TWI808695B - Milling real-time monitoring system - Google Patents

Abstract

A real-time monitoring system for milling. The function of the intelligent tool holder is to use the sensing capability and wireless transmission function of the body to transmit the sensing value to a background program through wireless transmission, and to establish the real-time monitoring system through an algorithm and a decoupling program. The real-time cutting model is used to predict the cutting parameters of the machining tool of the smart knife, calculate the actual machining physical quantity of the machining tool tip, and monitor the state of the machining tool in real time.

Description

Milling real-time monitoring system

The present invention relates to a milling processing monitoring system, in particular to establishing a model of a smart tool milling processing behavior and a sensing mechanism to realize higher-precision real-time monitoring of processing.

Currently, metal milling remains the most important machining process, providing the basis for all craft products. Over the decades, increasing demands on the milling process and the quality of its products have led to dramatic changes in the milling process.

In the process of metal processing, cutting tools play an extremely important role. The number of cutting tools is huge and the application is complex, making the use and management of cutting tools an important factor in reducing production costs and shortening production time. The development trend of modern factories is towards automatic and intelligent production methods. Therefore, real-time monitoring of processing status and real-time tool information can improve the utilization rate of machinery and equipment and the competitiveness of products.

As far as the current technology is concerned, it may be a way to design the sensing mechanism on the machine tool spindle or the worktable, so as to monitor the machining status of the tool with real-time dynamic force sensing signals. The sensor used is a strain gauge sensor. By monitoring specific parameters to feed back the cutting control method, the sensor is installed between the workbench and the workpiece to detect the cutting force, and the sensor is installed on the fixture to detect the rotating cutting force. However, the decoupling design of this type of monitoring technology is complicated, and a large number of algorithms are required for analysis. The accuracy is low, and the sensing of each axis is easy to interfere with each other.

The purpose of the present invention is to provide a real-time monitoring system for milling processing. For the smart knife handle with the sensing mechanism installed on the knife handle body, the sensed value is transmitted to the background process through wireless transmission. On the formula side, through the established algorithm and decoupling program, the value is converted into the desired physical quantity (torque, bending moment, axial force), which is used to detect the overall force of the tool during the machining process, and realize higher-precision machining real-time monitoring.

In order to achieve the above object, the present invention provides a real-time monitoring system for milling, comprising: a smart tool handle for milling, the smart tool handle is provided with a sensing module to sense the piezoelectric sensing value of the stress and strain generated by the machining tool under a corresponding load through mechanical sensing, and transmits it wirelessly through a transmission terminal;

The receiving end is used to receive the piezoelectric sensing value transmitted from the transmitting end, and the sensing value decoupling module converts the aforementioned piezoelectric sensing value into the force signal value at the sensing position; the cutting theoretical model is based on the basic mathematical model of the processing parameters, and is used to simulate the general trend of change in the cutting process; The state of the machining tool.

As a preferred mode, the sensing module includes a plurality of piezoelectric sensing elements embedded in the smart knife handle body to sense piezoelectric sensing values of stress and strain generated by the smart knife handle body under the load of the corresponding processing tool. The piezoelectric sensing elements are used for layered sensing of the bending moment load and the torque load of the processing tool, and the sensing elements are configured with two symmetrically embedded positions for each bending moment and torque sensing.

As a preferred manner, the processing parameters include cutting depth, cutting width, number of cutting edges, workpiece material and spindle speed. The actual machining physical quantities of the machining tool tip include torque, bending moment and axial force.

As an optimal way, the cutting theoretical model will have certain discrepancies between the theory and the real situation. Phase difference, so phase coupling with the sensing value decoupling module is required during the program calculation process to determine the cutting coordinate system. After the cutting coordinate system is established, the piezoelectric sensing value and coordinate system are imported as parameters into the real-time prediction model.

As a preferred mode, the real-time monitoring system for milling processing includes an experimental end. The experimental end measures the physical quantity outside the smart tool handle by an experimental method, and conducts a model verification with the actual machining physical quantity of the machining tool tip calculated by the real-time cutting model, and feeds the verification back to the training correction of the model parameters of the real-time cutting model.

As a preferred manner, the experimental end includes a dynamometer to measure physical quantities outside the smart knife handle, and converts the aforementioned physical quantities into parameters corresponding to the real-time cutting model through a conversion matrix.

The establishment of the processing dynamic monitoring system of the present invention includes three functional modules, which are the smart knife handle, the program end, and the experiment end. The function of the smart knife handle is to use the sensing ability and wireless transmission function of the main body to transmit the sensed value to the background program end through wireless transmission, and then convert the value into the desired physical quantity (torque, bending moment, and axial force) through the established algorithm and decoupling program. Establish a mathematical model for the milling behavior and sensing mechanism, and use the AI algorithm to correct the model parameters to achieve higher precision cutting force prediction. The model calculates the actual machining physical quantity of the tool tip and monitors the status of the tool in real time.

100: Wisdom Knife Handle

101:Spindle assembly part

102: clamping part

103: Connecting part

104: Processing tools

110:Sensing module

111: Piezoelectric sensing value

112: Piezoelectric sensing element

120: transmission end

130: shell

200: terminal

210: Receiver

220: Sensing value decoupling module

230: Processing parameters

231: depth of cut

232: Cutting width

233: Number of cutting edges

234: Workpiece material

235:Spindle speed

240: Cutting theory model

250: instant cutting model

260: Actual processing physical quantity

261: Torque

262: Bending moment (X)

263: Bending moment (Y)

264: axial force

300: Experimental end

310: Dynamometer

320: Transformation matrix

400: Model Validation

[Figure 1] is the block diagram 1 of the real-time monitoring system for processing in this case.

[Fig. 2] is the second block diagram of the real-time processing monitoring system in this case.

[Figure 3] is a schematic diagram of the structure of the algorithm in this case.

[Figure 4] is a schematic diagram of the smart knife handle in this case.

Embodiments of the present invention will be described in detail below and illustrated with accompanying drawings. In addition to these detailed In addition to the description, the present invention can also be widely implemented in other embodiments, and any easy replacement, modification, and equivalent change of any of the embodiments are included in the scope of the present invention, and the scope of the patent application shall prevail. In the description of the specification, many specific details are provided in order to enable readers to have a more complete understanding of the present invention; however, the present invention may still be practiced under the premise of omitting some or all of the specific details. Furthermore, well-known steps or elements have not been described in detail in order to avoid unnecessarily limiting the invention. The same or similar elements in the drawings will be denoted by the same or similar symbols. It should be noted that the drawings are for illustrative purposes only, and do not represent the actual size or quantity of components, and some details may not be fully drawn for the sake of simplicity of the drawings.

Please refer to Figures 1 to 3, which are the block diagram of the real-time processing monitoring system of this case and the structural diagram of the algorithm of this case, and Figure 4 is the schematic diagram of the smart knife handle of this case. The real-time monitoring system for milling processing in this embodiment includes: a smart knife handle 100 for milling processing. The smart knife handle 100 is provided with a sensing module 110 through mechanical sensing of the piezoelectric sensing value 111 of the stress and strain generated by the machining tool 104 under a corresponding load, and transmits it wirelessly through a transmission terminal 120.

In terms of implementation, the smart knife handle 100 includes a spindle assembly part 101 , a clamping part 102 and a connecting part 103 , wherein the machining tool 104 is connected to the end of the connecting part 103 . The spindle assembly part 101 is used to connect the spindle of a processing machine, such as a milling machine, a drilling machine, a lathe or a sawing machine. The clamping part 102 is connected to the spindle assembly part 101, and the clamping part 102 is used for clamping or changing tools in the tool magazine; the clamping part 102 is connected to the connecting part 103, and the connecting part 103 is connected to the processing tool 104, and the processing tool 104 is, for example, a milling cutter, a drill bit, a turning tool, a saw blade, etc.

In terms of implementation, the connecting portion 103 of the smart knife handle 100 is provided with a casing 130 covering the sensing module 110 and the transmission end 120 for wireless transmission. The sensing module 110 includes a plurality of piezoelectric sensing elements 112 embedded in the connecting portion 103 of the body of the smart knife handle 100, and uses mechanical sensing to sense the piezoelectric sensing value 111 of the stress and strain generated by the smart knife handle 100 under the load of the corresponding processing tool 104; these piezoelectric sensing elements 112 are used for bending moment load and The torque load is sensed in layers, and the piezoelectric sensing elements 112 are configured with two symmetrical embedded positions for each bending moment and torque sensing.

In terms of implementation and application, this case uses mechanical analysis to find out the position where the body of the smart knife handle 100 can generate the maximum stress and strain under the load of the corresponding tool tip 104 . As a result, it can be seen that the stress of the bending moment load is the largest near the main shaft assembly part 101. In order to decouple the force signal output of the piezoelectric sensing elements 112, this project adopts a layered design for the sensing of the bending moment load and torque load, and adopts a symmetrical design. For each bending moment (Mx, My) and torque (Tz), two piezoelectric sensing elements 112 with symmetrical embedded positions are arranged to improve the sensing accuracy of the force on the tool tip of the machining tool 104. It can also be known through mechanical analysis that when Mx or My bending moment is applied at the tip of the tool, although the piezoelectric sensing element 112 of the bending moment can output its corresponding voltage signal, the piezoelectric sensing element 112 of the torque will also be affected by it and output a part of the voltage signal. Therefore, in order to further obtain a better decoupling effect in practice, the positions of the embedded holes of the piezoelectric sensing elements 112 in the upper and lower groups are staggered from each other, for example, by 45 degrees, which can be used to improve the force coupling effect.

The real-time monitoring system for milling processing in this embodiment includes a programming terminal 200, which is installed in a computer device, and a processing parameter 230 can be input through a man-machine interface.

The receiving end 210 is a wireless transmission receiving end for receiving the piezoelectric sensing value 111 transmitted from the transmitting end 120, and the sensing value decoupling module 220 converts the aforementioned piezoelectric sensing value 111 into a force signal value at the sensing position. As mentioned above for the configuration of the piezoelectric sensing element 112, decoupling means that in the original multi-variable sensing system, it is necessary to build an appropriate mechanism to eliminate the mutual coupling between the variables in the system, so that each input will only affect the corresponding output, and each output is only controlled by the input, so that the original multi-variable system can be transformed into multiple Single input single output system. The sensing module in this case can directly generate the voltage signal output corresponding to the load by placing the polarization direction of the PZT piezoelectric sheet in the direction of the piezoelectric force under the load we want to be able to sense through piezoelectric sensing, without the need for a large number of decoupling calculations required by the strain gauge sensing system. Therefore, as long as the embedded position and deflection angle are designed, a good enough decoupling effect can be obtained.

The cutting theoretical model 240 is a basic mathematical model based on the machining parameters 230, and is used to simulate the general trend of change in the tool cutting process; the machining parameters 230 include cutting depth 231, cutting width 232, number of cutting edges 233, workpiece material 234 and spindle speed 235.

The real-time cutting model 250 is established based on the aforementioned force signal value and the simulation of the cutting theoretical model 240, and is used to predict the cutting parameters of the processing tool 104 of the smart handle 100, calculate the actual processing physical quantity 260 of the tool tip of the processing tool 104, and monitor the state of the processing tool in real time. The actual machining physical quantity 260 of the tool tip of the machining tool 104 includes a torque (Tz) 261 , a bending moment (X) 262 , a bending moment (Y) 263 and an axial force 264 .

In terms of implementation and application, because the cutting theoretical model 240 is based on the basic mathematical model of the machining parameters 230, the purpose is to simulate the general change trend in the cutting process, and because there will be a certain phase difference between the theory and the real situation, it is necessary to perform phase coupling in the programming terminal 200 to determine the cutting coordinate system (as shown in Figure 3). During the program calculation process, phase coupling with the sensing value decoupling module 220 is required to determine the cutting coordinate system. After the cutting coordinate system is established, the piezoelectric sensing value 111 and the coordinate system are imported as parameters into the real-time prediction model 250, and the cutting parameters are corrected according to the cutting time domain. The goal of the real-time prediction model 250 is to predict cutting parameters (K item: cutting force coefficient N/mm2, e item: correction coefficient N). The prediction of cutting parameters is to calculate the actual physical quantity (knife tip) from the sensor force signal (knife body). The two include factors such as tool handle material and processing interference, so it needs to be performed by means of a predictive model.

In terms of implementation and application, the real-time monitoring system for milling processing in this case includes an experimental terminal 300, which The experimental end 300 measures the physical quantity outside the smart knife handle 100 by means of an experiment, and conducts a model verification 400 with the real-time cutting model 250 to calculate the actual machining physical quantity of the machining tool 104 tip, and feeds the verification back to the training and correction of the model parameters of the real-time cutting model 250, corrects the algorithm parameters according to the model accuracy, and corrects the establishment of the real-time cutting model 250.

In terms of implementation, the experimental end 300 includes a dynamometer 310 to measure physical quantities outside the smart knife handle 100 , and converts the aforementioned physical quantities into parameters corresponding to the real-time cutting model 250 through a conversion matrix 320 .

The establishment of the processing dynamic monitoring system of the present invention includes three functional modules, which are the smart knife handle, the program end, and the experiment end. The function of the smart knife handle is to use the sensing ability and wireless transmission function of the main body to transmit the sensed value to the background program end through wireless transmission, and then convert the value into the desired physical quantity (torque, bending moment, and axial force) through the established algorithm and decoupling program. Establish a mathematical model for the milling behavior and sensing mechanism, and use the AI algorithm to correct the model parameters to achieve higher precision cutting force prediction. The model calculates the actual machining physical quantity of the tool tip and monitors the status of the tool in real time.

The embodiments disclosed above are only illustrative of the principles, features and effects of the present invention, and are not intended to limit the scope of the present invention. Anyone skilled in the art can modify and change the above embodiments without departing from the spirit and scope of the present invention. Any equivalent change and modification accomplished by using the content disclosed in the present invention should still be covered by the scope of the following patent application.

100: Wisdom Knife Handle

110:Sensing module

120: transmission end

200: terminal

210: Receiver

220: Sensing value decoupling module

230: Processing parameters

240: Cutting theory model

250: instant cutting model

260: Actual processing physical quantity

Claims (8)

A real-time monitoring system for milling processing, comprising: a smart tool handle for milling processing. The smart tool handle is provided with a sensing module to sense the piezoelectric sensing value of the stress and strain generated by the processing tool under a corresponding load through mechanical sensing, and transmits wirelessly through a transmission terminal; and a program terminal installed on a computer device, which can input a processing parameter through a man-machine interface. The piezoelectric sensing value transmitted from the terminal is converted into the force signal value at the sensing position by the sensing value decoupling module; the cutting theoretical model is a basic mathematical model based on the processing parameters, which is used to simulate the general trend of change in the cutting process; and the real-time cutting model is established based on the aforementioned force signal value and the simulation of the cutting theoretical model, and is used to predict the cutting parameters of the processing tool of the smart handle, calculate the actual processing physical quantity of the processing tool tip, and monitor the state of the processing tool in real time. The real-time monitoring system for milling processing as described in Claim 1, wherein the sensing module includes a plurality of piezoelectric sensing elements embedded in the smart knife handle body to sense the piezoelectric sensing values of the stress and strain generated by the smart knife handle body under the load of the corresponding processing tool. The real-time monitoring system for milling processing as described in Claim 2, wherein the piezoelectric sensing elements are used for layered sensing of the bending moment load and torque load of the processing tool, and each of the bending moment and torque sensing is configured with two symmetrical embedded position sensing elements. The real-time monitoring system for milling processing as described in Claim 1, wherein the processing parameters include cutting depth, cutting width, number of cutting edges, workpiece material and spindle speed. The real-time monitoring system for milling processing according to Claim 1, wherein the actual processing physical quantities of the cutting tool tip include torque, bending moment and axial force. The real-time monitoring system for milling processing as described in Claim 1, wherein the cutting theoretical model has a certain phase difference between the theory and the real situation, so it needs to be phase-coupled with the sensing value decoupling module during the program calculation process to determine the cutting coordinate system. After the cutting coordinate system is established, the piezoelectric sensing value and coordinate system are imported as parameters into the real-time prediction model. The real-time monitoring system for milling processing as described in claim 1, wherein the real-time monitoring system for milling processing includes an experiment end, the experiment end measures the physical quantity outside the smart tool handle by an experimental method, performs a model verification with the actual processing physical quantity of the machining tool tip calculated by the real-time cutting model, and feeds the verification back to the training correction of the model parameters of the real-time cutting model. The real-time monitoring system for milling processing as described in Claim 7, wherein the experimental end includes a dynamometer to measure physical quantities outside the smart tool handle, and converts the aforementioned physical quantities into parameters corresponding to the real-time cutting model through a conversion matrix.

Images