阅读说明：本技术 具有恒力自适应控制方法的五轴精密小龙门数控加工中心 (Five-axis precise small gantry numerical control machining center with constant-force adaptive control method ) 是由 李爱军 刘欢 李寿超 彭刚 刘国锦 于 2020-06-23 设计创作，主要内容包括：本发明公开了一种具有恒力自适应控制方法的五轴精密小龙门数控加工中心,在切削过程中,检测模块对加工中心的主轴电机电流和实际切削参数不断采样,经过数据处理单元的处理,提供切削力分析数据；恒力控制系统通过实际检测和分析切削过程中切削力的变化,优化进给速度,向数控加工中心的运动控制器发送调节指令；在运动控制器和伺服驱动器的调节下,执行伺服电机调整进给速度,从而使切削力维持在期望的水平。相比传统的小龙门数控加工中心,本发明能够通过对加工过程的实时监控,识别其动态变化特征,自动调节加工参数；根据切削量的变化优化切削进给速度,在保证刀具、机床以及人员安全的基础上,进行恒力平稳切削,实现高精、高效加工作业需求。(The invention discloses a five-axis precise small gantry numerical control machining center with a constant-force self-adaptive control method, wherein in the cutting process, a detection module continuously samples the current of a spindle motor and actual cutting parameters of the machining center and provides cutting force analysis data through the processing of a data processing unit; the constant force control system optimizes the feeding speed by actually detecting and analyzing the change of the cutting force in the cutting process, and sends an adjusting instruction to a motion controller of the numerical control machining center; under the regulation of the motion controller and the servo driver, the servo motor is executed to adjust the feed speed so as to maintain the cutting force at a desired level. Compared with the traditional small gantry numerical control machining center, the numerical control machining center has the advantages that the dynamic change characteristics of the machining process can be identified through real-time monitoring of the machining process, and the machining parameters can be automatically adjusted; the cutting feed speed is optimized according to the change of the cutting amount, the constant-force stable cutting is carried out on the basis of ensuring the safety of a cutter, a machine tool and personnel, and the requirements of high-precision and high-efficiency machining operation are met.)

1. A five-axis precise small gantry numerical control machining center with a constant-force self-adaptive control method is characterized in that the control method of the small gantry numerical control machining center comprises the following steps:

step 101, aiming at different cutting parameters, sampling the main shaft current under the idle running condition of a machine tool to obtain a current signal during idle cutting;

102, filtering a main shaft current signal, removing interference noise and obtaining a trend sequence; thereby obtaining the spindle current characteristic value I under the idle cutting state according to the mean calculation of the trend sequence sample data_q；

Step 103, in the machining process, sampling, filtering and mean value calculating processing are carried out on the spindle current to obtain a spindle current characteristic value I in the material cutting process_p；

104, removing the influence of the characteristics of the machine tool in the real cutting state, and obtaining the cutting force characteristic value I of the current cutting state_rI.e. I_r＝I_p-I_qWherein, the change characteristic of the cutting force is converted into the change characteristic of the cutting force characteristic in the machining process;

105, in the actual cutting process of the small gantry numerical control machining center, when the machining conditions of a cutter, a workpiece and a material are not changed, the cutting force characteristic value I only has a relation with the cutting amount H and the feeding speed w, and a mapping model of the cutting force characteristic value I can be expressed as: designing a calibration experiment to obtain a mapping relation among a cutting force characteristic value I, a cutting amount H and a feeding speed w, and establishing an initial function relation of control variables (I, H, w) according to sample data of the calibration experiment;

and 106, designing the following constant cutting force self-adaptive control rate according to the established system model f (·):

w_ref＝g(I_ref,H_real)

wherein, I_refPreset desired cutting force set point, H_realAnd in the current cutting depth, the feeding speed is adaptively optimized according to the current cutting amount so as to ensure the constant-force stable cutting in the cutting process.

Step 107, according to the current cutting force characteristic value I_realAnd the current feed speed w_realMaking an estimate of the depth of cut, i.e. H_real＝h(I_real,w_real)。

2. The five-axis precision small gantry numerically controlled machining center with constant-force adaptive control method according to claim 1, wherein said step 105 further comprises: and (3) carrying out a series of cutting experiments only changing the cutting amount or only changing the feeding speed under the standard working condition, sampling the main current in the experimental process, extracting the characteristic value, establishing a sample database omega, and carrying out system modeling after collecting enough samples.

3. The five-axis precision small gantry numerically controlled machining center with constant-force adaptive control method according to claim 2, wherein said step 106 further comprises: modeling the Radial Basis Function (RBF) neural network, and establishing the Radial Basis Function (RBF) neural network with two inputs and one output to describe mapping relations h (-) and g (-) with the number of neurons in the middle layer being 5, wherein the Radial Basis Function (RBF) neural network has three layers, namely an input layer, a hidden layer and an output layer, wherein the nonlinear transformation is performed from the input layer to the hidden layer, and the linear transformation is performed from the hidden layer to the output layer.

4. The five-axis precision small gantry numerical control machining center with the constant-force adaptive control method according to claim 3, wherein the activation function of the radial basis function neural network can be expressed as:

the neural network output can be expressed as:

for the mapping g (-), input (x) of the neural network₁,x₂) Is (I)_ref,H_real) The output y is w_ref(ii) a For the mapping h (·), the input (x) of the neural network₁,x₂) Is (I)_real,w_real) The output y is H_real。

5. The five-axis precision small gantry numerical control machining center with the constant-force adaptive control method according to claim 4, wherein a neural network is trained through the sample data of the step 105 to obtain mapping models of a cutting force characteristic value I, a cutting amount H and a feeding speed w at different moments, a dynamic database is constructed based on an immediate learning algorithm, when working conditions of a cutting process change, the immediate learning algorithm can determine a current neighborhood of a controlled parameter through a vector similarity selection criterion, and the neural network model is updated based on local neighborhood data.

6. The five-axis precision small gantry numerically controlled machining center with constant-force adaptive control method according to claim 5, wherein the step of building a dynamic database by the just-in-time learning algorithm further comprises: during the cutting operation of the numerical control machining center, the current control variable (I) is used_ref,H_real) Or (I)_real,w_real) Called query vector, the just-in-time learning algorithm evaluates the similarity between the query vector and the sample vector, and the similarity evaluation function is defined as follows:

wherein x is_cAnd x_iRepresenting a current query vector and a sample vector; arranging the sample vectors in the database according to the similarity, and selecting M sample vectors with higher similarity to form a neighborhood of the query vector, thereby updating the neural network model; and if the similarity between the sample vector and the query vector in the database is maintained at a lower level, adding the query vector into the database as a new sample vector, or else, discarding the new sample vector after updating the neural network model.

7. A constant-force self-adaptive cutting control method based on a small gantry numerical control machining center is characterized by comprising the following steps of:

step 201: detecting the cutting force change characteristic in real time, and detecting the cutting force in a direct detection mode and an indirect detection mode, wherein the direct detection mode is characterized in that a force transducer is arranged at the main shaft part, the indirect detection mode utilizes the corresponding relation between the main shaft current and the cutting force, a motor current signal is acquired in real time and filtered to obtain a characteristic signal of the cutting force, and the change characteristic of the cutting force is indirectly described by the change characteristic of the main shaft current signal;

step 202, constructing a dynamic sample database, constructing the sample database through an initialized calibration experiment and real-time sampling in the cutting process, and providing a basis for establishing a constant force self-adaptive control model and adjusting the feeding speed;

step 203, executing a constant force adaptive control strategy, respectively modeling two mapping functions of an evaluation function h (-) and a feed speed adjusting function g (-) of the current cutting amount by using a cubic Radial Basis Function (RBF) neural network, and executing a servo motor to adjust the feed speed under the regulation of a motion controller and a servo driver according to the adjustment strategy given later, so that the cutting force is maintained at a desired level.

8. The constant-force adaptive cutting control method based on the small gantry numerical control machining center according to claim 7, wherein the step 203 further comprises: the constant-force self-adaptive control system sends an adjusting instruction to a motion controller of the numerical control machining center to adjust the feeding speed at the optimal feeding speed according with the current working condition and maintain the cutting force as the preset constant force.

9. An electronic device, comprising:

a processor; and the number of the first and second groups,

a memory for storing executable instructions of the processor;

wherein the processor is configured to perform the method of constant force adaptive control of the five-axis precision small gantry numerically controlled machining center of any of claims 1-6 via execution of the executable instructions.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the method for constant-force adaptive control of a five-axis precision small gantry numerical control machining center of any one of claims 1 to 6.

Technical Field

The invention relates to the technical field of design of a control system of a numerical control machining center, in particular to a five-axis precise small gantry numerical control machining center with a constant-force self-adaptive control method.

Background

Compared with the traditional vertical machining center, the small gantry numerical control machining center has the advantages of good rigidity, light weight, heavy load, high machining precision, small limitation on the height of a main shaft and the like, is particularly suitable for machining various special-shaped precision parts, gradually replaces large and medium vertical machining centers in many application fields, and has wide application prospects. However, the application of the industries such as aerospace, automobile mold, rail transit and the like also puts higher and higher requirements on the machining precision, machining efficiency, intelligent degree and the like of the small gantry numerical control machining center, and the research on the small gantry numerical control machining center with high performance becomes one of the key problems for improving the equipment manufacturing level.

In the operation process of the small gantry numerical control machining center, the machining allowance always fluctuates and even changes suddenly. If the uniform cutting feed speed is always kept, the fluctuation and the sudden change of the machining allowance can cause great change of the cutting force, thereby influencing the machining precision and aggravating the abrasion of the cutter. In order to improve the stability, the machining precision and the machining efficiency of the machine tool, the numerical control machining center needs to automatically adjust the feeding speed according to the current working condition, so that the cutting force is kept near a constant expected value, and stable machining is realized. Therefore, a simple and effective control strategy needs to be explored to further improve the performance of the small gantry numerical control machining center.

Disclosure of Invention

Aiming at the defects of the prior art, the invention aims to provide a five-axis precise small gantry numerical control machining center with a constant-force self-adaptive control method. Firstly, establishing a mapping relation of 'spindle motor current characteristics-cutting force', extracting current characteristics according to the dynamic change condition of spindle current, and identifying the dynamic change of cutting force; meanwhile, a Radial Basis Function (RBF) neural network based on an instant learning algorithm establishes a mapping model of a cutting force characteristic value and cutting parameters (cutting depth and feeding speed), and the feeding multiplying power of the machine tool is adjusted in a self-adaptive mode according to the current working condition, so that the constant force self-adaptive control of a five-axis precise small gantry numerical control machining center is realized, and the machining precision and the machining efficiency are improved.

The present invention is directed to solving at least one of the problems of the prior art. Therefore, the invention discloses a five-axis precise small gantry numerical control machining center with a constant-force self-adaptive control method, which comprises the following steps:

step 101, aiming at different cutting parameters, sampling the main shaft current under the idle running condition of a machine tool to obtain a current signal during idle cutting;

102, filtering a main shaft current signal, removing interference noise and obtaining a trend sequence; thereby obtaining the spindle current characteristic value I under the idle cutting state according to the mean calculation of the trend sequence sample data_q；

Step 103, in the machining process, sampling, filtering and mean value calculating processing are carried out on the spindle current to obtain a spindle current characteristic value I in the material cutting process_p；

104, removing the influence of the characteristics of the machine tool in the real cutting state, and obtaining the cutting force characteristic value I of the current cutting state_rI.e. I_r＝I_p-I_qWherein, the change characteristic of the cutting force is converted into the change characteristic of the cutting force characteristic in the machining process;

105, in the actual cutting process of the small gantry numerical control machining center, when the machining conditions of a cutter, a workpiece and a material are not changed, the cutting force characteristic value I only has a relation with the cutting amount H and the feeding speed w, and a mapping model of the cutting force characteristic value I can be expressed as: designing a calibration experiment to obtain a mapping relation among a cutting force characteristic value I, a cutting amount H and a feeding speed w, and establishing an initial function relation of control variables (I, H, w) according to sample data of the calibration experiment;

and 106, designing the following constant cutting force self-adaptive control rate according to the established system model f (·):

w_ref＝g(I_ref,H_real)

wherein, I_refPreset desired cutting force set point, H_realAnd in the current cutting depth, the feeding speed is adaptively optimized according to the current cutting amount so as to ensure the constant-force stable cutting in the cutting process.

Step 107, according to the current cutting force characteristic value I_realAnd the current feed speed w_realMaking an estimate of the depth of cut, i.e. H_real＝h(I_real,w_real)。

Still further, the step 105 further comprises: and (3) carrying out a series of cutting experiments only changing the cutting amount or only changing the feeding speed under the standard working condition, sampling the main current in the experimental process, extracting the characteristic value, establishing a sample database omega, and carrying out system modeling after collecting enough samples.

Still further, the step 106 further comprises: modeling the Radial Basis Function (RBF) neural network, and establishing the Radial Basis Function (RBF) neural network with two inputs and one output to describe mapping relations h (-) and g (-) with the number of neurons in the middle layer being 5, wherein the Radial Basis Function (RBF) neural network has three layers, namely an input layer, a hidden layer and an output layer, wherein the nonlinear transformation is performed from the input layer to the hidden layer, and the linear transformation is performed from the hidden layer to the output layer.

Still further, the activation function of the radial basis function neural network may be expressed as:

the neural network output can be expressed as:

for the mapping g (-), input (x) of the neural network₁,x₂) Is (I)_ref,H_real) The output y is w_ref(ii) a For the mapping h (·), the input (x) of the neural network₁,x₂) Is (I)_real,w_real) The output y is H_real。

Further, the neural network is trained through the sample data of the step 105, that is, mapping models of the cutting force characteristic value I, the cutting amount H and the feeding speed w at different moments are obtained, a dynamic database is constructed based on an instant learning algorithm, when the working condition of the cutting process changes, the instant learning algorithm can determine the current neighborhood of the controlled parameter through a vector similarity selection criterion, and the neural network model is updated based on local neighborhood data.

Further, the building of the dynamic database by the just-in-time learning algorithm further comprises: during the cutting operation of the numerical control machining center, the current control variable (I) is used_ref,H_real) Or (I)_real,w_real) Called query vector, the just-in-time learning algorithm evaluates the similarity between the query vector and the sample vector, and the similarity evaluation function is defined as follows:

wherein x is_cAnd x_iRepresenting a current query vector and a sample vector; arranging the sample vectors in the database according to the similarity, and selecting M sample vectors with higher similarity to form a neighborhood of the query vector, thereby updating the neural network model; and if the similarity between the sample vector and the query vector in the database is maintained at a lower level, adding the query vector into the database as a new sample vector, or else, discarding the new sample vector after updating the neural network model.

The invention further discloses a constant-force self-adaptive cutting control method based on the small gantry numerical control machining center, which comprises the following steps of:

step 201: detecting the cutting force change characteristic in real time, and detecting the cutting force in a direct detection mode and an indirect detection mode, wherein the direct detection mode is characterized in that a force transducer is arranged at the main shaft part, the indirect detection mode utilizes the corresponding relation between the main shaft current and the cutting force, a motor current signal is acquired in real time and filtered to obtain a characteristic signal of the cutting force, and the change characteristic of the cutting force is indirectly described by the change characteristic of the main shaft current signal;

step 202, constructing a dynamic sample database, constructing the sample database through an initialized calibration experiment and real-time sampling in the cutting process, and providing a basis for establishing a constant force self-adaptive control model and adjusting the feeding speed;

step 203, executing a constant force adaptive control strategy, respectively modeling two mapping functions of an evaluation function h (-) and a feed speed adjusting function g (-) of the current cutting amount by using a cubic Radial Basis Function (RBF) neural network, and executing a servo motor to adjust the feed speed under the regulation of a motion controller and a servo driver according to the adjustment strategy given later, so that the cutting force is maintained at a desired level.

Still further, the step 203 further comprises: the constant-force self-adaptive control system sends an adjusting instruction to a motion controller of the numerical control machining center to adjust the feeding speed at the optimal feeding speed according with the current working condition and maintain the cutting force as the preset constant force.

The present invention further provides an electronic device comprising:

a processor; and the number of the first and second groups,

a memory for storing executable instructions of the processor;

wherein the processor is configured to execute the above method of constant force adaptive control of a five-axis precision small gantry numerically controlled machining center via execution of the executable instructions.

The invention further discloses a computer readable storage medium on which a computer program is stored, which when executed by a processor implements the method for constant force adaptive control of the five-axis precision small gantry numerical control machining center.

In the invention, the spindle current in the idle cutting and actual cutting processes is sampled in real time, and the processing such as filtering, mean value calculation and the like is carried out to establish the mapping relation between the spindle motor current characteristics and the cutting force; then, analyzing the relation between the cutting force characteristic value I and the cutting amount H and the feeding speed w, and constructing a neural network model of the cutting force characteristic value I; then, an initial database is constructed based on some calibration experiments, and a constant-force self-adaptive control neural network model is initialized; and finally, in a cutting experiment, a dynamic database is constructed based on an instant learning algorithm, a neural network model is updated, and the feeding speed is optimized based on the self-adaptive control rate, so that the machine tool can stably cut with constant cutting force.

In summary, compared with the prior art, the invention has the following beneficial effects:

1. the invention improves a control system of a five-axis precise numerical control machining center, and the designed control system can monitor the cutting process in real time, adaptively adjust machining parameters and realize constant-force stable cutting.

2. The method designs a constant-force self-adaptive control system based on the neural network, establishes a mapping model of a cutting force characteristic value I, a cutting amount H and a feeding speed w according to the good nonlinear system modeling capability of the neural network, can automatically determine the optimal feeding speed according to the cutting amount and the cutting force of the current machine tool in real time, and can improve the production efficiency and the cutting precision on the basis of ensuring the safety of equipment and personnel.

3. The invention constructs the dynamic database of the control system by using the instant learning algorithm, and can rapidly improve the modeling identification precision in real time in a limited sample space by using the local learning capability of the instant learning algorithm, thereby ensuring the control precision and the response performance of the constant-force self-adaptive control system.

Drawings

The invention will be further understood from the following description in conjunction with the accompanying drawings. The components in the figures are not necessarily to scale, emphasis instead being placed upon illustrating the principles of the embodiments. In the drawings, like reference numerals designate corresponding parts throughout the different views.

FIG. 1 is a flow chart of the operation of a constant force adaptive control strategy according to an embodiment of the present invention;

FIG. 2 is a control block diagram of a constant force adaptive control system of an embodiment of the present invention;

FIG. 3 is a schematic diagram of an RBF neural network according to an embodiment of the present invention;

FIG. 4 is a schematic structural view of a numerical control machining center of a small gantry of the present invention.

Detailed Description

The invention is further explained below with reference to the drawings and the examples. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not constitute a limitation on the invention.