JPS58107913A - Flow rate controlling method - Google Patents

Abstract

PURPOSE:To attain highly reliable flow rate control while ensuring the control response of a desired flow rate at all times, by adjusting fluctuation of a system with a control constant taken as a parameter automatically, taking nonlinear relation into consideration. CONSTITUTION:A proportional/integral control operation is performed by using control parameters KP, TI, and based on the result of operation, the flow rate is controlled with a flow rate control valve 5. In this case, an operation controller 9 calculates a dynamic characteristic value K of a flow path as a controlled system, based on the flow rate as process variables, a flow rate control valve opening detected at an opening detector 7, and their change, change in the number of driven pumps 3, and the function of them. Further, the control parameter is changed by a cross angular frequency preset and the characteristics K thus calculated.

Description

[Detailed Description of the Invention] Field of the Invention The present invention is directed to controlling the flow axis of a flow path system using a ratio-integral (PIJ'F) or ratio-integral-differential (PID) system. This invention relates to an improvement in a control method for a station.

Branch of the invention I Background and its problems At a flow rate of 1 to 1 line, the flow rate detection value is compared with the flow rate detection value, the integral and differential of the difference are multiplied by a certain constant, and the sum is calculated. The so-called proportional system, where is the output of the manipulated variable.
When the flow rate is controlled by the integral-derivative (PID) system, the control constants used for these -j(IYA)l have conventionally been treated as fixed values. However, when the flow rate is controlled by PID control using the conventional fixed ゛cut@j parameter, the relationship between the ηΩ amount and the flow rate control valve opening is non-linear, so it is not always possible to control the desired flow rate. It is impossible to answer and return.

Purpose of the Invention The present invention was made to solve the above-mentioned problems, and its purpose is to take into account the non-linear curvature relationship with the i-quantity control @j-5f opening W, and to Is the control response always the desired flow rate by automatically adjusting v@ as a constant gold parameter? It is an object of the present invention to provide a current control method that can perform flow control with a high level of comfort while ensuring +1-.

Outline of "Hatsuyu" In order to achieve the above object, the present invention detects the flow rate in the flow path thread and compares it with a preset flow rate @sen,
Based on this Hiwa Ayu Shu, control is performed using the H1 constant control parameter, and the flow rate is controlled by operating the flow rate side 1 ha 11 squares provided in the flow path system in the area of this performance. Wiiii
When +, the flow rate as a process quantity, the flow rate control valve opening m'', changes in these quantities, and the number of operating pumps installed in the flow path system are controlled according to the system configuration. Elevate the dynamic characteristic values of the target flow path system (process),
Based on the calculated dynamic characteristic value, the above-mentioned control side 1 automatically adjusts the adrenal gland (variable) to always provide a desirable control response for the flow rate.

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be explained below with reference to the following. FIG. 1 is a block diagram showing the configuration of a flow rate control system for a flow path yarn as a controlled object. In this example, the water supply system valve 1. ．． iF! This figure shows a pump discharge flow rate control system based on I-type.

3- In the figure, water in a water storage tank 1 is pumped up to a water receiving tank 4 through a pipe 2, a pump 3 operated at a constant speed, and a flow rate control valve 5.

On the discharge side of the pump 3, there is a flow extractor 6 for detecting the flow rate, an opening detector 7 for detecting the opening degree of the control valve 5, and an opening detector 7 for controlling the operation of the pump 3. A pump number control device f11 capable of detecting the number of pumps is installed at e. "The flow rate as a process quantity detected by these exchangers 6 and 7.11, the valve opening degree, and the number of operating units 4W are human-powered equipment 8.
The signals are respectively inputted to the arithmetic and control unit 9 from the electronic LTt lift etc. In the arithmetic control +H+t9, the open side 1@ calculation is performed based on the logic that has been applied in advance, and the compensation M- and the result are outputted as a control amount signal via the output device 10.

That is, the number control amount signal Nr of the pump 3 is sent to the number control device g11, and the flow rate control valve 50 opening control amount signal ΔU is sent to the number control device g11.
Each is output to the quantity control valve opening degree control device 12. Furthermore, each of the control devices 11 and 12 receives the control amount signals Nr and Δ
Based on U, the pump 3 and flow rate control valve 5, which are the objects to be controlled, are controlled by adjusting the manipulated variables 4-. Here, the number control loop basically exists as a separate loop, and for example, there is a method of controlling using a table in which a preset flow rate target value corresponds to the number of operating machines.

The main flow control routine is executed within the framework of the number of operating vehicles determined by this number control loop. Here, we take up the case where both loops are built in the same arithmetic and control unit Aonai, and when the first loop on the open side is executed, the @
The number of vehicles in operation is controlled by outputting operation signals for continuous operation or stop based on the calculation results. Thereby, the flow rate control valve 5 and the pump 3 are controlled in response to a change in the target flow rate, and as a result, the discharge flow rate of the flow rate control valve 5 is controlled.

The logic built into the arithmetic and control unit t9 will now be described. Its functions are broadly divided into two parts: the first is a linearization model within minute changes in the controlled object, and the second is an algorithm that determines control parameters from the W% value that is improved by that model.

(+) Linearization of control model The pressure average of the controlled object shown in Figure 2μ is expressed in the form of the following equation.

・・・・・・・・・・・・ 11+ Here, hI: Water level of the water tank (→ h2: Water level of receiving water ←) U: Flow control valve opening degree (p, u) fv(u): Loss coefficient of flow control valve q: Flow rate flowing through the pipe (-/excerpt) N: Number of pumps in operation (units) fp (N, q): Pump discharge pressure (-) 1 and 3: Pipe loss coefficient fO: Loss factor due to inertia uni time derivative (it Le, r! L Nibrandt IM Mei's constant We accept each.

In addition, in this formula (1), the third tomb and the 410th tomb on the right side
is small, and this it1 equation can be approximated as shown in equation (2).

b4 J = fy(u) q path-f r+ CN
, Q) -......-...12+As can be seen from equation (2), the n-th power of the flow iq and the flow rate control014I5P
There is a power loss coefficient fv(u) of 5, which is a non-linear relationship with respect to these variables. Therefore, in order to apply the control implant to this object, it is necessary to linearize the object.

Now, assuming that the control target range described above is near the Hiramata T point, the following equation holds true.

fi+oh! o=fv(uo)q(/'-fp(N
, qo) ・−−−−−−・−(3) (h, .+J
) −(h2(1+Δh,)=fv(uo×ΔuX
qo+Δq)rL-fp(N, qo to q)-141
Here, U: at when the process rarely reaches a certain equilibrium point
Control valve opening ΔU: Flow rate when the process rarely reaches a certain equilibrium point Deviation from the control valve opening Qo Near “Flow rate Mq when the process rarely reaches a certain equilibrium point” Deviation from the flow rate 7-h, .: Water level in the water tank when the process reaches a certain equilibrium point ΔhI: Deviation from the water level in the water tank when the process reaches the equilibrium point h2o: When the process reaches the equilibrium point Water level in the water tank Δh2: Deviation from the water level in the water tank when the process is at a certain equilibrium point Note that the other signals are similar to equation (1).

In this equation (4), ignoring the term ΔU:Δq, (31
When substituted into the formula, formula (5) is superior.

Δh, -Δhl=fv'(uO)q(!L・Δu+(rLfv(uo)q
6''-fp' (N+Qo month Δq...tilN Equation (5) is linear with respect to ΔU and Δq, and when shown in block form, it becomes as shown in FIG. 2.

1. Here, the dynamic characteristic values of the flow path system, that is, the process, are as follows.

8- Therefore, this model has been accepted as a proportional system. In the end, pump 30 cab tk Ns flow lid control side 1 valve 5
opening uo and flow rate q. If (hesi seconds) is superior as a process value, then fV l fV' (fV is the benefit of the differential of fV) and f'p (fp' is the steepness of the differential of fp) are functions, and taktaru is a constant. Since it can be set in advance, the dynamic temporality value K of the wandering control model is superior.

(1) Automatic adjustment method of pr constant parameters Here, we will use ratio-integral (PI) control (value-integral control may have an undesirable effect on noise) for a system where the object is proportional. Let us consider the case in which the ratio is 0 and the ratio is 0.) is applied, and the cause of the block is shown in Figure 3. However, in the figure, Qr is the flow end cap Δq(ym
The target values of m-shira, KP, and TI are PIitll# parameters, and S is a Lagrassi operator, and if Nr +ΔU in the figure is decorated on the calculation side, they are output as control signals from 9, respectively.

The open loop transfer function of such a thread is G(8) = Kp −K・−・(1+Tl5)・
・・・・・・・・・・・・ (7) T process Opening degree of valve 5 uo
and the flow rate Qo. Therefore, in order to keep the control response time constant against this variation, the present invention provides the following guidelines.

(Guideline 1) In order to control the control response time, KP/TI is calculated backwards so that the angular frequency IC is constant.

Since there is a term (Zupashi 2) (1+Trs), the number of stones is 3.
- Set it to be 51c or higher. For example, (3.5
1c = -) TI (Guideline 3) To ensure stability, the slope at around 1% 1'c of the gain characteristic in the frequency response of the open loop transfer function should be -20 dB/dec.

If we follow these guidelines 1 to 3, we can use the algorithm shown in Fig. 4 to set the PI control parameters and keep the above-mentioned cross angle frequency pc constant, taking into account IC = 肚饚 from equation (7) above. It becomes possible to automatically adjust so that it can be realized.

In this way, the flow rate detector 6 detects the flow rate in the flow path system that pumps water in the water storage tank 1 to the receiving water 41! 4 via the path 2, the pump 3, and the flow 1 control valve 5. This is detected and compared with a preset flow rate target value, and a predetermined control parameter KP is set based on the comparison result.

Relative/integral control of K P (1+T) using TI @
al: Total l: When controlling the flow rate by operating the flow rate control valve 5 provided in the flow path system based on the true result, the algorithm built in the arithmetic and control unit 9 is used as the siprocess amount. control based on the functional relationship between the flow rate, the opening degree of the flow control valve detected by the opening/crossing device 2, changes in these amounts, and changes in the number of operating pumps 3 provided in the flow path system. The dynamic characteristic value K of the flow path yarn (process) as an acid resistant is defined as (
6) From the formula, the above control @1 parameter K P (=a, 1θ,
TI (=subtraction) 105 61 1 1 row 191.

11- Therefore, by using the algorithm shown in Fig. 4, the control N parameter is automatically adjusted in response to system fluctuations, and the control response of the flow rate, which is variable even when the crossing angle frequency, lFc, is kept constant is neglected. You can do it like this. It's Kaya.
A simple algorithm means that it is highly efficient, and automatically adjusting control parameters will shorten product test and adjustment time.

Although northward/integral control was adopted in the above implementation, input signals from the process, especially current signals, are passed through a filter and inputted to the input device 8, thereby reducing noise in the process signals. , relative/integral/derivative control (PID control) can also be employed, and its effects can be greatly enhanced by using it in combination with the present invention.

In addition, in the above example, the application was applied to a flow channel system that includes a pump, but the valve opening (
(In the flow rate control system according to It can be shown by deleting the first term.

If speed control of pump 3 is possible, 1 of pump 3! ! By simply detecting one rotation, the dynamic characteristic values of the target process can be completely determined using a formula similar to formula 161.

1+ Achievements of the Invention As explained above, according to the present invention, a highly reliable flow rate open side 1 can be achieved by always securing and stopping the flow rate control@j response.
method can be provided.

[Brief explanation of drawings]

Section 1 is a block diagram showing the -11 aspect of the present invention, Fig. 2 is a drawing showing a linearized water supply model, Fig. 3 is a block diagram showing one embodiment of the present invention, and Fig. 4 is the invention FIG. 3 is a schematic flowchart of control parameter movement adjustment built in the king control device of the embodiment. 1...Water tank, 2...Pipe line, 3...Pump, 4...
...Water receiving, monk, 5...Flow rate control valve, 6...Flow rate detector, 7...Opening degree detector, 8...Manual power device, 9...
Monarchy (all device, 10... Output device, 11... Number control device, 12... υtE'W system side J valve opening control device. Supplier's representative Masujishi Suzue Takehiko 15- 65-

Claims (3)

[Claims] (1) Detect the flow rate in the flow path system, compare it with a preset flow rate target value, perform a calculation using predetermined control parameters based on the comparison result, and based on the calculation result, When controlling the flow rate by operating a flow control valve provided in the flow path system, the flow path thread (process This is a patented flow rate control method in which the dynamic characteristic m: of ) is calculated, and the calculated dynamic characteristic value is used to vary the ) former6kichi11@l parameter. (2) A flow rate control method according to claim 10, in which a pump whose speed can be controlled is selected, and the rotational speed of the pump and its change are added to calculate the operating characteristic value. (3) Detect the flow rate in the flow path system, compare it with the preset flow rate target value, and set the predetermined 1t based on the comparison result.
When control@1 operation is performed using the III control parameter and the flow rate is controlled by operating the flow rate control valve provided in the flow path thread based on the result of the operation, the flow rate as a process quantity, the flow rate ff1lJ control The dynamic temporal value of the flow path (process) as the control target was calculated based on the functional relationship between the valve opening degree, the changes in these voltages, and the number of operating pumps installed in the flow path system. A method according to the present invention is characterized in that the G control parameter is changed by h1 according to a dynamic temporal value.