JPH0435762B2 - - Google Patents

Description

[Detailed description of the invention] [Technical field to which the invention pertains] This invention relates to a sample value PID control having a function of identifying process dynamic characteristics during closed-loop control and automatically adjusting optimal control constants based on the identification results. The present invention relates to devices, and in particular to sample value PID control devices that can detect changes in process characteristics during operation and automatically readjust control constants.

[Prior art and its problems]

Conventional sample value PID controllers manually activated the autotuning function for control constants, which caused the following problems in readjusting control constants in response to changes in process characteristics during operation.

(1) Workers must constantly monitor changes in process characteristics, so human labor cannot be omitted. Furthermore, expensive equipment such as a recorder and a television monitor is required for monitoring.

(2) If operations are performed without noticing changes in process characteristics, product quality may deteriorate, plant efficiency may decrease,
This is a problem because it reduces safety.

(3) Even if a change in process characteristics is noticed, the automatic tuning function is activated manually, making it impossible to make adjustments quickly.

[Purpose of the invention]

In order to solve the above-mentioned problems, the present invention fully automates the online monitoring of changes in process characteristics during operation and readjustment of control constants, thereby preventing deterioration in product quality.
The purpose of the present invention is to provide a sample value PID control device that allows plant operation without deterioration of plant efficiency or safety, and that saves manpower.

[Summary of the invention]

The present invention includes a sample value PID control calculation unit that controls a process to be controlled by sample values, applies a predetermined identification signal into a control loop controlled by the sample value PID control calculation unit, and based on the sample value PID control calculation unit, In a sample value PID control device configured to automatically calculate and auto-tuning control constants of the sample value PID control calculation unit by identifying parameters of a dynamic characteristic model of the process, estimating the process. A plurality of possible characteristic changes in a plurality of cases are determined in advance as estimated characteristic changes, and a plurality of estimated dynamic characteristic models are obtained by changing the parameters of the dynamic characteristic model of the process after the identification is completed in accordance with each of the plurality of estimated characteristic changes. , an estimated error calculating means that is stored in advance in association with the estimated characteristic change, and calculates the actual model error of the process from the parameters of the dynamic characteristic model of the process after the completion of identification, and the operation signal and output signal of the process. and determining that an actual characteristic change of the process has occurred when the actual model error of the process detected by the characteristic change detecting means is larger than a threshold, Using the model, multiple model errors corresponding to the estimated characteristic changes of the process are calculated as estimated model errors, and from among the calculated estimated model errors, the estimated model error with the minimum value is selected. The control constant of the sample value PID control calculation unit is adjusted using the parameter of the estimated dynamic characteristic model corresponding to the estimated model error.

〔Effect of the invention〕

The effects of the sample value PID control device of the present invention will be described below.

(1) Detection of changes in process characteristics can be automatically monitored online, saving human labor.
Moreover, there is no need for expensive monitoring equipment.

(2) Since changes in process characteristics are not overlooked, high product quality and plant safety can always be maintained.

(3) The readjustment of control constants when process characteristics change can be automated, allowing prompt response. Moreover, manpower at this time can be saved.

[Example of the present invention]

An embodiment of the present invention will be described with reference to FIG.
FIG. 1 is a block diagram showing the configuration of a sample value PID control device according to the present invention.

Note that the sample value PID control device of the present invention is based on a device that is equipped with a sample value PID control calculation unit that controls a process to be controlled using sample values, and that is controlled by the sample value PID control calculation unit. Apply a predetermined identification signal within the loop,
For example, regarding the "sample value PID control device equipped with an auto-tuning function that automatically calculates the control constants of the sample value PID control calculation section by identifying the parameters of the process dynamic characteristic model based on this," Since it is described in detail in JP-A No. 57-39412 (Japanese Patent Publication No. 63-18202), detailed explanation thereof will be omitted in the present specification.

The process 1 to be controlled constitutes a closed loop system with the sample value control calculation unit 5 as shown in the figure. r in the figure,
t is the target value, e and t are the deviations, u and t are the operation signals,
y and t represent process outputs, and d and t represent disturbances that change the equilibrium point of the process. In addition( )
The letter t in the squares indicates a real-time signal, and k indicates a signal sampled by the sampler 4.

During operation, the process 1 generates an operation signal U(k) {Uo which is PID-calculated by the sample value control calculation unit 5 so that the deviation signal ^* _e (k) becomes smaller every sampling period k.
(k)=U(k)} is added and controlled via sample hold 2. The deviation signal e(t) is obtained by adding the target value r(t) and the process output y(t) at addition point 3.

The auto-tuning function that determines the control constants of the sample value control calculation unit 5 first uses the initial values of the proportional gain, integral time constant, and differential time constant (Kco, Tio . let

In this way, a persistent and exciting identification signal ^* _v o (k) that satisfies the identification condition of a closed-loop system process is added to the operation signal ^* _U o (k) and injected into the process as shown in the figure. be done.

At this time, the process input signal U(k) and the process output signal y sampled by the sampler 8
A Z-domain pulse transfer function is calculated by time-series processing of (k) in a pulse transfer function identification unit 9, which will be described later.The result is converted into an S-domain pulse transfer function in a transfer function calculation unit 10, and then The result is used to calculate a PID control constant in the sample value control constant calculation section 11 and set in the sample value control calculation section 5.

By sequentially performing the above processing every sampling period k, it is possible to determine the optimum control constants that match the characteristics of the process to be controlled.

Here, as the process identification progresses, the parameters of the transfer function in the S region become constant values, so the identification end determination section 13 determines this and stops auto-tuning.

In other words, the identification signal ^* _v o from the identification signal generator 7
The injection of (k) is stopped, and only the operation signal ^* _U o (k) which is subjected to pure control calculation is applied to process 1. Further, a pulse transfer function identification section 9, a transfer function calculation section 10, and a sample value control constant calculation section 11
is stopped, and the control constants Kc, Ti, and Td are held after the identification is completed.

Next, the premise of the present invention is to identify the process including the equilibrium points D and t of the process, and the pulse transfer function identifying section 9 is used for this purpose. A process model of the pulse transfer function identification section 9 is shown in FIG. As shown in the figure, a process model, a noise model, and a disturbance model that changes the equilibrium point are defined. Here, U(k) is the operation signal,
ε(k) is white noise, D is a DC signal, and Y(k) is a process output.

The process model shown in FIG. 2 can be identified as shown in the following equation.

Y(K)=- _n 〓 ⁱ⁼¹ a^iY(k-i)+ _n 〓 ⁱ⁼¹ b^iU(K-i) + _n 〓 ⁱ⁼¹ C^iε(k-i)+ε(K )＋D^ (1) D^=(1+ _n 〓 ⁱ⁼¹ a^i) Yo− _n 〓 ⁱ⁼¹ b^iUo (2) Here, Yo is the equilibrium point of the process output, and Uo is the equilibrium point of the process input. Show points.

Transforming equation (1), Y(K)=θ^ ^T (k-1) + ε(K) (3) θ^ ^T = [a^i,…, a^m, b^ ₁ ,…, b ^m, C^ ₁ , …, C^m, D^] (4) (K) = [−Y(k-1), …, −Y(km), U(kl
−1),...,U(klm),...,U(klm),ε(k
−1), ..., ε(km), 1] (5) Identification is performed.

The pulse transfer function G^p(Z) of the process is as follows.

The sample value PID control constant is the parameter a^i (i=
1...m), b^i (i=1...n).

Next, a method for detecting characteristic changes in the process of the present invention will be explained in detail. As mentioned above, after the process identification is completed, the parameters A^(Z ^-1 ), B^(Z ^-1 ), and D are known, and unless the characteristics of the process change or the equilibrium point of the process changes. It won't change unless there is. Note that since the process outputs and inputs of {Y(K)} and {U(K)} can be measured, A^(Z ^-1 ), (B^(Z ^-1 ), and D^ after the completion of identification) { ^* _Y (K)}, { ^* _U (
K)} is used to define the model error (model error is defined as follows to detect characteristic changes based on the difference between the predicted value estimated from the model and the actually observed control amount. This model error can be calculated (also called prediction residual, identification residual, etc., but referred to herein as model error).

η(K)=− ^* _Y (K)− _n 〓 ⁱ⁼¹ a^i ^* _Y (k−i) + _n 〓 ⁱ⁼¹ b^iU(k−i)+D (7) Characteristics in Fig. 1 The change detection unit 13 is a block that calculates equation (7).

If there is a characteristic change in the process during operation, a direct current component will occur in the model error, and the characteristic change can be monitored online.

Next, a method for readjusting the control constants when the characteristics of the process during operation change will be explained in detail.

The above equation (6) is calculated by the transfer function calculation unit 10 as S
can be converted into a pulse transfer function of the domain.

Gp(S)=1/g ₀ +g ₁ S+g ₂ S ² +g ₃ S ³ +... (8) Here, the parameters of S (g ₀ , g ₁ , g ₂ , g ₃ ,
) represents the gain and time constant of the process. For example, the gain is 1/g ₀ and the time constant is g ₁ /g ₀ .

Generally, if the characteristics of the process during operation change by about 10%, readjustment of the control constants is required.

Therefore, since changes in process characteristics are generally continuous, in the present invention, the process characteristics can be changed by ±10% only, when only the process gain is changed by ±10%, when only the time constant is changed by ±10%, and when the gain and time constant are changed by ±10%. A plurality of cases of process characteristic changes are predicted in advance, and a plurality of combinations of these characteristic changes are estimated to create a plurality of estimated process models.

In this way, the most characteristic feature of the present invention is to estimate a plurality of predicted characteristic changes of a process and construct a characteristic change model corresponding to the predicted characteristic changes in advance as a characteristic change model. Construct it as shown below.

In other words, the transfer function of the process when the characteristics change is estimated as follows.

G′p(S)=1/g ₀ ′+g ₁ ′S+g ₂ ′S+g ₃ ′S+… =Ke ^-Ls /(1+T ₁ S)(1+T ₂ S)(1+T ₃ S)(
9) Here, gain K, delay time L, time constant T ₁ ,
By estimating T ₂ and T ₃ , g ₀ ′, g ₁ ′, g ₂ ′,
Estimate g ₃ ′.

The characteristic estimation calculating section 15 shown in FIG.

Furthermore, by performing Z transformation from equation (9) to the Z domain as follows, the estimated parameters of the process are calculated.
Find A^′(Z ^-1 ) and B^′(Z ^-1 ).

In this way, characteristic change models of a plurality of processes are constructed using the estimated parameters obtained, and the model error of each is calculated in the same manner as the characteristic change detecting section 13.

The estimated error calculation unit 16 in FIG. 1 is a block that calculates the model error of the equation (8) obtained from the process model of the equation (8).

In this way, when the characteristics of the process in operation do not change, the model error of the characteristic change detection section is zero, and the estimated model error signal is deviated, so there is a direct current component.

However, if the characteristics of the process during operation change, for example, by 10% only in the gain, a DC component will occur in the characteristic change detection section, and the estimated model error in the case where the gain changes by 10% will be zero. , in other cases it does not become zero.

In other words, since the model error of the estimated case has become zero, the actual characteristic change of the process during operation can be identified using the model error of the estimated case. Furthermore, the control constant can be quickly determined based on equation (9) estimated for that case. Therefore, the control constant can be readjusted without repeating. The above explanation is shown in FIG. 3 in relation to model errors. It can be seen from the figure that cases of changes in process characteristics during operation can be quickly determined from model errors.

As described above, the present invention is a sample value PID control device that can adjust control constants in response to changes in process characteristics during operation by estimating changes in process characteristics and calculating model errors. .

In addition, if the characteristic change of the process changes more than estimated, the model error of the characteristic change detection section will be larger than the initial model error of the estimated characteristic change, so this will be detected and the sampled value PID controller will automatically The tuning function can be started automatically.

[Brief explanation of drawings]

Fig. 1 is a block diagram showing the configuration of the sample value PID control device of the present invention, Fig. 2 is an explanatory diagram of a process model for parameter identification used in the sample value PID control device of the present invention, and Fig. 3 is a block diagram showing the configuration of the sample value PID control device of the present invention. FIG. 3 is a characteristic change detection and an explanatory diagram thereof. 1...Process, 2...Sample hold, 5
... Sample value control calculation section, 7 ... Identification signal generation section, 9 ... Pulse transfer function identification section, 10 ... Transfer function calculation section, 11 ... Sample value control constant calculation section, 12 ... Identification end determination section , 13...Characteristic change detection unit, 14...Control unit, 15...Characteristic estimation calculation unit, 16...Estimation error calculation unit.

Claims (1)

[Claims] 1. A sample value PID control calculation unit that controls a process to be controlled using sample values, and a predetermined identification signal is applied to a control loop controlled by the sample value PID control calculation unit, and In the sampled value PID control device, the sampled value PID control device is configured to automatically calculate and auto-tuning a control constant of the sampled value PID control calculation unit by identifying parameters of the dynamic characteristic model of the process based on the above. An estimated dynamic characteristic model is created by defining in advance a plurality of characteristic changes of a plurality of cases that can be estimated as estimated characteristic changes, and changing the parameters of the process dynamic characteristic model after the completion of identification in accordance with the plurality of estimated characteristic changes. an estimation error calculation means that calculates a plurality of values and stores each in advance in association with the estimated characteristic change, and an actual model of the process from the parameters of the dynamic characteristic model of the process and the operation signal and output signal of the process after the completion of identification. a characteristic change detection means for calculating an error; and determining that an actual characteristic change of the process has occurred when the actual model error of the process detected by the characteristic change detection means is larger than a threshold; Using the estimated dynamic characteristic model, multiple model errors corresponding to the change in the estimated characteristics of the process are calculated as estimated model errors, and from among the calculated multiple estimated model errors, the estimated model error with the smallest value is selected. A sample value PID control device, wherein a control constant of the sample value PID control calculation unit is adjusted using a parameter of the estimated dynamic characteristic model corresponding to the estimated model error.