JPH0934503A - Adjustment method for pid controller - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide the adjustment method which can easily decide the PID parameter of optimum value to design a stable control system and also can automatically adjust a PID controller by the simulation of computer application by adjusting even the gain of an integration term when the PID parameter is adjusted by the critical sensitivity method. SOLUTION: When the PID parameter is decided by the critical sensitivity method the integration time and the differentiation time are set at the maximum value (T1 =∞) and the minimum value (TD=0) respectively and the proportional gain Kp is increased up to its maximum value from the minimum value in a normal application range. If the difference (y) is negative between the 1st and 2nd peaks, the sustained oscillation is decided. Thus a 1st step is prepared to set the proportional gain at Kpm of the relevant time point together with a 2nd step where the integration time is reduced down to its minimum value from the maximum value in the normal application range and then set at T1 M that produces the least error against the input with the porportional gain and the differentiation time are set at 0.2Kpm and the minimum value (TD=0) respectively. Then the differentiation time TD is set at the value obtained by multiplying the value T1 M by a certain coefficient.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for adjusting a PID controller using a limit sensitivity method.

[0002]

2. Description of the Related Art As is well known, a PID controller (controller; also called a controller) that combines three basic operations of proportional operation (P operation), integral operation (I operation), and derivative operation (D operation) is known. Since it is possible to improve steady state characteristics and quick response at the same time, it is widely used in various machines and equipment, plants and the like. At that time, P to ensure the best control performance
It is necessary to adjust the ID parameters (proportional gain Kp, integration time T _I , differential time T _D, etc.). The limiting sensitivity method is often used as the method for adjusting the PID controller. The limit sensitivity method is the steady oscillation state of the controlled variable (sustained oscillation) with the integral time set to the maximum value and the derivative time set to the minimum value to make the P operation practically only and the proportional gain gradually increased.
And the proportional gain Kpc and the vibration period Tc at this time are generated.
From the above, the PID parameter is determined according to the following equation.

Kp = Kpc / 1.7 T _I = 0.5Tc T _D = 0.125Tc

[0004]

However, when the PID parameter is determined by such a method, when the value is actually used in the plant, the proportional gain becomes a large value as compared with the case where it is adjusted by a person. The response could be oscillatory. This is because if the plant has complex characteristics,
This is because it is difficult to determine at what point in time the continuous vibration occurred.

Therefore, an object of the present invention is to solve the above problems and provide a method of adjusting a PID parameter which enables a stable control system to be designed.

[0006]

According to the present invention, when the PID parameter is determined by using the limit sensitivity method, the integration time is set to the maximum value (T _I = ∞) and the differentiation time is set to the minimum value (T _D = 0). Then, the proportional gain Kp is increased from the minimum value to the maximum value (for example, 0.05 to 25) of the normal use range, and if the difference between the first peak and the second peak is negative, continuous vibration is generated. The first step of determining the proportional gain to the value Kpm at that time, the proportional gain to 0.2 Kpm, the derivative time to the minimum value (T _D = 0), and the integral time to the maximum value in the normal use range. from to the minimum value (e.g., from 20 seconds to 1 second) will decrease, and two-step adjustment of the second stage of determining the value T _IM when the error is smallest with respect to the input of the integral time, derivative time T _{D Is} a coefficient in T _IM (for example, 1 /
20 = 0.05).

[0007]

1 and 2 show an embodiment of the present invention. FIG. 1 is a flowchart for obtaining PID parameters, and FIG. 2 is a control block diagram. In FIG. 2, 1
Is a PID controller, 2 is a plant to be controlled, Kp
Is proportional gain, T _I is integration time, T _D is derivative time, Gp
(S) is the transfer function of the plant 2.

The PID parameter is obtained by the procedure of the flowchart shown in FIG. That is, first, the proportional gain Kp is increased. At this time, the integration time T _I is the maximum and the differentiation time T _{D is}
Is set to the minimum (T _D = 0) (step S1). In the process of increasing the proportional gain Kp from 0.05 to 25 (from the minimum value to the maximum value in the normal use range), it is determined whether or not continuous vibration has occurred (S2). The criterion is shown in Fig. 3.
The difference y between the first and second peaks becomes negative as shown in, and the state is regarded as continuous oscillation. The proportional gain Kp is determined to the value Kpm at that time.

After that, the gain K _{I of the} integral term is adjusted.
For that purpose, the proportional gain Kp is set to Kp = 0.2 Kpm (S3). Also, the differential time T _D is the minimum (T _D = 0)
And In this state, the integration time T _I is reduced from 20 seconds to 1 second (from the maximum value to the minimum value in the normal use range) (S4). That is, the gain K _{I of the} integral term is changed. The value of the integration time T _I when the error with respect to the input during that time is the smallest is set to T _IM (S5 to S8).

After the two-step adjustment, the PID parameter is determined to the following value (S9).

Proportional gain Kp = Kpm Integration time T _I = T _IM Derivative time T _D = T _IM · A (A is a coefficient, and is selected to be about 1/20 = 0.05 or less. The smaller this coefficient is, , D movement is less involved, and PI movement is the main control.) Figures 4 and 5 show the vibration waveform during adjustment. 4 shows the case where the proportional gain Kp is being adjusted, and FIG. 5 shows the case where the gain K _{I is being} adjusted.

The above procedure (flow chart of FIG. 1)
Is running in a computer simulation. That is, automatic tuning is performed.

[0013]

As described above, according to the present invention, when the PID parameter is adjusted by the limit sensitivity method, the gain of the integral term is also adjusted in two steps, so that the parameter can be easily determined to the optimum value. It becomes possible to design a stable control system. Moreover, it can be automatically adjusted by computer simulation.

[Brief description of drawings]

FIG. 1 is a flowchart showing an embodiment of the present invention.

FIG. 2 is a control block diagram showing an embodiment of the present invention.

FIG. 3 is a waveform diagram illustrating criteria for determining continuous vibration according to an embodiment of the present invention.

FIG. 4 is a graph during Kp adjustment showing an embodiment of the present invention.

FIG. 5 is a graph during K _I adjustment showing one embodiment of the present invention.

[Explanation of symbols]

1 ... PID controller 2 ... controlled object (plant) Kp ... proportional gain T _I ... integral time T _D ... derivative time

Claims (1)

[Claims] 1. When determining a PID parameter using a limiting sensitivity method, the proportional gain Kp is usually set by setting the integration time to a maximum value (T _I = ∞) and the differentiation time to a minimum value (T _D = 0). Increase from the minimum value to the maximum value of the usage range of 1
If the difference between the 2nd peak and the 2nd peak is negative, it is considered as continuous oscillation, and the first step of determining the proportional gain to the value Kpm at that time, the proportional gain to 0.2 Kpm, and the derivative time to the minimum value (T _D = 0), the integration time is decreased from the maximum value to the minimum value in the normal use range, and the integration time is determined to the value T _IM when the error with respect to the input is the smallest. And the differential time T _D is determined to be a value obtained by multiplying T _IM by a certain coefficient, and the adjustment method of the PID controller.