阅读说明：本技术 一种半导体封装控制系统的控制方法 (Control method of semiconductor packaging control system ) 是由 田兴银 张勇 毛军 于 2021-07-07 设计创作，主要内容包括：本发明提供了一种半导体封装控制系统的控制方法,属于半导体技术领域。本发明首先基于PID/PI/PI控制位置环、速度环和电流环,在此基础上以控制对象、电流环和速度环作为一个控制整体,在位置环加入加速度和速度前馈提高系统响应能力,保证“目标跟踪特性”性能,同时也在位置环加入特定的干扰观测器,提高内部干扰和外部干扰的抑制能力,实现“外扰抑制特性”性能。(The invention provides a control method of a semiconductor packaging control system, and belongs to the technical field of semiconductors. The invention firstly controls a position loop, a speed loop and a current loop based on PID/PI/PI, on the basis, a control object, the current loop and the speed loop are taken as a control whole, acceleration and speed feedforward is added into the position loop to improve the system response capability, ensure the performance of target tracking characteristics, and meanwhile, a specific interference observer is added into the position loop to improve the inhibition capability of internal interference and external interference and realize the performance of external interference inhibition characteristics.)

1. A control method of a semiconductor package control system is characterized by comprising the following steps:

step 1, establishing a curing control kernel function G based on a linear motor mathematical model, a current loop and a speed loop_V(s)；

Step 2, controlling a kernel function G based on solidification_V(s) establishing a position loop transfer function G using PID control_p；

Step 3, establishing a transfer function G for adding velocity and acceleration feedforward control_ry(s)；

Step 4, establishing a transfer function added into the special interference observer;

step 5, integrating the speed acceleration feedforward, the position loop PID control and the disturbance observer model control, and establishing a linear motor displacement output y expression of the control system as follows:

wherein the content of the first and second substances,

G_V(s) is a solidification control kernel function;

G_n(s) is a control object nominal model; g_n(s)＝G_V(s)；

G_pIs a position loop transfer function;

K_afis the acceleration feedforward coefficient;

K_vfis a velocity feedforward coefficient;

K_ppis the proportionality coefficient of the position loop;

K_piis the integral coefficient of the position loop;

K_pdis the differential coefficient of the position loop;

s is a complex space variable after Laplace transformation, the physical meaning of the complex space variable is corresponding to real space time, and the unit is ms;

q(s) is the filter transfer function.

2. The control method of a semiconductor package control system according to claim 1, wherein a transfer function G incorporating velocity and acceleration feedforward control is established_ry(s) specifically the following:

in the formula:

G_V(s) is a solidification control kernel function;

G_pis a position loop transfer function;

K_afis the acceleration feedforward coefficient;

K_vfis the velocity feedforward coefficient.

3. The method of claim 1, wherein the step of establishing the transfer function for adding the dedicated disturbance observer comprises the steps of:

establishing an interference observer according to the principle of the interference observer;

according to the established disturbance observer, establishing a transfer function G from a reference input u to a system output y_uy(S), the transfer function is as follows:

according to the established interference observer, establishing a transfer function G from the low-frequency interference d of the system to the output y of the system_dy(S), the transfer function is as follows:

according to the established interference observer, a transfer function G from the high-frequency interference xi of the system to the output y of the system is established_ξy(S), the transfer function is as follows:

wherein the content of the first and second substances,

G_V(s) is a solidification control kernel function;

G_n(s) is a control object nominal model; g_n(s)＝G_V(s)；

Xi is system high frequency interference;

y is the system output;

q(s) is the filter transfer function.

4. The method of claim 3, wherein the filter Q(s) is a binomial low pass filter expressed as follows:

wherein the content of the first and second substances,

tau is a time constant and is the only adjustable parameter of Q(s);

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms.

5. The method of claim 3, wherein the step of establishing the disturbance observer according to the principle of the disturbance observer specifically comprises the steps of:

setting u as a reference input with a basic model of a disturbance observer, and the value is the obtained velocity acceleration look-ahead K_afs²、K_vfs²And position loop control output G_PID(s) sum;

setting d as system low-frequency interference, xi as system high-frequency interference, Q(s) as filter, G_V(s) for the established control kernel model,is a control object nominal model G_n(s) inverse, y being the system output;

get G_n(s)＝G_V(s)。

6. The method for controlling a semiconductor package control system according to claim 1, wherein the step of establishing the solidification control kernel function based on the linear motor mathematical model, the current loop and the speed loop specifically comprises the steps of:

establishing a mathematical model of a linear motor platform in a control system;

establishing a transfer function of a PI control current loop based on a linear motor platform mathematical model;

and establishing a transfer function of the speed loop controlled by the PI based on the transfer function of the current loop.

7. The method of claim 6, wherein the mathematical model of the linear motor stage in the control system is established as follows:

wherein the content of the first and second substances,

G_Mcontrolling the output of a linear motor platform in the system;

l is linear motor armature inductance (unit: mH);

r is the armature resistance (unit: omega) of the linear motor;

K_fis the linear motor force constant (unit: N/A);

m is the moving mass (unit: kg) of the linear motor;

b is the coefficient of viscous friction;

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms.

8. The control method of the semiconductor package control system according to claim 6, wherein a transfer function using a PI control current loop is established based on a linear motor platform mathematical model;

the PI control expression for the current loop is:

establishing a transfer function G using a PI control current loop_I(S) is:

wherein the content of the first and second substances,

K_ipis a proportionality coefficient for current loop control;

K_iiis an integral coefficient;

r is the armature resistance (unit: omega) of the linear motor;

l is linear motor armature inductance (unit: mH);

G_iis the output of the current loop;

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms.

9. The control method of the semiconductor package control system according to claim 6, wherein a transfer function using a PI control speed loop is established based on a transfer function of a current loop;

the PI control expression for the speed loop is:

establishing a transfer function G using a PI control speed loop_V(S) is:

wherein the content of the first and second substances,

K_vpa proportionality coefficient for speed loop control;

K_viis an integral coefficient;

G_vis the output of the speed loop;

B₂＝K_fK_vpK_ip；

B₁＝K_f(K_viK_ip+K_vpK_ii)；

B₀＝K_fK_viK_ii；

A₄＝mL；

A₃＝mR+BL+mK_ip；

A₂＝BR+mK_ii+BK_ip+K_f(K_e+K_vpK_ip)；

A₁＝BK_ii+K_f(K_ipK_vi+K_vpK_ii)；

A₀＝K_fK_viK_ii；

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms.

10. The control method of the semiconductor package control system according to claim 1, wherein the control kernel function G is based on solidification_V(s) establishing a position loop transfer function controlled by PID, specifically comprising the following steps:

the PID control expression for the position loop is:

establishing a transfer function G using a PID control position loop_PID(S) is:

wherein the content of the first and second substances,

K_ppis the proportionality coefficient of the position loop;

K_piis the integral coefficient of the position loop;

K_pdis the differential coefficient of the position loop;

G_pis the position loop output;

D₄＝K_ppB₂；

D₃＝K_ppB₂+K_pdB₁；

D₂＝K_ppB₁+K_piB₂+K_pdB₀；

D₁＝K_ppB₀+K_piB₁；

D₀＝K_piB₀；

C₆＝A₄；

C₅＝A₃；

C₄＝A₂+K_PdB₂；

C₃＝A₁+K_ppB₂+K_pdB₁；

C₂＝A₀+K_ppB₁+K_piB₂+K_pdB₀；

C₁＝K_ppB₀+K_PiB₁；

C₀＝K_piB₀；

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms.

Technical Field

The invention relates to the technical field of semiconductors, in particular to a control method of a semiconductor packaging control system.

Background

In the manufacturing of integrated circuits, the servo performance requirements of miniaturized, multi-pin, fine-pitch chips on the response capability, positioning accuracy, positioning time and the like of a motion control system of packaging equipment are extremely high, and a motion platform is often required to rapidly, stably and accurately run to a given target position in a very short time. However, in the process of high-speed and high-precision point-to-point motion of fast start-stop, external interference, mechanical resonance and non-linear factors of the control system influence the servo performance of the system to different degrees, and directly influence the positioning precision and the positioning time of the control system of the packaging equipment, so the servo control strategy of the system is the core difficulty of the chip packaging control system.

In chinese patent application CN102237101A, a sliding mode variable structure servo controller for NVD system optical disc drive and a servo control method thereof are provided, the servo controller has a sliding mode variable structure control operation module, and the servo control method is that the DSP module outputs a next time position command r(s) and a current position signal x(s) to the sliding mode variable structure control operation module, and outputs a voltage control signal u(s) through data processing. However, in the method, the phenomenon of 'buffeting' controlled by the sliding mode variable structure is not well proposed and solved, and the discontinuity of the switching surface is not well solved, so that the integral performance of operation is influenced.

In chinese patent application document CN105867113A, a servo control method related to satellite tracking is proposed, which applies a kalman filter algorithm to perform data smoothing on an acquired error signal, then assigns a fuzzy rule table by using a fuzzy PID algorithm, and adjusts PID operation parameters according to the fuzzy rule table to perform PID operation on the error signal after data smoothing processing, so as to obtain a servo drive signal, thereby effectively improving the automatic tracking accuracy and efficiency of a target signal. However, the method is applied to the control of semiconductor packaging equipment with higher requirements on real-time performance and precision, and the control and positioning precision and the high-speed and high-precision motion performance of quick start and stop of the method are slightly insufficient.

Based on the analysis of the literature, the servo control method is an important technical bottleneck of many control systems, however, the servo control methods in different occasions need to be improved to different degrees according to respective application characteristics.

When the moving mechanism and the electric system determine, whether the semiconductor packaging equipment can realize fast and accurate positioning control under high-acceleration high-speed dynamic operation completely depends on the performance of a control system, particularly the servo control performance of the system. From the perspective of servo control, high precision, high acceleration and short motion time are obviously a set of contradictions which are difficult to reconcile, when the motion distance is fixed, the high acceleration is required when the operation time is short, and the high acceleration can prolong the stability time of the system and reduce the positioning precision of the motion platform, and vice versa. Therefore, research on a control method of a control system of a high-speed and high-precision semiconductor packaging device is necessary.

Based on the high requirements of semiconductor packaging equipment on high speed, high acceleration and high precision movement, the servo control technology of an equipment control system is very important. The traditional PID control has the advantages of simple structure, convenient parameter tuning, good robustness and reliable work, is one of main control methods applied in the field of automation, but the traditional PID control also has certain limitation, is difficult to simultaneously consider the target tracking characteristic and the external disturbance suppression characteristic, and is difficult to realize extremely low following error to complete command movement while eliminating external disturbance in high-speed, high-acceleration and high-precision movement.

The prior art has the following defects in semiconductor packaging control:

1. in the prior art, the speed acceleration feedforward and other improved PID technologies are adopted, so that the problem of external interference in semiconductor packaging equipment cannot be solved well;

2. when intelligent algorithm control means such as a fuzzy algorithm, a neural network algorithm and the like are adopted in the prior art, the performance requirements of high speed, high acceleration, high precision, high frequency start-stop positioning and the like in semiconductor packaging equipment cannot be well met;

disclosure of Invention

Aiming at the problems in the prior art, the invention adopts a new improved algorithm aiming at the requirements of the semiconductor packaging equipment on high-speed, high-acceleration and high-precision motion on the basis of the traditional PID, and mainly solves the control technical problem of the linear motor of the semiconductor packaging equipment.

The invention realizes improvement based on the traditional PID principle, firstly, the control object (linear motor) and the current loop and the speed loop are solidified and unchanged, and the control object is taken as a new integral control object, and the control object is only the linear motor under the traditional PID control; secondly, on the basis of a new control object, the proposed speed and acceleration control model is added in parallel at the position ring control part, so that the system response capability is improved, and high-speed high-acceleration and low-tracking error are achieved; and the proposed interference observer model is added again to realize internal and external interference suppression and realize low tracking error and high-precision positioning. The control technical problem of the linear motor is solved based on the two means.

The invention provides a control method of a semiconductor packaging control system, which comprises the following steps:

step 1, establishing a curing control kernel function G based on a linear motor mathematical model, a current loop and a speed loop_V(s)；

Step 2, controlling a kernel function G based on solidification_V(s) establishing a position loop transfer function using PID control;

step 3, establishing a transfer function G for adding velocity and acceleration feedforward control_ry(s)；

Step 4, establishing a transfer function added into the special interference observer;

step 5, integrating the speed acceleration feedforward, the position loop PID control and the disturbance observer model control, and establishing a linear motor displacement output y expression of the control system as follows:

wherein the content of the first and second substances,

G_V(s) is a solidification control kernel function;

G_n(s) is a control object nominal model; g_n(s)＝G_V(s)；

G_pIs a position loop transfer function;

K_afis the acceleration feedforward coefficient;

K_vfis a velocity feedforward coefficient;

K_ppis the proportionality coefficient of the position loop;

K_piis the integral coefficient of the position loop;

K_pdis the differential coefficient of the position loop;

s is a complex space variable after Laplace transformation, the physical meaning of the complex space variable is corresponding to real space time, and the unit is ms;

q(s) is the filter transfer function.

Preferably, a transfer function G incorporating velocity and acceleration feedforward control is established_ry(s) specifically the following:

in the formula:

G_V(s) is a solidification control kernel function;

G_pis a position loop transfer function;

K_afis the acceleration feedforward coefficient;

K_vfis the velocity feedforward coefficient.

Preferably, the establishing of the transfer function to be added to the special disturbance observer specifically includes the following steps:

establishing an interference observer according to the principle of the interference observer;

according to the established disturbance observer, establishing a transfer function G from a reference input u to a system output y_uy(S), the transfer function is as follows:

according to the established interference observer, establishing a transfer function G from the low-frequency interference d of the system to the output y of the system_dy(S), the transfer function is as follows:

according to the established interference observer, a transfer function G from the high-frequency interference xi of the system to the output y of the system is established_ξy(S), the transfer function is as follows:

wherein the content of the first and second substances,

G_V(s) is a solidification control kernel function;

G_n(s) is a control object nominal model; g_n(s)＝G_V(s)；

Xi is system high frequency interference;

y is the system output;

q(s) is the filter transfer function.

Preferably, the filter q(s) is a binomial low-pass filter, expressed as follows:

wherein the content of the first and second substances,

tau is a time constant and is the only adjustable parameter of Q(s);

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms.

Preferably, according to the principle of the disturbance observer, establishing the disturbance observer specifically includes the following steps:

setting u as a reference input with a basic model of a disturbance observer, and the value is the obtained velocity acceleration look-ahead K_afs²、K_vfs²And position loop control outputG_PID(s) sum;

setting d as system low-frequency interference, xi as system high-frequency interference, Q(s) as filter, G_V(s) for the established control kernel model,is a control object nominal model G_n(s) inverse, y being the system output;

get G_n(s)＝G_V(s). In practical applications, the control kernel simulation model is usually taken to be very close to the nominal model, i.e. G_n(s)＝G_V(s)。

Preferably, the establishing of the curing control kernel function based on the linear motor mathematical model, the current loop and the speed loop specifically includes the following steps:

establishing a mathematical model of a linear motor platform in a control system;

establishing a transfer function of a PI control current loop based on a linear motor platform mathematical model;

and establishing a transfer function of the speed loop controlled by the PI based on the transfer function of the current loop.

Preferably, a mathematical model of a linear motor platform in the control system is established, and the formula is as follows:

wherein the content of the first and second substances,

G_Mcontrolling the output of a linear motor platform in the system;

l is linear motor armature inductance (unit: mH);

r is the armature resistance (unit: omega) of the linear motor;

K_fis the linear motor force constant (unit: N/A);

m is the moving mass (unit: kg) of the linear motor;

b is the coefficient of viscous friction;

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms.

Preferably, a transfer function of a PI control current loop is established based on a linear motor platform mathematical model;

the PI control expression for the current loop is:

establishing a transfer function G using a PI control current loop_I(S) is:

wherein the content of the first and second substances,

K_ipis a proportionality coefficient for current loop control;

K_iiis an integral coefficient;

r is the armature resistance (unit: omega) of the linear motor;

l is linear motor armature inductance (unit: mH);

G_iis the output of the current loop;

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms.

Preferably, a transfer function of a speed loop controlled by a PI is established based on a transfer function of a current loop;

the PI control expression for the speed loop is:

establishing a transfer function G using a PI control speed loop_V(S) is:

wherein the content of the first and second substances,

K_vpa proportionality coefficient for speed loop control;

K_viis an integral coefficient;

G_vis the output of the speed loop;

B₂＝K_fK_vpK_ip；

B₁＝K_f(K_viK_ip+K_vpK_ii)；

B₀＝K_fK_viK_ii；

A₄＝mL；

A₃＝mR+BL+mK_ip；

A₂＝BR+mK_ii+BK_ip+K_f(K_e+K_vpK_ip)；

A₁＝BK_ii+K_f(K_ipK_vi+K_vpK_ii)；

A₀＝K_fK_viK_ii；

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms.

Preferably, the kernel function G is controlled based on curing_V(s) establishing a position loop transfer function controlled by PID, specifically comprising the following steps:

the PID control expression for the position loop is:

the transfer function using the PID control position loop is established as follows:

wherein the content of the first and second substances,

K_ppis the proportionality coefficient of the position loop;

K_piis a positionThe integral coefficient of the loop;

K_pdis the differential coefficient of the position loop;

G_pis the position loop output;

D₄＝K_ppB₂；

D₃＝K_ppB₂+K_pdB₁；

D₂＝K_ppB₁+K_piB₂+K_pdB₀；

D₁＝K_ppB₀+K_piB₁；

D₀＝K_piB₀；

C₆＝A₄；

C₅＝A₃；

C₄＝A₂+K_PdB₂；

C₃＝A₁+K_ppB₂+K_pdB₁；

C₂＝A₀+K_ppB₁+K_piB₂+K_pdB₀；

C₁＝K_ppB₀+K_PiB₁；

C₀＝K_piB₀；

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms.

Compared with the prior art, the invention has the following beneficial effects:

1. the invention is based on PID/PI/PI control position loop, velocity loop and current loop, on this basis, the control object, current loop and velocity loop are used as a control whole, acceleration and velocity feedforward is added into the position loop to improve the system response capability, ensure the performance of target tracking characteristic, and meet the requirement of high-speed high-acceleration rapid positioning of the semiconductor packaging equipment on movement;

2. according to the invention, a specific interference observer is added in the position ring, so that the inhibition capability of internal interference and external interference is improved, the performance of external interference inhibition characteristic is ensured, and high-precision positioning is realized.

Drawings

FIG. 1 is a system control architecture diagram of one embodiment of the present invention;

FIG. 2 is a simplified system control architecture of one embodiment of the present invention;

FIG. 3 is a control system framework diagram of one embodiment of the present invention;

fig. 4 is a flowchart of a control method according to an embodiment of the invention.

In the figure:

the method comprises the following steps of 1-speed acceleration feedforward, 2-disturbance observer, 3-position loop control, 4-control kernel, 5-upper computer, 6-motion controller, 71-X axis servo driver, 72-Y axis servo driver, 73-Z axis servo driver, 81-X axis linear motor, 82-Y axis linear motor and 83-Z axis linear motor.

Detailed Description

The following detailed description of embodiments of the invention refers to the accompanying drawings.

As shown in fig. 3, one embodiment of a semiconductor package control system is shown. The control system comprises an upper computer 5, a motion controller 6, an X-axis servo driver 71, a Y-axis servo driver 72, a Z-axis servo driver 73, an X-axis linear motor 81, a Y-axis linear motor 82 and a Z-axis linear motor 83.

The upper computer 5 can be a PC, an upper computer control program runs on the PC, the upper computer 5 is connected with the motion controller 6 through a network cable, the motion controller 6 is respectively connected with the X-axis servo driver 71, the Y-axis servo driver 72 and the Z-axis servo driver 73 through network cables based on an EtherCat bus protocol, the X-axis servo driver 71 is connected with the X-axis linear motor 81 through a signal cable of analog quantity voltage, the Y-axis servo driver 72 is connected with the Y-axis linear motor 82 through a signal cable of analog quantity voltage, and the Z-axis servo driver 73 is connected with the Z-axis linear motor 83 through a signal cable of analog quantity voltage.

The X-axis servo driver 71, the Y-axis servo driver 72 and the Z-axis servo driver 73 all comprise 3 software functional modules which are respectively used for parameter management, state management and motor control; the parameter management function module is used for setting servo parameters of the servo driver, the state management function module is used for managing the program state of the servo driver, and the motor control function module is a servo driver core module and is used for realizing the motor control technology of the servo driver.

The servo parameters mainly refer to universal electronic gear ratio, control mode, filtering parameters and the like, and are set before use;

the state management only refers to the running state of the servo driver, such as a motion mode, a motion starting and stopping state, a motion alarm state, an internal system state and the like;

the servo driving software system is formed by three modules of parameter management, state management and motor control module, the motor control module and the parameter management module have parameter data interaction, and the motor control module and the state management module have state data interaction;

the control method of the semiconductor packaging control system provided by the invention is realized in motor control functional modules in an X-axis servo driver 71, a Y-axis servo driver 72 and a Z-axis servo driver 73.

The control method finally obtains linear motor displacement output y used for the control output of the servo driver to the linear motor, wherein r is an instruction value, y is an actual output value after a control algorithm, and s is a Laplace transformation intermediate variable which can be related to time.

The invention provides a control method of a semiconductor packaging control system, which comprises the following steps:

step 1, establishing a curing control kernel function G based on a linear motor mathematical model, a current loop and a speed loop_V(s)；

Step 2, controlling a kernel function G based on solidification_V(s) establishing a position loop transfer function using PID control;

step 3, establishing a transfer function G for adding velocity and acceleration feedforward control_ry(s)；

Step 4, establishing a transfer function added into the special interference observer;

step 5, integrating the speed acceleration feedforward, the position loop PID control and the disturbance observer model control, and establishing a linear motor displacement output y expression of the control system as follows:

wherein the content of the first and second substances,

G_V(s) is a solidification control kernel function;

G_n(s) is a control object nominal model; g_n(s)＝G_V(s)；

G_pIs a position loop transfer function;

K_afis the acceleration feedforward coefficient;

K_vfis a velocity feedforward coefficient;

K_ppis the proportionality coefficient of the position loop;

K_piis the integral coefficient of the position loop;

K_pdis the differential coefficient of the position loop;

s is a complex space variable after Laplace transformation, the physical meaning of the complex space variable is corresponding to real space time, and the unit is ms;

q(s) is the filter transfer function.

According to one embodiment of the present invention, a transfer function G incorporating velocity and acceleration feedforward control is established_ry(s) specifically the following:

in the formula:

G_V(s) is a solidification control kernel function;

G_pis a position loop transfer function;

K_afis added withA velocity feedforward coefficient;

K_vf is the velocity feedforward coefficient.

According to a specific embodiment of the present invention, establishing a transfer function for adding a dedicated disturbance observer specifically comprises the following steps:

establishing an interference observer according to the principle of the interference observer;

according to the established disturbance observer, establishing a transfer function G from a reference input u to a system output y_uy(S), the transfer function is as follows:

according to the established interference observer, establishing a transfer function G from the low-frequency interference d of the system to the output y of the system_dy(S), the transfer function is as follows:

according to the established interference observer, a transfer function G from the high-frequency interference xi of the system to the output y of the system is established_ξy(S), the transfer function is as follows:

wherein the content of the first and second substances,

G_V(s) is a solidification control kernel function;

G_n(s) is a control object nominal model; g_n(s)＝G_V(s)；

Xi is system high frequency interference;

y is the system output;

q(s) is the filter transfer function.

According to one embodiment of the invention, the filter q(s) is a binomial low-pass filter, expressed as follows:

wherein the content of the first and second substances,

tau is a time constant and is the only adjustable parameter of Q(s);

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms.

According to a specific embodiment of the present invention, according to the principle of the disturbance observer, the establishing of the disturbance observer specifically includes the following steps:

setting u as a reference input with a basic model of a disturbance observer, and the value is the obtained velocity acceleration look-ahead K_afs²、K_vfs²And position loop control output G_PID(s) sum;

setting d as system low-frequency interference, xi as system high-frequency interference, Q(s) as filter, G_V(s) for the established control kernel model,is a control object nominal model G_n(s) inverse, y being the system output;

get G_n(s)＝G_V(s). In practical applications, the control kernel simulation model is usually taken to be very close to the nominal model, i.e. G_n(s)＝G_V(s)。

According to a specific embodiment of the present invention, the establishing of the solidification control kernel function based on the linear motor mathematical model, the current loop and the speed loop specifically comprises the following steps:

establishing a mathematical model of a linear motor platform in a control system;

establishing a transfer function of a PI control current loop based on a linear motor platform mathematical model;

and establishing a transfer function of the speed loop controlled by the PI based on the transfer function of the current loop.

According to a specific embodiment of the present invention, a mathematical model of a linear motor platform in a control system is established, and the formula is as follows:

wherein the content of the first and second substances,

l is linear motor armature inductance (unit: mH);

r is the armature resistance (unit: omega) of the linear motor;

K_fis the linear motor force constant (unit: N/A);

m is the moving mass (unit: kg) of the linear motor;

b is the coefficient of viscous friction;

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms.

According to a specific embodiment of the invention, a transfer function adopting a PI control current loop is established based on a linear motor platform mathematical model;

the PI control expression for the current loop is:

establishing a transfer function G using a PI control current loop_I(S) is:

wherein the content of the first and second substances,

K_ipis a proportionality coefficient for current loop control;

K_iiis an integral coefficient;

r is the armature resistance (unit: omega) of the linear motor;

l is linear motor armature inductance (unit: mH);

G_iis the output of the current loop;

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms.

According to a specific embodiment of the present invention, a transfer function using a PI control speed loop is established based on a transfer function of a current loop;

the PI control expression for the speed loop is:

establishing a transfer function G using a PI control speed loop_V(S) is:

wherein the content of the first and second substances,

K_vpa proportionality coefficient for speed loop control;

K_viis an integral coefficient;

G_vis the output of the speed loop;

B₂＝K_fK_vpK_ip；

B₁＝K_f(K_viK_ip+K_vpK_ii)；

B₀＝K_fK_viK_ii；

A₄＝mL；

A₃＝mR+BL+mK_ip；

A₂＝BR+mK_ii+BK_ip+K_f(K_e+K_vpK_ip)；

A₁＝BK_ii+K_f(K_ipK_vi+K_vpK_ii)；

A₀＝K_fK_viK_ii；

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms.

According to one embodiment of the inventionImplementation, control kernel function G based on curing_V(s) establishing a position loop transfer function controlled by PID, specifically comprising the following steps:

the PID control expression for the position loop is:

the transfer function using the PID control position loop is established as follows:

wherein the content of the first and second substances,

K_ppis the proportionality coefficient of the position loop;

K_piis the integral coefficient of the position loop;

K_pdis the differential coefficient of the position loop;

G_pis the position loop output;

D₄＝K_ppB₂；

D₃＝K_ppB₂+K_pdB₁；

D₂＝K_ppB₁+K_piB₂+K_pdB₀；

D₁＝K_ppB₀+K_piB₁；

D₀＝K_piB₀；

C₆＝A₄；

C₅＝A₃；

C₄＝A₂+K_PdB₂；

C₃＝A₁+K_ppB₂+K_pdB₁；

C₂＝A₀+K_ppB₁+K_piB₂+K_pdB₀；

C₁＝K_ppB₀+K_PiB₁；

C₀＝K_piB₀；

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms.

Example 1

According to a specific embodiment of the present invention, a step of establishing a transfer function to which a dedicated disturbance observer is added in the control method of the semiconductor package control system of the present invention is explained in detail.

According to a specific embodiment of the present invention, establishing a transfer function for adding a dedicated disturbance observer specifically comprises the following steps:

according to the principle of the disturbance observer, the disturbance observer is established, and the method specifically comprises the following steps:

setting u as a reference input with a basic model of a disturbance observer, and the value is the obtained velocity acceleration look-ahead K_afs²、K_vfs²And position loop control output G_PID(s) sum;

setting d as system low-frequency interference, xi as system high-frequency interference, Q(s) as filter, G_V(s) for the established control kernel model,is a control object nominal model G_n(s) inverse, y being the system output;

get G_n(s)＝G_V(s). In practical applications, it is often desirable to control the kernel simulation model to be very close to the nominal model, i.e. G_n(s)＝G_V(s)。

According to the established disturbance observer, establishing a transfer function G from a reference input u to a system output y_uy(S), the transfer function is as follows:

according to the established interference observer, establishing a transfer function G from the low-frequency interference d of the system to the output y of the system_dy(S), the transfer function is as follows:

according to the established interference observer, a transfer function G from the high-frequency interference xi of the system to the output y of the system is established_ξy(S), the transfer function is as follows:

wherein the content of the first and second substances,

G_V(s) is a solidification control kernel function;

G_n(s) is a control object nominal model; g_n(s)＝G_V(s)；

Xi is system high frequency interference;

y is the system output;

q(s) is the filter transfer function; q(s) is a binomial low pass filter, expressed as follows:

wherein the content of the first and second substances,

tau is a time constant and is the only adjustable parameter of Q(s);

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms.

Example 2

According to an embodiment of the present invention, a step of establishing a solidification control kernel function based on a linear motor mathematical model, a current loop and a speed loop in the control method of the semiconductor package control system of the present invention is described in detail.

Based on linear electric motor mathematics mouldModel, current loop and speed loop, establishing a solidification control kernel function G_V(s) specifically comprises the steps of:

establishing a mathematical model of a linear motor platform in a control system, wherein the formula is as follows:

wherein the content of the first and second substances,

G_Mcontrolling the output of a linear motor platform in the system;

l is linear motor armature inductance (unit: mH);

r is the armature resistance (unit: omega) of the linear motor;

K_fis the linear motor force constant (unit: N/A);

m is the moving mass (unit: kg) of the linear motor;

b is the coefficient of viscous friction;

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms.

Establishing a transfer function of a PI control current loop based on a linear motor platform mathematical model; the PI control expression for the current loop is:

establishing a transfer function G using a PI control current loop_I(S) is:

wherein the content of the first and second substances,

K_ipis a proportionality coefficient for current loop control;

K_iiis an integral coefficient;

r is the armature resistance (unit: omega) of the linear motor;

l is linear motor armature inductance (unit: mH);

G_iis the output of the current loop;

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms. Establishing a transfer function of a speed loop controlled by a PI (proportional integral) based on a transfer function of a current loop;

the PI control expression for the speed loop is:

establishing a transfer function G using a PI control speed loop_V(S) is:

wherein the content of the first and second substances,

K_vpa proportionality coefficient for speed loop control;

K_viis an integral coefficient;

G_vis the output of the speed loop;

B₂＝K_fK_vpK_ip；

B₁＝K_f(K_viK_ip+K_vpK_ii)；

B₀＝K_fK_viK_ii；

A₄＝mL；

A₃＝mR+BL+mK_ip；

A₂＝BR+mK_ii+BK_ip+K_f(K_e+K_vpK_ip)；

A₁＝BK_ii+K_f(K_ipK_vi+K_vpK_ii)；

A₀＝K_fK_viK_ii；

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms.

Example 3

According to an embodiment of the present invention, the control method for the semiconductor package control system according to the present invention is based on a solidified control kernel function G_V(s), the steps of establishing a position loop transfer function using PID control are described in detail.

Control kernel function G based on solidification_V(s) establishing a position loop transfer function controlled by PID, specifically comprising the following steps:

the PID control expression for the position loop is:

establishing a transfer function G using a PID control position loop_PID(S) is:

wherein the content of the first and second substances,

K_ppis the proportionality coefficient of the position loop;

K_piis the integral coefficient of the position loop;

K_pdis the differential coefficient of the position loop;

G_pis the position loop output;

D₄＝K_ppB₂；

D₃＝K_ppB₂+K_pdB₁；

D₂＝K_ppB₁+K_piB₂+K_pdB₀；

D₁＝K_ppB₀+K_piB₁；

D₀＝K_piB₀；

C₆＝A₄；

C₅＝A₃；

C₄＝A₂+K_PdB₂；

C₃＝A₁+K_ppB₂+K_pdB₁；

C₂＝A₀+K_ppB₁+K_piB₂+K_pdB₀；

C₁＝K_ppB₀+K_PiB₁；

C₀＝K_piB₀；

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms.

Example 4

A control method of the semiconductor package control system of the present invention will be described in detail according to an embodiment of the present invention.

The invention provides a control method of a semiconductor packaging control system, which comprises the following steps:

step 1, establishing a solidification control kernel function based on a linear motor mathematical model, a current loop and a speed loop, and specifically comprising the following steps:

establishing a mathematical model of a linear motor platform in a control system, wherein the formula is as follows:

wherein the content of the first and second substances,

G_Mcontrolling the output of a linear motor platform in the system;

l is linear motor armature inductance (unit: mH);

r is the armature resistance (unit: omega) of the linear motor;

K_fis the linear motor force constant (unit: N/A);

m is the moving mass (unit: kg) of the linear motor;

b is the coefficient of viscous friction;

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms.

Based on a linear motor platform mathematical model, establishing a transfer function of a PI control current loop:

the PI control expression for the current loop is:

establishing a transfer function G using a PI control current loop_I(S) is:

wherein the content of the first and second substances,

K_ipis a proportionality coefficient for current loop control;

K_iiis an integral coefficient;

r is the armature resistance (unit: omega) of the linear motor;

l is linear motor armature inductance (unit: mH);

G_iis the output of the current loop;

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms. Establishing a transfer function adopting a PI control speed loop based on the transfer function of the current loop:

the PI control expression for the speed loop is:

establishing a transfer function G using a PI control speed loop_V(S) is:

wherein the content of the first and second substances,

K_vpa proportionality coefficient for speed loop control;

K_viis an integral coefficient;

G_vis the output of the speed loop;

B₂＝K_fK_vpK_ip；

B₁＝K_f(K_viK_ip+K_vpK_ii)；

B₀＝K_fK_viK_ii；

A₄＝mL；

A₃＝mR+BL+mK_ip；

A₂＝BR+mK_ii+BK_ip+K_f(K_e+K_vpK_ip)；

A₁＝BK_ii+K_f(K_ipK_vi+K_vpK_ii)；

A₀＝K_fK_viK_ii；

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms.

Step 2, controlling a kernel function G based on solidification_V(s) establishing a position loop transfer function controlled by PID, specifically comprising the following steps:

the PID control expression for the position loop is:

establishing a transfer function G using a PID control position loop_PID(S) is:

wherein the content of the first and second substances,

K_ppis the proportionality coefficient of the position loop;

K_piis the integral coefficient of the position loop;

K_pdis the differential coefficient of the position loop;

G_pis the position loop output;

D₄＝K_ppB₂；

D₃＝K_ppB₂+K_pdB₁；

D₂＝K_ppB₁+K_piB₂+K_pdB₀；

D₁＝K_ppB₀+K_piB₁；

D₀＝K_piB₀；

C₆＝A₄；

C₅＝A₃；

C₄＝A₂+K_PdB₂；

C₃＝A₁+K_ppB₂+K_pdB₁；

C₂＝A₀+K_ppB₁+K_piB₂+K_pdB₀；

C₁＝K_ppB₀+K_PiB₁；

C₀＝K_piB₀；

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms.

Step 3, establishing a transfer function G for adding velocity and acceleration feedforward control_ry(s) transfer function G for velocity and acceleration feedforward control_ry(s) specifically the following:

in the formula:

G_V(s) is a solidification control kernel function;

G_pis a position loop transfer function;

K_afis the acceleration feedforward coefficient;

K_vfis the velocity feedforward coefficient.

Step 4, establishing a transfer function added into the special disturbance observer, which specifically comprises the following steps:

according to the principle of the disturbance observer, the disturbance observer is established, and the method specifically comprises the following steps:

setting u as a reference input with a basic model of a disturbance observer, and the value is the obtained velocity acceleration look-ahead K_afs²、K_vfs²And position loop control output G_PID(s) sum, d is system low-frequency interference, xi is system high-frequency interference, Q(s) is filter, G_V(s) for the established control kernel model,is a control object nominal model G_nThe inverse of(s), y is still the system output, and in practical application, it is usually desirable to control the kernel simulation model to be very close to the nominal model, i.e. G_n(s)＝G_V(s)

According to the established disturbance observer, establishing a transfer function G from a reference input u to a system output y_uy(S), the transfer function is as follows:

according to the established interference observer, establishing a transfer function G from the low-frequency interference d of the system to the output y of the system_dy(S), the transfer function is as follows:

according to the established interference observer, a transfer function G from the high-frequency interference xi of the system to the output y of the system is established_ξy(S), the transfer function is as follows:

wherein the content of the first and second substances,

G_V(s) is a solidification control kernel function;

G_n(s) is a control object nominal model; g_n(s)＝G_V(s)；

Xi is system high frequency interference;

y is the system output;

q(s) is the filter transfer function; q(s) is a binomial low pass filter, expressed as follows:

wherein the content of the first and second substances,

tau is a time constant and is the only adjustable parameter of Q(s);

and S is a complex space variable after Laplace transformation, and the physical meaning of the complex space variable is corresponding to real space time with the unit of ms.

Step 5, integrating the speed acceleration feedforward, the position loop PID control and the disturbance observer model control, and establishing a displacement output y expression as follows:

wherein the content of the first and second substances,

G_V(s) is a solidification control kernel function;

G_n(s) is a control object nominal model; g_n(s)＝G_V(s)；

G_pIs a position loop transfer function;

K_afis the acceleration feedforward coefficient;

K_vfis a velocity feedforward coefficient;

K_ppis the proportionality coefficient of the position loop;

K_piis the integral coefficient of the position loop;

K_pdis the differential coefficient of the position loop;

s is a complex space variable after Laplace transformation, the physical meaning of the complex space variable is corresponding to real space time, and the unit is ms;

q(s) is the filter transfer function;

with the transfer function G established_V(s) controlling as a function of the solidification control kernel.

The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.