Intelligent control method and system of servo pipe bending machine

Abstract

The invention relates to the technical field of machining, and discloses an intelligent control method and system for a servo pipe bender, and the method comprises the steps: collecting an original voltage matrix and a temperature field matrix on a bent pipe corresponding to the servo pipe bender; identifying a real stress field and a stress change rate field of the bent pipe by using the original voltage matrix and the temperature field matrix; the real stress field and the stress change rate field are input into a torque measuring system, torque observed quantity of the servo pipe bending machine is output through the torque measuring system, and the torque measuring system comprises a convolution layer, an observer and a rigid constraint device; based on a servo driver in the servo pipe bending machine, the torque observed quantity is converted into a control signal of the servo pipe bending machine; and intelligent control processing of the servo pipe bending machine is executed through the control signal. According to the invention, the chained defects of basic data acquisition distortion, safety observation misalignment, control signal distortion and safety response lag can be reduced.

Description

Technical Field

[0001] The invention relates to an intelligent control method and system for a servo pipe bender, and belongs to the technical field of mechanical processing. Background Art

[0002] The servo tube bender is a high-precision tube forming device that drives the tube through a servo motor. Its core lies in the use of a numerical control system to perform closed-loop control of parameters such as bending angle, speed, and torque. In the fields of automotive manufacturing, aerospace, etc., this technology must meet two core requirements: the bending angle error must be controlled within ±0.1, and the tube ovality must not exceed 2% to avoid assembly failure. During the plastic deformation process of the tube, the yield risk must be monitored in real time to prevent breakage or permanent deformation.

[0003] At present, although the existing servo pipe bending machines have adopted CNC systems, there are still some technical defects. The traditional strain gauge layout is sparse, usually only 4-8 points, which cannot capture the microscopic strain gradient in the direction of the pipe diameter. The stress positioning error is large, and the temperature sensor is not integrated or a static compensation coefficient is used. As a result, heat-sensitive materials such as aluminum pipes undergo stress drift due to temperature rise during high-speed bending. The torque threshold is manually set, such as empirically setting 70% of the yield strength. The safe torque is not dynamically calculated based on the real-time stress field. The observer weight is adjustable, resulting in the risk of human intervention, and there is no rigid constraint mechanism. The permanent deformation rate of the pipe is high. The use of a universal conversion coefficient leads to a large deviation in the output torque of the driver. PWM modulation introduces current ripple, and the yield strength ratio is calculated based on the motor current rather than direct stress data. These problems result in a long measured response delay.

[0004] Therefore, the current intelligent control system of the servo pipe bending machine has chain defects such as distortion of basic data acquisition, inaccurate safety observation, distortion of control signals, and delayed safety response. Summary of the Invention

[0005] The present invention provides an intelligent control method and system for a servo pipe bender, the main purpose of which is to reduce chain defects such as basic data acquisition distortion, safety observation inaccuracy, control signal distortion, and safety response lag.

[0006] To achieve the above objectives, the present invention provides an intelligent control method for a servo pipe bender, comprising:

[0007] Collect the original voltage matrix and temperature field matrix on the bent pipe corresponding to the servo pipe bending machine;

[0008] Identifying the true stress field and stress change rate field of the bent pipe using the original voltage matrix and the temperature field matrix;

[0009] Inputting the true stress field and the stress change rate field into a torque measurement system to output an observed torque of the servo pipe bender through the torque measurement system, wherein the torque measurement system includes a convolution layer, an observer, and a rigid constraint;

[0010] Based on a servo driver in the servo tube bender, converting the torque observation into a control signal of the servo tube bender;

[0011] Intelligent control processing of the servo pipe bender is performed through the control signal.

[0012] Optionally, the collecting of the original voltage matrix and temperature field matrix on the bent pipe corresponding to the servo pipe bender includes:

[0013] identifying the bent and deformed outer surface of the bent pipe;

[0014] Covering the curved deformed outer surface with a 64-point piezoresistive strain gauge array in a preset grid array;

[0015] collecting the original voltage matrix on the curved deformed outer surface through the 64-point piezoresistive strain gauge array;

[0016] A MEMS temperature sensor is deployed at the center of every four strain gauges of the 64-point piezoresistive strain gauge array;

[0017] The temperature field matrix of the bent pipe is collected by the MEMS temperature sensor.

[0018] Optionally, the identifying the true stress field and stress change rate field of the bent pipe by using the original voltage matrix and the temperature field matrix includes:

[0019] According to the original voltage matrix, the uncompensated stress field of the bent pipe is calculated using the following formula:

[0020]

[0021] Among them, σ temp (x, y, t) represents the uncompensated stress field, x, y represent the spatial coordinates of the strain gauge on the bent pipe, t represents the time variable, V(x, y, t) represents the original voltage value in V in the original voltage matrix, G sens It represents the sensitivity coefficient of the strain gauge on the bent pipe, and E represents the elastic modulus of the material of the bent pipe;

[0022] According to the temperature field matrix, the uncompensated stress field is temperature compensated to obtain the true stress field:

[0023] σ real (x,y,t)=σ temp(x,y,t)-Eα[T(x,y,t)-T ref ]

[0024] Among them, σ real (x, y, t) represents the true stress field, T(x, y, t) represents the temperature value in degrees Celsius in the temperature field matrix, and T ref represents the reference temperature, α represents the thermal expansion coefficient of the material;

[0025] The stress change rate field of the true stress field is calculated using the following formula:

[0026]

[0027] in, represents the stress change rate field.

[0028] Optionally, outputting the torque observation value of the servo pipe bender through the torque measurement system includes:

[0029] Using the convolution layer in the torque measurement system to perform spatiotemporal convolution on the true stress field and the stress change rate field to obtain a convolution feature field;

[0030] Utilizing an observer in the torque measurement system to output an unconstrained torque observation corresponding to the convolution feature field;

[0031] The unconstrained torque observation quantity is rigidly constrained by utilizing a rigid constrainer in the torque measurement system to obtain a torque observation quantity.

[0032] Optionally, the method of performing spatiotemporal convolution on the true stress field and the stress change rate field using the convolution layer in the torque measurement system to obtain a convolution feature field includes:

[0033] In the convolution layer, a spatiotemporal convolution operation is performed between the true stress field and the stress change rate field to obtain a convolution feature field.

[0034] Optionally, the step of utilizing an observer in the torque measurement system to output an unconstrained torque observation corresponding to the convolution feature field includes:

[0035] In the observer, the fixed weight value corresponding to the bent pipe is calculated using the following formula:

[0036]

[0037] Among them, w(x,y) represents the fixed weight value, E represents the elastic modulus of the material of the bent pipe, ∈ y represents the yield strain of the material of the bent tube, x0, y0 represent the neutral axis coordinates of the servo tube bender, and x, y represent the spatial coordinates of the strain gauge on the bent tube;

[0038] According to the fixed weight value, the unconstrained torque observation corresponding to the convolution feature field is output using the following formula:

[0039]

[0040] in, represents the unconstrained torque observation, represents the convolution characteristic field, ΔA represents the physical area of ​​the strain gauge unit on the bent pipe, and t represents the time variable.

[0041] Optionally, the rigidly constraining the unconstrained torque observation using a rigid constraint in the torque measurement system to obtain the torque observation includes:

[0042] In the rigid restrainer, calculating the material yield strength of the bent pipe;

[0043] Based on the yield strength of the material, the unconstrained torque observation is rigidly constrained using the following formula to obtain the torque observation:

[0044]

[0045] Among them, τ out (t) represents the torque observation, represents the unconstrained torque observation, τ yield Represents the yield strength of the material, sign() represents the sign function, and t represents the time variable.

[0046] Optionally, the converting the torque observation into a control signal of the servo tube bender based on a servo driver in the servo tube bender includes:

[0047] Obtaining an inherent conversion coefficient of the servo drive;

[0048] Based on the inherent conversion coefficient, the torque observation is converted into a control signal of the servo pipe bender.

[0049] Optionally, the intelligent control processing of the servo pipe bender is performed by using the control signal, including:

[0050] Decoupling the control signal into a main torque instruction of a main servo motor in the servo tube bender and an auxiliary torque instruction of an auxiliary servo motor in the servo tube bender through a servo driver in the servo tube bender;

[0051] Compensating for gravity torque interference in the main torque instruction to obtain a compensated main instruction;

[0052] Setting the main servo motor to a torque control mode;

[0053] Based on the torque control mode, controlling the main servo motor to execute the compensation main instruction to obtain a main instruction execution result;

[0054] Setting the auxiliary servo motor to a position-torque hybrid mode;

[0055] Based on the position-torque hybrid mode, controlling the auxiliary servo motor to execute the auxiliary torque instruction to obtain an auxiliary instruction execution result;

[0056] Calculating the yield strength ratio of the bent pipe under the execution result of the main instruction and the execution result of the auxiliary instruction;

[0057] When the yield strength ratio is greater than a preset first threshold, starting the speed reduction control of the main servo motor to complete the intelligent control processing process of the servo pipe bender;

[0058] When the yield strength ratio is greater than a preset second threshold, the reverse retraction intervention of the dedicated retraction mechanism in the servo pipe bender is triggered to complete the intelligent control processing process of the servo pipe bender.

[0059] In order to solve the above problems, the present invention also provides an intelligent control system for a servo pipe bender, the system comprising:

[0060] A data collection module is used to collect the original voltage matrix and temperature field matrix on the bent pipe corresponding to the servo pipe bending machine;

[0061] A stress identification module, configured to identify a true stress field and a stress change rate field of the bent pipe using the original voltage matrix and the temperature field matrix;

[0062] an observation output module, configured to input the true stress field and the stress change rate field into a torque measurement system, so as to output the torque observation of the servo tube bender through the torque measurement system, wherein the torque measurement system includes a convolution layer, an observer, and a rigid constraint;

[0063] An observation quantity conversion module, configured to convert the torque observation quantity into a control signal of the servo pipe bender based on a servo driver in the servo pipe bender;

[0064] An intelligent control module is used to perform intelligent control processing of the servo pipe bender through the control signal.

[0065] Compared with the problems described in the background technology, the embodiment of the present invention realizes high-resolution synchronous sampling of stress and temperature on the surface of the pipe through the coordinated deployment of a 64-point piezoresistive strain gauge array and a MEMS temperature sensor. The strain gauge spacing is no more than one-tenth of the pipe diameter, which can capture microscopic strain gradients, thereby improving the stress positioning accuracy. The MEMS sensor at the center of every 4 strain gauges eliminates thermal gradient errors, and the effectiveness of temperature compensation will also be improved, ultimately providing distortion-free basic data for subsequent stress field calculations and laying the foundation for full-link accuracy. The embodiment of the present invention realizes self-consistent stress field reconstruction of physical dimensions by adopting dynamic temperature compensation of thermal expansion coefficient and calculation of stress change rate field, so that the stress error after compensation is reduced, which is especially suitable for heat-sensitive materials such as stainless steel / aluminum. On the other hand, the stress change rate field reflects the plastic deformation trend of the bent pipe in real time, and the deformation warning speed can also be obtained. Finally, by constructing a time-varying physical field model, input data that conforms to material mechanics is provided for subsequent torque observations. The embodiment of the present invention uses the spatiotemporal feature fusion of the convolutional layer, the fixed weight function of the observer, and the material bending of the rigid constraint. Three-level processing of yield strength generates safe and physically interpretable torque observables. The convolutional layer fuses the stress field and the rate-of-change field, improving the signal-to-noise ratio of key features. Fixed weights mitigate the risk of human intervention, enhancing observation stability, and limiting torque to no more than 80% of yield strength, reducing the permanent deformation rate of the pipe. Ultimately, the output is a dimensionally autonomous torque observable with clear safety boundaries. This embodiment of the present invention establishes a distortion-free control command transmission chain through linear conversion based on the servo drive's inherent conversion coefficients. The inherent conversion coefficients are adaptable to any drive, ensuring a unified control baseline. The linear conversion avoids PWM modulation distortion, reduces current command errors, and ultimately ensures shorter control signal execution delays. This embodiment of the present invention achieves closed-loop control for safe plastic forming through dual-motor coordination in a main motor torque mode and an auxiliary motor position-torque hybrid mode, gravity torque compensation, yield strength ratio monitoring, and dynamic intervention. Main motor gravity compensation eliminates posture errors, improving angular accuracy. Auxiliary motor position constraints reduce pipe ovality. The closed-loop safety monitoring of the yield strength ratio reduces pipe fracture rates, thereby achieving stability across operating conditions. Therefore, the intelligent control method and system of the servo pipe bender provided by the embodiment of the present invention can reduce the chain defects of basic data acquisition distortion, safety observation inaccuracy, control signal distortion, and safety response lag. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 A schematic flow chart of an intelligent control method for a servo pipe bender provided by one embodiment of the present invention;

[0067] Figure 2 A schematic flow chart of the material yield strength control method of the servo pipe bender provided in one embodiment of the present invention;

[0068] Figure 3 A schematic diagram of a module for implementing the intelligent control system of the servo pipe bender provided in one embodiment of the present invention.

[0069] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0070] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0071] The present invention provides an intelligent control method for a servo tube bender. The method can be executed by at least one of a server, a terminal, or other electronic device capable of executing the method provided by the present invention. In other words, the method can be executed by software or hardware installed on a terminal or server. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster.

[0072] Example 1:

[0073] Reference Figure 1 FIG. 1 is a flow chart of an intelligent control method for a servo pipe bender according to an embodiment of the present invention. In this embodiment, the intelligent control method for a servo pipe bender includes:

[0074] S1. Collect the original voltage matrix and temperature field matrix on the bent pipe corresponding to the servo pipe bending machine.

[0075] The embodiment of the present invention achieves high-resolution synchronous sampling of pipe surface stress and temperature through the coordinated deployment of a 64-point piezoresistive strain gauge array and a MEMS temperature sensor. The strain gauge spacing is no more than one-tenth of the pipe diameter, which can capture microscopic strain gradients, thereby improving stress positioning accuracy. The MEMS sensor at the center of every four strain gauges eliminates thermal gradient errors, and the effectiveness of temperature compensation is also improved. Ultimately, it provides distortion-free basic data for subsequent stress field calculations, laying a foundation for full-link accuracy.

[0076] In one embodiment of the present invention, the collection of the original voltage matrix and temperature field matrix on the bent tube corresponding to the servo tube bender includes: identifying the bent and deformed outer surface of the bent tube; covering the bent and deformed outer surface with a 64-point piezoresistive strain gauge array in a preset grid array; collecting the original voltage matrix on the bent and deformed outer surface through the 64-point piezoresistive strain gauge array; deploying a MEMS temperature sensor at the center of every four strain gauges in the 64-point piezoresistive strain gauge array; and collecting the temperature field matrix of the bent tube through the MEMS temperature sensor.

[0077] Among them, the servo tube bender is a tube bending equipment powered by a servo motor. It can achieve high-precision tube bending operations through precise servo motor control. The bent tube is the object processed by the servo tube bender and is a tubular material that needs to be bent. The bent deformed outer surface refers to the outer surface area of ​​the tube that undergoes plastic deformation during the bending process, such as the outer arc surface of the bent section of a stainless steel tube. The 64-point piezoresistive strain gauge array refers to a metal foil strain gauge arranged in an 8×8 grid (preset grid array), each piece is 5mm×3mm in size, covering the bending area of ​​the tube. The MEMS temperature sensor refers to a micro-electromechanical system temperature sensor. The original voltage matrix refers to the matrix composed of the original voltage collected by each piezoresistive strain gauge. The temperature field matrix refers to the matrix composed of the temperatures at different coordinates on the bent deformed outer surface collected by each MEMS temperature sensor.

[0078] S2. Identify the true stress field and stress change rate field of the bent pipe using the original voltage matrix and the temperature field matrix.

[0079] The embodiment of the present invention realizes self-consistent stress field reconstruction of physical dimensions by adopting dynamic temperature compensation of thermal expansion coefficient and calculation of stress change rate field, so that the stress error after compensation is reduced. It is particularly suitable for heat-sensitive materials such as stainless steel and aluminum. On the other hand, the stress change rate field reflects the plastic deformation trend of the bent pipe in real time, and the deformation warning speed can also be obtained. Finally, by constructing a time-varying physical field model, input data that conforms to material mechanics is provided for subsequent torque observation.

[0080] In one embodiment of the present invention, identifying the true stress field and stress change rate field of the bent pipe using the original voltage matrix and the temperature field matrix includes: calculating the uncompensated stress field of the bent pipe using the following formula based on the original voltage matrix:

[0081]

[0082] Among them, σ temp (x, y, t) represents the uncompensated stress field, x, y represent the spatial coordinates of the strain gauge on the bent pipe, t represents the time variable, V(x, y, t) represents the original voltage value in V in the original voltage matrix, G sens It represents the sensitivity coefficient of the strain gauge on the bent pipe, and E represents the elastic modulus of the material of the bent pipe;

[0083] According to the temperature field matrix, the uncompensated stress field is temperature compensated to obtain the true stress field:

[0084] σ real (x,y,t)=σ temp(x,y,t)-Eα[T(x,y,t)-T ref ]

[0085] Among them, σ real (x, y, t) represents the true stress field, T(x, y, t) represents the temperature value in degrees Celsius in the temperature field matrix, and T ref represents the reference temperature, α represents the thermal expansion coefficient of the material;

[0086] The stress change rate field of the true stress field is calculated using the following formula:

[0087]

[0088] in, represents the stress change rate field.

[0089] The strain gauge sensitivity coefficient refers to the voltage change caused by each unit microstrain, for example, 1V / με means that 1 microstrain produces a 1 volt voltage change. The thermal expansion coefficient of the material is derived from the material-related manual, for example, the α of stainless steel is 17.3×10 -6 / ℃, it should be noted that με, as the unit of strain, is a dimensionless quantity. It is the unit of strain, representing one millionth of strain, that is, 1με=10 -6 strain.

[0090] S3. Input the true stress field and the stress change rate field into a torque measurement system to output the torque observation value of the servo pipe bender through the torque measurement system, wherein the torque measurement system includes a convolution layer, an observer and a rigid constraint.

[0091] The embodiment of the present invention generates a safe and physically interpretable torque observation through three-level processing of spatiotemporal feature fusion of the convolution layer, fixed weight function of the observer, and material yield strength of the rigid constraint. The convolution layer fuses the stress field and the rate of change field to improve the signal-to-noise ratio of key features. The fixed weight avoids the risk of human intervention, improves the observation stability, limits the torque to no more than 80% of the yield strength, reduces the permanent deformation rate of the pipe, and ultimately outputs a torque observation with autonomous dimension and clear safety boundaries.

[0092] In one embodiment of the present invention, the outputting of the torque observation quantity of the servo pipe bender through the torque measurement system includes: using the convolution layer in the torque measurement system to perform spatiotemporal convolution on the real stress field and the stress change rate field to obtain a convolution feature field; using the observer in the torque measurement system to output the unconstrained torque observation quantity corresponding to the convolution feature field; and using the rigid constraint in the torque measurement system to rigidly constrain the unconstrained torque observation quantity to obtain the torque observation quantity.

[0093] In another embodiment of the present invention, the method of using the convolution layer in the torque measurement system to perform spatiotemporal convolution on the true stress field and the stress change rate field to obtain the convolution feature field includes: in the convolution layer, using the following formula to perform spatiotemporal convolution operation between the true stress field and the stress change rate field to obtain the convolution feature field:

[0094]

[0095] in, represents the convolution feature field, σ real (x,y,t) represents the true stress field, represents the stress change rate field, t represents the time variable, and * represents the spatiotemporal convolution operator.

[0096] In another embodiment of the present invention, the step of outputting the unconstrained torque observation corresponding to the convolution feature field using an observer in the torque measurement system includes calculating, in the observer, a fixed weight value corresponding to the bent pipe using the following formula:

[0097]

[0098] Among them, w(x,y) represents the fixed weight value, E represents the elastic modulus of the material of the bent pipe, ∈ y represents the yield strain of the material of the bent tube, x0, y0 represent the neutral axis coordinates of the servo tube bender, and x, y represent the spatial coordinates of the strain gauge on the bent tube;

[0099] According to the fixed weight value, the unconstrained torque observation corresponding to the convolution feature field is output using the following formula:

[0100]

[0101] in, represents the unconstrained torque observation, represents the convolution characteristic field, ΔA represents the physical area of ​​the strain gauge unit on the bent pipe, and t represents the time variable.

[0102] The material yield strain refers to the critical strain value at which the pipe begins to undergo permanent plastic deformation, and the material yield strain is a dimensionless value.

[0103] In another embodiment of the present invention, the method of rigidly constraining the unconstrained torque observation value using a rigid constraint in the torque measurement system to obtain the torque observation value includes: calculating the material yield strength of the bent pipe in the rigid constraint; and based on the material yield strength, rigidly constraining the unconstrained torque observation value using the following formula to obtain the torque observation value:

[0104]

[0105] Among them, τ out (t) represents the torque observation, represents the unconstrained torque observation, τ yield Represents the yield strength of the material, sign() represents the sign function, and t represents the time variable.

[0106] Among them, the calculation formula of the material yield strength is τ yield =σ y S, σ y Indicates the yield strength of the material (the limit value determined by the tensile test, which is related to τ yield Different, σ y The unit is Pa), S represents the section modulus, τ yield The unit is N·m, σ y It refers to the stress value corresponding to 0.2% plastic deformation in the tensile test. The calculation formula of section modulus is: D represents the outer diameter of the pipe, and d represents the inner flow hole diameter of the pipe.

[0107] See Figure 2 The figure shows a flow chart of the material yield strength of the intelligent control method of the servo pipe bender provided by one embodiment of the present invention. Figure 2 The limit value is σ y , section modulus is S, yield strength is τ yield .

[0108] S4. Based on a servo driver in the servo pipe bender, convert the torque observation into a control signal of the servo pipe bender.

[0109] The embodiment of the present invention establishes a distortion-free control instruction transmission chain through linear conversion based on the inherent conversion coefficient of the servo driver. The inherent conversion coefficient is adapted to any driver to ensure the uniformity of the control baseline. The linear conversion avoids PWM modulation distortion, reduces the current instruction error, and ultimately ensures that the control signal execution delay is shortened.

[0110] In one embodiment of the present invention, the step of converting the torque observation amount into a control signal of the servo pipe bender based on a servo driver in the servo pipe bender includes: obtaining an inherent conversion coefficient of the servo driver; and converting the torque observation amount into the control signal of the servo pipe bender using the following formula based on the inherent conversion coefficient:

[0111] u(t)=K drive τ out (t)

[0112] Where u(t) represents the control signal, τ out(t) represents the torque observation, K drive represents the inherent conversion coefficient, and t represents the time variable.

[0113] The inherent conversion coefficient is the driver current-torque conversion constant, and the calculation formula is: The inherent conversion coefficient mainly comes from the driver-related manual. The unit of the inherent conversion coefficient is N·m / A, τ rated Indicates the rated torque of the servo motor, I rated Indicates the rated current of the driver.

[0114] S5. Execute intelligent control processing of the servo pipe bender through the control signal.

[0115] The embodiment of the present invention realizes closed-loop control of safe plastic forming through dual-motor collaboration of the main motor torque mode and the auxiliary motor position-torque hybrid mode, gravity torque compensation, yield strength ratio monitoring and dynamic intervention. The gravity compensation of the main motor can eliminate posture errors and improve angle accuracy. The position constraint of the auxiliary motor reduces the ovality of the pipe. The safety monitoring closed loop of the yield strength ratio can reduce the pipe fracture rate, thereby achieving stability across working conditions.

[0116] In one embodiment of the present invention, the intelligent control processing of the servo pipe bender is performed by the control signal, including: decoupling the control signal into a main torque instruction of the main servo motor in the servo pipe bender and an auxiliary torque instruction of the auxiliary servo motor in the servo pipe bender through the servo driver in the servo pipe bender; compensating for gravity torque interference in the main torque instruction to obtain a compensated main instruction; setting the main servo motor to a torque control mode; based on the torque control mode, controlling the main servo motor to execute the compensated main instruction to obtain a main instruction execution result; setting the auxiliary servo motor to a torque control mode; Position-torque hybrid mode; based on the position-torque hybrid mode, controlling the auxiliary servo motor to execute the auxiliary torque instruction to obtain the auxiliary instruction execution result; calculating the yield strength ratio of the bent pipe under the main instruction execution result and the auxiliary instruction execution result; when the yield strength ratio is greater than a preset first threshold, starting the speed reduction control of the main servo motor to complete the intelligent control processing process of the servo pipe bender; when the yield strength ratio is greater than a preset second threshold, triggering the reverse retraction intervention of the special retraction mechanism in the servo pipe bender to complete the intelligent control processing process of the servo pipe bender.

[0117] Optionally, the decoupling of the control signal into a main torque instruction of a main servo motor in the servo tube bender and an auxiliary torque instruction of an auxiliary servo motor in the servo tube bender is achieved by using the following formula:

[0118]

[0119] τ 主 =H m (s)u(s),τ 辅 =H a (s)u(s)

[0120] Where u(s) is the Laplace transform of u(t), τ 主 is the main torque command, τ 辅 It is auxiliary torque command;

[0121] Please refer to Table 1 below for details:

[0122]

[0123]

[0124] Furthermore, the process of compensating the gravity torque interference in the main torque instruction to obtain the compensated main instruction is achieved by using the following formula:

[0125] τ′ 主 =τ 主 +mgRcosθ

[0126] Please refer to Table 2 below for details:

[0127]

[0128]

[0129] Furthermore, the main servo motor is set to a torque control mode: the torque control mode is to take the motor output shaft torque as the direct control target, and the driver is driven according to the command value τ cmd Adjustable current I q , ignoring the position / velocity error mode, based on the following formula:

[0130]

[0131] Among them, K t is the motor torque constant;

[0132] Furthermore, based on the torque control mode, the main servo motor is controlled to execute the compensation main instruction, and the main instruction execution result is obtained as follows: after the main servo motor is set to the torque control mode, the main servo motor is controlled to execute the compensation main instruction. The main instruction execution result obtained is the result of the pipe being controlled by the main servo motor. Further, the auxiliary servo motor is set to the position-torque hybrid mode as follows: the position-torque hybrid mode refers to receiving the position instruction θ at the same time. cmd and torque limit τ lim, which prioritizes ensuring that the motor moves to the target position but does not exceed the set upper limit on the output torque. The formula is as follows:

[0133]

[0134] Please refer to Table 3 below for details:

[0135]

[0136]

[0137] Furthermore, the yield strength ratio refers to the value measured after the pipe is subjected to the compensation main instruction and the auxiliary torque instruction. The calculation formula of the yield strength ratio is: σ actual is the real-time actual stress value on the surface of the pipe bending deformation zone, σ actual The unit is Pa (N / m 2 ), further, the process of controlling the auxiliary servo motor to execute the auxiliary torque instruction based on the position-torque hybrid mode and obtaining the auxiliary instruction execution result is similar to the process of controlling the main servo motor to execute the compensation main instruction based on the torque control mode and obtaining the main instruction execution result, which will not be repeated here. Further, when the yield strength ratio is greater than the preset first threshold, the speed reduction control of the main servo motor is started to complete the intelligent control processing process of the servo pipe bender; when the yield strength ratio is greater than the preset second threshold, the reverse retraction intervention of the special retraction mechanism in the servo pipe bender is triggered to complete the intelligent control processing process of the servo pipe bender: when η≥0.85, the speed reduction control is started, and when η≥0.95, the reverse retraction intervention is triggered, wherein the speed reduction control refers to the smooth reduction of the speed to a safe stop, and the reverse retraction intervention refers to the active retraction of the positioning component during the pipe pushing or bending process to prevent the front end of the pipe from being deformed due to collision or insufficient pushing resulting in length and short legs defects.

[0138] Compared with the problems described in the background technology, the embodiment of the present invention realizes high-resolution synchronous sampling of stress and temperature on the surface of the pipe through the coordinated deployment of a 64-point piezoresistive strain gauge array and a MEMS temperature sensor. The strain gauge spacing is no more than one-tenth of the pipe diameter, which can capture microscopic strain gradients, thereby improving the stress positioning accuracy. The MEMS sensor at the center of every 4 strain gauges eliminates thermal gradient errors, and the effectiveness of temperature compensation will also be improved, ultimately providing distortion-free basic data for subsequent stress field calculations and laying the foundation for full-link accuracy. The embodiment of the present invention realizes self-consistent stress field reconstruction of physical dimensions by adopting dynamic temperature compensation of thermal expansion coefficient and calculation of stress change rate field, so that the stress error after compensation is reduced, which is especially suitable for heat-sensitive materials such as stainless steel / aluminum. On the other hand, the stress change rate field reflects the plastic deformation trend of the bent pipe in real time, and the deformation warning speed can also be obtained. Finally, by constructing a time-varying physical field model, input data that conforms to material mechanics is provided for subsequent torque observations. The embodiment of the present invention uses the spatiotemporal feature fusion of the convolutional layer, the fixed weight function of the observer, and the material bending of the rigid constraint. Three-level processing of yield strength generates safe and physically interpretable torque observables. The convolutional layer fuses the stress field and the rate-of-change field, improving the signal-to-noise ratio of key features. Fixed weights mitigate the risk of human intervention, enhancing observation stability, and limiting torque to no more than 80% of yield strength, reducing the permanent deformation rate of the pipe. Ultimately, the output is a dimensionally autonomous torque observable with clear safety boundaries. This embodiment of the present invention establishes a distortion-free control command transmission chain through linear conversion based on the servo drive's inherent conversion coefficients. The inherent conversion coefficients are adaptable to any drive, ensuring a unified control baseline. The linear conversion avoids PWM modulation distortion, reduces current command errors, and ultimately ensures shorter control signal execution delays. This embodiment of the present invention achieves closed-loop control for safe plastic forming through dual-motor coordination in a main motor torque mode and an auxiliary motor position-torque hybrid mode, gravity torque compensation, yield strength ratio monitoring, and dynamic intervention. Main motor gravity compensation eliminates posture errors, improving angular accuracy. Auxiliary motor position constraints reduce pipe ovality. The closed-loop safety monitoring of the yield strength ratio reduces pipe fracture rates, thereby achieving stability across operating conditions. Therefore, the intelligent control method and system of the servo pipe bender provided by the embodiment of the present invention can reduce the chain defects of basic data acquisition distortion, safety observation inaccuracy, control signal distortion, and safety response lag.

[0139] Example 2:

[0140] like Figure 3 FIG. 1 is a functional module diagram of an intelligent control system of a servo pipe bending machine according to the present invention.

[0141] The intelligent control system 300 for a servo tube bender described in the present invention can be installed in an electronic device. Depending on the functionality implemented, the intelligent control system can include a data collection module 301, a stress identification module 302, an observation output module 303, an observation conversion module 304, and an intelligent control module 305. A module, also referred to as a unit, is a series of computer program segments that can be executed by an electronic device processor and perform a fixed function. These modules are stored in the electronic device's memory.

[0142] In the embodiment of the present invention, the functions of each module / unit are as follows:

[0143] The data collection module 301 is used to collect the original voltage matrix and temperature field matrix of the bent pipe corresponding to the servo pipe bender;

[0144] The stress identification module 302 is configured to identify the true stress field and stress change rate field of the bent pipe using the original voltage matrix and the temperature field matrix;

[0145] The observation output module 303 is used to input the true stress field and the stress change rate field into a torque measurement system to output the torque observation of the servo pipe bender through the torque measurement system, wherein the torque measurement system includes a convolution layer, an observer, and a rigid constraint;

[0146] The observed value conversion module 304 is used to convert the torque observed value into a control signal of the servo pipe bender based on the servo driver in the servo pipe bender;

[0147] The intelligent control module 305 is used to perform intelligent control processing of the servo pipe bender through the control signal.

[0148] In detail, the modules in the intelligent control system 300 of the servo pipe bender in the embodiment of the present invention are used in the same manner as above. Figure 1 The intelligent control method of the servo pipe bender described in the invention has the same technical means and can produce the same technical effects, so I will not go into details here.

[0149] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0150] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An intelligent control method for a servo pipe bender, characterized in that: The method comprises: Collect the original voltage matrix and temperature field matrix on the bent pipe corresponding to the servo pipe bending machine; Identifying the true stress field and stress change rate field of the bent pipe using the original voltage matrix and the temperature field matrix; Inputting the true stress field and the stress change rate field into a torque measurement system to output an observed torque of the servo pipe bender through the torque measurement system, wherein the torque measurement system includes a convolution layer, an observer, and a rigid constraint; Based on a servo driver in the servo tube bender, converting the torque observation into a control signal of the servo tube bender; Intelligent control processing of the servo pipe bender is performed through the control signal.

2. The intelligent control method of the servo pipe bender according to claim 1, characterized in that: The collecting of the original voltage matrix and temperature field matrix on the bent pipe corresponding to the servo pipe bender includes: identifying the bent and deformed outer surface of the bent pipe; Covering the curved deformed outer surface with a 64-point piezoresistive strain gauge array in a preset grid array; collecting the original voltage matrix on the curved deformed outer surface through the 64-point piezoresistive strain gauge array; A MEMS temperature sensor is deployed at the center of every four strain gauges of the 64-point piezoresistive strain gauge array; The temperature field matrix of the bent pipe is collected by the MEMS temperature sensor.

3. The intelligent control method of the servo pipe bender according to claim 1, characterized in that: The identifying the true stress field and stress change rate field of the bent pipe by using the original voltage matrix and the temperature field matrix includes: According to the original voltage matrix, the uncompensated stress field of the bent pipe is calculated using the following formula: Among them, σ temp (x, y, t) represents the uncompensated stress field, x, y represent the spatial coordinates of the strain gauge on the bent pipe, t represents the time variable, V(x, y, t) represents the original voltage value in V in the original voltage matrix, G sens It represents the sensitivity coefficient of the strain gauge on the bent pipe, and E represents the elastic modulus of the material of the bent pipe; According to the temperature field matrix, the uncompensated stress field is temperature compensated to obtain the true stress field: σ real (x,y,t)=σ temp (x,y,t)-Eα[T(x,y,t)-T ref ] Among them, σ real (x, y, t) represents the true stress field, T(x, y, t) represents the temperature value in degrees Celsius in the temperature field matrix, and T ref represents the reference temperature, α represents the thermal expansion coefficient of the material; The stress change rate field of the true stress field is calculated using the following formula: in, represents the stress change rate field.

4. The intelligent control method of a servo pipe bender according to claim 1, characterized in that: Outputting the torque observation value of the servo pipe bender through the torque measurement system includes: Using the convolution layer in the torque measurement system to perform spatiotemporal convolution on the true stress field and the stress change rate field to obtain a convolution feature field; Utilizing an observer in the torque measurement system to output an unconstrained torque observation corresponding to the convolution feature field; The unconstrained torque observation quantity is rigidly constrained by utilizing a rigid constrainer in the torque measurement system to obtain a torque observation quantity.

5. The intelligent control method of the servo pipe bender according to claim 4, characterized in that: The method of using the convolution layer in the torque measurement system to perform spatiotemporal convolution on the true stress field and the stress change rate field to obtain a convolution feature field includes: In the convolution layer, a spatiotemporal convolution operation is performed between the true stress field and the stress change rate field to obtain a convolution feature field.

6. The intelligent control method of a servo pipe bender according to claim 4, characterized in that: The step of utilizing the observer in the torque measurement system to output the unconstrained torque observation corresponding to the convolution feature field comprises: In the observer, the fixed weight value corresponding to the bent pipe is calculated using the following formula: Among them, w(x,y) represents the fixed weight value, E represents the elastic modulus of the material of the bent pipe, ∈ y represents the yield strain of the material of the bent tube, x0, y0 represent the neutral axis coordinates of the servo tube bender, and x, y represent the spatial coordinates of the strain gauge on the bent tube; According to the fixed weight value, the unconstrained torque observation corresponding to the convolution feature field is output using the following formula: in, represents the unconstrained torque observation, represents the convolution characteristic field, ΔA represents the physical area of ​​the strain gauge unit on the bent pipe, and t represents the time variable.

7. The intelligent control method of a servo pipe bender according to claim 4, characterized in that: The method of rigidly constraining the unconstrained torque observation by using the rigid constraint in the torque measurement system to obtain the torque observation comprises: In the rigid restrainer, calculating the material yield strength of the bent pipe; Based on the yield strength of the material, the unconstrained torque observation is rigidly constrained using the following formula to obtain the torque observation: Among them, τ out (t) represents the torque observation, represents the unconstrained torque observation, τ yield Represents the yield strength of the material, sign() represents the sign function, and t represents the time variable.

8. The intelligent control method of a servo pipe bender according to claim 1, characterized in that: The method of converting the torque observation quantity into a control signal of the servo pipe bender based on a servo driver in the servo pipe bender includes: Obtaining an inherent conversion coefficient of the servo drive; Based on the inherent conversion coefficient, the torque observation is converted into a control signal of the servo pipe bender.

9. The intelligent control method of a servo pipe bender according to claim 1, characterized in that: The intelligent control processing of the servo pipe bender is performed by the control signal, including: Decoupling the control signal into a main torque instruction of a main servo motor in the servo tube bender and an auxiliary torque instruction of an auxiliary servo motor in the servo tube bender through a servo driver in the servo tube bender; Compensating for gravity torque interference in the main torque instruction to obtain a compensated main instruction; Setting the main servo motor to a torque control mode; Based on the torque control mode, controlling the main servo motor to execute the compensation main instruction to obtain a main instruction execution result; Setting the auxiliary servo motor to a position-torque hybrid mode; Based on the position-torque hybrid mode, controlling the auxiliary servo motor to execute the auxiliary torque instruction to obtain an auxiliary instruction execution result; Calculating the yield strength ratio of the bent pipe under the execution result of the main instruction and the execution result of the auxiliary instruction; When the yield strength ratio is greater than a preset first threshold, starting the speed reduction control of the main servo motor to complete the intelligent control processing process of the servo pipe bender; When the yield strength ratio is greater than a preset second threshold, the reverse retraction intervention of the dedicated retraction mechanism in the servo pipe bender is triggered to complete the intelligent control processing process of the servo pipe bender.

10. An intelligent control system for a servo pipe bender, characterized in that: The system comprises: A data collection module is used to collect the original voltage matrix and temperature field matrix on the bent pipe corresponding to the servo pipe bending machine; A stress identification module, configured to identify a true stress field and a stress change rate field of the bent pipe using the original voltage matrix and the temperature field matrix; an observation output module, configured to input the true stress field and the stress change rate field into a torque measurement system, so as to output the torque observation of the servo tube bender through the torque measurement system, wherein the torque measurement system includes a convolution layer, an observer, and a rigid constraint; An observation quantity conversion module, configured to convert the torque observation quantity into a control signal of the servo pipe bender based on a servo driver in the servo pipe bender; An intelligent control module is used to perform intelligent control processing of the servo pipe bender through the control signal.

Images