JPS5898686A - Device for controlling discharge pressure or flow amount of pump in constant - Google Patents

Abstract

PURPOSE:To uniform a loop gain and improve the response property of the device by a method wherein a square-law function generator is provided between an adjusting meter and an invertor in the system controlling the discharge pressure or the amount of the pump. CONSTITUTION:The output of the adjusting meter 8 is squared by the square-law function generator 12 and is introduced into the invertor 9. The gain between the adjusting meter 8 and the invertor 9 is proportional to the output N of the adjusting meter 8, therefore, when the revolving number N is reduced, the gain is reduced and when the revolving number N is increased, the gain is increased. According to this method, when the gain is summed with the gain of the pump controlled in the pressure or the amount of flow, it becomes constant. Therefore, if the adjusting gain is determined at any arbitrary operating point, the response property at the other operating point will never be deteriorated and there is no anxiety that a cycling phenomenon is caused.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an improvement in a control device that performs constant control of pump discharge pressure or flow rate.

Constant control of pump discharge pressure or flow rate is conventional.

This was done using feedback control as shown in Figure 1.

That is, a pressure or flow rate detector 5 is provided in the discharge pipe 4 of the pump 2 driven by the electric motor IK via the coupling 3,
The detected value of + is converted into an electrical signal proportional to the pressure or flow rate by the converter 6, and the deviation from the set value by the pressure or Fi flow rate setting device 7 is set to a total of 8 to the PIDli meter 8. The output is given to the inverter 9 (variable speed drive device) as a rotation speed command, and the electric motor 1 is driven by the next wave number and voltage according to the command, thereby realizing constant discharge pressure or constant discharge flow rate control. It is something.

Note that the P gain (proportional gain) of the PIDii4 meter 8 can be set in advance, but it is a constant value of 4 during operation.

Further, the pressure gauge or flow meter used as the detector 5 converts pressure or flow rate into an electrical signal, and since its input/output ratio is generally constant, the gain is constant.

Further, since the ratio of the output frequency to the input of the speed change command to the inverter 94b is constant, the gain is constant.

Since the ratio of the rotational speed to the input frequency of the electric motor 1 can be considered to be almost constant, the gain is constant.

For pumps, in pressure or flow control systems.

It is known that the rate of change in pressure or flow rate with respect to rotational speed is the gain of the pump in the control loop, and changes depending on the operating condition 1ill (flL flow or pressure).

Therefore, the PID controller of the conventional device 8. Inverter 9, electric motor 1. Pump 2 and detection i! Loop gain including 5 (
(referring to the closed-loop gain of feedback control) will change depending on the load condition, i.e., fItit or the magnitude of the pressure, even during constant pressure or constant flow operation.However, until now, no change has been made. Conventionally, when trying to determine the P gain of the PID controller 8, the P gain was determined by first determining the P gain over the entire range of flow rate (or pressure) used, and then It was determined based on the safest (meaning the smallest) P gain.

Seventh point, there was a drawback that the response was poor at the flow rate (or pressure) where the loop gain was the smallest.

In order to save you the trouble of cycling, if you set the P gain only for a certain flow rate (or pressure) K, the loop gain will become too high for other flow rates (or pressures) and you will not be able to cycle. Wake up.

There was a risk of 11 km of equipment being damaged.

As a result of Hinoki's experimental research, we have found that the cause of this is due to the pump characteristics and load characteristics, so by canceling them out, we can improve responsiveness in all usage ranges. This has succeeded in reducing the variable type. In other words, in this pressure and flow rate control system, the gain of the pump changes in inverse proportion to the number of revolutions, and when the number of revolutions decreases, the gain increases, and when the number of revolutions increases, the gain decreases, so the number of revolutions increases. Square function generator 1 that has the effect of lowering the gain when the number of revolutions decreases and increasing the gain when the number of rotations increases.
By installing the controller between the controller 8 and the inverter 9, we succeeded in making the loop gain uniform and creating a control system with good responsiveness.

Embodiments 2.3 of the present invention will be described below.

FIG. 2 is a block diagram of an embodiment of the present invention, in which a square function generator 12 is provided between the PID controller 8 and the inverter 9 in the conventional device shown in FIG. Therefore, the first
The same parts as those of the conventional device shown in the figure are given the same reference numerals.

In order to facilitate understanding of the features and effects of the present invention, the operation of the conventional device shown in FIG. 1 will be explained below.

First, a case where pressure is controlled will be explained.

It is well known that the pump characteristics are well approximated by the following formula % formula % (where H: pressure N: rotation speed Q: flow rate a, b, C: constant, and many of the characteristics of a centrifugal pump are expressed as (1 ) is small, and can be ignored and approximated by equation (2).

H=aN”-bQ”・・・・・・・・・・・・・・・
(21) On the other hand, it is known that the load characteristic is expressed by the following formula in which the resistance loss pressure is proportional to the square of the flow rate. However, α is a constant. (2, (31 For example, the equation is a curve such as N1.J (Load pressure loss) in Figure 3, and if the pressure is controlled to be constant, the operating point is AI, A2 t-connection H = Hm in the figure.
Follow the line. (However, H4 is the set pressure), which is explained in Fig. 1, but since it is a very common control method, it will be omitted here.

Here, the pump gain, ie, the rate of change of Ho with respect to N, will be clarified with respect to the operating point.

In Figure 3, when the operating point is M1, if the rotational speed N changes, A1-
Move along the resistance characteristic curve of B2. The operating points on the curves M1 to B2 are represented by the H-N relationship and become a curve that passes through point A1 on the diagram. This is obtained by eliminating Q from the +21 and 133 equations, resulting in a quadratic curve like a linear equation.

1(=+ HtI Nm/ (b+αn...)
・・・・・・・・・ (◆On the other hand, in Fig. 3, when the load characteristic is M2~B10 curved armor, the number of revolutions is N2, and the operating point is point A2 where H=Hm.Mu2~ The operating point on the B10 curve is a quadratic curve passing through M2 in FIG. 4 in the H-N relationship, and the operating point in the H-Hs relationship is point A2 in the same figure.

Point 1 in Figure 4. The slope of A2 is the pump gain in the control loop at each operating point when H=Hs is controlled.

The pump gain in the control loop is the rate of change in pressure with respect to rotational speed.

M1 in Figure 4. The slope of the two points, dj (/dN), is graphed as shown in Figure 6. From Figures 4 and 5, dH/dN when pressure is controlled is inversely proportional to the magnitude of N. It can be seen that

These matters can be explained using equations (21 to (4)) as follows.

(The relationship between H and N is shown by Equation 41, but when the load characteristics are constant, that is, when α is constant (differentiating the peak by N) (Equation 51 is obtained).

d H/ls = 2 aαN/(b+α) (51 On the other hand, if we solve for α from equation (4), we get equation (6).

α= bH/(aNLH) (11J
Here, by setting H=Hs, equation (1) is obtained.

α()I=H,)=bHa/(aN”-us)
(J) This means that when H=Hm is controlled, N
The difference is that the constant α indicating job #characteristics can be found. (5) and (formula 77) (formula 81 is dH/dN
(H=Hs)=2Hs/N (83 is obtained. This is the gain of the pump when it is controlled as H=H*, that is, the rate of change in pressure p with respect to the rotation speed, and the curve shown in Fig. 5 mentioned earlier. is consistent with

Equation (8) also shows that dH/dN (kl-14m) is proportional to the size of the set boulder, but this means that the ml of H = Hi connecting Mu1 to A2 in Figure 4 is This is also consistent with the fact that the slope of each point increases or decreases by moving it up or down.

As described above, when the pressure car leg side #t is turned on, the gain of the pump 2 changes according to the load condition 11, and it is inversely proportional to the rotation speed, so the 11-node 8 of this control system When determining the gain, it must be determined at a point where N is the smallest, so that dH/dN is the largest, and,
If the adjustment gain is determined in this way, the load will fluctuate and N will increase, resulting in a decrease in dH/dNf, that is, the pump gain, resulting in poor responsiveness.

On the other hand, I thought that the adjustment gain was determined at the lowest NO point, but if the center is often operated at a small N point, if the operating point comes to that point, it may result in an excessive gain. The friend control system causes cycling, H, Q, N
Each of these will vibrate and damage the equipment.

However, since it was generally known that dH/dN varied depending on the operating point, the adjustment gain was set to a small value for safety reasons, resulting in poor response.

Although pressure control has been described above, the same can be said about flow rate control. This will be explained only mathematically. +
21, +31 Equation (9) is obtained by eliminating H from the equation.

Q −/Ja/ (α+b)·N (9) When this is differentiated with respect to N, equation (10) is obtained.

d Q / dN “Zan X 4th edition (10) On the other hand (9
) and solve for α, we get (11), Q=Q%
and α=(aN"-bQ")/Q"
(11), we get (12).

α(Q=Qs) = (aN寞−bQ:)/cP,
(12) This means that when controlled as Q=Ql, if N is known, the constant α indicating the load characteristics can be found. (13) is obtained from (10) and (12). When this is controlled as Q=Q island, d Q/ dN (Q= Q s) = Q s/N
(13) This is the gain of the pump, that is, the rate of change of the flow rate with respect to the rotational speed, and indicates that it is inversely proportional to the rotational speed and proportional to the set value, as in the case of pressure control.

These can be similarly explained using the tubes shown in FIG. 6 and FIG. 1, which are shown in comparison with explanatory diagrams for pressure control.

As mentioned above, in the pressure/flow control system, the pump gain (dH/dN or dQ/d
N) changes and the number of revolutions decreases, the gain increases,
The present invention focuses on this point and creates a uniform loop by adding something to the control system that lowers the gain when the number of revolutions goes down and increases it when the number of revolutions goes up. This is a book that has a control system with good responsiveness in terms of gain.

That is, in the embodiment shown in FIG. 2, this is realized by guiding the output of the controller 8 to the inverter 9 through the square function generator 1z.

In the case of the conventional device shown in Fig. 1, the output of the A cylinder needle 8 is the N command and the inverter 9 ON setting input until that time.
The gain of the inverter 90 input signal with respect to the output signal of 111jittl is N/N=1.

On the other hand, in the case of the present invention shown in the second @, the output of the controller 8 is raised to the power of t-2 and is input to the inverter 90, so #
The gain of the input signal of the inverter 9 with respect to the output signal of the 14-section total 8 is N! /N = N. That is, since the gain from the controller 8 to the inverter 9 is proportional to N, if the number of revolutions N decreases, the gain decreases, and if the number of revolutions N increases, the gain increases.

In other words, by doing this, when summed up with the gain of the pump with pressure self-flow control, (1) NX2 in the case of pressure control
Ha/N=2Hs (Nx for Ill flow rate control
It becomes constant as Qa/N2Qa.

Therefore, even if the adjustment gain is determined at an arbitrary operating point,
Responsiveness will not deteriorate at other operating points, and there is no risk of cycling.Therefore, at any operating point, it is only necessary to adjust using a certain adjustment gain adjustment method, and set the gain to a smaller value with a margin. No need to adjust.

Above, we have assumed that the input signal Cc is the pump ON, but pumps are often used in places where N is relatively large, and motor slippage does not become extremely large, so this method is not suitable. Figure 8 shows a different embodiment of the invention.

As can be seen from equations (S and (13) above, the pump gain is proportional to the set value, so a variable gain amplifier 13 having a gain of 1 which is inversely proportional to the set value is added to the embodiment shown in FIG. 2. Nine things.

The input side resistance of the inverting amplifier 14 of this variable gain amplifier 13Fi operational amplifier is a variable resistor R, and this is linked with the pressure (flow rate) setting device 1', so that when the set value becomes smaller, the resistance of the variable gain amplifier 13 is also reduced, and the setting When the value is increased, the resistance of the variable gain amplifier 13 is also increased. The reason for this configuration is that 1 inverting amplifier 1
This is because the gain of 4 is determined by the two resistors, R and R, and the gain is CH. In addition, in Figure 1, the inverting amplifier 14
, so another inverting amplifier 15 is used.

Similar to the embodiment shown in FIG. 8, FIG. 8 further includes a companion divider 16 whose gain is inversely proportional to the set value, and the output of the square function generator 12 is converted to pressure (flow rate). ) is designed to be divided by the output signal of the setting device 7°.

As described above, with the device of the present invention, if the adjustment gain is determined at any operating point, the same responsiveness can be obtained at other operating points as well, so there is no need to set it on the safe side, and the responsiveness is maintained over the entire range. can be improved. Therefore, it is possible to reduce the amount of variation in the control amount and obtain extremely good operating performance, making it excellent in practice as a constant control device for pump discharge output or flow rate.

[Brief explanation of drawings]

Fig. 1 is a block diagram of a conventional device, Fig. 2 is a block diagram of an embodiment of the present invention, Figs. 3 to 7 are diagrams for explaining problems with the conventional device, and Fig. 3 is a pump Load characteristic diagram, Figure 4 is pressure @ flow rate characteristic diagram. Figure 5 is a graph of rotation speed vs. pressure change rate, Figure 6 is a graph of rotation speed vs. flow rate, Figure 7 is a graph of rotation speed vs. pressure change rate, and Figure 8 is a graph of rotation speed vs. pressure change rate.
9 and 9 are block diagrams of embodiments of the present invention, respectively. 1...Electric motor, 2...B/P. 3...Coupling, 4...Discharge pipe. 5...Detector, 6...Converter, 7...
Setting device, 8...PID controller. 9...Inverter, 12... Square function generator.

Claims (2)

[Claims] (1) A pump driven by an electric motor controlled by a variable speed drive, a detector for its discharge 1IIVc high pressure or Fi flow rate, and a negative feedback signal of this detector and a set value for the pressure or flow rate. and a controller that adjusts the control signal using the deviation signal of
A constant control device for pump discharge pressure or flow rate, characterized in that a square function generator is provided between the adjustment needle and the variable speed drive device. (2) A pump driven by an electric motor controlled by a variable speed drive, a pressure or flow rate detector provided on its discharge side, and a negative feedback signal of this detector and a pressure or 15! and an adjustment needle that adjusts a control signal using a deviation signal from a set value of a setter as an input signal, and a pump discharge pressure or flow rate constant control device that controls the variable speed drive device by the output of the adjustment needle. In addition, a square function generator, a variable gain amplifier and an inverting amplifier 15 having a gain inversely proportional to the setting value are provided between the adjustment needle and the variable speed drive device in conjunction with the setting device. A constant control device for pump discharge pressure or flow rate. (31) A pump driven by an electric motor controlled by a variable speed drive, a pressure or flow rate detector installed on its discharge side, a negative feedback signal of this detector and a setting value of a pressure or flow rate setter. and an adjustment needle that adjusts the control signal using a deviation signal between the pump discharge pressure and r
! In the constant control device for the amount a, a square function generator and a divider for dividing the output signal of the function generator by the output signal of the setting device are provided between the adjustment needle and the variable speed drive device. A constant control device for the pump discharge pressure or flow rate to reduce the amount of mold.