JP5723642B2 - Distribution pressure control system - Google Patents

Description

  The present invention relates to a water distribution pressure control device in a case where purified water is pumped from a distribution area to a consumer at the end through a water distribution pipe network. Considering disturbances to the distribution system such as modeling errors in the pipe resistance model used for distribution pressure control, sudden water demand in the event of a fire, and water interchange in the distribution section, etc. The present invention relates to a water distribution pressure control device capable of precisely controlling the terminal pressure.

  In patent document 1, in a distribution pressure control device for supplying purified water by a pump from a distribution pipe network of a water supply system arranged through a pipeline from a distribution area to a terminal consumer, in a distribution block to be controlled, Pipe resistance is modeled based on actual process data of inflow rate, discharge pressure, terminal pressure, and demand volume, and control performance degradation caused by demand fluctuations and aging of pipeline network processes can be suppressed using the model. A distribution pressure control device is provided.

  In Patent Document 2, in order to precisely control the end pressure in response to an interchange between water distribution areas or in case of an abnormality such as a fire, the state of the water distribution pipe network is simulated using real-time process data. It realizes water distribution control that can be set by automatically calculating the optimum operation amount for the operation point including the injection point.

JP2009-209523 JP 2006-104777 A

  In Patent Document 1, it is possible to maintain the control accuracy corresponding to the process characteristic change due to aging, but the modeling error of the pipe resistance model is not taken into consideration, and the terminal pressure is precisely controlled to the limit lower limit value. There is a problem that it is difficult to do. This consumes extra pump energy. In addition, there is a problem that it is difficult to maintain control performance in response to sudden disturbances such as the flow rate of fire extinguishing plugs, which is different from the demands of homes and factories.

  In Patent Document 2, it is possible to control the end pressure by capturing pressure fluctuations in the water distribution system using a pipe network model, but because the calculation load of the pipe network calculation is large and the control cycle is large, the flow rate of the digestive plug etc. There is a problem that it is difficult to control in response to various flow rate changes. That is, there is a problem that the terminal pressure deviates from the target value more than expected.

  Therefore, it is possible to model the pipe resistance model of the distribution block to be controlled, calculate its modeling error, and based on the pipe resistance model considering the pressure modeling error, the end pressure may be higher than the target value An object of the present invention is to provide a water distribution pressure control device that enhances the performance.

Distribution pressure control system
Discharge pressure measured by the discharge pressure measuring instrument installed between the distribution pipe network and the pump, and end pressure installed between the water pipe of the distribution destination that receives water supply from the distribution pipe network and the distribution pipe network Based on the end pressure measured by the measuring instrument and the flow measured by the flow measuring instrument installed between the distribution pipe network and the pump, the distribution pipe network reflects the influence of the modeling error at a predetermined level. A pipe resistance model generation unit for generating a pipe resistance model of
A pressure loss calculation unit for calculating a pressure loss amount of water pressure generated in the distribution pipe network based on a pipe resistance model and a distribution flow rate pattern that the distribution pressure control system has in advance;
A target discharge pressure calculation unit that receives the target value of the terminal pressure and calculates the target discharge pressure based on the pressure loss amount and the target value of the terminal pressure;
A rotation speed control unit that controls the rotation speed of the pump so as to achieve the target discharge pressure.

According to the present invention, the pipe resistance model of the water distribution block to be controlled is modeled and its modeling error is calculated. Based on the pipe resistance model taking into account the modeling error, the end pressure is calculated even in the worst case. It is possible to control the water distribution pressure so that becomes more than the target value. In addition, sudden demand (disturbance) that is different from normal demand patterns such as digestive plug flow is measured with a flow sensor to quickly determine sudden demand, and a target discharge pressure with a cycle shorter than the original using a pipe resistance model. By calculating, water distribution pressure can be precisely controlled. In addition, it is possible to precisely control the water distribution pressure by constructing an original pipe resistance model that takes into account the sudden demand model (disturbance) that is different from the original demand pattern.

It is a figure which shows the structural example of a distribution pressure control system. It is a figure which shows the example of the data stored in a database. It is a graph showing the example of a relationship between a flow volume and a pressure loss. It is a flowchart which shows the example of a distribution pressure control process. It is a figure which shows the example of a demand pattern. It is a figure which shows the example of the flow volume-head characteristic of a pump. It is a figure which shows an example of the control performance when not considering the dispersion | variation in an estimated value. It is a figure which shows the other structural example of a distribution pressure control system. It is a figure which shows an example of the change of the terminal pressure when the digestive plug flow volume occurs and when it is not so It is a flowchart which shows another example of a distribution pressure control process. It is a figure which shows the other structural example of a distribution pressure control system. It is a figure which shows an example of the coefficient value of a pipe line resistance model, and the standard error of a coefficient. It is a flowchart which shows another example of a distribution pressure control process.

  Embodiments of the present invention will be described with reference to the drawings.

  The first embodiment will be described with reference to FIGS.

  FIG. 1 is a configuration diagram of a water distribution pressure control system according to the first embodiment. The control system includes a distribution pipe network 1, a distribution reservoir 11, a first pressure sensor 2 that measures discharge pressure, a second pressure sensor 3 that measures terminal pressure, a flow rate sensor 4 that measures distribution flow rate, and pumps 8 and 9. 10, rotational speed sensors 5, 6, 7 for measuring the pump rotational speed, DB (database) 101 for storing measurement time-series data, pipe resistance model identification means 102, measured values by the above-mentioned various sensors, pipe resistance model The water supply pressure control device 100 controls the pump rotational speed so as to realize the target terminal pressure with the target terminal pressure as an input.

  The first pressure sensor 2 is a sensor that measures the pressure (discharge pressure) of water distributed from the pump to the water distribution network 1 and is installed between the pumps 8, 9, 10 and the water distribution network 1. The second pressure sensor 3 is a sensor for measuring a water distribution pressure (terminal pressure) to a water pipe of a supply destination (also called a water distribution destination) that receives water supply from the water distribution pipe network 1. It is installed at the boundary with the previous water pipe. The flow rate sensor 4 is a sensor that measures the flow rate of water distributed from the pump to the water distribution network 1, and is installed between the pumps 8, 9, 10 and the water distribution network 1.

  Each of the DB 101, the pipe resistance model identification means 102, and the water distribution pressure control device 100 is a computer having a processor, a memory, and a storage device such as an HDD. That is, the DB 101 acquires measurement values from the various sensors described above by the processor executing a program stored in the memory, and saves the measurement values as DB data in a storage device. The pipe resistance model identification means 102 accesses the DB 101 by the processor executing a program stored in the memory, obtains measured values of various sensors, models the pipe resistance, and estimates modeling errors. Calculate the value. In the water distribution pressure control apparatus 100, various means described later included in the water distribution pressure control apparatus 100 are realized by the processor executing various programs stored in the memory.

  Note that the DB 1010, the pipe resistance model identification means 102, and the water distribution pressure control device 100 may be configured by different computers or by the same computer.

  The water distribution pressure control apparatus 100 includes a demand prediction unit 103, a pressure loss calculation unit 104, a target discharge pressure calculation unit 105, and a rotation speed control unit 107. The pressure sensors 2 and 3 are respectively installed at the entrance and the end of the water distribution pipe network, and measure the discharge pressure and the end pressure, respectively. The flow sensor 4 is installed at the entrance of the distribution pipe network and measures the distribution flow rate.

  The DB 101 stores the measurements of various sensors at a predetermined time, that is, the values of the field flow rate, the discharge pressure, and the terminal pressure. An example is shown in FIG. In this example, data is measured and stored every 3 hours. It is also possible to improve the accuracy of pipe resistance model identification described later by shortening the measurement cycle.

  The pipe resistance identification means 102 uses the data stored in the database to model the pipe resistance model and estimate the level of modeling error. Here, the pipe resistance model is given by the following equation and is stored in the memory of a computer constituting the pipe resistance identifying means 102.

P = Pe + h + k · Q ^α (1)
here,
P: Discharge pressure (m)
Pe: Terminal pressure (m)
h: Elevation of discharge pressure measurement point (m)-Elevation of end pressure measurement point (m)
k: Constant Q: Water distribution flow rate (m ³ / s)
α: Constant (uses values of 1.85 and 2.0)
h is a known real number, and is set in advance in a memory of a computer constituting the pipe resistance identifying means 102. Since time series data of P, Pe, and Q exist in the DB 101, the pipe resistance identification unit 102 can estimate (calculate) the constant k by the least square method using the equation (1). Here, α can also be estimated together with k as an unknown parameter. When calculating α, the pipe resistance identifying means 102 applies the least square method after calculating the logarithm of both sides of the equation (1).

  The pipe resistance identification means 102 can estimate the standard error σk representing the variation of the coefficient estimated value together with the estimated value k0 of the coefficient k by the least square method. This represents how much the coefficient k varies around the estimated value (expected value) as a standard deviation.

  For example, the probability that the coefficient k falls between k0-2σk and k0 + 2σk is about 95%. This is shown in a graph in FIG. In the graph, the pressure loss H is (P-Pe). A range surrounded by two curves represented by dotted lines is a 95% confidence interval. It is the case of the above curve, that is, the field where k = k0 + 2σk, that the end pressure decreases most. Therefore, if the control is performed using the pipe resistance model in the case of k = k0 + 2σk, the terminal pressure can be kept substantially higher than the target value.

  FIG. 7 shows a case where control is performed using a pipe resistance model that does not consider variation in coefficient k, and control is performed using a pipe resistance model when k = k0 + 2σk in consideration of variation in coefficient k. Is a comparison. If the variation of the coefficient k is not taken into account, there is a possibility that it will often fall below the target pressure. On the other hand, if the control is performed using the pipe resistance model in the case of k = k0 + 2σk in consideration of the variation of the coefficient k, 97. Since the actual terminal pressure will exceed the target value with a probability of 5% (the probability that the actual terminal pressure will be below the target value is 2.5%), the possibility of controlling the terminal pressure above the target pressure is increased. . Furthermore, energy consumption of the pump can be minimized by setting the target pressure of the model value to the lower limit.

  Now, an outline of the process of the water distribution pressure control device for realizing this control will be described with reference to FIGS.

  In the demand prediction means 103 in FIG. 1, for example, the distribution pressure control device 100 stores the demand pattern data for each season and day of the week stored in the storage device (that is, the distribution flow rate pattern data, for example, the demand pattern in FIG. 5). To predict future demand. For example, if the control cycle of the control device is 5 minutes, the current demand Q0 and the demand Q2.5 2.5 minutes ahead are searched. When the current distribution flow rate measurement value is Q, the demand forecast amount Qf is calculated by the following equation.

Qf = Q-Q0 + Q2.5 (2)
Next, the pressure loss calculation means 104 calculates the pressure loss H based on the following equation.

H = h + (k0 + 2σk) Qf ² (3)
The target discharge pressure calculation means 105 receives the input of the target end pressure Pe0, adds the pressure loss to this, and calculates the target discharge pressure P0 by the following equation.

P0 = Pe0 + h + (k0 + 2σk) Qf ² (4)
The discharge pressure control means 106 and the rotation speed control means 107 control the pump rotation speed by a signal to each pump so that the calculated target discharge pressure matches the measured discharge pressure. First, the discharge pressure control means 106 determines the target rotation speed N0. FIG. 6 shows a performance curve of a pump illustrating the performance characteristic data of the pump that the water distribution pressure control device 100 has in its storage device. The vertical axis represents the flow rate Q and the horizontal axis represents the head H. ing. Performance curves for single pump operation, two pump operation, and three pump operation are drawn. This can be expressed as follows:

H = fi (Q, N) (5)
Here, H: head, Q: flow rate, N: number of pump revolutions, f: function expressing performance curve, i: constants of 1, 2, 3 correspond to the number of pumps operated.

  Here, it is necessary to determine the number of pumps to be operated, which is performed based on the predicted flow rate Qf. For example, the discharge pressure control means 106 determines the number of operating pumps under the constants Q1 and Q2 (Q1 <Q2) as follows.

1 unit operation when Qf <Q1, 2 units operation when Q1 ≦ Qf <Q2 (6)
When Q2 ≦ Qf, the following equation is obtained by solving the three-unit operation (5) with respect to the rotational speed.

N = gi (H, Q) (7)
Using this equation, the target rotational speed N0 is calculated by the following equation.

N0 = gi (P0, Qf) (8)
Here, P0: target discharge pressure, Qf: predicted demand, N0: pump target rotational speed Next, the pump rotational speed control means 107 controls the signal to the pump so that the measured rotational speed matches the target rotational speed. Next, based on FIG. 4, the process flow of the above solid pressure control apparatus is demonstrated. This process is executed, for example, at a cycle of 5 minutes. The control cycle can be set to any length such as lengthening or shortening by setting in the water distribution pressure control device. If the control cycle is lengthened, the control performance is degraded, but if it is shortened, the calculation load increases. Therefore, it is desirable to set a cycle that can balance the trade-off between the two. The process shown in FIG. 4 is executed by the water distribution pressure control apparatus 100.

  First, in step 401, the demand prediction means 103 calculates a predicted demand from demand pattern data based on the equation (2). In step 402, the pressure loss calculation means 104 calculates the pressure loss based on the equation (3). In step 403, the target discharge pressure calculation means 105 calculates the target discharge pressure based on the equation (4). In step 404, the rotational speed control means 107 determines the number of operating pumps using the determination formula described in the formula (6). In step 405, the rotational speed control means 107 calculates the target rotational speed of the pump based on the equation (8). Finally, in step 406, the rotation speed control means 107 controls a signal to the pump so that the measured rotation speed matches the target rotation speed. The process ends here.

  As described above, according to the first embodiment, the terminal pressure can be maintained at a target value or higher with a high probability as shown in FIG. Moreover, the energy consumption of the pump can be minimized by setting the target discharge pressure to the lower limit.

  Next, a second embodiment will be described with reference to FIGS. In this embodiment, there is provided a pressure control method capable of maintaining the terminal pressure at a target value even when sudden demand such as digestive plug flow occurs.

  FIG. 8 shows an overall configuration diagram of the water distribution pressure control system. The difference from the first embodiment is that the water distribution pipe network 1 is provided with digestive plugs 801 and 802, the sensors 803 and 804 for measuring the flow rate of the digestive plugs are provided, and a temporary provision in the water distribution pressure control device. A processing determination means 805 is provided. Other processes are the same as those in the first embodiment.

  The temporary process determination means 805 determines whether or not a temporary control process is necessary when the digestive plug flow rate is generated, and activates the water distribution pressure control process shown in FIG. 4 when the temporary control process is necessary. . The process by the temporary process determination means 805 is executed at a considerably shorter period, for example, 100 ms than the control period (5 minutes) of the water distribution pressure control process shown in FIG.

  Assuming that the flow rate q1 (t) of the digestive plug 1 and the flow rate q2 (t) of the digestive plug 2 at time t, the temporary processing determination unit 805 determines whether or not the following conditions are satisfied.

| Q1 (t) −q1 (t−0.1) |> a certain threshold value,
Or (9)
| Q2 (t) −q2 (t−0.1) |> a certain threshold value These conditions are used to determine the occurrence of the digestive plug flow and the moment when it stops.
As long as these determination conditions are satisfied, the water distribution pressure control apparatus 100 executes the water distribution pressure control process of FIG. 4 at a cycle of 100 ms. As a result, it is possible to prevent a rapid decrease or increase in the terminal pressure by adapting when sudden demand such as a fire occurs or when the demand disappears.

FIG. 10 shows a flow of this temporary processing. The process shown in FIG. 10 is executed with a startup period of 100 ms.
In step 1001, the temporary processing determination means 805 calculates the changes Δq1 and Δq2 in the digestive plug flow based on the following equations.

Δq1 = | q1 (t) −q1 (t−0.1) | (10)
Δq2 = | q2 (t) −q2 (t−0.1) | (11)
In step 1002, the temporary processing determination unit 805 determines whether or not the amount of change is equal to or greater than a predetermined threshold value. If it is less than the predetermined threshold, the process ends. In step 1003, the temporary process determination means 805 starts the water distribution pressure control process shown in FIG. However, the demand forecast 401 is not executed. Instead of the demand pattern data, the demand prediction means 103 acquires the latest flow rate measurement value from the flow rate sensor 4 and sets this value as Qf. Thereafter, each means calculates various state quantities based on this Qf.

FIG. 9 shows the effect of the second embodiment. In the prior art, the terminal pressure is greatly reduced immediately after the generation of the digestive plug flow, but this can be prevented by using the method of the present invention.
In addition, although a present Example has shown the case where there are two digestive plugs, the case where there are three or more digestive plugs can be handled by the same process.

  In the present embodiment, the fire hydrant is used as an example of the sudden demand for water. However, the fire hydrant may not be used as long as it generates a demand based on the demand pattern.

  Next, Embodiment 3 will be described with reference to FIGS. In this embodiment, the end pressure is controlled with higher accuracy by using a more precise model considering the digestive plug flow rate as the pipe resistance model.

  FIG. 11 shows an overall configuration diagram of the water distribution pressure control system. The difference from FIG. 8 is that a digestive plug flow rate measurement value is added to the input of the database. In consideration of these digestive plug flow rates, the pipeline resistance model identifying means 1102 in this embodiment constructs a pipeline resistance model. The database 1101 stores the flow rate for each hydrant in addition to the original flow rate and pressure.

  The following model is used as a pipe resistance model.

P = Pe + h + m1 · Q ² + m2 · Q · q + m3 · q ² (12)
Where P: discharge pressure (m)
Pe: Terminal pressure (m)
h: Elevation of discharge pressure measurement point (m)-Elevation of end pressure measurement point (m)
k: Constant
Q: Distribution flow rate excluding digestive plug flow rate (original demand) (m ³ / s)
q: digestive plug flow (m ³ / s)
The model is refined by providing the hydrant flow variable q separately from the original demand Q.

  The pipe resistance model identification means 1102 estimates the coefficients m1, m2, m3 by the least square method using the time series of pressure and flow rate stored in the database 1101, and standard errors σm1, σm2, σm3 is calculated. q uses the flow rate for each hydrant, and calculates the coefficient value and standard miscalculation for each hydrant as shown in FIG. Using these calculated values, the pressure loss calculating means 1103 calculates the pressure loss H by the following equation.

H = h + (m10 + 2 · σm1) Q ² + (m20 + 2 · σm2) Q · q
+ (M20 + 2 · σm2) q ²
(13)
FIG. 13 shows a flow of control processing corresponding to a sudden change in demand, such as a fire hydrant flow rate. The processing in steps 1001 and 1002 is the same as the processing in steps 1001 and 1002 in FIG. 10 and is executed by the temporary processing determination unit 805.

  In step 1301, the table of FIG. 12 is searched to obtain the coefficient value and standard error corresponding to the fire hydrant in which the flow rate is generated. Here, the table shown in FIG. 12 is a coefficient calculated by the least square method by the pipe resistance model path means 1102, and the water distribution pressure control device 100 acquires the latest coefficient value from the pipe resistance model path means 1102 and stores it in the memory. Stored in

  In step 1302, the pressure loss calculation means 1103 calculates the pressure loss using the equation (12). The subsequent processing in steps 403 to 406 is equivalent to the processing of the same step number in FIG. However, as in the second embodiment, the demand prediction unit 103 sets the latest flow rate measurement value acquired from the flow rate sensor 4 as Qf instead of the demand pattern data. In addition, this process is performed in parallel with the water distribution pressure control (FIG. 4) performed every 5 minutes normally performed.

  As described above, according to the third embodiment, the terminal pressure can be precisely controlled with respect to the sudden demand such as the flow of the fire hydrant.

  In addition, when a plurality of digestive plugs are activated at the same time, a pipeline resistance model can be constructed by the same method to cope with it.

  In addition, this embodiment can be used for precise control of the terminal pressure when there is water accommodation by replacing the flow rate of the digestive plug, which is a disturbance, with the accommodation flow rate of the water distribution section.

100 water distribution pressure control device 101DB
102 Pipe resistance model identification means 103 Demand prediction means 104 Pressure loss calculation means 105 Target discharge pressure calculation means 106 Discharge pressure control means 107 Rotational speed control means

Claims (6)

A distribution pressure control system for controlling the distribution pressure from the pump to the distribution pipe network when distributing water from the distribution area to the distribution pipe network via the pump,
Installed between the discharge pressure measured by the discharge pressure measuring instrument installed between the water pipe network and the pump, and the water pipe of the water distribution destination receiving the water supply from the water pipe network and the water pipe network Based on the terminal pressure measured by the terminal pressure measuring device and the flow rate measured by the flow measuring device installed between the water distribution pipe network and the pump, the parameters of the pipe resistance model of the water distribution network A pipe resistance model generation unit that calculates a standard error representing a variation in parameters and generates a pipe resistance model;
Based on the pipe resistance model, the standard error, and a water distribution flow rate pattern that the water distribution pressure control system has in advance, a pressure loss of water pressure that occurs in the water distribution pipe network when the parameter changes based on the standard error A pressure loss calculation unit for calculating a range of the amount;
A target discharge pressure calculation unit that receives a target value of the terminal pressure and calculates a target discharge pressure based on the range of the pressure loss amount and the target value of the terminal pressure;
A water distribution pressure control system comprising: a rotation speed control unit that controls the rotation speed of the pump so as to achieve the target discharge pressure. The water distribution pressure control system according to claim 1,
The pipe resistance model generation unit generates a pipe resistance model reflecting a standard error generated by a least square method based on the discharge pressure, the terminal pressure, and the flow rate. system. The water distribution pressure control system according to claim 1,
The pressure loss calculation unit calculates a pressure loss amount at predetermined intervals, the target discharge pressure calculation unit calculates a target discharge pressure, and the rotation speed control unit controls the rotation speed of the pump. Distribution pressure control system. The water distribution pressure control system according to claim 3,
Further, based on the flow rate measured by the flow rate measuring instrument installed at a predetermined position of the water distribution pipe network, it is determined whether or not the amount of change in the unit time of the flow rate is greater than or equal to a predetermined value. When the amount of change in unit time is greater than or equal to a predetermined value, the pressure loss calculation unit is installed between the pipe resistance model, the water distribution network, and the pump regardless of the period. A temporary processing unit that calculates a pressure loss amount based on a flow rate measured by a flow rate measuring device, causes the target discharge pressure calculation unit to calculate a target discharge pressure, and causes the rotation speed control unit to control the rotation speed of the pump; Water pressure control system characterized by. The water distribution pressure control system according to claim 3,
Furthermore, based on the flow rate measured by the flow rate measuring device installed at a predetermined position of the water distribution pipe network, when the amount of change in unit time of the flow rate is greater than or equal to a predetermined value,
In the pipe resistance model generation unit, the flow rate measured by the flow rate measuring instrument installed at the predetermined position, the flow rate measured by the flow rate measuring instrument installed between the water distribution pipe network and the pump, and A pipeline resistance model is generated based on the discharge pressure and the terminal pressure,
Regardless of the period, the pressure loss calculation unit is made to calculate the pressure loss amount based on the pipe flow resistance model and the flow rate measured by the flow meter installed between the distribution pipe network and the pump,
A water distribution pressure control system comprising: a temporary processing unit that causes the target discharge pressure calculation unit to calculate a target discharge pressure and causes the rotation speed control unit to control the rotation speed of the pump. A water distribution pressure control system according to claim 4,
The water distribution pressure control system according to claim 1, wherein the predetermined position of the water distribution pipe network where the flow measuring device is installed is a water distribution port to a fire hydrant installed in the water distribution pipe network.

Images