JPH0934557A - Water distribution terminal pressure controller - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a water distribution terminal pressure controller which can secure the desired water distribution terminal pressure in response to its demand factor and with the desired response when the terminal pressure is controlled at a fixed level. SOLUTION: A target discharge pressure value calculation means 9 calculates the correction value of the target discharge pressure value by the fuzzy inference based on the deviation between the water distribution terminal pressure and its set value and also the discharge flow rate of a water distribution pump, so that the terminal pressure is kept at its set value. The target discharge pressure value is corrected based on the calculated correction value. A water distribution pump rotational frequency control means 10 controls the rotational frequency of the distribution pump, so that the discharge pressure of the pump is set at the target discharge pressure value corrected by the means 9. Thus it is possible to apply the proper control even to a nonlinear change characteristic that is defined between the water distribution terminal pressure and the discharge flow rate of the pump.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a water distribution end pressure control device for controlling the water distribution end pressure of a water distribution pipe network so that the response can be changed according to the demand.

[0002]

2. Description of the Related Art Generally, tap water supplied to each home is
Water is distributed from a distribution pipe buried under the road near each home. The entire distribution pipe in this prescribed area is called the distribution network. Then, the water sent from the river is stored in a distribution reservoir via a purification plant, and the stored water is sent to a distribution pipe network by a distribution pump.

By the way, in order to comfortably use tap water in each home, the pressure of tap water is required to be as constant as possible. Therefore, it is necessary to control the pressure in the water distribution network, that is, the pressure at the water distribution end to be constant, and it is desired that the control response be a predetermined response depending on the amount of demand. Conventionally, so-called estimated end pressure control or PI control has been mainly performed for the water distribution end pressure control.

Estimated terminal pressure control is to estimate the pressure loss Rf in the water distribution network when the discharge pressure and the discharge flow rate of the water distribution pump are PO and Q, respectively, and express it as P1 = PO-Rf · Q. This is a control for adjusting the discharge pressure PO so that the distribution end pressure P1 is constant. The PI control is a discharge pressure of a water distribution pump calculated by proportional-plus-integral calculation based on a deviation between a water distribution end pressure detected by a pressure gauge installed at a predetermined position in the water distribution network and a set value of the water distribution end pressure. A target value is obtained, and thereby the discharge pressure is controlled so as to follow the discharge pressure target value.

[0005]

However, in the estimated terminal pressure control, there is an error between the estimated value and the actual value of the pressure loss Rf, and in that case, the desired water distribution terminal pressure cannot be obtained. Further, according to the designed parameter, the PI control can be controlled only with the same control response when the demand amount is large or small. That is, the control response cannot be changed depending on the discharge flow rate of the water distribution pump, which is the demand amount.

An object of the present invention is to provide a water distribution end pressure control device capable of obtaining a desired value of the water distribution end pressure with a desired response in accordance with the magnitude of demand when performing constant control of the water distribution end pressure. It is to be.

[0007]

According to a first aspect of the present invention, the discharge pressure target value of the water distribution pump is corrected by fuzzy inference based on the deviation between the water distribution end pressure and the set value of the water distribution end pressure and the discharge flow rate of the water distribution pump. A discharge pressure target value calculating means for calculating the amount and correcting the discharge pressure target value based on this correction amount, and a water distribution so that the discharge pressure of the water distribution pump becomes the discharge pressure target value corrected by the discharge pressure target value calculating means. The water distribution pump rotation speed control means for controlling the rotation speed of the pump is provided.

According to a second aspect of the present invention, in the first aspect of the present invention, the discharge pressure target value calculating means uses fuzzy inference based on the deviation between the distribution end pressure and the distribution end pressure set value and the discharge flow rate of the distribution pump. Based on both the first correction amount obtained and the second correction amount obtained by fuzzy inference based on the change rate of the deviation between the distribution end pressure and the set value of the distribution end pressure and the discharge flow rate of the distribution pump. The discharge pressure target value is corrected.

According to a third aspect of the present invention, in the second aspect of the invention, the discharge pressure target value calculating means includes a first correction amount and a second correction amount.
The correction amount of is multiplied by a predetermined gain.

According to the first aspect of the present invention, the discharge pressure target value calculating means uses the fuzzy inference based on the deviation between the distribution end pressure and the set value of the distribution end pressure and the discharge flow rate of the distribution pump to determine the distribution end pressure. The correction amount of the discharge pressure target value is calculated so as to be maintained at the terminal pressure setting value, and the discharge pressure target value is corrected based on this correction amount. The water distribution pump rotation speed control means controls the rotation speed of the water distribution pump so that the discharge pressure of the water distribution pump becomes the discharge pressure target value corrected by the discharge pressure target value calculation means. As a result, it is possible to appropriately control the non-linear change characteristic of the water distribution end pressure and the discharge flow rate.

According to the invention of claim 2, in addition to the operation of the invention of claim 1, the discharge pressure target value calculating means is based on the deviation between the distribution end pressure and the set value of the distribution end pressure and the discharge flow rate of the distribution pump. Fuzzy inference based on the first correction amount, the change rate of the deviation between the distribution end pressure and the set value of the distribution end pressure, and the discharge flow rate of the distribution pump. The discharge pressure target value is corrected based on both the second correction amount and the second correction amount. In this case, the first correction amount corresponds to the control operation signal of the I element (integral element) in PI control, and the second correction amount corresponds to the control operation signal of the P element (proportional element) in PI control. It can be regarded as control and designed. As a result, it is possible to perform the same control as the PI control while making a non-linear change in the discharge flow rate, which is impossible with the conventional PI control.

According to the invention of claim 3, in addition to the operation of the invention of claim 2, the discharge pressure target value calculating means multiplies the first correction amount and the second correction amount by predetermined gains. Thereby, the ratio of the I element and the P element can be changed.

[0013]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of the present invention. In FIG. 1, the purified water in the distribution reservoir 1 is distributed to the distribution pipe network 2 by n distribution pumps P1 to Pn. A discharge pressure gauge 3 and a discharge flow meter 4 are provided between the water distribution pumps P1 to Pn and the water distribution network 2, and a water distribution terminal pressure gauge 5 is provided at a predetermined position of the water distribution network 2.

The central management control room is provided with a human interface 6 which is an operating device such as a keyboard, and an operator can operate the human interface 6 while watching the display on the CRT display device 7. ing. The command signal from the human interface 6 is output to the arithmetic unit 8. The arithmetic unit 8 is
It has a discharge pressure target value calculation means 9, a water distribution pump rotation speed control means 10, a first fuzzy rule table 11 and a second fuzzy rule table 12.

In the discharge pressure target value calculation means 9, fuzzy inference is performed by the first fuzzy rule table 11 based on the deviation between the distribution end pressure and the set value of the distribution end pressure and the discharge flow rate of the distribution pump. A first correction amount is calculated. On the other hand, based on the change rate of the deviation between the distribution end pressure and the set value of the distribution end pressure and the discharge flow rate of the distribution pump, fuzzy inference is performed by the second fuzzy rule table 12, and thereby the second correction amount is calculated. To do.
Then, both of these correction amounts are added to the discharge pressure target value, the discharge pressure target value is corrected, and the corrected discharge pressure target value is output. Further, the water distribution pump rotation speed control means 10
The discharge pressure of the water distribution pump is the discharge pressure target value calculation means 9
The rotation speed of the water distribution pump is controlled so that the discharge pressure target value corrected in step 1 is reached.

That is, the discharge pressure target value of the water distribution pump P is calculated by fuzzy inference from the water distribution end pressure set value set by the operator and the measured water distribution end pressure. At this time, in the fuzzy inference, the correction amount of the discharge pressure target value is calculated based on the information of the discharge flow rate and the water distribution end pressure. The previous manipulated variable is added to this calculation result to obtain a new corrected discharge pressure target value, and the pressure of the water distribution pump is controlled by PI calculation.

Next, the calculation of the correction amount of the discharge pressure target value by the fuzzy inference in the discharge pressure target value calculation means 9 will be described. The input of fuzzy inference is the deviation between the distribution end pressure and the distribution end pressure set value (distribution end pressure deviation), the change rate of the deviation between the distribution end pressure and the distribution end pressure set value (change in distribution end pressure deviation). Rate), and the discharge flow rate of the water distribution pump. Based on these pieces of information, in the fuzzy inference, as shown in FIG.
The correction amount of the discharge pressure target value is calculated by the first and second fuzzy rule tables 12.

As shown in FIG. 2, in the first fuzzy rule table 11, the first correction amount of the discharge pressure target value is calculated from the distribution end pressure deviation and the discharge flow rate. In the second fuzzy rule table 12, a second correction amount of the discharge pressure target value is calculated based on the change rate of the distribution end pressure deviation and the discharge flow rate. Then, the adding means 14 adds the first correction amount from the first fuzzy rule table 11 and the second correction amount from the second fuzzy rule table 12 to correct the final discharge pressure target value. Calculate the amount. In this form, the first rule table 11 is I in PI control.
The second rule table 12 corresponds to the control (integral control) and the second rule table 12 corresponds to the P control (proportional control), and can be designed by considering it as the non-linear PI control.

Next, a method of fuzzy inferring the correction amount of the discharge flow rate target value will be described. FIG. 3 (a) shows a temporal transition of the water distribution end pressure, and FIG. 3 (b) shows a temporal transition of the discharge flow rate. Now, as shown in Fig. 3 (a), the deviation between the distribution end pressure set value Hsv and the distribution end pressure Hpv (t-dt) at time t-dt is e (t-dt), and the distribution at time t The deviation between the terminal pressure setting value Hsv and the water distribution terminal pressure Hpv (t) is e (t). In addition, as shown in FIG.
Let Qpv (t) be the discharge flow rate at.

At this time, the deviation e (t) between the water distribution end pressure set value Hsv and the water distribution end pressure Hpv (t) at the control cycle dt and the change rate de (t) of the deviation are obtained by the following equations.

E (t) = Hsv-Hpv (t) (1) de (t) = e (t) -e (t-dt) = {Hsv-Hpv (t)}-{Hsv-Hpv (t -dt)} = Hpv (t-dt) -Hpv (t) (2) Distribution end pressure deviation e (t), deviation change rate de (t) and discharge flow rate Qpv (t), which are the inputs of fuzzy inference. Is normalized as shown below during actual inference.

E (t) = Hsv-Hpv (t) (3) de (t) = e (t) -e (t-1) (4) E (t) = e (t) / Hemax ... (5) dE (t) = de (t) / Hdemax… (6) Qm = (Qmax + Qmin) / 2… (7) Qg = (Qmax-Qmin) / 2… (8) Q (t) = ( Qpv (t) -Qm) / Qg where e (t): Deviation between the distribution end pressure setting value and the distribution end pressure [mA
q] de (t): Deviation rate [mAq] Hsv: Distribution end pressure set value [mAq] Hpv (t): Distribution end pressure [mAq] Hemax: Distribution end pressure deviation range [mAq] Hdmax: Distribution end pressure Rate of deviation change [mAq] Qmax: Maximum discharge flow rate [m3 / h] Qmin: Minimum discharge flow rate [m3 / h] Qm: Average of Qmax and Qmin [m3 / h] Qg: Difference between Qmax and Qmin Absolute value [m3 / h] Qpv (t): Actual discharge flow rate [m3 / h] Q (t): Normalized discharge flow rate E (t): Deviation between distribution end pressure setting value and distribution end pressure Normalized value dE (t): Normalized value of deviation change rate t: Deviation E (t), deviation rate of deviation between distribution end pressure set value and distribution end pressure normalized by the formula above The correction amount of the discharge pressure target value is calculated from dE (t) and the discharge flow rate Q (t).

FIG. 4 (a) shows the membership functions that define the deviation between the water distribution end pressure set value and the water distribution end pressure, which are input variables, the deviation change rate, and the discharge flow rate. Also,
FIG. 4 (b) shows the membership functions that define the first correction amount of the discharge pressure target value and the second correction amount of the discharge pressure target value, which are output variables.

PB, PM, P of each membership function
The meanings of the symbols S, Z, NS, NM and NB are shown in FIG.
The details of the first fuzzy rule table 11 are shown in FIG.
The details of the second fuzzy rule table 12 are shown in FIG.

The first fuzzy rule table 11 shown in FIG.
The second fuzzy rule table 12 of FIG. 7 shows that a fast control response can be obtained in the daytime when the demand of water distribution is large, that is, in the daytime when the discharge flow rate is large, and it can be slowly controlled in the nighttime when the demand of water distribution is small. It is designed to obtain the controlled response. For example, in the example where the water distribution terminal pressure fluctuates as shown in FIG. 3, the water distribution end pressure becomes smaller than the set value at time t, and the discharge flow rate Qpv (t) is large (the water distribution demand is large).
Therefore, as can be seen from FIG. 5, the state of the water distribution end pressure deviation is PB, and the state of the discharge flow rate of the water distribution pump is PB. Therefore, as can be seen from the first fuzzy rule table 11 shown in FIG. 6, the first correction amount of the discharge pressure target value must be in the PB state, and the first correction of the discharge pressure target value is performed. A fuzzy inference result that the amount is positive and large is obtained. That is, the rule R1 of the first fuzzy rule table 11 in FIG. 6 is fired.

Similarly, the rate of change of pressure deviation at the water distribution end is negative and large. That is, since the water distribution end pressure is rapidly decreasing, a fuzzy inference result that the correction amount of the discharge pressure target value in the second fuzzy rule table 12 is increased by a positive value is obtained at this time. That is, the rule R1 of the second rule table 12 of FIG. 7 is fired. Therefore, in the fuzzy inference, it is inferred that the second correction amount of the discharge pressure target value is positive and large.

Based on the fuzzy inference value obtained by the above fuzzy inference, the corrected discharge pressure target value hsv (t),
The first correction amount dh1 (t) and the second correction amount dh2 (t) at that time
Is expressed by the following equation.

Hsv (t) = hsv (t-dt) + K1 * dh1 (t) + K2 * dh2 (t) (10) dh1 (t) = [Hinf1] * dhmax1 (11) dh2 (t) = [Hinf2] * dhmax2… (12) where hsv (t): Target discharge pressure value [mAq] dh1 (t): 1st correction amount of target discharge pressure value [mAq] dh2 (t): Discharge pressure Second correction amount of target value [mAq] dhmax1: Range of correction amount of discharge pressure target value 1 [mAq] dhmax2: Range of correction amount of discharge pressure target value 2 [mAq] K1: First correction of discharge pressure target value Volume gain K2: Gain of second correction amount of discharge pressure target value dt: Control cycle [Hinf1]: Fuzzy inference value (Normalized value of first correction amount of discharge pressure target value) [Hinf2]: Fuzzy inference Value (normalized value of the second correction amount of the discharge pressure target value) In the equation (10), the gain K1 of the first correction amount of the discharge pressure target value and the gain of the second correction amount of the discharge pressure target value K2
Can be set as needed according to the necessity.
Can be weighted.

As described above, in this embodiment, the discharge pressure target value computing means 9 distributes the distribution end pressure set value from the human interface 6 and the distribution end pressure which is the detection value from the distribution end pressure gauge 5. The end pressure deviation is obtained, and the fuzzy inference value is obtained using the first fuzzy rule table 11 from the combination of the distribution end pressure deviation and the discharge flow rate from the discharge flow meter 4. Further, the discharge pressure target value calculation means 9 obtains the change rate of the water distribution end pressure deviation, and from the combination of the change rate of the water distribution end pressure deviation and the discharge flow rate from the discharge flow meter 4, the second fuzzy rule table 12 is obtained. Is used to find the fuzzy inference value.

Then, the first fuzzy rule table 1
The first correction amount of the discharge pressure target value obtained based on the fuzzy rule inference value obtained from 1 and the discharge pressure target obtained based on the fuzzy inference value obtained from the second fuzzy rule table 12 The correction amount of the discharge pressure target value is calculated by the sum of the value and the second correction amount, and the discharge pressure target value is corrected every inference cycle. Distribution pump rotation speed control means 1
0 controls the number of rotations of the water distribution pumps P1 to Pn in a feedback manner so that the deviation between the discharge pressure target value corrected in this way and the discharge pressure detection value from the discharge pressure gauge 3 becomes zero.

Here, the gain K1 of the first correction amount is K1 = 1,
When the gain K2 of the second correction amount 2 is set to K2 = 0, the first correction amount by the first fuzzy rule table 11 becomes the correction amount of the discharge pressure target value, which is equivalent to the invention according to claim 1. . That is, the second fuzzy rule table 12 is deleted. Also, the gain K1 of the first correction amount is set to K1 =
If the gain K2 of 0.5 and the second correction amount 2 is K2 = 0.5, the weight of the first correction amount by the first fuzzy rule table 11 and the second correction amount by the second fuzzy rule table 12 becomes It is an even correction amount of the discharge pressure target value,
It is equivalent to the invention described in claim 2. Then, the gain K1 of the first correction amount and the gain K2 of the second correction amount 2 are arbitrarily weighted and set as needed, whereby the invention of claim 3 is obtained.

[0032]

As described above, according to the present invention, the discharge flow rate and the water distribution end pressure are detected, and the amount of correction of the discharge pressure target value is determined by the fuzzy rule table which inputs the discharge flow rate and the water distribution end pressure deviation. Since the target value of discharge pressure can be corrected for each inference cycle by calculation, in order to perform constant control of the water distribution end pressure, it is necessary to control the water distribution end pressure of a desired value with a desired response according to the demand amount. You can

Further, the first fuzzy rule table having the discharge flow rate and the water distribution end pressure deviation as input and the second fuzzy rule table having the discharge flow rate and the change rate of the water distribution end pressure deviation as the input respectively. By calculating the correction amount of the target value of the discharge pressure based on the sum of the fuzzy calculation outputs of, the control with higher accuracy can be performed.

Further, since these correction amounts can be respectively multiplied by a predetermined gain, weighting can be performed, so that a wide range of control is possible.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the invention.

FIG. 2 is an explanatory diagram showing a relationship between a first fuzzy rule table and a second fuzzy rule table in FIG.

FIG. 3 is a characteristic diagram showing a characteristic example of a water distribution end pressure and a discharge flow rate.

FIG. 4 is an explanatory diagram showing a membership function used in fuzzy calculation by the discharge pressure target value calculation means of the present invention.

FIG. 5 is an explanatory diagram showing the meaning of symbols in the membership function of the present invention.

FIG. 6 is an explanatory diagram of a first fuzzy rule table according to the present invention.

FIG. 7 is an explanatory diagram of a second fuzzy rule table according to the present invention.

[Explanation of symbols]

 1 Reservoir 2 Water distribution network 3 Discharge pressure gauge 4 Discharge flowmeter 5 Water distribution end pressure gauge 6 Human interface 7 CRT display device 8 Calculation device 9 Discharge pressure target value calculation means 10 Water distribution pump rotation speed control means 11 First fuzzy Rule table 12 Second fuzzy rule table P1 to Pn Water distribution pump

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification number Reference number within the agency FI Technical display location G05D 27/02 G05D 27/02 29/00 29/00 (72) Inventor Masanaga Niiyama Tokyo Fuchu Toshiba Town Fuchu Factory, Toshiba Corporation (72) Inventor Tatsuo Ashiki 1-1-1 Shibaura, Minato-ku, Tokyo Inside Toshiba Head Office

Claims (3)

[Claims] 1. The purified water in a distribution reservoir is supplied to a distribution pipe network by a distribution pump, and the discharge pressure of the distribution pump is controlled to reach its discharge pressure target value in response to fluctuations in the demand for the purified water. In a water distribution terminal pressure control device for controlling the water distribution terminal pressure of the water distribution network at a constant level, fuzzy inference based on a deviation between the water distribution terminal pressure and a set value of the water distribution terminal pressure and a discharge flow rate of the water distribution pump. And a discharge pressure target value calculating means for calculating a correction amount of the discharge pressure target value of the water distribution pump and correcting the discharge pressure target value based on the correction amount, and calculating the discharge pressure target value of the discharge pressure of the water distribution pump. Water distribution pump rotation speed control means for controlling the rotation speed of the water distribution pump so that the discharge pressure target value corrected by the means is provided. . 2. A first correction amount obtained by fuzzy inference based on a deviation between the water distribution end pressure and a set value of the water distribution end pressure and the discharge flow rate of the water distribution pump, the discharge pressure target value calculating means; Based on both the rate of change of the deviation between the water distribution end pressure and the set value of the water distribution end pressure and the second correction amount obtained by fuzzy inference based on the discharge flow rate of the water distribution pump, the discharge pressure target value is calculated. The water distribution end pressure control device according to claim 1, wherein the water distribution end pressure control device is modified. 3. The water distribution according to claim 2, wherein the discharge pressure target value calculation means is configured to multiply the first correction amount and the second correction amount by predetermined gains, respectively. Terminal pressure control device.