the artificial intelligence model.
  • the accelerometer (300) and the data acquisition system (400) provided in one embodiment of the invention are adapted to continuously collect data together.
  • the impact hammer (200), which performs the measurement, is not operational.
  • the trained models obtained after training the artificial intelligence model continuously collect data during machining with the help of the accelerometer (300) and the data acquisition system (400). With the said collected data, the models evaluate the material removal process for the chatter detection. In case of chatter in the system, stable material removal parameters but at higher speeds are recommended without reducing efficiency.
  • the control unit (500) provided in one embodiment of the invention is adapted to perform pre-processing.
  • pre-processing is performed in order for the artificial intelligence models to provide more successful results.
  • signals obtained or measured artificially, i.e. by simulation are pre-processed.
  • the preprocessing enables the said data to be decomposed into intrinsic mode functions (IMF) by EEMD (Ensemble Empirical Decomposition Method).
  • IMF intrinsic mode functions
  • EEMD Endsemble Empirical Decomposition Method
  • the entropies of the said intrinsic modes are calculated and ranked according to the amount of increase.
  • the most increasing intrinsic modes are selected as much as the number of vibration modes.
  • the selected intrinsic modes are decomposed.
  • the said bands are independently converted into images by Hibert-Huang Transform. These images are used to train the artificial intelligence model.
  • the said embodiment enables chatter to be detected substantially without the need to collect any physical data for the training of the artificial intelligence model.
  • control unit (500) and the data acquisition system (400) provided in one embodiment of the invention are preferably integrated.
  • the data acquisition system (400) is provided as a module which is located in the control unit (500) and integrated into the control unit (500).
  • a computer is used as the control unit (500) provided in one embodiment of the invention.
  • the computer i.e. the control unit (500)
  • the computer is a well-equipped computer with simple or different modules adapted to store the data of the machine tool and the data obtained from the data acquisition system (400), and to create a model by processing the said data with an artificial intelligence.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Automation & Control Theory (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Manufacturing & Machinery (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Human Computer Interaction (AREA)
  • Mechanical Engineering (AREA)
  • Numerical Control (AREA)

Abstract

The present invention relates to a chatter detection system (100) for detecting self-excited chatter which is generated by the dynamic interactions between the cutting tool and the workpiece during manufacturing, particularly in machining industry.

Description

A SYSTEM FOR DETECTING MACHINE TOOL CHATTER
Field of the Invention
The present invention relates to a chatter detection system for detecting self-excited chatter which is generated by the dynamic interactions between the cutting tool and the workpiece during manufacturing, particularly in machining industry.
Background of the Invention
The shaping of the raw material to be used in the manufacturing process, in which the design to be manufactured is predetermined, on the machine tools suitable for the manufacturing process, with the operations of the specified cutting tool machines is generally called machining.
The product manufactured and processed by machining is called a workpiece. Machining is realized by creating tension on the workpiece upon movement of the cutting tool and the workpiece in relation to each other.
The process of shaping metal, plastic, wood and similar materials by removing material from the surface or inner parts thereof is called machining. The said material removal process is applied to the workpiece by the cutting tool. The whole process is called the machining process. The waste materials that are removed from the workpiece are called chips.
There are different types of machining such as facing, chamfering, step turning, threading. Machining is frequently preferred due to its advantages such as the ability to process different materials, having a precise margin of error, ability to manufacture products with different dimensions and geometries, and having fewer corner and surface defects.
Dynamic interactions occurring between the cutting tool and the workpiece during machining cause chatter. Chatter causes decrease of the workpiece surface quality and shortening of the cutting tool life. Chatter occurs when the material removal rate increases and the cutting force decreases. Variation of the material removal rate and variation of the shear angle cause chatter. Chatter causes negative effects on the workpiece in the manufacturing process such as decreased precision, surface defects, and damage to the cutting tool.
During machining, material is removed-separated from the workpiece. The vibration occurring during separation of the materials, which are removed from the workpiece and called chips, independent of the machine tool and the external environment is called chatter.
Machining is used in almost every industry, especially in aviation, defense industry, automotive and biomedical products. The machining method is frequently used in the production of high value-added products. The surface quality of the materials affects the application performance and the service life of the materials. It is extremely important to prevent surface defects of the materials manufactured by the machining method especially in the aviation and automotive industries.
In an application in the state of the art, there are methods used to reduce chatter. The said application enables to relatively reduce the chatter by using low cutting speeds. The machining method with low cutting speeds reduces the chatter but increases the manufacturing time. Especially in applications requiring mass production, the slow manufacturing time has a significant impact on production and procurement processes. In another application in the state of the art, artificial intelligence and system analysis methods are used for chatter detection. In the said application, only the chatter that occurs during the production of certain parts on specified machine tools is detected. The biggest disadvantage of the said application is that it cannot detect the chatter that occurs during the production of different parts and when different machine tools are used. Frequently new parts are manufactured in production facilities where custom production is carried out and R&D production centers in which new inventions are produced. During the production of the said new parts, the application in the state of the art is insufficient for detection of the chatter.
In another application in the state of the art, artificial intelligence is used to detect the chatter. The artificial intelligence in the said application learns with multiple data inputs. Numerous different parts are manufactured using a large number of machine tools for the artificial intelligence to learn the chatter detection. This increases the time for the artificial intelligence to predict the chatter and requires more materials to be produced for optimal results.
By means of the present application, a system for chatter detection is provided which collects chatter data using an impact hammer and an accelerometer and has a low error margin upon processing the said data and decomposing it into its modes by EEMD and IMF methods during preprocessing, and which is applicable to existing machine tools.
Objects of the Invention
It is an object of the present invention to provide a system for chatter detection, which enables to theoretically generate chatter data by means of analytical and numerical models. Another object of the present invention is to provide a system for chatter detection, which collects chatter data using an impact hammer and accelerometer during the machining process.
A further object of the present invention is to provide a system for chatter detection, in which the data is used to train the artificial intelligence module provided therein by using Hilbert-Huang Transform.
Another object of the present invention is to provide a system for chatter detection, which enables to detect chatter by putting the data obtained during machining through the same processes.
Yet another object of the present invention is to provide a system for chatter detection, which performs mode decomposition by EEMD and IMF methods during preprocessing.
Summary of the Invention
The machine tool chatter detection system of the present invention is used particularly in machine tools in the machining process to detect the chatter that occurs due to the dynamic interactions between the cutting tool and the workpiece during manufacturing.
The method of manufacturing by removing material from a workpiece using a cutting tool is called machining. Machining is frequently preferred in industries such as aviation, aerospace, automotive and biomedical industries.
A system for detecting machine tool chatter, which is developed to achieve the objects of the present invention and is defined in the first claim and the other claims dependent thereon, comprises a control unit; an impact hammer adapted to measure the forces occurring between the cutting tool and the workpiece; an accelerometer adapted to measure the acceleration occurring between the cutting tool and the workpiece; a data acquisition system adapted to digitally collect the data obtained from the impact hammer and accelerometer.
The machine tool chatter detection system of the present application is used to detect the chatter that occurs during machining. Dynamic interactions between the cutting tool and the workpiece during machining cause chatter formation.
The chatter occurring between the cutting tool and the workpiece in the machine tools increases the surface defects of the manufactured material. In some cases, the chatter damages the cutting tool and thus the machine tool. This causes production to slow down and the cost to increase.
The machine tool chatter detection system of the present application enables to detect the chatter occurring in the machine tool by training an artificial intelligence model.
In the machine tool chatter detection system of the present application, the data on the chatter in the machine tool is theoretically generated and processed using analytical and numerical models. An impact hammer and an accelerometer collect chatter data during machining.
In the machine tool chatter detection system of the present application, the chatter data obtained in the machine tool is used for artificial intelligence training and generation of artificial intelligence module by using Hilbert-Huang Transform.
The machine tool chatter detection system of the present application performs a preprocessing. The said preprocessing is performed by decomposition into the intrinsic mode functions by EEMD. In addition, the data obtained by processing is transferred to the machining system and the chatter state is determined. Detailed Description of the Invention
A system for detecting machine tool chatter developed to achieve the objects of the present invention is illustrated in the accompanying figures, in which;
Figure 1 is a schematic view of the chatter detection system of the present invention.
Figure 2 is a schematic view of the chatter detection system of the present invention.
The components in the figures are numbered individually and the reference numbers corresponding thereto are given below:
100. Chatter detection system
200. Impact hammer
300. Accelerometer
400. Data acquisition system
500. Control unit
A chatter detection system (100) used in machine tools in order to detect the chatter caused by the dynamic interactions occurring between the cutting tool and the workpiece especially during machining, basically comprises the following: at least one impact hammer (200) adapted to measure the forces between the cutting tool and the workpiece on the machine tool, which cause the chatter, at least one accelerometer (300) adapted to measure the acceleration between the cutting tool and the workpiece on the machine tool, which causes the chatter, a data acquisition system (400) used to enable the data obtained by the impact hammer (200) and accelerometer (300) to be digitally collected, stored and transferred to the control unit (500), at least one control unit (500), positioned on or away from the machine tool, having an artificial intelligence or comprising models generated by an artificial intelligence, and adapted to perform the following technical elements;
• theoretical generation of the chatter data using analytical and numerical models,
• using the chatter data in artificial neural network training using Hilbert-Huang Transform,
• decomposition into EEMD and IMF modes during preprocessing.
The chatter detection system (100) of the present application is used in production carried out by machining, especially in the aviation, automotive, medical and defense industries.
The chatter detection system (100) of the present application is used in machine tools that perform machining. The process in which a cutting tool removes material from a material to be manufactured, i.e. a workpiece, is called machining. The said process is carried out on a machine tool. The cutting tool removes material from the workpiece. The said material removal causes a dynamic interaction between the cutting tool and the workpiece. The dynamic interaction occurring between the cutting tool and the workpiece causes chatter. Chatter causes defects on the surface of the workpiece and damage to the workpiece. Chatter also causes damage to the cutting tool and therefore to the machine tool. It is extremely important to detect chatter in order to reduce manufacturing cost and facilitate production.
The chatter detection system (100) of the present application comprises at least one impact hammer (200), at least one accelerometer (300), at least one data acquisition system (400) and at least one control unit (500). The impact hammer (200) provided in one embodiment of the invention is adapted to measure the force generated due to the dynamic interactions. The said impact hammer (200) measures the direction and magnitude of the forces applied to the workpiece on the machine tool. The data obtained by the impact hammer (200) is transmitted to the control unit (500).
The accelerometer (300) provided in one embodiment of the invention is adapted to measure the direction and magnitude of the acceleration that occurs due to the dynamic interactions. The said accelerometer (300) measures the direction and magnitude of the acceleration applied to the workpiece on the machine tool. The data obtained by the accelerometer (300) is transmitted to the control unit (500).
The data acquisition system (400) provided in one embodiment of the invention is used to collect data digitally. The said data acquisition system (400) is called DAQ. The data obtained by means of the impact hammer (200) and the accelerometer (300) are collected digitally by the data acquisition system (400). The data obtained by the data acquisition system (400) is transmitted to the control unit (500) in order to train the artificial neural networks.
The control unit (500) provided in one embodiment of the invention is adapted to store and process the chatter data. The control unit (500) is adapted to theoretically generate the chatter data using analytical and numerical models. The said chatter data is processed by analytical and numerical models in the control unit (500). Analytical and numerical dynamic models are solved. Artificial neural networks are obtained by analytical and numerical model solutions. At the same time, stability lobe diagrams are obtained. Artificial neural networks and stability lobe diagrams are automatically labeled.
The control unit (500) provided in one embodiment of the invention is adapted to store and process the chatter data. The control unit (500) uses the obtained chatter data in artificial neural network training using Hilbert-Huang Transform. The said labeled data is processed and converted into two-dimensional images called “HHT images”. The said data set is used to train a convolutional artificial neural network (CNN).
The control unit (500) provided in one embodiment of the invention is adapted to store and process the chatter data. The control unit (500) enables to determine the chatter by putting the processed data through the same machining process. The artificial neural networks trained using the Hilbert-Huang Transform are transferred to the machining environment. In the said actual machining environment, the actual chatter data is obtained. The actual chatter data from the actual environment is labeled and converted into two-dimensional images called "HTT images". The actual data is compared by the artificial neural networks trained by the data obtained from the analytical and numerical solutions. The artificial neural network model distinguishes between the stable or chatter state of the said machining process.
In one embodiment of the invention, the chatter detection system (100) of the present application is characterized such that its components are in communication with each other. In the said embodiment, particularly the impact hammer (200), the accelerometer (300) and the data acquisition system (400) are dynamically characterized.
In one embodiment of the invention, the data obtained using the data acquisition system (400) is processed. In the said embodiment, the frequency response function of the data obtained by the data acquisition system (400) and the system is modeled. The said models generate vibration signals that appropriately simulate the material removal process performed during machining. Thereupon, stability lobe diagrams are plotted for automatic labeling. The said stability lobe diagram automatically generates a plurality of signal samples required for training the artificial intelligence model. The accelerometer (300) and the data acquisition system (400) provided in one embodiment of the invention are adapted to continuously collect data together. In the said embodiment, the impact hammer (200), which performs the measurement, is not operational. The trained models obtained after training the artificial intelligence model continuously collect data during machining with the help of the accelerometer (300) and the data acquisition system (400). With the said collected data, the models evaluate the material removal process for the chatter detection. In case of chatter in the system, stable material removal parameters but at higher speeds are recommended without reducing efficiency.
The control unit (500) provided in one embodiment of the invention is adapted to perform pre-processing. In the said embodiment, pre-processing is performed in order for the artificial intelligence models to provide more successful results. During pre-processing, signals obtained or measured artificially, i.e. by simulation, are pre-processed. The preprocessing enables the said data to be decomposed into intrinsic mode functions (IMF) by EEMD (Ensemble Empirical Decomposition Method). The entropies of the said intrinsic modes are calculated and ranked according to the amount of increase. The most increasing intrinsic modes are selected as much as the number of vibration modes. The selected intrinsic modes are decomposed. The said bands are independently converted into images by Hibert-Huang Transform. These images are used to train the artificial intelligence model. The said embodiment enables chatter to be detected substantially without the need to collect any physical data for the training of the artificial intelligence model.
The control unit (500) and the data acquisition system (400) provided in one embodiment of the invention are preferably integrated. In the said embodiment, the data acquisition system (400) is provided as a module which is located in the control unit (500) and integrated into the control unit (500). A computer is used as the control unit (500) provided in one embodiment of the invention. In the said embodiment, the computer, i.e. the control unit (500), is a well-equipped computer with simple or different modules adapted to store the data of the machine tool and the data obtained from the data acquisition system (400), and to create a model by processing the said data with an artificial intelligence.

Claims

CLAIMS A chatter detection system (100) used in machine tools in order to detect the chatter caused by the dynamic interactions occurring between the cutting tool and the workpiece especially during machining, comprising at least one impact hammer (200) adapted to measure the forces between the cutting tool and the workpiece on the machine tool, which cause the chatter, at least one accelerometer (300) adapted to measure the acceleration between the cutting tool and the workpiece on the machine tool, which causes the chatter, and characterized by a data acquisition system (400) used to enable the data obtained by the impact hammer (200) and accelerometer (300) to be digitally collected, stored and transferred to a control unit (500), at least one control unit (500), positioned on or away from the machine tool, having an artificial intelligence or comprising models generated by an artificial intelligence, and adapted to perform the following technical elements;
• theoretical generation of the chatter data using analytical and numerical models,
• using the chatter data in artificial neural network training using Hilbert- Huang Transform,
• decomposition into EEMD and IMF modes during preprocessing. A chatter detection system (100) according to Claim 1, characterized by at least one data acquisition system (400) adapted to digitally collect the data obtained by means of the impact hammer (200) and the accelerometer (300) and to transmit the obtained data to the control unit (500) in order to train the artificial neural networks. A chatter detection system (100) according to Claim 1, characterized by at least one control unit (500) adapted to process the chatter data by analytical and numerical models, to obtain artificial neural networks by analytical and numerical model solutions and to label the artificial neural networks and stability lobe diagrams. A chatter detection system (100) according to Claim 1, characterized by at least one control unit (500) which uses the obtained chatter data in artificial neural network training using Hilbert-Huang Transform, and which is adapted to convert the labeled data into two-dimensional images called “HHT images” and to train the convolutional artificial neural network (CNN) with the said images. A chatter detection system (100) according to Claim 1, characterized by at least one impact hammer (200), at least one accelerometer (300) and at least one data acquisition system (400) which are dynamically adapted such that they are in communication with each other. A chatter detection system (100) according to Claim 1, characterized by at least one data acquisition system (400) adapted to model the frequency response function of the obtained data and the machine tool, to enable the said models to generate vibration signals that appropriately simulate the material removal process performed during machining, and to plot stability lobe diagrams for automatic labeling which automatically generate a plurality of signal samples required for training the artificial intelligence model. A chatter detection system (100) according to Claim 1, characterized by the accelerometer (300) and the data acquisition system (400) adapted to continuously collect data during machining with the trained models obtained after training the artificial intelligence model, to evaluate the obtained data and, in case of chatter in the machine tool, to recommend stable material removal parameters but at higher speeds without reducing efficiency. A chatter detection system (100) according to Claim 1, characterized by at least one control unit (500) adapted to perform preprocessing by decomposition of the obtained data into intrinsic mode functions (IMF) by EEMD in order for the artificial intelligence models to provide more successful results, and thereby enabling the chatter to be detected substantially without the need to collect any physical data for the training of the artificial intelligence model.
9. A chatter detection system (100) according to Claim 1, characterized by at least one data acquisition system (400) which is located in the control unit (500) and integrated into the control unit (500). 10. A chatter detection system (100) according to Claim 1, characterized by at least one control unit (500) which is a well-equipped computer with simple or different modules adapted to store the data of the machine tool and the data obtained from the data acquisition system (400), and to create a model by processing the said data with an artificial intelligence.
PCT/TR2023/051053 2022-10-03 2023-09-29 A system for detecting machine tool chatter WO2024076326A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
TR2022/015107 TR2022015107A1 (en) 2022-10-03 A SYSTEM FOR MACHINE TOOL CHATTER VIBRATION TESTING
TR2022015107 2022-10-03

Publications (1)

Publication Number Publication Date
WO2024076326A1 true WO2024076326A1 (en) 2024-04-11

Family

ID=90608479

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/TR2023/051053 WO2024076326A1 (en) 2022-10-03 2023-09-29 A system for detecting machine tool chatter

Country Status (1)

Country Link
WO (1) WO2024076326A1 (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2013036912A (en) * 2011-08-10 2013-02-21 Jtekt Corp Chattering vibration detection device
CN104898565A (en) * 2014-03-05 2015-09-09 麦克隆·阿杰·查米莱斯股份公司 Improved database for chatter predictions
CN111624947A (en) * 2019-02-27 2020-09-04 发那科株式会社 Chattering determination device, machine learning device, and system
JP2020160830A (en) * 2019-03-27 2020-10-01 ブラザー工業株式会社 Numerical control device, machine tool, control program, and storage medium
MY184771A (en) * 2016-10-12 2021-04-21 Univ Malaysia Pahang Chatter suppression in a high-speed milling machine

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2013036912A (en) * 2011-08-10 2013-02-21 Jtekt Corp Chattering vibration detection device
CN104898565A (en) * 2014-03-05 2015-09-09 麦克隆·阿杰·查米莱斯股份公司 Improved database for chatter predictions
MY184771A (en) * 2016-10-12 2021-04-21 Univ Malaysia Pahang Chatter suppression in a high-speed milling machine
CN111624947A (en) * 2019-02-27 2020-09-04 发那科株式会社 Chattering determination device, machine learning device, and system
JP2020160830A (en) * 2019-03-27 2020-10-01 ブラザー工業株式会社 Numerical control device, machine tool, control program, and storage medium

Similar Documents

Publication Publication Date Title
CN108227634B (en) Machine learning device, CNC device, and machine learning method
Teti et al. Advanced monitoring of machining operations
CN110153801A (en) A kind of cutting-tool wear state discrimination method based on multi-feature fusion
Dornfeld Application of acoustic emission techniques in manufacturing
Peng et al. A new method for the prediction of chatter stability lobes based on dynamic cutting force simulation model and support vector machine
CN102929210A (en) Control and optimization system for feature-based numerical control machining process and control and optimization method therefor
CN104741638A (en) Turning cutter wear state monitoring system
CN103192292B (en) Numerical control machine error identification and separation method based on processing workpiece curved surface morphology information
CN112192319A (en) Tool wear monitoring method and system of unsupervised model
Yamazaki et al. Autonomously proficient CNC controller for high-performance machine tools based on an open architecture concept
Klein et al. Quality prediction of honed bores with machine learning based on machining and quality data to improve the honing process control
Unver et al. Exploring the potential of transfer learning for chatter detection
CN112733298B (en) Machining performance evaluation method of series-parallel robot at different poses based on spiral hole milling
CN111390648A (en) Turning tool abrasion judging method based on antagonistic neural network
Maeda et al. Method for automatically recognizing various operation statuses of legacy machines
WO2024076326A1 (en) A system for detecting machine tool chatter
CN111487924B (en) Cutter damage analysis method based on multi-source heterogeneous data of production line
Codjo et al. Honeycomb core milling diagnosis using machine learning in the industry 4.0 framework
Zuperl et al. A cyber-physical system for smart fixture monitoring via clamping simulation
CN109759628B (en) Engine cylinder block top surface milling flutter prediction method based on dynamic meshing tooth number
Fomin et al. An approach to the construction of a nonlinear dynamic model process cutting for diagnosis condition of tools
Park et al. Prediction of the CNC tool wear using the machine learning technique
Pourmostaghimi et al. Vibration based Assessment of Tool Wear in Hard Turning using Wavelet Packet Transform and Neural Networks.
Arslan et al. Automated machine tool prognostics for turning operation using acoustic emission and learning vector quantization
KR102512875B1 (en) Tool path calibration model generation method and tool path calibration method using the same

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23875337

Country of ref document: EP

Kind code of ref document: A1