JP6350420B2 - Learning type tension controller - Google Patents

Description

  The present invention relates to a learning type tension control device for a process line.

  FIG. 2 is a diagram illustrating a configuration example in which a tension control device for a non-ferrous metal process line is applied to a loop car. 1 is a strip made of a metal material (ferrous and non-ferrous metal), 2 is a bridle roll, and 4 is a loop car. The loop car 4 is driven by a motor 5. A tension detector 6 measures the tension of the strip 1.

The target tension T _AIM given to the strip 1 is converted into the target torque Q _AIM by the tension / torque conversion circuit 7. The mechanical torque Q _L and the acceleration / deceleration torque Q _F are added to the target torque Q _AIM and introduced into the torque control circuit 8 as a torque reference Q _REF . The torque control circuit 8 controls the torque of the motor 5 via a power converter (not shown) based on the torque reference Q _REF .

The tension deviation ΔT between the target tension T _AIM and the actual tension _TFBK detected by the tension detector 6 is introduced into the amplifier 11, performs a proportional / integral calculation and outputs a correction signal, and the tension deviation ΔT is zero. It is controlled to become.

  In addition, the applicant has recognized patent document 1 as a thing relevant to this invention. Patent Document 1 discloses a tension control device that does not require feedforward compensation.

Japanese Patent No. 4904939

The mechanical loss torque Q _L described above is a mechanical loss torque such as friction between the loop car 4 and the motor 5, and is stored in the function table for each speed reference SP _REF . Furthermore, acceleration and deceleration torque Q _F described above are torque required at the time of acceleration or deceleration is calculated based on the sum J of the moment of inertia of Rupuka 4 and the motor 5 (mechanical inertia). In the above-mentioned tension control device, Mekarosutoruku Q _L and the moment of inertia J is stored as a constant. Therefore, even if it matches the actual product at the beginning, the mechanical loss torque compensation amount and the acceleration / deceleration torque compensation amount are actually increased when the machine changes with time (that is, the mechanical loss increases due to mechanical wear or the inertia decreases due to roll wear). The deviation gradually increases with respect to the appropriate value required for. As a result, under-compensation or over-compensation was caused, and the amount corresponding to the difference from the target tension _TAIM was a factor that deteriorated the tension control accuracy.

  As described above, the tension deviation ΔT is introduced into the amplifier 11 and is subjected to proportional / integral calculation to output a correction signal and controlled so that ΔT becomes zero. However, a delay occurs due to the feedback control system. Not enough.

  The present invention has been made to solve the above-described problems, and is a learning type that can reduce an error expansion due to a change with time of mechanical loss torque and acceleration / deceleration torque used for tension control and can reduce deterioration of tension control accuracy. An object is to provide a tension control device.

In order to achieve the above object, the present invention is a learning type tension control device used in a process line including a roll for conveying a material and a motor for driving the roll,
A tension detector for detecting the actual tension of the material;
Mechanical loss compensation circuit that outputs mechanical loss torque according to the speed reference;
An acceleration / deceleration compensation circuit that outputs acceleration / deceleration torque based on the moment of inertia of the roll and the motor and the speed reference;
The corrected target tension is corrected to a target torque so that a tension deviation between the target tension and the actual tension becomes zero, and the target torque is converted into a target torque, the mechanical torque output from the mechanical loss compensation circuit, and the target torque A control circuit for controlling the motor based on the torque reference corrected by the acceleration / deceleration torque output from the acceleration / deceleration compensation circuit;
A conversion circuit that divides the tension deviation by the actual radius of the roll to convert it into torque deviation;
When the roll is at a constant speed, the torque deviation is regarded as a mechanical loss torque error, and the mechanical loss torque is updated based on the mechanical loss torque error according to the speed reference stored in the mechanical loss compensation circuit. Circuit,
A calculation circuit that calculates an inertia moment error from the torque deviation and the speed reference when the roll is accelerating or decelerating;
And an inertia moment update circuit that updates the inertia moment stored in the acceleration / deceleration compensation circuit based on the inertia moment error.

  According to the learning type tension control device according to the present invention, even if the machine data changes due to the change of the machine over time in the process line, the speed reference is made based on the tension deviation between the target tension and the actual tension. Mecharos torque and moment of inertia can be updated. Therefore, according to the present invention, it is possible to reduce an error increase due to a change with time of mechanical loss torque and acceleration / deceleration torque used for tension control, and to reduce deterioration of tension control accuracy.

It is a block diagram for demonstrating the system configuration | structure which concerns on Embodiment 1 of this invention. It is a figure which shows the structural example which applied the tension control apparatus of the nonferrous metal process line to the loop car.

Embodiment 1 FIG.
[System Configuration of Embodiment 1]
FIG. 1 is a configuration diagram for explaining a system configuration according to Embodiment 1 of the present invention.
The system shown in FIG. 1 includes a process line and its tension control device. The process line includes a plurality of rolls that convey the strip 1 that is a material. The strip 1 is a metal material (ferrous and non-ferrous metal). In FIG. 1, bridle rolls 2 and 3 and a loop car 4 are depicted as a plurality of rolls. The loop car 4 is connected to the motor 5 via a gear box (not shown). The motor 5 drives the loop car 4 to rotate.

The tension control device includes a tension detector 6 that detects an actual tension (actual value of tension) of the strip 1 and a detector 17 that detects a roll radius (actual radius) of the loop car 4. The tension control device includes a basic control circuit. Control circuit converts the target tension _{T AIM} and the corrected target tension by correcting the target tension _{T AIM} as tension deviation ΔT becomes zero and the actual tension _{T FBK} to the target torque _{Q AIM,} the target torque _{Q AIM,} The motor 5 is controlled based on the torque reference Q _REF corrected by the mechanical loss torque Q _L output from the mechanical loss compensation circuit 9 and the acceleration / deceleration torque Q _F output from the acceleration / deceleration compensation circuit 10.

A specific configuration of the control circuit described above will be described. The target tension T _AIM is a target value of the tension applied to the strip 1. The target tension T _AIM is converted into the target torque Q _AIM by the tension / torque conversion circuit 7. The mechanical torque Q _L and the acceleration / deceleration torque Q _F are added to the target torque Q _AIM and introduced into the torque control circuit 8 as the torque reference Q _REF . The torque control circuit 8 controls the torque of the motor 5 via a power converter (not shown) based on the torque reference Q _REF .

Further, the tension deviation ΔT between the target tension T _AIM and the actual tension T _FBK detected by the tension detector 6 is introduced into the amplifier 11, performs proportional / integral calculation, outputs a correction signal, and the tension deviation ΔT is calculated. Feedback control is performed so that it becomes zero.

Here, the mechanical loss torque Q _L is a mechanical loss torque such as friction between the loop car 4 and the motor 5. The mechanical loss compensation circuit 9 stores a function table that defines the mechanical loss torque Q _L for each speed reference SP _REF . The mechanical loss compensation circuit 9 outputs the mechanical loss torque Q _L corresponding to the speed reference SP _REF from the function table.

Furthermore, acceleration and deceleration torque Q _F is a torque required for acceleration and deceleration, and the sum J of the moment of inertia of Rupuka 4 and the motor 5, the speed reference SP _REF by multiplying the acceleration obtained by differentiating equation (1) It is obtained and output from the acceleration / deceleration compensation circuit 10.

Even if the mechanical loss torque Q _L stored in advance in the mechanical loss compensation circuit 9 and the inertia moment J stored in advance in the acceleration / deceleration compensation circuit 10 match the actual product, the mechanical change over time (ie, mechanical wear (bearing and As the mechanical loss increases (including gearbox wear) and the inertia decreases due to roll wear, the mechanical loss torque compensation amount and acceleration / deceleration torque compensation amount gradually increase from the actual required values. Become. As a result, under-compensation or over-compensation occurs, and that amount becomes a difference from the target tension _TAIM , which causes a deterioration in tension control accuracy.

  Therefore, the tension control device according to the first embodiment of the present invention is provided with a learning function for updating mechanical loss torque compensation and acceleration / deceleration compensation in accordance with changes over time of the machine. Hereinafter, main features of the learning type tension control device according to the first embodiment of the present invention will be described.

  The tension control device includes a tension / torque conversion circuit 12 that converts a tension deviation ΔT into a torque deviation ΔQ. Specifically, the torque deviation ΔQ is calculated by dividing the tension deviation ΔT by the roll radius (actual radius) of the loop car 4. The actual radius is detected by the detector 17.

The tension controller determines whether the torque deviation ΔQ is the mechanical loss torque error ΔQ _L or the acceleration / deceleration torque error ΔQ _F from the speed reference SP _REF when the torque deviation ΔQ is detected. Prepare.

A method for identifying the mechanical loss torque error ΔQ _L will be described. When the loop car 4 is operating at a constant speed, the output of the acceleration / deceleration compensation circuit 10 is zero (dSP _REF / dt = 0), and the torque deviation ΔQ obtained by converting the tension deviation ΔT is expressed as a mechanical loss torque error ΔQ _L. Can be considered. Therefore, the determination circuit 13 determines the torque deviation ΔQ as the mechanical loss torque error ΔQ _L in the speed reference SP _REF . Thereafter, when the roll of the loop car 4 is at a constant speed, the mechanical loss torque update circuit 15 converts the mechanical loss torque Q _L corresponding to the speed reference SP _REF stored in the function table of the mechanical loss compensation circuit 9 to the mechanical loss torque error ΔQ. Update based on _L. For example, the mechanical loss torque Q _L is updated with a value Q _L ′ obtained by adding or subtracting the mechanical loss torque error ΔQ _L to the mechanical loss torque Q _L.

Next, a method for identifying the acceleration / deceleration torque error ΔQ _F and a method for calculating the moment of inertia J will be described. When the loop car 4 changes from constant speed operation to acceleration or deceleration, the torque deviation ΔQ obtained by converting the tension deviation ΔT can be regarded as an acceleration / deceleration torque error ΔQ _F. Therefore, the determination circuit 13 determines that the torque deviation ΔQ is an acceleration / deceleration torque error ΔQ _F. When the roll of the loop car 4 is accelerating or decelerating, the J calculating circuit 14 calculates the moment of inertia error ΔJ from the equation (2) using the torque deviation ΔQ and the speed reference SP _REF at that time.

  The inertia moment update circuit 16 updates the inertia moment J stored in the acceleration / deceleration compensation circuit 10 based on the inertia moment error ΔJ. For example, the value is updated with a value J ′ obtained by adding or subtracting the inertia moment error ΔJ to the inertia moment J.

As described above, according to the tension control device according to the present embodiment, even if the machine data changes due to the change of the machine over time, the tension control device according to the present embodiment is based on the tension deviation ΔT between the target tension T _AIM and the actual tension T _FBK. Thus, the mechanical loss torque Q _L and the moment of inertia J corresponding to the speed reference SP _REF can be updated. For this reason, it is possible to reduce an increase in error due to a change with time of mechanical loss torque and acceleration / deceleration torque used for tension control, and to reduce deterioration of tension control accuracy.

(Modification)
By the way, in the system of Embodiment 1 described above, the data update method in the mechanical loss torque update circuit 15 and the inertia moment update circuit 16 is not limited to the method described above. For example, past data may be stored and updated using an average value of new data and past data. Moreover, you may update based on the average value of the number of passed coils. Moreover, you may update using the following relational expression, new data = old data (Q _L or J) x (1-α) + measurement data (Q _L 'or J') x α. Α is a weight and is set in advance.

  Moreover, in the system of Embodiment 1 mentioned above, although the material was the strip 1, the material is not limited to a metal material. The material may be paper. That is, the present invention can also be applied to a tension control device in a paper machine.

  Moreover, in the system of Embodiment 1 mentioned above, although the roll was the roll of the loop car 4, it is not limited to the roll of the loop car 4. Any roll driven by a motor may be used.

  In the first embodiment described above, the roll of the loop car 4 is the “roll” in the present invention, the basic control circuit described above is the “control circuit” in the present invention, the “tension / torque conversion circuit 12”. The J calculation circuit 14 corresponds to the “conversion circuit” in the present invention and the “calculation circuit” in the present invention.

DESCRIPTION OF SYMBOLS 1 Strip 2, 3 Bridle roll 4 Loop car 5 Motor 6 Tension detector 7 Tension / torque conversion circuit 8 Torque control circuit 9 Mechanical loss compensation circuit 10 Acceleration / deceleration compensation circuit 11 Amplifier 12 Tension / torque conversion circuit 13 Judgment circuit 14 J calculation circuit 15 Mechanical loss torque update circuit 16 Inertia moment update circuit 17 Detector Q _AIM target torque Q _F acceleration / deceleration torque Q _L mechanical loss torque Q _REF torque reference SP _REF speed reference T _AIM target tension T _FBK actual tension ΔJ Inertia moment error ΔQ Torque deviation ΔQ _F Acceleration / deceleration torque error ΔQ _F Inertia moment error ΔQ _L Mecharos torque error ΔT Tension deviation

Claims (1)

A learning type tension control device used in a process line comprising a roll for conveying a material and a motor for driving the roll,
A tension detector for detecting the actual tension of the material;
Mechanical loss compensation circuit that outputs mechanical loss torque according to the speed reference;
An acceleration / deceleration compensation circuit that outputs acceleration / deceleration torque based on the moment of inertia of the roll and the motor and the speed reference;
The corrected target tension is corrected to a target torque so that a tension deviation between the target tension and the actual tension becomes zero, and the target torque is converted into a target torque, the mechanical torque output from the mechanical loss compensation circuit, and the target torque A control circuit for controlling the motor based on the torque reference corrected by the acceleration / deceleration torque output from the acceleration / deceleration compensation circuit;
A conversion circuit that divides the tension deviation by the actual radius of the roll to convert it into torque deviation;
When the roll is at a constant speed, the torque deviation is regarded as a mechanical loss torque error, and the mechanical loss torque is updated based on the mechanical loss torque error according to the speed reference stored in the mechanical loss compensation circuit. Circuit,
A calculation circuit that calculates an inertia moment error from the torque deviation and the speed reference when the roll is accelerating or decelerating;
An inertia moment update circuit that updates the inertia moment stored in the acceleration / deceleration compensation circuit based on the inertia moment error;
A learning type tension control device comprising:

Images