Design method of servo spot welding clamp controller and controller

Abstract

The invention discloses a design method of a servo spot welding clamp controller and the controller, and relates to the technical field of resistance spot welding, and the design method comprises the following steps: constructing a permanent magnet synchronous motor mathematical model; taking a control target that the permanent magnet synchronous motor can quickly track the given speed of an admittance controller under an unknown interference condition, and establishing a state equation of the permanent magnet synchronous motor by adopting integral sliding mode control; wherein the state equation of the permanent magnet synchronous motor comprises a system friction item, and an RBF neural network is adopted to carry out online self-adaptive compensation of system friction; calculating the output quantity of the controller based on the permanent magnet synchronous motor state equation and the approaching rate function; wherein the approach rate function adopts an improved exponential approach rate. According to the method, sliding mode control and the RBF neural network are combined, the influence of friction and the like on the welding quality is effectively reduced, and point welding pliers can be controlled more accurately to conduct spot welding work.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of resistance spot welding, in particular to a design method of a servo spot welding clamp controller and the controller. BACKGROUND

[0002] Resistance spot welding is a key process for vehicle body assembly, which can efficiently connect chassis, doors, body frames and other components to ensure the strength and safety of the vehicle body structure. The working principle of resistance spot welding is to use the resistance heat generated by the current passing through the contact points of the welding parts to heat the metal at the contact points to a molten state to form a welding spot, thereby realizing the connection of the welding parts. The resistance spot welding process is roughly divided into four stages: a, pre-pressing stage - place the welding parts between the electrodes and apply a certain pressure to make the welding parts tightly contact to ensure that the current can pass uniformly; b, welding stage - turn on the welding current, the current is conducted to the welding part contact point through the electrode, and since the resistance at the contact point is relatively large, a large amount of heat will be generated when the current passes through this place, which makes the metal at the contact point rapidly heated to molten state to form a liquid nucleus; c, pressure maintaining stage - after the welding current is turned off, the electrode continues to maintain the pressure for a period of time, so that the liquid nucleus cools and solidifies under the action of pressure to form a solid welding spot; d, releasing stage - release the electrode and take out the welding parts to complete a resistance spot welding process.

[0003] As a control method with strong robustness, sliding mode control has been applied in servo system control to a certain extent. However, there are still some problems in existing sliding mode control methods. Chattering is an inevitable phenomenon in sliding mode control, which will make the movement of the spot welding clamp unstable and affect the application of the spot welding pressure, resulting in uneven welding spot quality. At the same time, in dealing with system friction disturbance, the existing sliding mode control method usually uses a fixed gain to offset the influence of friction. However, due to the nonlinearity and time-varying nature of friction, the fixed gain cannot achieve good compensation effect in different working conditions, and it is difficult to effectively reduce the disturbance of friction on system performance. SUMMARY

[0004] In view of the deficiencies of the prior art, the purpose of the present application is to provide a design method of a servo spot welding clamp controller and the controller, which combines sliding mode control with RBF neural network to effectively reduce the influence of friction and other factors on welding quality and more accurately control the spot welding clamp for spot welding work.

[0005] In order to achieve the above-mentioned purpose, the present application is realized by the following technical scheme: In a first aspect, the embodiments of the present application provide a design method of a servo spot welding clamp controller, comprising: constructing a permanent magnet synchronous motor mathematical model; The permanent magnet synchronous motor can quickly track the admittance controller given speed under the condition of unknown disturbance as a control target, integral sliding mode control is adopted, and the permanent magnet synchronous motor state equation is established; wherein the permanent magnet synchronous motor state equation contains system friction term, and RBF neural network is adopted for online adaptive compensation of system friction. The controller output is calculated based on the permanent magnet synchronous motor state equation and the approach rate function; wherein the approach rate function adopts an improved exponential approach rate.

[0006] As a further implementation manner, the controller output is represented as: ; Wherein, x 2 is a system state variable, is a system friction term; the function , the coefficient term , the disturbance term ; is calculated by , is given by the admittance controller, is the bearing viscous coefficient of the permanent magnet synchronous motor, is the moment of inertia of the permanent magnet synchronous motor, is the mechanical angular velocity, is the flux of the rotor permanent magnet, is the spot welding torque, is the gravity torque.

[0007] As a further implementation manner, the gravity torque satisfies: ; By placing the spot welding clamp in the vertical direction, the gravity torque is calculated according to the torque difference of the permanent magnet synchronous motor when the electrode holder moves uniformly in the positive and negative directions .

[0008] As a further implementation manner, the RBF neural network is used for approximation of the system friction term : ; Wherein, is the ideal weight value of the RBF neural network, is the approximation error thereof, h is the output of the hidden layer; The network input is the mechanical angular velocity of the permanent magnet synchronous motor, and the actual output of the network is represented as: ; wherein, is the estimated weight value of the network, and T represents the sampling time.

[0009] As a further implementation, a Lyapunov function is defined, and derivation is performed to obtain an adaptive rate: ; wherein, .

[0010] As a further implementation, an integral type sliding mode control is adopted, and system state variables are selected: ; wherein, .

[0011] As a further implementation, the permanent magnet synchronous motor state equation is: .

[0012] As a further implementation, the improved exponential approach rate is expressed as: ; Parameters and are expressed as: ; wherein parameters .

[0013] As a further implementation, when a permanent magnet synchronous motor mathematical model is constructed, saturation of a stator and rotor core is ignored, it is determined that a magnetic circuit is linear, and direct and quadrature axis inductance parameters are constant; core and winding eddy current loss and hysteresis loss are not taken into account; a rotor has no damping winding; a magnetic field generated by the rotor and three-phase winding is distributed in a sinusoidal rule in an air gap.

[0014] In a second aspect, an embodiment of the present application further provides a servo spot welding clamp controller, which is designed based on the design method.

[0015] The present application has the following beneficial effects: The present application indirectly realizes spot welding clamp pressure control through speed control by using an admittance controller to realize force and motion collaborative adaptation; a permanent magnet synchronous motor speed controller is designed by using a sliding mode control related theory to realize servo spot welding clamp pressure control, and a gain of an approach rate is effectively reduced by combining with a RBF neural network, so that a chattering level is reduced, various friction disturbances in a spot welding process are reduced, and thus a spot welding clamp is robustly controlled to perform spot welding operation; secondary spot welding can be effectively avoided, spot welding quality and efficiency are improved, and spot welding flexible operation is realized. BRIEF DESCRIPTION OF DRAWINGS

[0016] The drawings constituting a part of the specification of the present application are used to provide further understanding of the present application, and the schematic embodiments of the present application and the description thereof are used to explain the present application, and do not constitute improper limitation on the present application.

[0017] Figure 1 is a schematic diagram of a servo welding tongs system structure; Figure 2 is a flow chart of a design method according to one or more embodiments of the present application; Figure 3 is a schematic diagram of a control system simulation according to one or more embodiments of the present application; Figure 4 is a schematic diagram of an RBF neural network.

[0018] Wherein, 1, electrode holder, 2, electrode handle, 3, electrode cap, 4, transmission mechanism, 5, servo motor. DETAILED DESCRIPTION It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the present application. Unless otherwise indicated, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0019] Example 1: As shown in Figure 1 , the servo welding tongs system includes a servo motor 5, a transmission mechanism 4, an electrode holder 1, an electrode handle 2, etc. The electrode handle 2 has an electrode cap 3 installed at the end, and the electrode holder 1 is connected to the top end of the electrode handle 2. The electrode holder 1 is connected to the servo motor 5 through the transmission mechanism 4, and the servo motor 5 is connected to a controller. The servo motor 5 in this embodiment is a permanent magnet synchronous motor. This embodiment provides a design method for a servo spot welding tongs controller, which is based on a sliding mode variable structure control combined with an admittance control strategy to design a servo welding tongs controller, including: Building a mathematical model of a permanent magnet synchronous motor; Taking the ability of the permanent magnet synchronous motor to quickly track the given speed of the admittance controller under unknown disturbance conditions as the control target, an integral sliding mode control is adopted to establish a permanent magnet synchronous motor state equation. The system friction term is included in the permanent magnet synchronous motor state equation, and an RBF neural network is used for online adaptive compensation of the system friction. Based on the permanent magnet synchronous motor state equation and the approach rate function, the controller output is calculated. The approach rate function uses an improved exponential approach rate.

[0020] This embodiment uses the method of sliding mode control to design a permanent magnet synchronous motor speed controller to achieve high-precision control of the spot welding tongs pressure. At the same time, it also combines an RBF neural network to effectively reduce the gain of the approach rate, reduce the level of chattering, and thus reduce various friction disturbances in the spot welding process, so as to accurately control the spot welding tongs to perform spot welding work.

[0021] Specifically, as shown in Figure 2 , the method includes the following steps: Step 1: Construct a permanent magnet synchronous motor (PMSM) mathematical model.

[0022] In this embodiment, the following assumptions are made: ignore the rotor core saturation, assume that the magnetic circuit is linear, the direct and quadrature axis inductance parameters are constant; do not consider the core and winding eddy current loss and hysteresis loss; the rotor has no damping winding; the magnetic field generated by the rotor and the three-phase winding is sinusoidal in the air gap.

[0023] The voltage balance equation of the three-phase stator winding of the PMSM is:

[0024] In formula (1), , , are the fluxes generated by the three-phase stator winding respectively, , , are the three-phase stator resistances respectively. According to the relevant coordinate transformation principle, the voltage equation of the PMSM based on the synchronous rotating d-q coordinate system can be obtained as:

[0025] In formula (2), as shown in Figure 3 , and represent the d-axis and q-axis stator currents respectively, is the flux of the rotor permanent magnet, is the electrical angular velocity, is the equivalent stator resistance, and represent the d-axis and q-axis stator inductances respectively. For a hidden pole PMSM, the stator inductance .

[0026] The dynamic equation of the PMSM in the spot welding clamp is:

[0027] Wherein:

[0028] In the above formula, is the bearing viscous coefficient of the PMSM, is the moment of inertia of the PMSM, is the number of rotor pole pairs of the PMSM, is the mechanical angular velocity. According to the force analysis, is the friction torque corresponding to the motor end of the mechanism friction of the servo spot welding clamp, is the welding torque corresponding to the motor end of the spot welding pressure measured during the spot welding process, The gravity of the moving part of the spot welding tong mechanism corresponds to the gravity torque of the motor end.

[0029] Step 2: For high-precision pressure control of the resistance spot welding process, a speed controller for the permanent magnet synchronous motor is designed based on the sliding mode variable structure theory.

[0030] Sliding mode control (SMC) makes the system state slide along the sliding mode surface through switching of the control quantity, and has strong robustness when the system is disturbed.

[0031] Further, step 2.1: for the control of the servo spot welding tong, the control target is that the servo motor can quickly track the speed given by the admittance controller under the condition of unknown disturbance, that is, .

[0032] In order to achieve the above control target and eliminate steady-state error, the embodiment adopts integral sliding mode control, and selects the system state variable:

[0033] According to formula (5), the state equation of the controlled object PMSM can be obtained as:

[0034] Where the function , the coefficient term , the disturbance term , the system friction term , the control quantity , is the mechanical position of PMSM. The spot welding torque can be measured by the force sensor installed on the electrode, is the rate of change of the given speed, that is, the given acceleration, which can be approximately calculated by , is given by the admittance controller.

[0035] During the operation of the spot welding tong, the friction caused by the transmission mechanisms such as the reduction mechanism and the roller screw, and the sliding friction suffered by the electrode holder during the movement constitute the system friction term , which is unknown. Since the moving part of the spot welding tong is affected by gravity (G) , it changes with the change of the posture of the spot welding tong, and the gravity is regarded as a disturbance term in the embodiment, which is bounded (G ).

[0036] Through force analysis, it can be known that the gravity term hardly changes in a single welding cycle, therefore, the gravity of the sliding mechanism of the spot welding tong can be approximately obtained by experimental method: the spot welding tong is placed in the vertical direction, and the gravity torque G the size of the sliding surface.

[0037] Step 2.2: For the tracking control problem, the typical sliding function construction method is:

[0038] In equation (7), the parameter satisfies the Hurwitz stability condition, i.e. The size of determines the speed of convergence of the state variable. To ensure that the system reaches the sliding surface from an arbitrary position point in the state space ), the sliding function must satisfy the following conditions:

[0039] where .

[0040] Step 2.3: The sliding state change includes two processes: approaching motion and sliding motion. The process of the system approaching the sliding surface from an arbitrary initial state is called approaching motion, i.e. Equation (8) gives the condition for reaching the sliding surface, but it does not limit the trajectory of the approaching motion. The method of approaching rate can improve the dynamic quality of the approaching motion.

[0041] Assume that the approaching rate function is i.e.

[0042] Differentiate equation (7) and substitute it into equation (9) to get:

[0043] Then, substitute equation (6) into equation (10) to get:

[0044] According to equation (11), the controller output can be calculated, which is expressed as:

[0045] The selection of the approaching rate function is usually to satisfy the Lyapunov asymptotic stability condition. The typical constant approaching rate expression is:

[0046] Obviously, the constant approaching rate satisfies the inequality. The time for the system to reach the sliding surface from the initial state is:

[0047] The traditional constant reaching rate has large chattering, which affects the stability of the system. In order to reduce the output chattering of the controller, the embodiment provides an improved exponential reaching rate, and the general form of the exponential reaching rate is as follows:

[0048] In the exponential reaching rate, in order to ensure fast approaching while weakening chattering, should dynamically increase and reduce . Therefore, the embodiment sets the expressions of parameters and as follows:

[0049] Wherein, the parameter .

[0050] In the new reaching rate of the embodiment, it can be analyzed that when the initial state of the system is far away from the sliding surface ( ), the reaching rate is:

[0051] Selecting , the reaching speed calculated by formula (17) is much greater than the constant reaching rate.

[0052] In the reaching motion process, it can be obtained that:

[0053] That is:

[0054] When reaching the sliding surface , from 0 to , integrating the above formula will obtain:

[0055] Since always holds, selecting , according to the above inequality, it can be obtained that:

[0056] Compared with the constant reaching rate, the new reaching rate can significantly reduce the time for the system state to reach the sliding surface, so as to achieve the purpose of fast response of the system.

[0057] On the other hand, when the system state reaches the sliding surface and tends to the equilibrium point ( ), the reaching rate is:

[0058] As can be seen from formula (22), when the approaching speed , the system state will eventually converge to the equilibrium point , which is not possible with constant approaching rate. The discrete form of equation (22) is:

[0059] where, is the sampling time. The discrete form of equation (21) can be obtained in the same way:

[0060] Step 3: For the non-linear disturbance such as friction in the servo welding tongs system, RBF neural network is introduced for dynamic compensation.

[0061] According to the expression of the controller output , the system friction term is caused by various mechanical contacts and friction, which is characterized by unknown variation, difficult modeling, and is a nonlinear function of speed . For the friction disturbance in the welding process, the RBF neural network is used for online adaptive compensation of the system friction.

[0062] The RBF neural network has good generalization ability, simple network structure, and can approximate any nonlinear function under a compact set and arbitrary precision. As shown in equation (23), Figure 4 is the input of the RBF network, is the output of the hidden layer, is the weight of the hidden layer, are the number of input layer, hidden layer, and output layer, respectively. The input-output relationship of the RBF network using Gaussian kernel function can be represented as:

[0063] where is the coordinate vector of the Gaussian basis function center point of the i-th neuron in the hidden layer, is the width of the Gaussian basis function. In this embodiment, the RBF neural network is used for approximation of the system friction term

[0064] , then:

[0065] is the ideal weight of the RBF neural network, is the approximation error. Taking the mechanical angular velocity of the servo motor as the network input , let be the actual output of the network, be the estimated weight of the network, then:​​

[0066] Take The estimation error of the system friction term is:

[0067] At this time, the expression of the control variable should be modified as:

[0068] where is the disturbance term estimation, so:

[0069] Define the Lyapunov function:

[0070] where , so . Taking the derivative of the above equation gives:

[0071] From the above equation, the adaptive rate is:

[0072] Since , take , let , always holds, so , so the Lyapunov function converges, . According to the above adaptive rate, update the weights of the RBF neural network, and the output of the network will approach the actual friction , and compensation to the control variable can effectively reduce the gain of the reaching rate and reduce the level of chattering.

[0073] The embodiment first realizes the "force and motion collaborative adaptation" of the system through the admittance controller, indirectly controls the spot welding clamp pressure through speed control; a PMSM speed controller is designed using the sliding mode control related theory to realize the control of the servo spot welding clamp pressure, and the gain of the reaching rate is effectively reduced by combining the RBF neural network to reduce the level of chattering, so as to reduce the interference of various frictions in the spot welding process, thereby robustly controlling the spot welding clamp to perform spot welding work. The embodiment can effectively avoid secondary spot welding, improve the quality and efficiency of spot welding, and realize flexible spot welding operation.

[0074] Embodiment 2: The embodiment provides a servo spot welding clamp controller, which is designed based on the design method in embodiment 1.

[0075] The above descriptions are only the preferred embodiments of the present application, and are not intended to limit the present application. The present application can have various changes and modifications for those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A design method for a servo spot welding clamp controller, characterized in that, include: Construct a mathematical model for a permanent magnet synchronous motor; With the goal of enabling the permanent magnet synchronous motor to quickly track the speed given by the admittance controller under unknown disturbance conditions, integral sliding mode control is adopted to establish the state equation of the permanent magnet synchronous motor. Among them, the state equation of the permanent magnet synchronous motor includes a system friction term, and an RBF neural network is used for online adaptive compensation of system friction. The controller output is calculated based on the state equation of the permanent magnet synchronous motor and the approach rate function; wherein, the approach rate function adopts an improved exponential approach rate.

2. The design method of a servo spot welding clamp controller according to claim 1, characterized in that, The controller output is expressed as: ； in, x 2 represents the system state variable. For the system friction term; function coefficient term Interference items ; Depend on Calculations show that Provided by the admittance controller, The bearing viscosity coefficient of a permanent magnet synchronous motor. This represents the moment of inertia of the permanent magnet synchronous motor. For mechanical angular velocity, For the magnetic flux of the rotor permanent magnet, For spot welding torque, This is the gravitational torque.

3. The design method of a servo spot welding clamp controller according to claim 2, characterized in that, The gravitational torque satisfies: ; By placing the spot welding clamp in the vertical direction, the gravitational torque is calculated based on the torque difference of the permanent magnet synchronous motor when the electrode grip rod moves at a constant speed in both directions. .

4. A design method for a servo spot welding clamp controller according to claim 1 or 2, characterized in that, Applying RBF neural networks to the system friction term Approaching: ； in, For the ideal weights of the RBF neural network, To approximate the error, h This is the output of the hidden layer; With the network input being the mechanical angular velocity of the permanent magnet synchronous motor, the actual output of the network is expressed as: ;in, Here, T represents the estimated weights of the network, and T represents the sampling time.

5. The design method of a servo spot welding clamp controller according to claim 4, characterized in that, Define the Lyapunov function and take its derivative to obtain the adaptive rate: ； in, .

6. The design method of a servo spot welding clamp controller according to claim 2, characterized in that, Integral sliding mode control is adopted, and the system state variables are selected as follows: ； in, .

7. The design method of a servo spot welding clamp controller according to claim 6, characterized in that, The state equation of the permanent magnet synchronous motor is: 。 8. The design method of a servo spot welding clamp controller according to claim 1, characterized in that, The improved exponential convergence rate is expressed as: ； parameter and Represented as: ； Where parameters .

9. The design method of a servo spot welding clamp controller according to claim 1, characterized in that, When constructing the mathematical model of the permanent magnet synchronous motor, the saturation of the stator and rotor cores is ignored, the magnetic circuit is assumed to be linear, and the inductance parameters of the quadrature and direct axes remain unchanged; the eddy current loss and hysteresis loss of the core and windings are neglected; the rotor has no damped windings; the magnetic field generated by the rotor and the three-phase windings is distributed sinusoidally in the air gap.

10. A servo spot welding clamp controller, characterized in that, Designed based on the design method described in any one of claims 1-9.

Images