JPH0772902A - 2-DOF PID adjustment device - Google Patents

Abstract

(57) [Summary] [Object] The present invention makes the integration operation two degrees of freedom simple and eliminates mutual interference of the two degrees of freedom coefficient, thereby making it possible to intuitively grasp the influence of adjustment. It is in. [Configuration] Control amount PV _n from control target and target value SV _n
PI or PID (P: proportional, I: so that the deviation from the effective target value SV _n0 obtained by performing the compensation calculation from
Integral, D: derivative) adjustment operation is performed, and the obtained adjustment operation signal is applied as the operation signal MV _n to the controlled object 7-1, in order to obtain the effective target value from the target value of the controlled variable, Gain 2 degrees of freedom coefficient α (α: 0 to 1
Constant) and integration time two-degree-of-freedom coefficient β (β: 0 to
2 is used, and the transfer function C _sv (s) between the target value and the operation signal is C _sv (s) = K _p.
Let {α + 1 / [T _I · s (1 + βT _I · s)]} 2
It is a PID adjustment device with a degree of freedom. However, K _p : proportional gain, T _I : integration time, s: Laplace operator.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a two-degree-of-freedom PI or two-degree-of-freedom PID (P: proportional, I: integral, D: derivative) adjusting device used in various process instrumentation control systems and the like. In particular, it relates to an improvement of a two-degree-of-freedom PID adjustment device that optimizes both the disturbance suppression characteristic and the target value tracking characteristic at the same time.

[0002]

2. Description of the Related Art PI or PID control devices have been widely used in all industrial fields since the history of control, and control systems in each industrial field are no longer available without PI or PID control devices.

Various adjusting calculation systems have been adopted for conventional adjusting devices, but with the passage of time, the analog adjusting calculation system has been changed to a digital adjusting calculation system.
The trend will remain the same in the future, and it will be the basis of the plant operation control system.

The basic formula of this PID adjustment calculation is represented by the following formula. MV (s) = K _c {1+ (1 / T _I · s) + [(T _D · s) / (1 + η · T _D · s)]} × E (s) (1) where (1 ), MV (s): operation signal, E (s):
Deviation signal, K _c : proportional gain, T _I : integration time, T _D :
Derivative time, s: Laplace operator, (1 / η): differential gain.

By the way, the PID of the equation (1) is called a one-degree-of-freedom PID adjustment method, and only one set of PID parameters can be set. However, in an actual control system, the values of the disturbance suppression optimum PID parameter and the target tracking optimum PID parameter are greatly different, and if the PID parameter is adjusted so as to optimize the disturbance suppression characteristic, the target value tracking characteristic is greatly overshot. Then, the characteristic becomes oscillatory, and conversely, if the target value tracking characteristic is optimized, the disturbance suppression characteristic deteriorates. That is, these two characteristics cannot be optimized at the same time, and there is a trade-off relationship.

Therefore, in this type of PID adjusting device, the appearance of a technique capable of simultaneously optimizing the disturbance suppression characteristic and the target value tracking characteristic has been desired. However, in 1963, Issac M. Two Degrees of Freedom PID algorithm allows two sets of PID parameters to be independently set by Horowits.
Algorithm: The basic concept of 2 DOF PID) has been announced.

Thereafter, such 2DOF PID has begun to be put into practical use, and has recently made a great contribution to the sophistication of plant operation control systems. The two degrees of freedom up to now are two degrees of freedom of proportional (P) adjustment calculation operation and integral (I) adjustment calculation operation.
The configuration of the ID adjustment device will be described with reference to FIG.

FIG. 5 is a diagram showing the functional blocks of the most advanced one of the conventional 2DOF PID adjustment devices in which the two terms of PI have two degrees of freedom. The adjusting calculation method of this adjusting device has a configuration in which a compensation calculating means is inserted into the target value of the measured value differential preceding one-degree-of-freedom PID so that the second term of PI has two degrees of freedom.

Specifically, the target value SV _n is (1 + αβT
_It is led to the target value filter means 1 having a transfer function of _I · s) / (1 + βT _I · s), and the target value calculation signal obtained here is introduced into the deviation calculation means 2. In this deviation calculating means 2, the deviation between the target value calculation signal and the control amount PV _n detected by the control amount detecting means 8 from the controlled object 7-1 is obtained and then applied to the PI adjusting means 3, where The obtained PI adjustment calculation output is sent to the subtracting means 4. Further, the control amount PV _n detected by the control amount detecting means 8 is sent to the D (differential) adjusting means 5, where the D adjusting operation output is obtained by the differential adjusting operation, and then is similarly sent to the subtracting means 4. It

In the subtraction means 4, the D adjustment operation output is subtracted from the PI adjustment operation output, and the subtraction output is multiplied by the proportional gain K _p in the proportional gain means 6 and applied to the process 7 as the operation signal MV _n. , Target value SV _n = control amount PV _n
It is controlled so that

[0011]

Therefore, in the adjusting device as described above, two degrees of freedom of the proportional gain and the integration time are considered, but when actually expressed by the transfer function of the control system, the following is obtained. It looks like this. In other words, PV → MV
The transfer function between SV and MV is C _pv (s), and the transfer function between SV and MV is C
_{If sv} (s), then C _pv (s) = MV _n / PV _n = -K _p {1 + 1 / (T _I · s) + (T _D · s) / (1 + ηT _D · s)} (2 ) C _SV (s) = MV _n / SV _n = K _P {α + [1 / (T _I · s) − (1−α) (β−1) / (1 + βT _I · s)]} (3) It is represented by. Where α is a proportional gain two-degree-of-freedom coefficient,
β is an integration time two-degree-of-freedom coefficient. Therefore, this FIG.
The response equation is calculated from: PV _n = {[C _SV (s) · G _p (s)] / [1-C _pv (s) · G _p (s)]} · SV _n + {G _p (s ) / [1-C _pv (s) · G _p (s)]} · D _n (s) (4), the former part of this equation (4) is the term for the SV change, and the latter part is the disturbance. This is a term relating to D.

Therefore, as is clear from the above equation, from the above equation (4), K _P ,
After determining T _I and T _D , if α and β are determined based on the equations (4) and (3) so that the target value tracking characteristics are optimal, the proportional action and the integral action will be 2 The degree of liberalization can be achieved.

However, although the integral operation occupies a very important position in the PID operation, as is clear from the above equation (3), (1) a mechanism for providing two degrees of freedom in the integral operation. Is complicated,
It is difficult to understand. (2) In addition to the integration time two-degree-of-freedom coefficient β, the proportional gain two-degree-of-freedom coefficient α is involved in the two-degree-of-freedom integration operation, and the two coefficients interfere with each other and are extremely adjusted. Is difficult. (3) Furthermore, since the integration time 2 liberalization coefficient is included in the denominator and the numerator, it is difficult to intuitively understand the effect of varying β, and it is very complicated.

In order to widely spread the two-degree-of-freedom PID control and improve the controllability of the plant, these problems must be overcome by all means. The present invention has been made in view of the above circumstances, and completely separates the two degrees of freedom of the proportional operation and the two degrees of freedom of the integral operation, and makes the two degrees of freedom of the integral operation simple and easy to understand. It is an object of the present invention to provide a two-degree-of-freedom PID adjustment device that facilitates adjustment of the two-degree-of-freedom coefficient, and particularly makes it easy to intuitively understand the adjustment of the integration-time two-degree-of-freedom coefficient.

[0015]

In order to solve the above-mentioned problems, the inventions according to claims 1 and 2 are effective obtained by performing compensation calculation from a controlled variable from a controlled object and a target value of this controlled variable. PI or PI so that the deviation from the target value becomes zero
D (P: proportional, I: integral, D: derivative) adjustment calculation is performed,
In an adjusting device that applies the obtained adjustment signal as an operation signal to the controlled object, in order to obtain an effective target value from the target value of the controlled variable, a two-degree-of-freedom coefficient α of proportional gain and two-degree-of-freedom integration time are obtained. A two-degree-of-freedom PID adjusting device provided with compensation calculation means using a coefficient β, and configured such that a transfer function C _sv (s) between the operation signals from the target value is expressed by the following equation.

C _sv (s) = K _p {α + 1 / [T _I · s (1 + βT _I · s)]} (5) Then, as the compensation calculation means, H (s) = α {1-1 / (1 + T _I · s)} + 1 / [(1 + T _I · s) (1 + βT _I · s)]} (6) The transfer function H (s) is obtained, and the obtained effective target value is measured. It is used as a target value of the value differential preceding type PID adjusting means.

Next, the invention according to claim 3 uses the control amount from the controlled object and the effective target value obtained by performing the compensation calculation from the target value of this control amount, to determine PI or PID.
(P: Proportional, I: Integral, D: Derivative) Adjusting calculation is performed, and an obtained adjusting signal is applied as an operation signal to the control target in an adjusting device. In order to obtain the effective target value which is the target value of the means, F (s) = 1 / {(1 + T _I · s) (1 + βT _I · s)} (7) Compensation calculation with the transfer function F (s) A two-degree-of-freedom PID adjusting device for adding and synthesizing a unit proportional adjustment signal of a target value obtained by multiplying the target value of the control amount by a two-degree-of-freedom coefficient α of a proportional gain to the measured value differential adjusting means. is there.

Further, the invention according to claim 4 uses the control amount from the controlled object and the effective target value obtained by performing the compensation calculation from the target value of this control amount, so as to use PI or PID.
(P: Proportional, I: Integral, D: Derivative) In an adjusting device that performs adjustment calculation and applies the obtained adjustment signal to the control target as an operation signal, I-PD from the target value of the control amount.
In order to obtain an effective target value which is the target value of the adjusting means, E (s) = 1 / (1 + βT _I · s) (8) Compensation calculation means having a transfer function E (s) is provided, and
A unit proportional adjustment signal obtained by subtracting the control amount from a value obtained by multiplying the target value of the control amount by a two-degree-of-freedom coefficient α of proportional gain,
It is a two-degree-of-freedom PID adjusting device that performs additive synthesis on the I-PD adjusting means.

However, in the above equations (5) to (8), K _p : proportional gain, T _I : integration time, s: Laplace operator, α: proportional gain two-degree-of-freedom coefficient (between 0 and 1) Constant), β: two-degree-of-freedom coefficient of integration time (a constant between 0 and 2).

[0020]

Therefore, in the inventions corresponding to claims 1 to 4, the two degrees of freedom coefficients α and β are completely separated by taking the above means, and either one of the coefficients is The other does not change even if changed, and since the coefficient β is inserted only in the denominator, the two-degree-of-freedom coefficient β for integration time
When you adjust the, you can intuitively grasp.

[0021]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS First, the basic structure of the device of the present invention is described.
Embodiments of the invention relating to the present invention will be described. This device is designed to solve the above-mentioned conventional problems by using a proportional gain 2
It is premised that the degree-of-freedom coefficient α and the two-degree-of-freedom integration time coefficient β do not interfere with each other, whereby α and β can be independently varied without being influenced by each other.

Further, the integration time two-degree-of-freedom coefficient β is α
Even if it is completely independent from, if β itself exists in the numerator and denominator, the variable result by β cannot be grasped intuitively.

Therefore, the present inventor pursued the ideal form while accumulating various experiments including whether or not it is sufficiently applicable to the control system while taking the above points into consideration. It has been found that a configuration having a transfer function as shown below as the transfer function Csv (s) of is ideal.

[0024] _{C sv (s) = K p} {α + 1 / [T I · s (1 + βT I · s)]} ... (9) where, K _p: proportional gain, T _I: integral time, s: Laplace operator , Α: two-degree-of-freedom coefficient of proportional gain (constant between 0 and 1), β: two-degree-of-freedom coefficient of integration time (constant between 0 and 2).

Therefore, with the configuration having the transfer function of the equation (9), the two-degree-of-freedom coefficients α and β are completely separated from each other, and even if α or β is changed, the other is completely changed. In addition, since β enters in the form of multiplying the first-order delay element T _I · s of the denominator, a very simple adjustment configuration in which only the β is changed and only the integral term in the latter stage is changed. This means β
It is possible to intuitively understand the effect of changing.

Next, an embodiment of the invention according to claim 2 will be described with reference to FIG. The apparatus of this embodiment is premised on the transfer function C _sv (s) of the equation (9), and the target value SV
_In order to obtain the effective target value SV _n0 which is the target value of the measured value differential preceding PID adjusting means from _n , the compensation calculating means 10 is improved, and the others are the same as those in the conventional apparatus shown in FIG. The same reference numerals are given and the description of the overlapping components is omitted.

Therefore, the device of the present invention will be described focusing on the part that is particularly different from the conventional device. That is, in the compensation calculation means 10, the input line of the target value SV _n is branched into two, the target value SV _n of one line thereof is introduced into the subtraction means 11, and the target value SV of the other input line is introduced.
_n is introduced into the first-order delay means 12 having an integration time T _I as a time constant. The first-order delay means 12 delay-compensates the target value SV _n for the required time and outputs it, and the first-order delay using the subtraction means 11 and βT _I (β: coefficient of two degrees of freedom of integration time) as time constants. Supply to means 13.

The subtracting means 11 subtracts the output of the first-order delay means 12 from the target value SV _n, and after the obtained subtraction output is multiplied by the proportional gain 2 degrees of freedom coefficient α in the coefficient means 14,
It leads to the addition means 15. In this adding means 15, the multiplication value of the proportional gain 2-degree-of-freedom coefficient and the output of the first-order delay means 13 are added and combined to obtain the effective target value SV _n0 of the measured value differential preceding PID adjusting device of 1 degree of freedom. It is configured to supply to the deviation calculation means 2.

Therefore, with the configuration of this embodiment, not only the transfer function C _sv (s) becomes a transfer function similar to the above equation (9), but the compensation calculation means 10 will be described below (1).
The transfer function H (s) is as shown in equation (0). Specifically, when expressed by a transfer function, C _sv (s) = H (s) · {1 + 1 / (T _I · s)} · K _p = K _p {α + 1 / [T _I · s (1 + βT _I · s)]} H (s ) = αT I · s / (1 + T I · s) +1 / {(1 + T I · s) · (1 + βT I · s)} = α {1-1 / (1 + T I · s )} + 1 / {(1 + T _I · s) (1 + βT _I · s)} (10) That is, this equation (10) becomes the transfer function H (s) of the compensation calculation means 10. Therefore, the target value SV _n in FIG.
→ While the transfer function between the operation signals MV _n maintains the ideal form of the equation (9), as is clear from the equation (10),
The proportional gain 2 degrees of freedom coefficient α and the integration time 2 degrees of freedom coefficient β can be completely separated, and the integration time 2 degrees of freedom coefficient β exists only in the denominator.

Next, an embodiment of the invention according to claim 3 will be described with reference to FIG. In the figure, the same parts as those in FIG. 5 are designated by the same reference numerals, and detailed description thereof will be omitted.

That is, in this apparatus, the input line of the target value SV _n is branched into two, and the target value SV _n of one of the input lines is calculated by the coefficient means 21 by the proportional gain two-degree-of-freedom coefficient α.
Is multiplied by to obtain a unit proportional adjustment signal, and the subtraction means 4 performs addition synthesis.

The target value SV _n of the other input line is the first-order delay means 22 having the time constant of the integration time T _I and βT _I.
An effective target value, which is the target value of the measured value differential preceding PID adjusting device, is obtained via the first-order delay means 23 having a time constant of (β: integration time two degrees of freedom coefficient) and supplied to the deviation calculation means 2. This is the configuration.

Therefore, according to the configuration of this embodiment, the transfer function C between the target value SV _n → the operation signal MV _n
sv (s) is C _sv (s) = MV _n / SV _n = K _p {α + [1 / (1 + T _I · s)] · [1 / (1 + βT _I · s)] ・ [(1 + T _I · s ) / (T _I · s)]} = K _p {α + 1 / [T _I · s (1 + βT _I · s)]} (11) Therefore, the equation (11) becomes a transfer function equal to the equation (9).

Therefore, also in the case of the configuration as shown in FIG. 2, it can be realized in an ideal manner as well, and the effective target value which is the target value of the measured value differential preceding PID adjusting device can be obtained from the target value SV _{n at} this time. The transfer function F (s) of the compensation calculation means of FIG.
As is clear from the above, F (s) = 1 / {(1 + T _I · s) · (1 + βT _I · s)} (12) Also in this case, the integration time two-degree-of-freedom coefficient β is completely separated from the proportional gain two-degree-of-freedom coefficient α, and the integration time two-degree-of-freedom coefficient β exists only in the denominator.

Further, an embodiment of the invention according to claim 4 will be described with reference to FIG. In this device, an input line of a target value SV _n is branched into two, and the target value SV _n of one of the input lines is introduced into a subtracting means 32 after being multiplied by a proportional gain two-degree-of-freedom coefficient α in a coefficient means 31. Here, the control amount PV _n is subtracted from the coefficient multiplication value to obtain a unit proportional adjustment signal.

The target value SV _n of the other input line is βT
_An effective target value, which is a target value of the I-PD adjusting means, is obtained via the first-order delay means 33 having a time constant of _I (β: integration time two degrees of freedom coefficient) and supplied to the deviation calculating means 2. After calculating the deviation, the deviation is supplied to the integrating means 34.

Then, the unit proportional adjustment signal and the integrated output of the integrating means 34 are added and synthesized by the subtracting means 4, and the output of the D adjusting means 5 is subtracted from the added synthetic value, and the garden result is proportional to the gain means 6 Is applied to the process 7 as an operation signal by multiplying the proportional gain K _p by the control amount PV _n.
= Controlled to be the target value SV _n.

The target value SV _n → in the structure of this embodiment
The transfer function C _sv (s) between the operation signals MV _n is C _sv (s) = MV _n / SV _n = K _p {α + 1 / [T _I · s (1 + βT _I · s)]} (13) The same ideal configuration as the equation (9) is obtained.

The transfer function E (s) of the compensation calculation means for obtaining the effective target value which is the target value of the I-PD adjusting means at this time is
As is clear from FIG. 3, E (s) = 1 / (1 + βT _I · s) (14)

In the conventional apparatus, however, the compensation operation means to be added depends on only the direction in which the PI has two degrees of freedom, and does not care whether the integration operation is easy to understand or easy to adjust. However, in the embodiment of each claim in the present apparatus, the proportional and integral 2
From the reverse idea of setting the ideal form of the degree of freedom as in the equation (9) and finding the compensation calculation means for realizing this, the configuration of each compensation calculation means has been realized. is there.

Next, let us consider the integration term having two degrees of freedom. Now, the ratio U (s) of the basic integration = K _p / T _I · s and the two-degree-of-freedom integration = K _I / {T _I · s (1 + βT _I · s)}
, U (s) = {1 / [T _I · s (1 + βT _I · s)]}
When adjustment such that β ≧ 0 for / (1 / T _I · s),
The initial and final values are as follows.

[0042]

[Equation 1]

This relationship is illustrated in FIG. That is, by changing β (β ≧ 0), the integral operation can be changed equivalently, and the configuration is simple and easy to understand, and the integration operation can be made to have two degrees of freedom. The present invention can be variously modified and implemented without departing from the scope of the invention.

[0044]

As described above, according to the present invention, P
The 2-degree-of-freedom integration operation for the 2-degree-of-freedom ID control is made extremely simple, and the mutual interference between the proportional gain 2-degree-of-freedom coefficient α and the integration time 2-degree-of-freedom coefficient β can be eliminated. By providing the compensation calculation means, (1) the two degrees of freedom of the integral operation can be extremely simplified and can be realized in an easy-to-understand form. (2) In the two-degree-of-freedom integration operation, the involvement of the proportional gain two-degree-of-freedom coefficient α can be eliminated, and the adjustment becomes simple. (3) Furthermore, the involvement of the 2-degree-of-freedom coefficient β in the integration operation is limited to the denominator, and it is possible to intuitively understand when adjusting the 2-degree-of-freedom coefficient β.

As described above, the device of the present invention completely overcomes the conventional defects, and makes it easy to understand the two-degree-of-freedom PID adjusting device.
It is easy to adjust and can be applied to a wide range.
As a result, the application to the plant control system in the future will be expanded, the controllability of the entire plant can be revolutionized, and by extension, it will make a great contribution to the industry.

[Brief description of drawings]

FIG. 1 is a configuration diagram showing an embodiment of the invention according to claim 2;

FIG. 2 is a configuration diagram showing an embodiment of the invention according to claim 3;

FIG. 3 is a configuration diagram showing an embodiment of the invention according to claim 4;

FIG. 4 is an integration operation output characteristic diagram when the integration time two-degree-of-freedom coefficient β is changed.

FIG. 5 is a configuration diagram of a conventional two-degree-of-freedom PID adjustment device.

[Explanation of symbols]

2 ... deviation calculating means, 3 ... PI adjusting means, 4 ... subtracting means,
5 ... D adjusting means, 6 ... proportional gain means, 7 ... process,
7-1 ... Control object, 10 ... Compensation calculation means, 12, 13, 2
2, 23, 33 ... First-order delay means, 14, 21, 31, ... Coefficient means, 34 ... Integrating means.

Claims (4)

[Claims] 1. PI or PID (P: proportional, I: integral,) so that a deviation between a controlled variable from a controlled object and an effective target value obtained by performing a compensation calculation from a target value of this controlled variable becomes zero. D: Derivative) Adjusting operation is performed, and the obtained adjusting signal is applied as an operation signal to the controlled object. In order to obtain an effective target value from the target value of the controlled variable, a two-degree-of-freedom coefficient of proportional gain is used. Compensation calculation means using α and a two-degree-of-freedom coefficient β of integration time is provided, and the transfer function C _sv (s) between the target signal and the operation signal is configured as follows. 2 degree of freedom PID adjustment device. C _sv (s) = K _p {α + 1 / [T _I · s (1 + βT _I ·
s)]} where K _p : proportional gain, T _I : integration time, s: Laplace operator, α: proportional gain 2 degrees of freedom coefficient (a constant between 0 and 1), β: integration time of 2 degrees Degree factor (constant between 0 and 2) 2. The compensation calculation means is H (s) = α {1-1 / (1 + T _I · s)} + 1 / [(1
+ T _I · s) (1 + βT _I · s)]} and the obtained effective target value is used as the target value of the measured value differential preceding PID adjusting means. The two-degree-of-freedom PID adjustment device according to claim 1. 3. PI or PID (P: proportional, I: integral, D: derivative) using a control amount from a controlled object and an effective target value obtained by performing compensation calculation from the target value of this control amount.
In an adjustment device that performs an adjustment calculation and applies the obtained adjustment signal as an operation signal to the controlled object, in order to obtain an effective target value that is the target value of the measured value differential preceding PID adjusting means from the target value of the control amount. , F (s) = 1 / {(1 + T _I · s) (1 + βT _I · s)} is provided with a compensation calculation means having a transfer function F (s), and the target value of the control amount is set to two degrees of proportional gain. A two-degree-of-freedom PID, wherein a unit-proportional adjustment signal of a target value multiplied by a conversion coefficient α is added and synthesized to the measured value differential adjustment means
Adjustment device. However, in the above equation, T _I : integration time, s: Laplace operator, α: proportional gain 2-degree-of-freedom coefficient (0 to 0)
(Constant between 1), β: 2 degree of freedom coefficient of integration time (0 to 2)
Constant) 4. A PI or PID (P: proportional, I: integral, D: derivative) using a control amount from a controlled object and an effective target value obtained by performing a compensation calculation from a target value of this control amount.
In an adjusting device that performs adjustment calculation and applies the obtained adjustment signal as an operation signal to the controlled object, in order to obtain an effective target value that is a target value of the I-PD adjusting means from the target value of the control amount, E ( s) = 1 / (1 + βT _I · s) A compensation calculation means having a transfer function E (s) is provided, and the control is performed from a value obtained by multiplying the target value of the control amount by a two-degree-of-freedom coefficient α of the proportional gain. A unit proportional adjustment signal with a reduced amount,
A two-degree-of-freedom PID adjustment device characterized by performing additive synthesis on the I-PD adjustment means. However, in the above equation, T _I : integration time, s: Laplace operator, α: proportional gain 2-degree-of-freedom coefficient (a constant between 0 and 1), β: integration time 2-degree-of-freedom coefficient (0 to 2) Constant)