JP2588202B2 - PID control device - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a PID control device having two types of control operation units in a positive direction and a negative direction.

(Prior Art) Conventionally, a PID control device provided with two kinds of control functions of a positive direction and a negative direction with respect to a set value, and makes a decision on selection of these controls according to the sign (positive or negative) of the deviation value It is provided (for example, see JP-A-60-163101).

(Problems to be Solved by the Invention) However, in this type of conventional PID control device, when the measured value repeats up and down near the set value during the transient response, the output in the positive direction and the output in the negative direction are repeatedly output alternately. As a result, vibration occurs. In addition, it was necessary to set appropriate PID parameters manually.

(Means for Solving the Problem) The PID control device of the present invention calculates a proportional term, an integral term, and a derivative term based on a forward PID calculation formula using a PID coefficient suitable for the forward calculation. A positive-direction PID control calculation unit that outputs a result of addition of the calculated terms as a calculation value;
When the operation value output from the ID control operation unit is positive, the positive direction judgment unit that outputs the operation value as a control amount to the control target, and is suitable for negative direction operation for each of the proportional, integral, and differential terms -Direction PI that outputs the addition result of each term calculated based on the negative-direction PID calculation formula using the calculated PID coefficient as a calculation value
A D control operation unit; and a negative direction determination unit that outputs the operation value as a control amount to the control target when the operation value output from the negative direction PID control operation unit is negative.

Further, the PID control device of the present invention, in the above configuration,
PID suitable for positive direction calculation based on two types of dead time and amplitude values measured from the waveform at the time of positive direction output and the waveform at the time of negative direction output obtained by controlling the control target on / off.
The coefficient and the PID coefficient suitable for the calculation in the negative direction are calculated respectively.

(Operation) The positive direction PID control calculation unit calculates the positive direction PID calculation by using a proportional term,
PID suitable for positive direction operation for each of the integral and derivative terms
The calculation is performed based on a positive direction PID calculation formula using coefficients (proportional gain, integration time, and differentiation time), and the addition result of each calculated term is output as a calculation value. The negative-direction PID control calculation unit performs a negative-direction PID calculation using a PID coefficient (proportional gain, integration time, and differentiation time) suitable for the calculation in the negative direction for each of the proportional term, the integral term, and the derivative term. The calculation is performed based on the direction PID calculation expression, and the addition result of the calculated respective terms is output as a calculation value.

When the operation value output from the forward direction PID control operation unit is positive, the forward direction determination unit outputs the operation value to the control target as a control amount. In addition, when the operation value output from the negative direction PID control operation unit is negative, the negative direction determination unit outputs the operation value to the control target as a control amount.

That is, the PID control device of the present invention independently operates the PID operation results in the positive and negative directions (that is, the operation results in the positive and negative directions do not affect the output in each direction). This determines the positive output and the negative output.

In addition, a PID coefficient suitable for the calculation in the positive direction and a PID coefficient suitable for the calculation in the negative direction are separately calculated. That is, the proportional gain, the integration time, and the differentiation time suitable for the calculation in the positive direction are two types of waste measured from the waveform at the time of the positive output and the waveform at the time of the negative output, which are obtained by turning on and off the controlled object. It is calculated based on the time and the amplitude value. The integral time and derivative time suitable for the negative direction calculation are two types of dead time and amplitude measured from the waveform at the time of positive output and the waveform at the time of negative output, which are obtained by turning on and off the control target. It is calculated based on the value. Further, the proportional gain suitable for the calculation in the negative direction is calculated from the ratio between the ON time and the OFF time in the positive direction.

As described above, since the optimal values of the PID coefficients in the positive and negative directions are automatically set independently of each other, the optimal control is always performed.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing a configuration of a PID control device according to the present invention.

The PID control device 1 controls the control target 3 measured by the detector 2.
The positive direction of the output ratio y _{n +} and negative for the control target 3 in accordance with a predetermined PID calculation expression to eliminate the deviation e _n between the measurement data (actually measured value x _n) and the set value X ₀ to a preset corresponding to the operation state of the This is a device for calculating and outputting the output rate y _{n- in the} direction. Set value x
The setting of ₀ is made by the setting block 4 and the setting value x
₀ the deviation e _n between the actually measured value x _n and is input to the control block 5 for PID calculation. In the control block 5, the output rates y _{n +} and y _n− are calculated using the PID parameters determined by the coefficient setting block 6 and output to the control target 3. This PID parameter is a value identified in a limit cycle that switches between on and off at the timing when the measured value reaches the set value during the operation of the control target 3. The identification means includes an identification execution block 7 for instructing a switch between a normal PID control and an identification, an identification block 8 for executing a calculation process required for the identification, and a PID parameter determined by the identification result. The control block 5 is constituted by the coefficient setting block 6 for setting a control operation expression.

2 and 3 are block diagrams showing the internal configuration of the control block 5. FIG. 2 shows the state of PID control, and FIG. 3 shows the state (limit cycle) of on / off control at the time of identification. ing.

At the time of the PID control shown in FIG.
a and the following formulas (1) and (2)
The output rate y _{n +} in the positive direction calculated according to the following equation (3),
The output rate y _n− in the negative direction calculated according to (4) is output.

Forward PID control Negative direction PID control However, y _n = output index e _n = deviation k _P = proportional gain T _I = integral time T _D = derivative time T _e = calculation cycle (subscript + - positive direction, a negative direction) In this example, two integration time by the sign of the value Sigma] e _n of the integration buffer T _{I +,} positive PID control T _I-, negative PI
Cases are classified in D control. That is, in the forward direction PID control, the sign of the value of the integral buffer, T _{I +} (positive direction integration time) to the integration time constant when the Σe _n ≧ 0, Σe _n <
If 0 using T integration time constant _I- (negative direction integration time), the negative PID control, the sign of the value of the integral buffer, T _{I +} to the integration time constant when the Sigma] e _n ≧ 0 ( positive integration time), T the integration time constant when the Sigma] e _n <0
_I- (Negative integration time) is used. in this way,
By using different integration time constants depending on the value of the integration buffer, the transition of the integration term in the positive or negative PID control operation can be performed when the output changes from the positive output to the negative output and vice versa. Delivery
This can be performed smoothly without causing an abrupt change in output.

The output rate y _{n +} is output and given to the control target 3 if its value is positive in the positive direction determination block 11a, and the output rate y _n- is output and output if its value is negative in the negative direction determination block 11b. Set value for each object 3
PID control is performed so as to eliminate the deviation e _n between x ₀ and the measured value x _n .

As described above, the PID control device of the present invention independently calculates the positive and negative operation results (that is, the positive and negative operation results do not affect the output in each direction). The output rate y _{n +} in the positive direction and the output rate y _{n- in the} negative direction are determined.

On the other hand, in the on-off state (limit cycle) at the time of identification shown in FIG.
In a, y _{n +} of the following equation (5) is output, and y _n− of the following equation (6) is output in the negative-direction on / off control block 12b and provided to the control target 3.

FIG. 4 shows an example of the waveform of the measured value _xn observed in this limit cycle.

Hereinafter, a method of identifying the PID parameter will be described with reference to this waveform.

From the waveform of the measured values x _n shown in FIG. 4, the dead time L ₁ and the inclination R ₁ when the control amount in the form of the output ratio y _{n +} or the y _n-is shifted from OFF to ON on the control target 3, the dead time L ₂ and the inclination R ₂ when shifted from oN to oFF, to measure the amplitude a. At this time, the following equation (7) holds.

A = R ₁ L ₁ + R ₂ L ₂ (7) = RL (8) where L is a weighted average in consideration of the degrees of contribution of the _two dead times L ₁ and L _{2 to} the control system. Is the combined dead time calculated by the following equation (9).

Then, the values of a and b are changed according to the importance of L ₁ and L ₂ . Generally, a = b = 1 Is used.

Thus, using the arithmetic expressions of the above expressions (8) and (10), the optimal adjustment expression of Ziegler-Nichols is given by the following expressions (11) to
It is represented by (13).

k _{P +} = 1.2 / RL (11) T _I = 2 L (12) T _D = 0.5 L (13) Here, the above equation (8) is substituted into (11), and k _{P +} = 1.2 / A ... (11) 'is obtained.

Further, since the ratio of the proportional gain matches the ratio of the positive ON time T _ON and the OFF time T _{OFF in the ON /} OFF control, k _P− / k _{P +} = T _OFF / T _ON (14) _P− = T _OFF / T _ON × k _{P +} (15) is obtained.

In the above embodiment, the synthetic dead time L is set to
By fitting optimally adjustable Nichols, but it calculates the positive and negative directions of the PID coefficients as the same value, the time of operation in the positive direction of the PID coefficients using dead time L ₁ in the positive direction, the negative direction of the time calculation of PID coefficients using the negative direction of the dead time L _2, integral time and derivative time of the positive and negative directions of the may be calculated separately. In this case, the calculation coefficient 2 for the integration time of Ziegler-Nichols and the calculation coefficient 0.5 for the differentiation time are changed.

(Effects of the Invention) As described above, according to the present invention, the positive-direction PID control calculation unit performs the positive-direction PID calculation on the proportional term, the integral term, and the derivative term, each of which is suitable for the positive-direction calculation. The calculation is performed based on a forward PID calculation formula using coefficients, and the result of addition of the calculated terms is output as a calculation value. In addition, the negative direction PID control calculation unit converts the negative direction PID calculation into PID coefficients (proportional gain, proportional gain,
(Integral time, derivative time) based on a negative PID calculation formula, and outputs the addition result of the calculated terms as a calculation value. When the operation value output from the positive direction PID control operation unit is positive, the positive direction judgment unit outputs the operation value to the control target as a control amount.
When the calculation value output from the control calculation unit is negative, the calculation value is output to the control target as a control amount. The positive output and the negative output can be determined (without the positive and negative operation results providing the output in each direction). Therefore, even if the measured value repeats up and down in the vicinity of the set value, no vibration occurs because the outputs in the positive and negative directions are not repeatedly output.

Since the PID coefficient suitable for the calculation in the positive direction and the PID coefficient suitable for the calculation in the negative direction are calculated separately, optimal control can be performed in any state.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of a PID control device according to the present invention, FIGS. 2 and 3 are block diagrams showing an internal configuration of a control block, and FIG. 2 is a diagram showing a state of PID control. FIG. 3 is a diagram showing a state at the time of identification, and FIG. 4 is a diagram exemplifying a waveform of an actually measured value at the time of identification. 1. PID control device 2. Detector 3. Control target 6. Coefficient setting block 8. Identification block

 ──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-62-111311 (JP, A) JP-A-60-163101 (JP, A) JP-A-60-168201 (JP, A) JP-A 48-48 61875 (JP, A)

Claims (2)

(57) [Claims] A positive output which outputs, as an operation value, an addition result of each term calculated based on a positive direction PID operation formula using a PID coefficient suitable for a positive direction operation for each of a proportional term, an integral term, and a differential term. A direction PID control operation unit, a positive direction determination unit that outputs the operation value as a control amount to a control target when the operation value output from the positive direction PID control operation unit is positive, a proportional term, an integral term, and a derivative. A negative direction PID control operation unit that outputs, as an operation value, an addition result of each item calculated based on a negative direction PID operation expression using a PID coefficient suitable for a negative direction operation for each of the terms; A PID control device comprising: a negative direction judging unit that, when a calculation value output from a calculation unit is negative, outputs the calculation value as a control amount to the control target. 2. A positive-direction calculation based on two types of dead time and amplitude values measured from a waveform at the time of a positive output and a waveform at the time of a negative output obtained by turning on and off a control target. 2. The method according to claim 1, wherein a suitable PID coefficient and a PID coefficient suitable for calculation in the negative direction are calculated.
PID controller.