JPH0498502A - Control system for number of pumps with fuzzy inference - Google Patents

Abstract

PURPOSE:To control the number of pumps without requiring a large set level interval by incarnating the judgements and procedures of a skilled operator with a controller and attaining the quick follow-up of the start/stop actions of pumps to the sudden increase or decrease of the estimated inflow. CONSTITUTION:An industrial personal computer 1 stores the water level setting data, the water inflow estimation data, etc., and the arithmetic processing for control of the number of pumps. A fuzzy inference controller 2 stores the rules and the membership functions and performs a fuzzy inference. The computer 1 judges whether the estimated water inflow Q is optimum or not when the water inflow estimation data is inputted to the computer 1. If so, the computer 1 checks whether the optimum time TM is long enough or not. If so, the present number of pumps is maintained. If the inflow Q is not optimum and the improper time is long enough, the number of pumps is controlled.

Description

[Detailed Description of the Invention] A. Industrial Application Field The present invention relates to a pump number control method for commanding an increase/decrease in the number of pumps in operation based on a prediction of inflow water into the well and the well water level in a water sampling well such as a water supply system. In particular, it relates to a system for controlling the number of pumps using fuzzy inference.

B. Summary of the Invention The present invention uses water level setting data and inflow water volume prediction data in a pump number control method that commands an increase or decrease in the number of pumps in operation based on a prediction of inflow water into the well and the well water level in a water sampling well of a water supply system, etc. It is equipped with an industrial PC that performs storage and number control calculation processing, a fuzzy inference controller that is set with rules and membership functions based on fuzzy inference, and input/output devices for interpersonal and system systems, and the operating order is set in advance. By controlling the number of multiple pumps, it is possible to realize this idea without requiring experienced operators all the time, and it is possible to quickly start and stop the pumps in response to sudden increases and decreases in the amount of inflow water. It is possible to provide a technology that does not require setting level intervals to be very wide even if there are a large number of pumps.

C1 Conventional Technology The method of controlling the number of pumps currently employed in pump wells such as water supply systems measures the water level of the pump well and sets the number of pumps in operation according to the water level.

The number of pumps in operation needs to be increased or decreased according to changes in the amount of inflow water. For this reason, a change in the amount of water is detected by a change in the pump well water level, and the number of operating pumps is changed to control the pump well water level within a certain range.

FIG. 9 is a configuration diagram showing an example of a conventional pump number control circuit. In the figure, 91 is a pump well, 92 is a water level gauge, 93 is a converter, 94 is an alarm setting device, 95 is an operation order switching circuit,
Reference numeral 96 denotes a pump operation circuit, which manually inputs a signal from a water level meter 92 that measures the water level of the pump well 91 to an alarm setting device 94 via a converter 93, converts it into a pump operation signal, and controls the pump by an operation order switching circuit 95. Control the number of vehicles in operation. The alarm setting device 94 is set as shown in FIG. 10, for example. In the figure, for example, the NO.1 pump starts when the water level in the pump well reaches L5 or higher, and operates until the water level drops to L2 or lower. When the amount of inflow water exceeds the capacity of the NO,l pump and the water level rises further and reaches L6,
Start the NO.2 pump. Pump N003 is operated similarly.

D3 Problems to be Solved by the Invention The conventional pump number control method described above has the advantage of being simple, but when the amount of inflow water fluctuates frequently, it is difficult to start/start the pumps.
This is not desirable as the number of stops increases. Also, depending on the relationship between the amount of inflow water and the pump capacity, the pump may start/stop more frequently even if the change in water amount is small, so it may not be suitable if the amount of inflow water increases due to repeated installations. be.

That is, the conventional pump number control method has the following problems.

(1) If the amount of inflow water is predicted accurately and the operator is an experienced operator, it is possible to set the optimal number of operating units, but it is not possible to always have an experienced operator on hand to set the number of operating units. It's fleeting.

(2) The number of pumps in operation is determined by the water level in the pump well, so if the volume of the pump well is small compared to the discharge capacity of the pump, the start/stop of the pumps will follow the sudden increase or decrease in the amount of inflow water. Can not.

(3) When controlling the number of pumps based on water level, if there are many pumps, it is necessary to have a wide interval between HWL (high set water level) and LWL (low set water level), resulting in large fluctuations in pump well water level. Furthermore, if the interval is narrowed, control cannot be achieved unless the detection accuracy of the water level gauge is very good, and even if the detection accuracy is cleared, the frequency of starting/stopping of the pump increases, which is not preferable.

The present invention was created in view of these problems, and
This idea can be implemented without requiring Heteran operators to be present all the time, and the amount of inflow water can rapidly increase.

It is an object of the present invention to provide a pump number control method that allows the start/stop of the pumps to quickly follow the decrease in the number of pumps, and that does not require setting level intervals to be very wide even if the number of pumps is large.

A means for solving the above-mentioned problems in the present invention is a method for controlling the number of pumps that commands an increase or decrease in the number of pumps in operation based on the water level and inflow water volume of wells and distribution reservoirs of the water system.
An industrial computer that accumulates water level setting data and inflow water volume prediction data and performs arithmetic processing for controlling the number of units, a fuzzy inference controller that is set with rules and maintenance functions based on fuzzy inference, and input/output devices for personnel and systems. A pump number control method is used to control the number of pumps in which the operating order is set in advance.

F0 Effect The present invention is a pump number control method that determines the number of pumps in operation by predicting the amount of inflow water and determining whether to increase or decrease the number of pumps using fuzzy reasoning.

Fuzzy inference is a theory that has gained attention with the advancement of AI (artificial intelligence) technology. In Al, in a certain field, it is possible to construct an automatic diagnosis system similar to that of a human by storing the required knowledge base and thinking and judgment rules in a computer, but it is not possible to accept ambiguous expressions like a human. It is not possible to do so, but it must be expressed logically. In order to compensate for this shortcoming and introduce ambiguity into computers, fuzzy control is being emphasized. Don't use fixed numbers.

Specifically, membership functions are used as the knowledge base, and dedicated inference rules are provided.

In the present invention, the current water level and the water level fluctuation prediction are each set in, for example, five stages as a maintenance function, and the inference rule is based on the matrix configuration of these, and the following steps are performed.

(1) Predict the amount of inflow water.

(2) Determine whether the predicted amount of inflow water is within the optimal range.

(3) When the predicted amount of inflow water is within the optimal range, the number of pumps will remain the same if the optimal time is long enough.

(4) If the predicted inflow water amount is outside the optimal range, increase or decrease the number of pumps if the unsuitable time is long enough.

(5) If the optimal time or unsuitable time is not long enough, increase the number of pumps by fuzzy reasoning.

Determine whether to decrease or maintain the status quo.

(6) If it is decided to increase or decrease the number of pumps, issue the command. Increase or decrease the number of machines one by one.

G. Embodiments Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a configuration diagram of an embodiment of the present invention.

In the figure, 1 is an industrial computer that stores water level setting data and inflow water volume prediction data and performs arithmetic processing for controlling the number of units, 2 is a fuzzy inference controller that is set with rules and membership functions based on fuzzy inference, and 3 is a personal computer. Machine
interface, 4 is a process personal output device, and 5 is a LAN (Locai Area N) that connects them.
etwork).

The industrial PC 1 is used to accumulate data such as water level setting data and inflow prediction data, and perform calculation processing for controlling the number of units.
Equipped with a LAN interface. The fuzzy inference controller 2 stores rules and member functions and performs fuzzy inference. The man-machine interface 3 is a human input/output device equipped with a CRT, keyboard, and mouse, and performs various settings, system maintenance, and result display. The process individual output device 4 receives inflow prediction data and current water level data, and outputs an instruction to increase by one or decrease by one.

The above-mentioned pump number control device can be incorporated into a monitoring control device or a local control device for common use.In this embodiment, as shown in FIG. n units) 22.

FIG. 3 is a flowchart showing an example of the pump number control method of the present invention using the above device. In the figure, the flow starts when inflow prediction data is manually entered into the industrial personal computer 1, and the personal computer 1 determines whether the predicted inflow water amount Q is within the optimum range or not. The optimal range is, where q is the amount of water discharged from each pump, and k is the current number of pumps in operation.
As shown in FIG. 4(a), it holds true when (k-1)q≦Q≦(k+1)q. Also, if there are two types of pumps, large and small, and there is one pump with a small discharge water volume (qs), the optimal range is a: 0 units, as shown in Figure 4 (b). b: Optimal range when there is one small pump; C: Optimal range when there are two large pumps; d: Optimal range when there are two large pumps; e: Optimal range when there are three large pumps. Optimal range, f~:
This is the optimal range when there are 4 or more large pumps.

Now, if the predicted amount of inflow water is within the optimal range, then it is checked whether the optimal time TM is sufficiently long. What is the optimal time TM?
If the bottom area of the pump well is S and the current water level is h, then M LWL≦h+51: (Q k q)/S codt≦H
The longest time for which WL is established, that is, if the current number of units is operated and the predicted inflow water @Q is correct, it means the time that the water level is maintained between LWL and HWL.

A reference time To is set in advance, and TM≧T.

If so, the optimal time is sufficiently long, and T M < T o
If so, the optimal time is short. If the optimal time is long enough, the number of pumps will remain the same.

Using the same concept, unsuitable time and its reference time are also set, and if the predicted inflow water amount Q is outside the optimal range, the number of pumps is increased or decreased if the unsuitable time is long enough.

If the optimal time or unsuitable time is not long, it is determined by fuzzy inference whether to increase or decrease the number of pumps or maintain the current status.

As already mentioned, fuzzy inference does not involve numerical specification. In this example, the current water level and water level fluctuation prediction are respectively NB, SS, M, and M as membership functions.
Set to 5 ambiguous stages of PS and PB, and between them
Set up inference rules as shown in a matrix in a table.

Table 1 NB NS NS Water level fluctuation prediction NB NS M PS PB NBI NBI NB'NSI M N B'NB' N sl M S NB: NS PS'PB NS PSIPBiPB The above table shows the IP-THEN rules in a matrix. For example, if IF ~ current water level is PS (slightly close to the control upper limit) and water level fluctuation prediction is PB (water level will rise considerably), then THEN ~ number of pumps is PB (needs to increase). shows 5x5=25 rules.

Hereinafter, fuzzy inference in the present invention will be explained in more detail.

FIG. 5 is an explanatory diagram of membership functions.

As shown in Figure (a), the water level is HH (upper limit).

Set H (upper control limit), L (lower control limit), and LL (lower limit), and set the water level control range between H and mustard.

The membership function of the current water level for this is shown in the same figure (
As shown in b), it is quite close to the NB/control lower limit.

NS: Slightly close to the lower control limit.

M: Medium.

PS: Slightly close to the management upper limit.

PB It's pretty close to the management limit.

In addition, the membership function for water level fluctuation prediction is shown in the same figure (c
), the NB water level will drop considerably.

NS, the water level will drop slightly.

M: The water level does not change.

PS The water level will rise slightly.

PB, the water level will rise considerably.

On the other hand, the membership function for increasing or decreasing the number of pumps is as shown in Figure (d): NB: The number of pumps should decrease or decrease.

NS: There is a slight need to reduce the number of pumps.

M: The number of pumps can be maintained as is.

PS: There is a slight need to increase the number of pumps.

PB: The number of pumps needs to be increased.

It is.

FIGS. 6 and 7 are explanatory diagrams of inference rules.

FIG. 6(a) is a diagram for examining the predicted amount of inflow water. In the figure, the straight line indicates the discharge amount q of the pump, and the curved line indicates the estimated estimated amount of inflow. The upper dotted line indicates the upper limit for managing the current number of devices, and the lower dotted line indicates the lower limit for managing the current number of devices. If the current time is 0, it is predicted that the predicted inflow will exceed the management upper limit between time tI and time t2, so it is necessary to increase the number of vehicles at time t1, and to decrease the number at time t2.

As shown in the flowchart of FIG. 3, the time T is the time from when a start or stop command is issued until the operation is completed, and the L2 time is set early in consideration of this. FIG. 6(b) is a water level time characteristic diagram corresponding to the above.

FIG. 7 shows fuzzy inference for the predicted inflow water amount and water level settings as shown in FIG. When −t 1, the seventh
In the current water level chart shown in Figure (a), if the water level is h' between LWL and HWL, the degree of necessity for changing the number of pumps (0 to 1
) is obtained by the line of intersection with the triangle whose apex is PS and the line of intersection with the triangle whose apex is PB. In addition, if the water level fluctuation prediction diagram shown in Fig. 7(b) indicates a fluctuation h' between -h and +h, similarly, the intersection line with the triangle with M as the apex and the triangle with PS as the apex. The intersection line of is obtained, and the necessity of changing the number of pumps (-1 to
+1) is shown on the plus side, as shown by the hatched area in FIG. 7(c). If the result of this inference is 0.75% or more, it will be determined that there is an increase of 11 units, and if it is less than -0.75%, it will be determined that there will be a decrease of 11 units.

Figures 7(d), (e), and (f) show the fuzzy inference when t=t2, and as shown in Figure 7(f), the inference results are shown on the negative side, and it is determined that one vehicle is reduced. Ru.

As described above, when it is decided to increase or decrease the number of pumps or to maintain the status quo, the flow branches to each direction, and when increasing or decreasing the number of pumps, it is assumed that it is done one by one, and a command to increase or decrease by one is issued.

In this embodiment, control of the number of pumps to remove inflow water has been shown, but the control method of the present invention can also be applied to control of the number of pumps 83 that send water from a water purification reservoir 81 to a water distribution reservoir 82, as shown in FIG. Applicable. In that case, HWL and LWL may be changed, and inflow prediction may be replaced with water distribution prediction. The control method of the present invention relies on some type of water volume prediction and that the prediction is accurate to some extent. However, in the case of rainwater, prediction methods such as prediction using a rainwater runoff analysis model or pump well There are predictions based on water level changes, and in the case of water distribution there is demand forecasting, and by combining these predictions, the effectiveness will increase. Further, the functions of the present invention can be incorporated into a supervisory control device or a local control device.

This example clearly has the following effects.

(1) Since the pump operation know-how of experienced operators is expressed by rules and membership functions, it is possible to set the number of pumps close to how humans think.

(2) The pump can be started/stopped quickly (with the shortest delay) in response to a sudden increase or decrease in the amount of inflow water.

(3) Even if there are a large number of pop-up devices, the number can be controlled without making the distance between HWL and LWL very wide.

(4) It is applicable to both fixed-speed pumps and variable-speed pumps, and even if the capacities of the plurality of pumps are the same, they can be controlled in six combinations with different sizes.

(5) Fuzzy inference rules and membership functions are CR
It can be easily added and changed using T, keyboard, and mouse, and various settings such as To and S can also be made using CRT operations.

(6) The volume of pump wells, the capacity and number of pumps, and the characteristics of inflow patterns can be easily expressed using fuzzy inference rules and membership functions, and can be packaged as general-purpose software that enables flexible system maintenance. .

H1 Effects of the Invention As explained above, according to the present invention, the idea can be realized without requiring a full-time operator, and the start/stop of the pump can be quickly followed in response to a sudden increase or decrease in the amount of inflow water. Therefore, it is possible to provide a pump number control method that does not require setting level intervals to be so wide even if there are many pumps.

[Brief explanation of drawings]

Fig. 1 is a block diagram of an embodiment of the present invention, Fig. 2 is a block diagram of an application example of the present invention, Fig. 3 is a flowchart of an embodiment of the present invention, and Fig. 4 is an explanatory diagram of the optimum control range. , FIG. 5 is an explanatory diagram of membership functions, FIGS. 6 and 7 are explanatory diagrams of inference rules, FIG. 8 is a configuration diagram of another application example of the present invention, and FIG. 9 is a configuration diagram of a conventional example. , FIG. 10 is an explanatory diagram of the pump operating water level. 1...Industrial PC, 2...Fano inference controller, 3...Man-machine interface, 4...
・Process personal output device, 5...LAN, 21-・Pump well, 22...Pump, 81...Water purification pond, 82...
・Water reservoir, 83...Pump, 9I...Pump well, 9
2... Water level gauge, 93-... Converter, 94-... Alarm setting device, 95... Operation order switching circuit, 96... Pump operation circuit. that's all. 1 other person (a) (b) (a) (b) (C) (d) (a) (b) (a> (cl) (b) (e) (c) (f)

Claims (1)

[Claims] (1) In a pump number control method that commands an increase or decrease in the number of pumps in operation based on the water level of a water system's wells or distribution reservoirs and the amount of inflow water, industrial use that accumulates water level setting data and inflow water volume prediction data and performs calculation processing to control the number of pumps. It is equipped with a personal computer, a fuzzy inference controller in which rules and membership functions based on fuzzy inference are set, and input/output devices for humans and systems, and is capable of controlling the number of pumps whose operating order is set in advance. Pump number control method using fuzzy inference.