Intelligent milling system for low-rigidity cylindrical part and self-adaptive machining method

Abstract

The invention discloses an intelligent milling system for a weak-rigidity cylindrical part and a self-adaptive machining method, and relates to the technical field of mechanical cutting machining.The intelligent milling system comprises a six-degree-of-freedom robot, a milling device arranged at the tail end of the six-degree-of-freedom robot, an inner supporting floating tool for clamping the weak-rigidity cylindrical part and an industrial personal computer for controlling the machining system; the robot control cabinet is used for controlling the six-degree-of-freedom robot to move in space, detecting the global deformation of the wall surface of the weak-rigidity cylindrical part through the inner support floating tool, and constructing and utilizing a finite element simulation-recurrent neural network global wall surface deformation prediction model to predict the wall surface deformation. And the cutting amount of the cutter is adjusted in real time for dynamic compensation in the machining process. The machining precision can be improved, machining errors are reduced, the tool utilization rate is increased, and the production cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of mechanical cutting processing, and in particular to an intelligent milling processing system and an adaptive processing method for weak-rigidity cylindrical parts. Background Art

[0002] Weak rigidity cylindrical parts are widely used in aerospace, shipbuilding, automobile industry, energy equipment and other fields. They are mainly formed by casting, and the outer wall precision of cast cylindrical parts is poor. In order to ensure that the cylindrical parts have good structural strength, meet the assembly precision of the cylindrical parts, and improve the production quality and consistency of the cylindrical parts, the outer wall needs to be milled and other cutting processes.

[0003] At present, the processing of the outer wall of weak-rigidity cylindrical parts mainly adopts manual operation, which has low processing efficiency and high labor intensity. At the same time, the processing parameters are adjusted by manual experience, resulting in poor processing accuracy, poor repeatability and poor product production consistency. The weak rigidity of the cylindrical parts also makes it easy to deform during the processing, resulting in large processing errors. Since the outer wall of this type of cylindrical parts is complex and thin, it is difficult to predict the deformation. In order to reduce the influence of the weak rigidity of the cylindrical parts, a multi-pass milling process is usually used to fully remove the excess to perform the outer wall cutting of the weak-rigidity cylindrical parts. However, this method is inefficient and causes great loss to the tool, and cannot meet the processing requirements of cylindrical thin-walled castings. In addition, the processing fixtures of weak-rigidity cylindrical parts are mostly customized fixtures, and they are clamped externally, which can only be applied to a specific process and specific workpiece, with low utilization rate. The outer wall clamping method is also prone to large workpiece deformation, resulting in poor processing accuracy, poor repeatability and poor product production consistency. In short, the poor rigidity and easy deformation of weak-rigidity cylindrical parts during processing are the main reasons for their processing errors. A complete processing system and processing method are urgently needed to achieve high-precision processing of weak-rigidity cylindrical parts.

[0004] Therefore, technicians in this field are committed to developing an intelligent milling system and an adaptive processing method for weak-rigidity cylindrical parts. Summary of the invention

[0005] In view of the above-mentioned defects of the prior art, the technical problem to be solved by the present invention is how to improve the milling accuracy of the outer wall of a weak-rigidity cylindrical part.

[0006] To achieve the above-mentioned objectives, the present invention provides an intelligent milling processing system for weak-rigidity cylindrical parts, comprising a six-degree-of-freedom robot, a milling device arranged at the end of the six-degree-of-freedom robot, an internal support floating tooling for clamping weak-rigidity cylindrical parts, an industrial computer for controlling the processing system, and a robot control cabinet for controlling the spatial movement of the six-degree-of-freedom robot. The global deformation of the wall of the weak-rigidity cylindrical part is detected by the internal support floating tooling, a finite element simulation-recurrent neural network global wall deformation prediction model is constructed and used to predict the wall deformation, and the tool cutting amount is adjusted in real time to perform dynamic compensation during the processing process.

[0007] Furthermore, the inner support floating tooling includes auxiliary support modules that are evenly distributed along the circumferential direction and in contact with the inner wall of the weak-rigidity cylindrical member, the number of the auxiliary support modules is six, a displacement sensor is arranged at the end of the auxiliary support module, and a communication control module is arranged inside the inner support floating tooling, and the displacement sensor is an infrared displacement sensor.

[0008] Furthermore, the internal support floating tooling also includes an internal support floating tooling base, a gear transmission pair consisting of a first radial transmission gear and a second radial transmission gear arranged on the internal support floating tooling base, an auxiliary support module guide rail platform is arranged on the upper surface of the first radial transmission gear, the auxiliary support modules are evenly distributed circumferentially along the auxiliary support module guide rail platform, and the auxiliary support modules are synchronously radially moved along the auxiliary support module guide rail through the transmission of the gear pair and the thread pair.

[0009] Furthermore, the auxiliary support module includes an auxiliary support module slider which is arranged on the auxiliary support module guide rail platform and moves radially along the auxiliary support module guide rail through a gear pair and a thread pair, and an auxiliary support module end floating groove surface pressure block which is connected to the auxiliary support module slider through an auxiliary support module floating groove surface pressure block slider connecting pin, the groove surface of the floating groove surface pressure block at the end of the auxiliary support module contacts the inner surface of the weak-rigidity cylindrical part, and the displacement sensor is arranged on the floating groove surface pressure block at the end of the auxiliary support module.

[0010] Furthermore, a spring is provided at the contact end between the floating groove surface pressing block at the end of the auxiliary support module and the inner surface of the weak rigidity cylindrical member.

[0011] Furthermore, the inner support floating tooling is arranged on a processing platform.

[0012] Furthermore, the inner support floating fixture also includes a clamping hole for clamping a weak rigidity cylindrical part.

[0013] Furthermore, the milling device includes a milling spindle and a tool clamped on the milling spindle.

[0014] Furthermore, an adaptive processing method using the weak-rigidity cylindrical part intelligent milling processing system comprises the following steps:

[0015] Step 1: Adaptive clamping and positioning of weak rigidity cylindrical parts;

[0016] Step 2: Construct a finite element simulation-recurrent neural network global wall deformation prediction model, collect training sets, and train and verify the wall time series deformation prediction recurrent neural network;

[0017] Step three, perform cutting processing. According to the slight deformation of the weak-rigidity cylindrical part, the real-time deformation of the processing position is predicted through the recurrent neural network global wall deformation prediction model, and the tool cutting amount is adjusted to perform dynamic compensation for processing.

[0018] Furthermore, step one also includes the following steps:

[0019] Step 1.1 Use the auxiliary support module to perform adaptive centering clamping on the weak rigidity cylindrical part;

[0020] Step 1.2: setting a suitable floating amount according to the error characteristics of the weak-rigidity cylindrical component;

[0021] Step 2 also includes the following steps:

[0022] Step 2.1 Import the digital model of the weak rigidity cylindrical part into the industrial computer, perform simulation analysis through finite element simulation software, and use the output simulation results to generate a network model data set;

[0023] Step 2.2: Match the deformation amount through the characteristic points of the outer wall of the weak-rigidity cylindrical part to obtain a global deformation result that matches the actual clamping coordinate system;

[0024] Step 2.3: Repeat steps 2.1-2.2 to obtain a training set and a validation set of sufficient size to train and validate the recurrent neural network global wall deformation prediction model;

[0025] Step three also includes the following steps:

[0026] Step 3.1 The industrial computer outputs the machining trajectory of the outer wall of the weak-rigidity cylindrical part through a digital model, and controls the machining end effector to perform outer surface machining according to the machining path;

[0027] Step 3.2 The displacement sensor detects the slight deformation displacement change of the weak rigidity cylindrical part and outputs it to the industrial computer through the communication control module;

[0028] Step 3.3 predicts the real-time deformation of the processing position through the recurrent neural network global wall deformation prediction model;

[0029] Step 3.4 generates tool cutting compensation according to the predicted deformation and performs machining dynamic compensation.

[0030] Compared with the prior art, the present invention has the following advantages: it can predict the wall deformation during the milling process of the outer wall of the cylindrical part, so as to adjust the tool in real time, complete dynamic compensation during the processing, and thus improve the processing accuracy; it can effectively reduce the deformation of the cylindrical part caused by clamping the fixture, and thus reduce the processing error; the same cylindrical part tooling can process the outer walls of cylindrical parts of multiple sizes, which can improve the tooling utilization rate, reduce production costs, and enhance the level of production flexibility.

[0031] The concept, specific structure and technical effects of the present invention will be further described below in conjunction with the accompanying drawings to fully understand the purpose, characteristics and effects of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 This is a schematic structural diagram of an implementation method of the present invention;

[0033] Figure 2 This is a schematic diagram of a floating inner support tooling structure according to an embodiment of the present invention;

[0034] Figure 3 The present invention is a flowchart of a processing control method according to an embodiment of the present invention.

[0035] Among them: 1-six-degree-of-freedom robot, 2-milling device, 3-inner support floating tooling, 301-inner support floating tooling base, 302-auxiliary support module guide rail platform, 303-auxiliary support module slider, 304-auxiliary support module end floating groove surface pressure block, 305-auxiliary support module floating groove surface pressure block slider connecting pin, 306-displacement sensor, 307-first radial transmission gear, 308-second radial transmission gear, 309-clamping hole, 310-communication control module, 4-weak rigidity cylindrical part, 5-industrial computer, 6-robot control cabinet, 7-processing platform. DETAILED DESCRIPTION

[0036] The following describes the preferred embodiments of the present invention with reference to the drawings in the specification, so that the technical content is clearer and easier to understand. The present invention can be embodied in many different forms of embodiments, and the protection scope of the present invention is not limited to the embodiments mentioned in the text.

[0037] In the drawings, components with the same structure are indicated by the same numerical reference numerals, and components with similar structures or functions are indicated by similar numerical reference numerals. The size and thickness of each component shown in the drawings are arbitrarily shown, and the present invention does not limit the size and thickness of each component. In order to make the illustration clearer, the thickness of the components is appropriately exaggerated in some places in the drawings.

[0038] like Figure 1As shown, the intelligent processing system for milling the complex outer wall of a weak-rigidity cylindrical part of the present invention comprises a six-degree-of-freedom robot 1, a milling device 2 installed at the end of the six-degree-of-freedom robot 1, a processing platform 7 corresponding to the milling device 2, an inner support floating fixture 3 installed on the upper surface of the processing platform 7, a weak-rigidity cylindrical part 4 to be processed is clamped on the processing platform 7 through the inner support floating fixture 3, an industrial computer 5 that controls the entire processing system, performs finite element simulation (FES: Finite Element Simulation), recurrent neural network (RNN: Recurrent Neural Network) training, and data transmission and processing, and a robot control cabinet 6 that controls the spatial movement of the six-degree-of-freedom robot 1. The industrial computer 5 controls the spatial movement freedom of the six-degree-of-freedom robot 1 through the robot control cabinet 6 to realize the milling processing of the outer wall of the weak-rigidity cylindrical part 4 by the milling device 2.

[0039] like Figure 2 As shown, the inner support floating fixture 3 includes an inner support floating fixture base 301 installed on the upper surface of the processing platform 7, and the inner support floating fixture 3 is provided with a clamping hole 309 for clamping the weak rigidity cylindrical member 4. A gear transmission pair consisting of a first radial transmission gear 307 and a second radial transmission gear 308 is provided on the upper part of the inner support floating fixture 3, and an auxiliary support module guide rail platform 302 is installed on the upper surface of the first radial transmission gear 307. Six auxiliary support modules are evenly installed on the auxiliary support module guide rail platform 302 along the circumferential direction. Through the transmission of the gear pair and the thread pair, the six auxiliary support modules can be adjusted by synchronous movement along the auxiliary support module guide rail.

[0040] Among them, the auxiliary support module includes an auxiliary support module slider 303 which is arranged on the auxiliary support module guide rail platform 302 and moves radially along the auxiliary support module guide rail through the gear pair and the thread pair, and an auxiliary support module end floating groove surface pressure block 304 which is connected to the auxiliary support module slider 303 through the auxiliary support module floating groove surface pressure block slider connecting pin 305. In this way, the auxiliary support module slider 303 can drive the auxiliary support module end floating groove surface pressure block 304 to move, and then the groove surface of the floating groove surface pressure block 304 at the end of the auxiliary support module can contact the inner surface of the weak rigidity cylindrical part 4, and the groove surface of the floating groove surface pressure block 304 at the end of the auxiliary support module can increase the contact area with the weak rigidity cylindrical part 4, effectively reduce the clamping pressure, and reduce the weak rigidity deformation of the weak rigidity cylindrical part 4. A spring is provided at the contact end between the floating groove surface pressure block 304 at the end of the auxiliary support module and the inner surface of the weak-rigidity cylindrical part 4, which can provide a pre-tightening force for the workpiece clamping and is used for floating positioning support for the weak-rigidity cylindrical part 4, which is beneficial to maintaining stable clamping and utilizing the vibration absorption of the spring to reduce vibration during the processing.

[0041] The inner support floating tooling 3 also includes a displacement sensor 306 installed on the floating groove surface pressure block 304 at the end of the auxiliary support module. The displacement sensor 306 is used to detect the slight displacement changes between the floating groove surface pressure block and the front end of the auxiliary support module, that is, when the tool contacts the weak rigidity cylindrical part 4, the slight deformation and displacement of the weak rigidity cylindrical part 4 due to the cutting force. The displacement sensor 306 is preferably an infrared displacement sensor, which is arranged in the communication control module 310 inside the inner support floating tooling 3, and is used to output the detected slight displacement changes to the industrial computer.

[0042] The milling device 2 comprises a milling spindle and a tool clamped on the milling spindle. The milling device 2 serves as an end effector and is used to perform milling on the outer wall of the weak-rigidity cylindrical component 4 .

[0043] When clamping the weak rigidity cylindrical part 4, the six auxiliary support modules are synchronously moved inward or outward in the radial direction through the transmission of the gear pair and the thread pair to adjust the automatic centering of the weak rigidity cylindrical part 4 to adapt to the clamping of weak rigidity cylindrical parts 4 of different sizes to be processed. Since the casting error will cause the thickness of the weak rigidity cylindrical part 4 to be uneven, the floating amount can be set according to the error characteristics of the weak rigidity cylindrical part 4 to be processed to ensure the clamping force of the weak rigidity cylindrical part 4. The appropriate floating amount can also suppress the vibration during the processing, increase the cutting amount, and reduce the processing deformation.

[0044] When processing the weak rigidity cylindrical part 4, the inner support floating tooling 3 provides an outward support force for the outer wall of the weak rigidity cylindrical part 4 in the normal direction, which can offset the deformation of part of the outer wall caused by the cutting force when the tool contacts the tool, reduce the error caused by the weak rigid deformation of the weak rigidity cylindrical part 4, and improve the processing accuracy. Due to the floating setting of the auxiliary support module and the weak rigid deformation of the weak rigidity cylindrical part 4 itself, when the tool contacts the weak rigidity cylindrical part 4, due to the action of the cutting force, the weak rigidity cylindrical part 4 produces a slight displacement and deformation. The displacement sensor 306 detects the slight displacement change of the front end of the floating groove surface pressing block 304 at the end of the auxiliary support module, and outputs it to the central processor of the industrial computer 5 through the communication control module 310 as the input of the global wall deformation prediction model of the recurrent neural network. The prediction model will output the deformation of the workpiece of the next step length, and the tool cutting amount is adjusted in real time through the industrial computer 5 to perform dynamic error compensation during the processing process, thereby effectively improving the wall processing accuracy.

[0045] like Figure 3 As shown, the adaptive processing method based on the above-mentioned intelligent processing system for milling complex outer walls of weak-rigidity cylindrical thin-walled parts includes the following steps:

[0046] Step 1: Adaptive clamping and positioning of weak rigidity cylindrical parts

[0047] First, the weak-rigidity cylindrical parts of different sizes to be processed are placed on the inner support fixture of the weak-rigidity cylindrical parts, and the auxiliary support module is adjusted by the gear pair and the thread pair to realize the adaptive centering clamping of the weak-rigidity cylindrical parts. Then, the appropriate floating amount is set according to the error characteristics of the weak-rigidity cylindrical parts.

[0048] Step 2: Construct a finite element simulation-recurrent neural network global wall deformation prediction model, collect training sets, and train and verify the wall time series deformation prediction recurrent neural network.

[0049] Import the digital model of the weak rigidity cylindrical part into the industrial computer, and use the finite element simulation software to set the simulation conditions such as finite element meshing, material parameters, and boundary constraints. Apply a certain force to the weak rigidity cylindrical part to make it displace to a certain extent, output the simulation results, and generate a network model data set. This step can also be performed through the proxy model to perform offline finite element simulation and result output to improve model training efficiency and reduce labor consumption.

[0050] The simulation digital model coordinate system and the actual clamping coordinate system are matched through the characteristic points of the outer wall of the weak-rigidity cylindrical part to obtain a global deformation result that matches the actual clamping coordinate system. The position signal of the displacement sensor extracted from the global deformation result is used as the input variable of the wall deformation prediction model, and the global deformation result is used as the output label of the wall deformation prediction model. Repeat the above simulation process to obtain a training set and a validation set of sufficient size, and train and validate the recurrent neural network global wall deformation prediction model.

[0051] Step 3: Execute cutting processing. According to the slight deformation of the weak rigidity cylindrical part, the real-time deformation of the processing position is predicted through the global wall deformation prediction model of the recurrent neural network, and the cutting amount of the tool is adjusted to perform dynamic compensation for processing.

[0052] Align the robot workpiece coordinate system and tool coordinate system. The industrial computer outputs the machining trajectory of the outer wall of the weak-rigidity cylindrical part through the digital model, and controls the machining end effector to perform outer surface machining according to the machining path. The displacement sensor detects the slight deformation and displacement change of the weak-rigidity cylindrical part, and outputs it to the industrial computer through the communication control module. The central processor of the industrial computer predicts the real-time deformation of the machining position through the global wall deformation prediction model of the recurrent neural network, generates the cutting compensation amount according to the predicted deformation amount, adjusts the tool cutting amount in real time, and performs dynamic compensation during the machining process.

[0053] The present invention can predict the wall deformation during the milling process of the outer wall of the cylindrical part, so as to adjust the tool in real time, complete dynamic compensation during the processing, realize adaptive processing of weak rigidity cylindrical parts, improve processing accuracy, improve production efficiency, and reduce labor production costs; it can improve the stability of clamping, effectively reduce the deformation of the cylindrical part caused by the clamping of the clamp, effectively suppress the vibration during the processing, increase the cutting amount, and thus reduce the processing error and improve the processing accuracy; the same cylindrical part tooling can position and clamp cylindrical parts of multiple sizes to perform outer wall processing, improve the flexible production level of weak rigidity cylindrical parts, improve the utilization rate of the clamp, and save production costs.

[0054] The preferred specific embodiments of the present invention are described in detail above. It should be understood that ordinary technicians in the field can make many modifications and changes based on the concept of the present invention without creative work. Therefore, all technical solutions that can be obtained by technicians in the technical field based on the concept of the present invention through logical analysis, reasoning or limited experiments on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. An intelligent milling system for weak-rigidity cylindrical parts, characterized in that: The invention comprises a six-degree-of-freedom robot, a milling device arranged at the end of the six-degree-of-freedom robot, an inner support floating tooling for clamping weak-rigidity cylindrical parts, an industrial computer for controlling the processing system, and a robot control cabinet for controlling the spatial movement of the six-degree-of-freedom robot. The global deformation of the wall of the weak-rigidity cylindrical part is detected by the inner support floating tooling, a finite element simulation-recurrent neural network global wall deformation prediction model is constructed and used to predict the wall deformation, and the tool cutting amount is adjusted in real time to perform dynamic compensation during the processing.

2. The intelligent milling system for weak-rigidity cylindrical parts according to claim 1, characterized in that: The inner support floating tooling includes auxiliary support modules that are evenly distributed along the circumferential direction and in contact with the inner wall of the weak-rigidity cylindrical member. The number of the auxiliary support modules is six. A displacement sensor is arranged at the end of the auxiliary support module, and a communication control module is arranged inside the inner support floating tooling. The displacement sensor is an infrared displacement sensor.

3. The intelligent milling system for weak-rigidity cylindrical parts according to claim 2, characterized in that: The inner support floating tooling also includes an inner support floating tooling base, a gear transmission pair consisting of a first radial transmission gear and a second radial transmission gear arranged on the inner support floating tooling base, an auxiliary support module guide rail platform is arranged on the upper surface of the first radial transmission gear, the auxiliary support modules are evenly distributed circumferentially on the auxiliary support module guide rail platform, and the auxiliary support modules are synchronously radially moved along the auxiliary support module guide rail through the transmission of the gear pair and the thread pair.

4. The intelligent milling system for weak rigidity cylindrical parts according to claim 3, characterized in that: The auxiliary support module includes an auxiliary support module slider which is arranged on the auxiliary support module guide rail platform and moves radially along the auxiliary support module guide rail through a gear pair and a thread pair, an auxiliary support module end floating groove surface pressure block connected to the auxiliary support module slider through an auxiliary support module floating groove surface pressure block slider connecting pin, the groove surface of the floating groove surface pressure block at the end of the auxiliary support module contacts the inner surface of the weak rigidity cylindrical part, and the displacement sensor is arranged on the floating groove surface pressure block at the end of the auxiliary support module.

5. The intelligent milling system for weak-rigidity cylindrical parts according to claim 4, characterized in that: A spring is arranged at the contact end between the floating groove surface pressing block at the end of the auxiliary support module and the inner surface of the weak rigidity cylindrical member.

6. The intelligent milling system for weak-rigidity cylindrical parts according to any one of claims 1 to 5, characterized in that: The inner support floating tooling is arranged on the processing platform.

7. The intelligent milling system for weak-rigidity cylindrical parts according to any one of claims 1 to 5, characterized in that: The inner support floating fixture also includes a clamping hole for clamping a weak rigidity cylindrical member.

8. The intelligent milling system for weak-rigidity cylindrical parts according to any one of claims 1 to 5, characterized in that: The milling device comprises a milling spindle and a tool clamped on the milling spindle.

9. An adaptive processing method using the intelligent milling processing system for weak-rigidity cylindrical parts according to any one of claims 1 to 5, characterized in that: The steps include: Step 1: Adaptive clamping and positioning of weak rigidity cylindrical parts; Step 2: Construct a finite element simulation-recurrent neural network global wall deformation prediction model, collect training sets, and train and verify the wall time series deformation prediction recurrent neural network; Step 3: Execute cutting processing. According to the slight deformation of the weak-rigidity cylindrical part, the real-time deformation of the processing position is predicted through the recurrent neural network global wall deformation prediction model, and the tool cutting amount is adjusted to perform dynamic compensation for processing.

10. The adaptive processing method of the intelligent milling processing system for weak rigidity cylindrical parts according to claim 9, characterized in that: Step one also includes the following steps: Step 1.1 Use the auxiliary support module to perform adaptive centering clamping on the weak rigidity cylindrical part; Step 1.2: setting a suitable floating amount according to the error characteristics of the weak-rigidity cylindrical component; Step 2 also includes the following steps: Step 2.1 Import the digital model of the weak rigidity cylindrical part into the industrial computer, perform simulation analysis through finite element simulation software, and use the output simulation results to generate a network model data set; Step 2.2: Match the deformation amount through the characteristic points of the outer wall of the weak-rigidity cylindrical part to obtain a global deformation result that matches the actual clamping coordinate system; Step 2.3: Repeat steps 2.1-2.2 to obtain a training set and a validation set of sufficient size to train and validate the recurrent neural network global wall deformation prediction model; Step three also includes the following steps: Step 3.1 The industrial computer outputs the machining trajectory of the outer wall of the weak-rigidity cylindrical part through a digital model, and controls the machining end effector to perform outer surface machining according to the machining path; Step 3.2 The displacement sensor detects the slight deformation displacement change of the weak rigidity cylindrical part and outputs it to the industrial computer through the communication control module; Step 3.3 predicts the real-time deformation of the processing position through the recurrent neural network global wall deformation prediction model; Step 3.4 generates tool cutting compensation according to the predicted deformation and performs machining dynamic compensation.

Images