JPS60251402A - Process controlling method - Google Patents

Abstract

PURPOSE:To control various process controls at a point of inflection with stable and high reliability by finding out the gain g1 of an output smaller than a reference value on a curve and the gain g2 of an output larger than that independently and adjusting these gain values so that g1-g2=0 is satisfied. CONSTITUTION:In the dissolved oxygen (DO) concentration control of an active sludge process, parameter application means 28, 29 adjust the gain g1 obtained when the DO concentration is smaller than the reference value and the gain g2 obtained when larger than the reference value respectively. A moving average operating part 30 calculates the reference value from the moving average value of process output for a fixed period. The difference between the gains g1, g2 obtained through the means 28, 29, the operation part 30, etc. is integrated by an integrator 32 and these gains g1, g2 are controlled so as to be equal with each other by properly selecting the integration constant of the output Cr. Consequently, the set point Cr of the DO concentration reaches a value corresponding to a specific inflection point of the process and is inputted to a DO control part as the set point. Thus, various process outputs can be controlled at the inflection point with stable and high reliability.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field to Which the Invention Pertains] The present invention relates to a method for controlling the output of various industrial processes in which a static characteristic curve representing an input relationship has an inflection point near the inflection point.

[Prior art and its problems]

Among the various processes in which the input-output relationship is nonlinear, there are many that have inflection points in their static characteristics.

For example, in the process of continuously neutralizing water containing acids, the pH changes suddenly at the equivalence point of alkali, so if the alkali injection rate is input and the pH is the output, the static characteristics between the two will change at the equivalence point. It becomes a curve with an inflection point. Such a relationship is also seen in redox processes, and the redox potential (ORP
) places an inflection point at the equivalence point of oxidizing and reducing substances. Processes that show an inflection point are not limited to the chemical processes mentioned above; for example, when detecting a sludge interface using changes in the amount of ultrasonic and light transmitted in a solid-liquid separation tank, the amount of transmission can also be detected. Because of the sudden change at the interface, there is an inflection point in the characteristic curve of the permeation rate with water depth.

From the above examples, the common feature of inflection points is that in many cases, inflection points can be viewed, in a broad sense, as the breaking point of a process that responds to rapid changes in state. This shows that when trying to control a process to its limit point, if the inflection point can be detected reliably and the process can be controlled to this inflection point, control will be extremely easy and effective. .

On the other hand, the following two methods are mainly used in the past to control the process output to the inflection point.

■Predetermine the output value at the inflection point and control the output so that it reaches that value.

The input is adjusted so that the first derivative of the output with respect to zero human power is maximized or the second derivative is zero.

Among these methods, method (2) is the most reliable if the relationship between the inflection point and the output value is always constant; in this case, the output can be controlled at a constant value without particularly considering the inflection point. However, it is better to consider that such cases are rather rare. For example, the above-mentioned pH or ORP inflection point is affected by coexisting substances, and if the characteristics of the electrode change, the inflection point also changes.

This also applies to the detection of the sludge interface; there are many factors that can cause changes in the position of the inflection point, such as changes in sludge density, changes in the light source, and dirt on the light receiving surface. Therefore■
This method cannot be expected to be effective.

Next, method (2) using differentiation has the most serious noise problem. To find the differential coefficient of the output with respect to the input, we must find the time derivatives of the input and output and find their ratio. However, since the differential operation uses a single high-pass filter, noise is magnified, and it is extremely difficult to obtain effective information from such a signal. Furthermore, for process control,
Factors that make differential calculation difficult include the inability to vary input and output over a wide range, the output response to changes in input with a delay, and the time constant varying depending on conditions.

[Purpose of the invention]

An object of the present invention is to provide a method capable of eliminating the above-mentioned entry points and controlling various process outputs to inflection points with stability and high reliability.

[Key points of the invention]

The present invention provides a reference value for the process output, and the slope of the input-output static characteristic curve when the output is smaller than this reference value, that is, the gain yl, and the slope of the input-output static characteristic curve when the output is larger than this reference value, that is, This is achieved by on-line identification of the gain 92 and manipulating the inputs such that gl and I2 are equal.

[Embodiments of the invention]

The present invention will be described in detail below based on examples.

First, we will describe the basic idea of the present invention, and then we will show specific embodiments.The inventors first need to solve the problem of output time delay, and for that purpose, we have We considered a method of approximating a model equation that includes a dynamic characteristic term and searching for an inflection point using this model equation. However, in reality, in order to modernize a nonlinear relationship that has an inflection point, the model equation must of course be nonlinear, and therefore the number of parameters increases and the identification calculation method becomes complicated, so the method is to approximate the characteristic curve with a model equation. judged to be impractical.

Therefore, instead of approximating the curve with a function, the inventors
Focusing on the fact that the objective is to detect inflection points, we have carefully examined the minimum requirements for detecting inflection points. As a result, a reference value is set for the output, and the gain 11 when the output is smaller than the reference value and the gain y2 when the output is larger than the reference value are calculated, and l! It was concluded that it is best to determine that the reference value corresponds to the inflection point when 1=f12, and this fact was applied to the method of the present invention.

For example, a moving average value of the output over a certain period of time can be used as the reference value. FIG. 1 shows, for example, an input-output relationship curve having an inflection point, and in FIG. This means that the static characteristic curve is approximated by a broken line whose breaking point is the reference point P. However, the difference between the gain g1 when the output is smaller than point P and the gain g2 when the output is larger than point P is only the magnitude relationship between the output value and the reference value P, and the approximation interval is not fixed, so 11 and 92 changes depending on the fluctuation range of the output.

In this way, the above method cannot approximate the characteristic curve with a function, but in order to detect the inflection point, the only question is whether gl and 92 are equal, so in order to control the output to the inflection point, It can be used for. Moreover, since this method uses a linear model, the parameter identification calculation is simple, and reliability is improved by using the difference between gl and g2. In other words, if the process output contains noise or if the process input has feedback from the output, a bias will often occur between the identified value and the true value of the parameter, but these effects are due to yl and I2. This is because the effect of the bias can be removed by the method of the present invention that uses the difference between gl and g2 since it acts on all the machines.

Furthermore, the present inventors developed a model norm adaptation system (Mode
Reference Adaptive System
em, hereinafter abbreviated as MRA S. ) was found to be suitable for carrying out the present invention. MRAS automatically adjusts the parameters of the model so that the process output and model output match when used for meter identification using an actual process and a model that approximates its dynamic characteristics.

An example in which the method of the present invention is applied to controlling dissolved oxygen (hereinafter abbreviated as Do) concentration in an activated sludge process using MRAS will be specifically described below.

must be adequately aerated to maintain DO concentrations at appropriate levels. However, from the viewpoint of power saving, etc., it is desirable to operate the grosses with as little amount of air as possible without deteriorating the properties of the fine particles.

In the activated sludge process, when the DO concentration C is measured at a certain point in the aeration tank and the relationship with the aeration air flow rate f is calculated, a curve shown in FIG. 2 is obtained. As the aeration air flow rate f increases, the DO concentration C suddenly increases at a certain point, so an inflection point Q appears on the static characteristic curve shown in FIG. The present inventors discovered that the minimum necessary amount of air can be maintained by controlling the DO concentration C to match the value of C at the inflection point Q, and in order to control the output to the inflection point. The present invention was applied to.

FIG. 3 is a system diagram showing an overview of the equipment configuration and functions of the activated sludge process to which the present invention is applied. In FIG. 3, the flow of water and air is shown by solid lines, and the electrical signal system is shown by broken lines, and the direction of flow is shown by arrows.

In FIG. 3, air sent from an aeration 1 passes through a flow meter 2 and is aerated from the bottom of an aeration tank 3 through an aeration pipe 4. In addition, raw water that has been sequentially treated by a device not shown passes through a flow meter 5 and flows into an aeration tank 3. After the IJIJ16 substances are decomposed, it is stored in a final sedimentation tank 6, and the supernatant water is discharged as secondary treated water. The precipitated sludge is returned to the aeration tank 3.

On the other hand, the electric signal is transmitted from D located inside the aeration tank 3.
The output of the o-sensor 7 is converted into a transmission signal by a signal converter 8, and is input to an arithmetic unit 9 together with the signals of the aeration air amount and inflow water amount measured by the flowmeters 2 and 5, respectively. The calculation device 9 performs control calculation according to the present invention and inputs a target air amount as a set value to the controller 10. The rotation speed of the aeration blower 1 is adjusted by the inverter 11 by inputting an operation signal from the controller 10 to the inverter 11 for controlling the aeration air amount to a set value.

FIG. 4 shows a block diagram for performing control calculations in the configuration of FIG. 3. In FIG. 4, numeral 21 represents the process shown in FIG. 3, where the input is the aeration air flow rate f and the output is the DO concentration C. 22
is the first-order lag element of the time constant Tf that filters f,
23 is a first-order lag element with a time constant ■ that filters C, 24 is a proportional element whose gain g is adjustable, and 25 is a time constant T
is an adjustable first-order delay element, and 26 is the fraction of 2 of the output xm of 25.
A divider 27 is used to calculate the coefficient, and 27 is a parameter adaptation mechanism that adjusts T to suit the actual process.

Is 28 the gain when the DO concentration is lower than the reference value? 29 is an adaptive mechanism for adjusting the gain I2 when the DO concentration is greater than the reference value; 30 is a signal Cf obtained by filtering the process output;
A moving average calculation unit 31 is Cf for calculating the moving average d of
and d, and depending on their magnitude relationship, switch S1.
I2. A comparator for opening and closing I3, 32 is I2-,
an integrator that integrates pl and outputs it as a set value Cr of the DO concentration; 33 is a target air amount calculation unit to which the difference ε between Cr and C and the inflow water amount signal W are inputted, and which controls C to Cr;
Reference numeral 34 denotes a sample hold section that holds for a certain period of time a signal obtained by adding identification noise N to the reference value fb of the air flow signal outputted from 33 in order to smooth the identification process.

A portion 35 surrounded by a broken line consisting of 24 and 25 represents an adaptive model.

The above control calculations are roughly divided into two parts. That is, there is a Do control section for controlling the process output C below the one-dot chain line shown in FIG. 4 to the set value Cr, and a parameter identification and Cr adjustment section above the one-dot chain line.

Of these two parts, the former performs feedback control so that the output C matches the set value Cr, and the latter performs gl
and I2 are calculated, and before I2-1□ becomes zero, ('r is adjusted.

Next, the control operation in this embodiment will be explained in order starting from the parameter identification part. The input f and output C of the process are required to identify the parameters. First, f and C are filtered by first-order delay elements 22 and 23 to obtain respective smoothed signals. −7 Next, subtract the smoothed signal from f and C to obtain U and y, respectively. In the case of an activated sludge process, the relationship between U and y can be expressed by a first-order lag model, so the adaptive model 35 constituting the MRA S can also be a first-order lag model consisting of 24 and 25. The role of this MRAS is to adapt the adaptive model 3 so that the process signal y and the output from 25 match.
5, the time constant T and gain y are automatically adjusted. Automatic adjustment of T and y is performed by parameter adaptation mechanisms 27, 28 and 29. Of these, 27 are T adaptation mechanisms, in which the difference e between y and xm and the time differential Am of xm are input, and T is automatically adjusted according to an integral adaptation algorithm. In equation (1), TO is the initial value of T, and kT is the adaptive gain of T. The gain g identification operation in this embodiment is different from the normal case. That is, in the method of the present invention, when the signal Cf obtained by filtering the output is smaller and larger than its moving average d, the respective gains g1 and I
The identification mechanism is divided into two, 28 and 29, so that 28 and 29 can be identified separately. The identification algorithm is. 1!1o and g in equations (2) and (3)
2o is the initial value of gl and 92, respectively, and kt is the adaptive gain. Furthermore, in the present invention, only one of the two adaptive mechanisms 28 and 29 is operated depending on the magnitude relationship between Cf and 6, and the gain J of the adaptive model 35 is the gain ( yl or I2
), it is possible to separately identify the gain g1 when Cf is smaller than E and the gain I2 when Cf is larger than δ.

Three changeover switches S, which are opened and closed by a comparator 31, perform switching operations for such a series of operations.
S2 and I3. In addition, in this embodiment, in order to ensure the identification operation, the output of the adaptive model 35 is made to match the process signal y at the time when the three switches are switched and when the air flow rate f is updated. . Comparator 3
1 and the dashed arrow leading from the sample and hold unit 34 to the adaptive model 35 indicate this operation.

The difference between I2 and 11 obtained in the above manner is calculated by the integrator 32.
If we integrate with reaches a value corresponding to an inflection, point, on the characteristic curve representing . It was calculated like this.

Cr is input as a set value to the DO control unit, and feedback control is performed so that the DO concentration C becomes equal to the set value Cr. In order to perform this control operation, in this embodiment, in addition to normal PI control, The response to sudden changes in the inflow load is improved by auxiliary operation using the inflow water volume W.
Then, an identification noise N is added to the reference value fb of the air amount requested by the control calculation so that the identification process is carried out smoothly, and this signal is held for a certain period of time to obtain the actual air amount f.
becomes.

The activated sludge process was actually controlled using the equipment configuration shown in FIG. 3 and the control method shown in FIG. 4, and the results obtained are shown in FIGS. 5(a) to 5(d). Figure 5(a)~
(d) is a diagram showing changes in various data obtained over the course of 100 hours.

The vertical axis of each of these diagrams is normalized by dividing by an appropriate full value (FS), and the FS value is marked on each diagram.

Figure 5(,) shows the change in time constant T, Figure 5(b) shows the change in gain 11 when the output is smaller than the reference value, and the change in I2 when the output is larger than the reference value, and Figure 5(c) shows the change in the time constant T. ) represents the change in the set value Cr of the Do density color, and Figure 5 (ψ represents the change in the air volume and inflow water volume W, respectively. Among these, the effect of the present invention is particularly clear in Figure 5 (c). ), and according to Figure 5(c), Cr including control of the activated sludge process
= 0.5 ml//l and was carried out for about ioo hours, but immediately after the start of control CR decreased slightly, but 5
It starts to increase after ~6 hours and reaches a nearly constant value after about 40 hours. After that, Cr is 1.5-2 mg/l K
The DO concentration C is maintained and controlled within a range of approximately ±Im, 9/l, centered on Cr. Comparing the value of Cr after reaching a certain value and the value of C corresponding to the inflection point Q of the curve shown in FIG. It can be seen that this has been fully achieved.

In the above examples, the present invention is applied to an activated sludge process, and the output is 9. Although we have described the effectiveness of controlling near the inflection point by adjusting it so that -412, the method of the present invention is not limited to activated sludge processes, It is expected that this method will have a great effect on various processes that have an inflection point, and can be applied to process control in various fields.

〔Effect of the invention〕

As explained above in the embodiments, the present invention is designed to control output from a reference point on the curve in order to control the output near the inflection point for a process that has an inflection point in the characteristic curve showing the input/output relationship. Since this method calculates the gain 11 of the smaller output and the gain h of the larger output than the reference point separately and adjusts them so that F2-y□=0, it is not necessary to approximate the characteristic curve with a function 21 It can be determined only by the magnitude relationship between 92 and 92, so parameter identification calculation using a linear model is easy. ! By using the difference between 1 and I2, it is possible to remove the influence of disturbances such as noise that commonly act on both. Furthermore, the present invention is advantageous in terms of practicality because it is suitable for on-line parameter identification to be carried out using, for example, MRAS.

From the above, the method of the present invention enables a stable control operation with significantly higher reliability than the conventional constant value control method or differential method. In particular, this method fully satisfies the purpose of process control in which improvements in stability and reliability are desired, and it can be said to be a method of extremely high practical value.

[Brief explanation of the drawing]

Figure 1 shows input/output static characteristic curves with inflection points and I! 1
FIG. 2 is a diagram showing the relationship between aeration air [f and Do density color in the activated sludge process, and FIG. 3 is a diagram showing the equipment configuration of the activated sludge process to which the present invention is applied. System diagram showing functions. Figure 4 is a block diagram showing control calculations according to the method of the present invention, and Figures 5 (a) to (d) show changes over time in various data obtained by controlling the activated sludge process according to the method of the present invention. In each case, (a) is the time constant T. (b) is 11 and g2. (c) is the DO density color setting value Cr
+'*Hanawa is a diagram showing changes in air resistance f and inflow water amount W. 1... Aeration Proa, 2...
Air flow meter. 3... Aeration tank, 4...
Diffusion 'f, 5... Inflow water flow meter, 6...
・Final sedimentation tank, 7...DO sensor, 8...
...Signal converter, 9...Arithmetic device, 10...
...Controller, 11...Inverter, 21...
...Activated sludge process, 22, 23.25...
...1st order lag element, 24... Proportional element, 26.
......Divider, 27, 28.29...Parameter adaptation mechanism, 30...Moving average calculation unit, 3
1... Comparator, 32... Integrator, 33
...Target air amount calculation section, 34...Sample hold section, 35.Old...adaptive model, C...
...DIII, f... Aeration air amount, llx... Gain when the output is smaller than the reference value. y2... Gain when the output is larger than the reference value, xm... Output of the adaptive model, y...
Process (No. i, P... Output reference point, Q...
...Inflection point, W...Inflow water volume. Figure 11''j2 Figure

Claims (1)

[Claims] Engineering) The static characteristic curve representing the relationship between input and output has a gain (11) of inflection and a gain (g) when the output is larger than the reference value.
2) A process control method characterized in that the output is controlled near the inflection point by identifying and adjusting the output level so that 91 and 12 are equal. Scope of Claims In the method according to claim 1, the reference value is a moving average value of output over a certain period of time.3) In the method according to claim 1 or 2, , using a model reference adaptation system (MRAS) for parameter identification having an adaptation mechanism for identification (28) for 11 and an adaptation m (29) for identification of I22, and when the output is smaller than the reference value, only adaptation *# (28) is used. is activated and MRAS
11 is input to the gain ω of the adaptive model (35) of MRA S, and when the output is larger than the reference value, only the adaptive mechanism (29) operates and the gain &
) is a process control method characterized in that LJ'1 and g2 are identified by inputting g2. 4) In the method according to any one of claims 1 to 3, when the input of the process suddenly changes or the output of the process passes the reference value, the MRAS
A process control method characterized in that an output signal OIm) of an adaptive model (35) of the above is matched with a process signal 0) excluding a DC component.