Water operation planning system and water operation planning method

Abstract

To enable a detailed guidance by determining an appropriate water amount.SOLUTION: A water operation plan system 1 executed by a computer in a water supply facility includes: a first treatment part 41 for determining whether or not a water supply facility is in a non-steady state different from a normal operation state on the basis of at least one or more of a mount of water distributed to consumers, a raw water turbidity, disaster information of the water supply facility, and natural disaster forecasting; a second treatment part 42 for setting an operation target and an operation restriction of the water supply facility corresponding to the non-steady state in the case that the first treatment part determines that the water supply facility is in the non-steady state; a third treatment part 32 for deriving an operation plan problem corresponding to the non-steady state on the basis of the operation target and the operation restriction that are set by the second treatment part, and a demand forecasting amount calculated on the basis of water distribution amount results; and a fourth treatment 33 for solving the derived operation plan problem to determine a water amount of the water supply facility.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a water management planning system and a water management planning method, and in particular to a water management plan for unsteady situations. [Background technology]

[0002] Systems have been put into practical use to support the operation of water facilities in the water intake and transmission systems when non-stationary events occur, such as natural disasters, such as typhoons, heavy rains, and earthquakes, accidents caused by deterioration, and power outages. For example, Patent Document 1 discloses a water facility maintenance support device that records the operating conditions and worker activity levels in past cases in a DB (database), searches the DB when in a non-stationary state (natural disaster, power outage, breakdown, etc.), runs a simulation under the conditions (including actions) of the searched case, and if the target operating conditions, including water quality, are met, presents the conditions to the user (worker). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2017-91032 A Summary of the Invention [Problem to be solved by the invention]

[0004] The technology described in Patent Document 1 has a problem that it cannot present a similar case to the user unless there is a similar case in the past. Even if there is a similar case in the past, it may not be possible to present it to the user because it does not necessarily provide a satisfactory solution for the current state. In addition, when limited to water operation, there is a problem that it cannot necessarily provide precise guidance for unsteady states because it does not formulate and solve a planning problem to present future operation.

[0005] An object of the present invention is to determine an appropriate amount of water for an unsteady event, thereby enabling precise guidance. [Means for solving the problem]

[0006] A preferred example of the water operation planning system according to the present invention is a water operation planning system for a water facility, which is executed by a computer having a processor that executes a program and a memory unit that stores data, a first processing unit that determines whether the water facility is in an unsteady state different from a normal operating state based on at least one or more of the amount of water distributed to consumers, raw water turbidity, damage information on the water facility, and a natural disaster forecast; a second processing unit that, when the first processing unit determines that the water facility is in an unsteady state, sets an operational target and an operational constraint for the water facility corresponding to the unsteady state; A third processing unit that derives an operation planning problem corresponding to the unsteady state based on the operation target and the operation constraint set by the second processing unit, and a demand forecast calculated based on a water distribution amount record; and The derived operation planning problem is solved to determine the amount of water for the water facility. do A fourth processing unit; A water operation planning system having the above features. The present invention can also be understood as a water management planning method in the water management planning system. Effect of the Invention

[0007] According to the present invention, by formulating and solving a water operation plan that corresponds to unsteady events, it is possible to determine the appropriate water volume and provide precise guidance. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram showing the overall configuration of a water supply system. [Diagram 2] FIG. 2 is a diagram showing the water system of the water supply facility 3. [Diagram 3] FIG. 2 is a diagram illustrating a functional configuration of an operation plan planning program. [Figure 4] FIG. 2 is a diagram showing a functional configuration of a constraint condition / objective function determination processing unit 31. [Diagram 5]FIG. 13 is a flowchart showing a water volume change determination process (1). [Figure 6] FIG. 13 is a flowchart showing a water volume change determination process (2). [Figure 7] FIG. 13 is a flowchart showing a water volume change determination process (3). [Figure 8] FIG. 13 is a flowchart showing a water volume change determination process (4). [Figure 9] FIG. 13 is a diagram illustrating an example of a water volume change determination table. [Figure 10] FIG. 13 is a diagram illustrating an example of a constraint condition conversion table. [Figure 11] FIG. 13 is a diagram illustrating an example of a management table of objective functions and constraint conditions in a normal state; [Figure 12] FIG. 11 is a flowchart of an operation mode determination process. [Figure 13] FIG. 13 is a diagram showing patterns of fluctuations in water distribution volume. [Figure 14] FIG. 1 is a diagram illustrating an example of a short-term demand forecast. [Figure 15] FIG. 13 is a flowchart illustrating an operation plan formulation process. [Figure 16] FIG. 13 is a diagram showing an example of a pattern of water volume and water level according to an operation plan. [Figure 17] FIG. 11 is a diagram illustrating a functional configuration of a steady-state operation plan formulation program in the second embodiment. [Figure 18] FIG. 11 is a diagram illustrating a functional configuration of an operation plan planning program in a second embodiment. [Figure 19] FIG. 13 is a diagram illustrating an example of an operation planning management table. BEST MODE FOR CARRYING OUT THEINVENTION

[0009] Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings. Example 1

[0010] FIG. 1 shows an overall view of a water supply system. The water supply system is a system in which a water operation planning system 1, a monitoring and control device 2, a waterworks facility 3, and an external server 4 are connected via a network 5, and are capable of mutual data communication.

[0011] As shown in FIG. 2, the water supply facility 3 takes in water from a water source A at a water intake plant 220 and sends it to a water purification plant 221 for purification. The purified water is temporarily stored in a purified water reservoir 222, and is then sent to a distribution reservoir 225 through a pump 223 and a pipeline 224. Purified water is also sent from a water supplier B to the purified water reservoir 222, the distribution reservoir 225, etc. The water from the distribution reservoir 225 is supplied to consumers through a water distribution network 226. The water intake plant 220 and the water purification plant 221 are equipped with sensors for measuring raw water turbidity, flow meters are installed in each pipeline 224, and water level gauges and flow meters for measuring the amount of water distributed are installed in the purified water reservoir 222 and the distribution reservoir 225, and the measurement data is sent to the monitoring and control device 2 at a predetermined cycle. The monitoring and control device 2 controls and manages the flow rate of each pipeline, pump operation, and the water level of the distribution reservoir based on the measurement data. In Figure 2, an example of two systems is shown, one above the dotted line (system 1) and one below it (system 2). Here, there is an interchange path from system 1 to system 2, and water can only be sent from system 1 to system 2.

[0012] The water management planning system 1 of this embodiment acquires information such as water level and water distribution amount required for future planning such as water volume from the monitoring and control device 2. The external server 4 is a server such as the Japan Meteorological Agency. The water management planning system 1 acquires meteorological information (weather forecast, temperature information) required for water demand prediction from the external server 4 and uses it for water management planning.

[0013] The water management planning system 1 is realized by a computer that executes a program and processes data for water management planning. The computer has a storage unit 13 such as a semiconductor memory or a hard disk that stores the program and data, a communication interface (I / F) 14 that exchanges data with the monitoring control device 2 and the external server 4 via the network 5, a processing unit 15 that is a processor that executes the program, an input device 16 that is a man-machine interface such as a keyboard or a mouse, and a display device 17 that displays data. In this embodiment, the processing unit 15 executes the operation plan planning program 11 stored in the storage unit 13, stores the processing results in a database (DB) 12, and displays them on the display device 17 as necessary. When the program is executed, data input from the input device 16 and data stored in the DB 12 in the storage unit 13 are referenced and data is processed. The DB 12 stores various data, but in this embodiment, in particular, it stores a water volume change judgment table 90 (FIG. 9), a constraint condition conversion table 100 (FIG. 10), a normal time management table 110 (FIG. 11), an operation plan planning management table 190 (FIG. 19), and the like.

[0014] <Operational Planning Program 11> FIG. 3 shows the functional configuration of the operational plan formulation program 11. The operational planning program 11 is configured to include a processing unit 31 that performs constraint condition / objective function setting processing, a processing unit 32 that performs calculation processing of the operational planning period, the operational planning period, and the demand forecast amount, and a processing unit 33 that performs the operational planning processing.

[0015] The processing unit 31 performs processing to derive operational constraints and operational targets during non-steady state. The processing unit 32 performs processing to derive operational planning cycles, operational planning periods, and demand forecast amounts during non-steady state. The processing unit 33 formulates an operational planning problem for dealing with non-steady events based on the derived operational constraints (constraint equations for the operational planning problem), operational targets (objective functions for the operational planning problem), operational planning cycles, operational planning periods, and demand forecast amounts, and solves the problem to derive future water volume plan values ​​during non-steady state. The derivation results are displayed on the display device 17 and presented to the user (water operations manager).

[0016] Each of the processing units 31-33 will be described in detail below.

[0017] <Constraint and objective function setting processing unit 31> The processing of the constraint condition / objective function setting processor 31 will be described with reference to FIGS. Fig. 4 shows the functional configuration of the constraint condition / objective function setting processor 31. The constraint condition / objective function setting processor 31 is composed of a water volume change determination processor 41 and a constraint condition / objective function conversion processor 42. The water volume change determination processor 41 judges whether the water facility is in a non-steady state and determines the water volume change that should be taken in the non-steady state. The constraint condition / objective function conversion processor 42 determines the operation constraints (constraint conditions of the operation planning problem) and operation targets (objective function of the operation planning problem) that correspond to the judgment of the water volume change determination processor 41. These processors will be described in detail below.

[0018] <Water volume change determination processing unit 41> First, the water volume change judgment processing unit 41 will be specifically described. The water volume change judgment is based on one or a combination of the following: (1) water volume increase (maintaining the stored volume), (2) water volume increase (increasing the stored volume), (3) water volume decrease (including stopping), and (4) system pipe water exchange. The judgment processes for (1) to (4) are each executed based on the processing flows shown in Figs. 5 to 8. They will be described below.

[0019] (1) Increasing the water volume (maintaining the storage volume) is determined when the water distribution network is damaged due to an accident caused by deterioration or a natural disaster, causing a sudden increase in the water distribution volume. The process flow for making this determination is shown in Figure 5. This process is executed at a predetermined time interval, for example, every 10 minutes.

[0020] In FIG. 5, in step 501 (calculation of normal water distribution rate of change), the rate of change in the water distribution rate is calculated using the latest water distribution rate and the water distribution rate 10 minutes ago (or 1 hour ago, etc.). In step 502 (calculation of index A), index A is calculated by subtracting the water distribution rate of change in normal time (non-unsteady state) (normal rate of change) from the rate of change in the water distribution rate. The normal rate of change is a rate of change calculated every minute based on the time series of water distribution rate in normal time, and is stored in DB12 in advance. In step 503, it is determined whether index A is equal to or greater than a predetermined value (e.g., 0.05, 0.1, etc.). If the result of the determination is equal to or greater than the predetermined value, the process proceeds to step 504, and if the result is less than the predetermined value, the process ends. In step 504, it is determined that the water volume is to be increased (reserved volume is maintained), and the result of the determination is set to 1 in the determination flag in the row of (1) water volume increase (reserved volume is maintained) in the water volume change determination table 90 held in DB12. The judgment flag is normally set to 0, so its initial value is 0. If there are multiple systems, judgment is performed for each system. According to the example of Fig. 2, the judgment results based on the water distribution amount for each of the two systems are recorded in the water amount change judgment table 90.

[0021] (2) An increase in water volume (increase in storage volume) is determined when damage to water facilities is expected due to a typhoon, heavy rain, etc., or when filtration is stopped for maintenance and inspection. Damage to water facilities may occur when facilities become unusable due to flooding, or when water purification is stopped due to high turbidity. The process flow for making this determination is shown in Figure 6. This process is also performed every 10 minutes.

[0022] In FIG. 6, in step 601, it is determined whether or not a user has instructed to increase the amount of stored water, whether or not the alert level set by the Meteorological Agency is high (whether or not flooding of waterworks facilities is expected due to heavy rain), and whether or not the predicted turbidity value is equal to or greater than a predetermined value. If any one of these conditions is met, the process proceeds to step 602, and if not, the process ends. Here, the user's instruction to increase the amount of stored water is acquired from the input device 16, and the alert level set by the Meteorological Agency is acquired from the external server 4 via the communication I / F 14. In addition, the predicted turbidity value being equal to or greater than a predetermined value means that if the situation continues, the turbidity will exceed the specified value in a few hours and water purification will no longer be possible. In step 602, a water volume increase (increase in stored water volume) is determined, and the judgment flag for (2) water volume increase (increase in stored water volume) in the water volume change judgment table 90 is set to 1.

[0023] (3) Water volume reduction is determined when water quality is polluted (high turbidity due to heavy rain, etc.), or when the water purification plant is forced to shut down due to wind and flood damage or earthquakes (this includes shutting down). The process flow for making this decision is shown in Figure 7. This process is also executed every 10 minutes.

[0024] In FIG. 7, in step 701, it is determined whether the measured or predicted raw water turbidity value is equal to or greater than a predetermined value, whether the turbidity of the sedimentation treated water is equal to or greater than a predetermined value, and whether the user has made a decision to stop. The user's decision to stop is acquired from the input device 16. If any one of these decisions is met, the process proceeds to step 702, and if not, the process ends. In step 702, a water volume reduction decision is made, and the (3) water volume reduction decision flag in the water volume change decision table 90 is set to 1. In step 703, the user is inquired about the water volume reduction amount and reduction period, and the values ​​acquired from the input device 16 are stored in DB12, and the process ends.

[0025] (4) System pipe water interchange is determined when the amount of purified water at some water purification plants is reduced (stopped) due to high turbidity or other reasons (including cases where facility operations are stopped due to damage) or when the amount of water distributed in some systems is increased due to pipeline damage. In other words, when there are multiple systems, the decision is made based on whether (1) the amount of water should be increased (storage volume should be maintained), (3) the amount of water should be reduced, or a combination of both. The process flow for making this decision is shown in Figure 8. This process is also performed in 10-minute cycles.

[0026] In Figure 8, in step 801, it is determined whether there is a water volume increase judgment (including maintaining the storage volume) or a water volume decrease judgment. If any one of these judgments is met, proceed to step 802, and if not, end the processing. In step 802, it is determined that system pipe water is to be accommodated, and the (4) system pipe water accommodation flag in the water volume change judgment table 90 is set to 1.

[0027] The four judgments made by the water volume change judgment processor 41 have been described above. These four judgments may be made individually, or multiple judgments may be made simultaneously. For example, (1) and (3), (1) and (4), (3) and (4), (1), (3) and (4), etc. may be made simultaneously. Here, since an increase and a decrease cannot be made simultaneously, if (1) and (3) are judged to be simultaneously in the same system, (3) is given priority for processing. After the flag becomes 1, the corresponding processing (processing (1) to (4)) is stopped until the user cancels the judgment. The flag is set to 0 by the user's cancellation instruction.

[0028] <Constraint condition / objective function conversion processing unit 42> Next, the constraint condition / objective function conversion processing unit 42 will be specifically described. This process is a process for determining the operation constraints (constraint conditions of the operation planning problem) and the operation targets (objective functions of the operation planning problem) corresponding to the above four judgments. This process is executed using the constraint condition conversion table 100 (FIG. 10).

[0029] Before describing the processing of the processing unit 42, the objective function and constraint conditions in the operation planning problem in normal times (steady state) will be described with reference to FIG. 11. FIG. 11 shows a normal time management table. The normal time management table 110 manages the objective function, constraint conditions, planning cycle, planning time interval, planning period, and demand forecast amount used in formulating the operation planning problem in normal times. A typical objective function in normal times is power consumption (1 day) or power cost (1 day) + flow rate fluctuation suppression. The main constraint conditions are upper and lower limit constraints (inequalities) of each pipeline flow rate, upper and lower limit constraints (inequalities) of each reservoir water level, upper and lower limit constraints (inequalities) of each filtration rate, future distribution water amount = demand forecast amount (equation), the water level after 24 hours returns to the current water level (or the water level at 6 am falls near a specified value), and equality constraints (flow rate conservation law) on each pipeline flow rate. Demand forecasting is performed, for example, in one-hour increments for 24 hours from the current time to the same time the next day. For this, data such as actual water distribution volume, weather forecasts, and temperature forecasts are used, and a predicted time series is obtained by statistically processing this data. The future flow rate of each pipeline is obtained by solving an operation planning problem using these objective functions and constraint conditions. In other words, using mathematical programming, the future flow rate of each pipeline that minimizes the objective function under these constraint condition equations is obtained. In the operation planning problem, time is handled in hourly increments, and the period is up to one day in advance. Under normal circumstances, this operation planning problem is solved to obtain the future water volume, which is then displayed to the user. In contrast, during unsteady conditions, constraints and objective functions corresponding to unsteady events are set.

[0030] Returning to the explanation of the constraint condition / objective function conversion processor 42, the constraint conditions corresponding to each of the water volume judgments (1) to (4) are obtained using a constraint condition conversion table 100.

[0031] (1) The constraint condition conversion table 100 is used as the constraint condition corresponding to the increase in water volume (maintaining the storage volume). In this case, the constraints and objective functions of the normal state management table 110 are modified. That is, the constraint conditions are the normal state constraints modified as follows: The upper and lower limit constraints on the reservoir water level will remain unchanged. Set the upper limit constraint for filtration flow rate to the operational upper limit. - The water level recovery constraint after 24 hours (or the water level recovery constraint at 6am) will be disabled.

[0032] With this constraint, if the water distribution volume is greater than the filtration volume, the reservoir water level will gradually drop, and the operation planning problem solving method may not be able to find a solution. In this case, it is necessary to relax the water level lower limit constraint (for example, adjust the lower limit constraint by decreasing it) until a solution is found. To avoid this problem, it is possible to abolish the water level lower limit inequality constraint and instead set the following objective function: · Objective function = power cost (power consumption) + flow rate fluctuation suppression term +Σ constant | water level (t) - target water level | (sum of deviations for the next 24 hours. In operation mode 3 described later, it becomes the sum of deviations for a shorter period of time, such as 4 to 6 hours in the future.) In the above objective function formula, the first term (electricity cost) and the second term (flow rate fluctuation suppression term) can be eliminated, and the objective function can be configured with only the third term (sum of absolute values ​​of water level deviations). This enables operations that place more importance on maintaining the water level (storage volume).

[0033] By making the sum of the deviations between the future water level and the target water level over time part of the objective function, it is possible to realize a future water level as close as possible to the target water level without ending up with no solution.

[0034] (2) As the constraint conditions related to the increase in water volume (increase in storage volume), the constraints and objective functions in the normal state management table 110 are modified using the constraint condition conversion table 100. That is, the constraint conditions used are the normal state constraints modified as follows: - Raise the lower limit of the water level from a specified time in the future (for example, from 12 hours later) to near the upper operational limit. Set the upper limit constraint for filtration flow rate to the operational upper limit. Disable the water level recovery constraint after 24 hours (or the water level recovery constraint at 6am).

[0035] In this case, if the time to raise the lower limit of the water level constraint is too early, the operation planning problem solving method may not be able to find a solution. In this case, it is necessary to adjust the timing of the raising. To avoid this, the lower limit of the water level inequality constraint can be abolished and instead the following objective function can be set. · Objective function = Power cost (power consumption) + Σ constant · | Water level (t) - Target water level (upper water level) | (The sum of deviations for the future 24 hours. In operation mode 3 described later, it becomes the sum of deviations for a shorter period of time, such as 4 to 6 hours in the future.) In the above objective function formula, the first term (electricity cost) may be eliminated, and the objective function may be configured with only the second term (sum of absolute values ​​of water level deviation). This allows operations to place more emphasis on increasing storage volume.

[0036] By making the sum of the deviations between the future water level and the target water level over time part of the objective function, it is possible to realize a future water level as close as possible to the target water level without ending up with no solution.

[0037] (3) As the constraint conditions related to the reduction in water volume (including suspension), the constraints and objective functions in the normal state management table 110 are modified using the constraint condition conversion table 100. That is, the constraint conditions used are the normal state constraints modified as follows: Lower the upper and lower limits of the filtration flow rate (or water intake volume). (Set according to turbidity level) The amount of decrease and the period for which the decrease is performed are set by the user (the value acquired in step 703 in FIG. 7).

[0038] (4) As constraint conditions related to inter-system water interchange, the constraints and objective functions in the normal-time management table 110 are modified using the constraint condition conversion table 100. That is, the constraint conditions are the normal-time constraints modified as follows:

[0039] The decrease in filtration flow rate and the increase in distribution volume of system 2 are handled by increasing the filtration volume of system 1. In other words, the constraints are amended as follows: Set the upper limit of the filtration flow rate of system 1 to the upper limit of operation. Set the upper limit of the flow rate of the interchange path from System 1 to System 2 to the upper operational limit. This is addressed by increasing the flow rate from system 1 to system 2.

[0040] A decrease in the filtration flow rate and an increase in the distribution volume of system 1 are addressed by increasing the filtration volume of system 2. In other words, the constraints are amended as follows: Set the upper limit of the filtration flow rate of system 2 to the upper limit of operation. Set the lower limit of the flow rate of the interchange path from System 1 to System 2 to the lower limit of operation (e.g., 0). This is done by reducing the flow rate from system 1 to system 2. Since system 1 does not have the capacity to transfer water to system 2, it responds by reducing the amount of water transferred.

[0041] If the water supplier responds to a decrease in filtration volume or an increase in water distribution volume of either or both of System 1 and System 2 by increasing the supply volume, the constraints shall be amended as follows: Set the upper limit of the flow rate of the interchange path from System 1 to System 2 to the upper operational limit. Set the lower limit of the flow rate of the interchange path from System 1 to System 2 to the lower limit of operation (e.g., 0). Set the upper limit constraint on the water supplier's supply volume to the maximum value that can be supplied.

[0042] Through the above processing, a decision is made to change the water volume in response to an event (unsteady event) based on the measured values ​​of water distribution volume and turbidity, natural disaster information from the Japan Meteorological Agency, etc. (the above four decisions (1) to (4)), and the corresponding operation constraints (constraint conditions for the operation planning problem) and operation targets (objective function for the operation planning problem) are found. The processing of the constraint condition / objective function setting processor 31 in Fig. 4 has been described in detail above.

[0043] The constraint condition equation and objective function corresponding to the current state obtained by the above processing are written in the objective function and constraint condition equation columns of the operation planning management table 190 (FIG. 19). The operation planning management table 190 is a management table used in the operation planning process 33 (processing by the operation planning processing unit 33), and manages data related to the current (latest) objective function, constraint condition, operation mode, planning cycle, planning period, and demand forecast amount. In the initial state, the objective function and constraint condition equation for steady state are written in this management table 190. As will be described later, the operation planning process 33 is executed based on the data of each item registered in the operation planning management table 190.

[0044] <Operation plan creation cycle, operation plan creation period, demand forecast amount calculation processing unit 32> Next, the operation planning cycle, the operation planning period, and the processing of the demand forecast amount calculation processing unit 32 will be described with reference to Figs. 12 to 14 and 19.

[0045] In this embodiment, it is determined which of the three operation modes (operation mode 1, operation mode 2, operation mode 3) the current state belongs to based on the measured water distribution amount, and further, the operation plan creation cycle, operation plan creation period, and demand forecast amount corresponding to the operation mode are calculated. FIG. 12 shows a flowchart of the operation mode determination process (processing by the operation mode determination process program). In this embodiment, the water distribution pattern is classified into three cases as shown in FIG. 13. Operation modes 1, 2, and 3 correspond to cases 1, 2, and 3, respectively. Case 1 is a pattern that forms the water distribution amount during normal times. Case 2 is a pattern in which a burst water leakage occurs due to a single pipeline accident, causing a temporary sudden increase in the water distribution amount. Case 3 is a pattern in which multiple burst water leakage occurs in the pipeline network due to an earthquake, causing a sudden increase in the water distribution amount, and forming a water distribution pattern different from the original.

[0046] The operation mode determination process in FIG. 12 is executed in the same cycle as the water volume change determination process (1)-(4) (FIGS. 5 to 8), that is, in a 10-minute cycle. In step 1201, index A and index B are calculated. These indexes are defined for operation mode determination. Index A is calculated by calculating the water volume change rate calculated from the water volume at the latest time and the water volume 10 minutes ago (or a predetermined time before, such as 1 hour ago) and the water volume change rate under normal conditions, and taking the difference between them. The method for calculating the water volume change rate under normal conditions is as described above. A sudden increase in water volume can be determined using index A. Index A may be calculated as (CD) / D (or simply CD) from the demand volume forecast value D (corresponding to the water volume forecast value) corresponding to the same time as the water volume measurement value C at the latest time. Index B is calculated as the correlation coefficient of the time series of the water volume measurement value from the current time to 12 hours (or 24 hours) ago and the time series of the demand volume forecast value (corresponding to the water volume forecast value). Demand forecasts are made at 10-minute intervals, and the time series of predicted demand values ​​can be found by storing the predicted value closest to the current time (the predicted value 10 minutes ahead) from the 10-minute demand forecast along with the time, and then recalling it. Indicator B can be used to determine whether the current water distribution pattern (demand pattern) has deviated from the original pattern. The method described above is used for demand forecasting. This general method only obtains a time series of demand (=water distribution) in 1-hour (or 30-minute) increments. Using simple linear interpolation (interpolation, extrapolation) or spline interpolation, this is converted into a time series in 10-minute increments, and used to calculate Indicators A and B.

[0047] In step 1202, it is determined whether index A is equal to or greater than a predetermined value (a value of approximately 0.05 or 0.1 is used) or whether operation mode 2 or 3 has already been determined. The former determines whether there has been a sudden increase in the amount of water distributed. The latter, once operation mode 2 or 3 has been determined, operation mode 2 or 3 is maintained even if no sudden increase in water distribution is confirmed in the latest data. This is because in operation mode 2, the sudden increase in water distribution is temporary, and operation mode 3 may also decrease after a sudden increase in water distribution. If either of these conditions is met, proceed to step 1203, and if they are not met, proceed to step 1204.

[0048] In step 1204, the operation mode is determined to be 1, i.e., the normal operation mode, with the operation plan creation cycle (time interval for operation plan creation) being the normal 1 hour (or 30 minutes), the operation plan creation period being 24 hours, and the demand forecast amount being a value calculated using the general prediction method described above. Data relating to the operation plan creation cycle, operation plan creation period, and demand forecast amount determined as the determined operation mode are written into the items of current operation mode, operation plan creation cycle, operation plan creation period, and demand forecast amount in the operation plan creation management table 190. Note that in the example of Fig. 19, the determination result of operation mode 1 is written, but the contents change depending on the latest determination result.

[0049] In step 1203, it is determined whether index B is equal to or less than a predetermined value (for example, 0.8 or 0.9). If this condition is met, it is assumed that the current water distribution pattern is deviated from the normal pattern, and the process proceeds to step 1206. If not, the process proceeds to step 1205.

[0050] Step 1205 is a judgment of operation mode 2. In this step, the operation plan creation cycle (time interval for operation plan creation) corresponding to this judgment is set to 10 minutes, the operation plan creation period is set to 24 hours, the same as in normal times, and the demand forecast amount is set to the demand forecast amount by the normal demand forecasting method (demand forecast amount time series) + constant value α. A new forecast time series is calculated and utilized by adding a fixed value α to each component of the future demand forecast amount time series by the conventional method. α is calculated by calculating the difference between the water distribution amount at the latest time and the demand forecast amount at the same time. As α, the difference calculated in the past may be stored and the average value of the difference time series may be used. The operation plan creation cycle, operation plan creation period, and demand forecast amount determined as the judgment operation mode are written into the items of the current operation mode, operation plan creation cycle, operation plan creation period, and demand forecast amount in the operation plan creation management table 190.

[0051] Step 1206 is a judgment of operation mode 3. In this step, the operation plan creation cycle corresponding to this judgment is set to 10 minutes, the operation plan creation period is set to a value shorter than the normal planning period of 24 hours, for example, 4 to 6 hours, and the demand forecast amount is set to a predicted value by a model constructed by applying the actual water distribution amount (for the past 2 to 12 hours) to a predetermined mathematical model (for example, an ARIMA model, a Gaussian process regression model, etc.). The operation plan creation cycle, operation plan creation period, and demand forecast amount determined as the judgment operation mode are written in the items of current operation mode, operation plan creation cycle, operation plan creation period, and demand forecast amount in the operation plan creation management table 190. The operation plan creation cycle is shortened to 10 minutes in order to ensure the accuracy of planning for a sudden increase in the water distribution amount. The maximum amount of change in water distribution volume per hour under normal circumstances, L, and the maximum amount of change in water distribution volume per hour when there is a sudden increase in water distribution volume, M, are calculated as the planning cycle for operation mode 3, which is 1 hour x L / M. A value close to this calculated value can also be selected and used from values ​​in 10-minute increments (10 minutes, 20 minutes, 30 minutes, 40 minutes, ...).

[0052] Furthermore, when a Gaussian process regression model is used, it is possible to calculate the expected value (1401) at a predetermined time in the future as well as a confidence interval (1402, 1403) that indicates the range of prediction error, as shown in Fig. 14. It is also possible to select a period (e.g., 5 hours) in which the ratio of the prediction error to the expected value is within 10%, for example, as the operation plan formulation period. As described above, the operation plan formulation cycle, operation plan formulation period, and processing of the demand forecast amount calculation processing unit 32 have been described in detail.

[0053] <Operation planning processing unit 33> The processing of the operation plan creation processing unit 33 will be described with reference to Figs. 15 and 16. Fig. 15 is a flowchart of the operation plan creation processing. This processing is executed by an operation plan creation processing program. In this processing, the processing of operation mode 1 is executed in a 1-hour cycle, and the processing of operation mode 2 and operation mode 3 is executed in a 10-minute cycle. In the latest operation mode, data for each item can be acquired by referring to the operation plan creation management table 190. In step 1501, data on the operation constraint (constraint of the operation plan problem) corresponding to the current event (steady or non-steady), operation target (objective function for the operation plan problem), operation plan creation period, operation plan creation period, and demand forecast amount are extracted from the operation plan creation management table 190. In step 1502, the operation plan problem is solved using the data obtained in step 1501, and the future water volume of each pipeline and the water level of each water reservoir for the current event (steady or non-steady) are calculated and obtained. In step 1503, the calculation results of the water volume and water level are displayed on the display device 17. An example of a pattern showing some of the water volume and water level display results is shown in Figure 16. In this example, the filtration volume was increased to accommodate a sudden increase in water distribution volume, and the reservoir water level was maintained.

[0054] As described above, the operational objectives, operational constraints, operational planning cycle, operational planning period, and demand forecast amount corresponding to non-steady events are derived, and an operational planning problem corresponding to the non-steady events is formulated and solved to determine the appropriate water volume, enabling precise guidance. Example 2

[0055] A second embodiment will be described with reference to FIGS. 17 is a functional configuration diagram of a normal operation planning program. A normal operation planning program 1701 is composed of a demand forecast processing unit 1702 and an operation plan creation processing unit 1703. The processing contents of each processing unit are as described in the first embodiment. If a program having such functions has already been realized, this function can be used to create an operation plan for the non-normal state according to the first embodiment.

[0056] FIG. 18 shows an operation planning system incorporating this function and capable of handling both stationary and non-stationary events. In stationary mode, a planned water volume value is obtained using a stationary mode operation planning function 1701. In non-stationary mode, the processing of the processing unit 31 that performs the constraint condition / objective function setting process and the processing unit 32 that performs the operation plan generation cycle, operation plan generation period, and demand forecast amount calculation process according to the first embodiment are executed to obtain the constraint condition and objective function of the operation plan problem, the operation plan generation cycle, operation plan generation period, and demand forecast amount. The obtained data is input to the stationary mode operation planning processing unit 1701, and the processing result of the stationary mode operation planning unit 1701 is displayed on the display device 17, so that it can be presented to the user as a planned water volume value corresponding to the non-stationary event. This has the advantage of reducing the man-hours required for developing the system.

[0057] Although the preferred embodiment has been described above, the present invention is not limited to the above embodiment and can be implemented with various modifications and applications. For example, the tables such as the operation plan creation management table 190 disclosed in the above embodiment only need to link the managed data, and do not necessarily have to be configured as tables. Also, instead of calling them tables, they may simply be called, for example, operation plan creation management information, operation plan creation management data structure, etc. [Explanation of symbols]

[0058] 1: Water operation planning system 2: Monitoring and control device 3: Water supply facility 4: External server 5: Network 11: Operation planning program 12: Database 13: Memory unit 14: Communication I / F 15: Processing unit 16: Input device 17: Display device 31: Processing unit that performs constraint and objective function setting processing 32: A processing unit that performs operation planning cycle, operation planning period, and demand forecast calculation processing 33: Processing section for carrying out operation plan creation processing 41: Water volume change judgment processing unit 42: Constraint condition / objective function conversion processing section 90: Water volume change judgment table 100: Constraint conversion table 110: Normal management table 190: Operation planning management table

Claims

1. A water operation planning system for a water facility, which is executed by a computer having a processor that executes a program and a memory unit that stores data, a first processing unit that determines whether the water facility is in an unsteady state different from a normal operating state based on at least one or more of the amount of water distributed to consumers, raw water turbidity, damage information on the water facility, and a natural disaster forecast; a second processing unit that sets an operation target and an operation constraint of the water facility corresponding to the unsteady state when the first processing unit determines that the water facility is in the unsteady state; a third processing unit that derives an operation planning problem corresponding to the unsteady state based on the operation target and the operation constraint set by the second processing unit, and a demand forecast calculated based on a water distribution amount record; and A fourth processing unit that solves the derived operation planning problem and determines the water volume of the water facility; having The third processing unit calculates a water distribution change based on a water distribution change rate calculated from the water distribution amount and a correlation coefficient calculated from the water distribution amount measurement value and the water distribution amount prediction value, A first mode for forming a water distribution amount during normal times; - The second mode is when a single pipeline accident causes a burst leak, causing a temporary surge in water supply. - The third mode is when an earthquake causes multiple burst leaks in the pipeline network, causing a sudden increase in water distribution volume, resulting in a water distribution volume that differs from the original amount. The operation plan is classified into three operation modes, and the operation plan period, operation plan period, and demand forecast amount are determined for each operation mode. The fourth processing unit solves the operation planning problem by using the determined operation planning cycle, the operation planning period, and the demand forecast amount. Water operations planning system.

2. A water operation planning system for a water facility, which is executed by a computer having a processor that executes a program and a memory unit that stores data, a first processing unit that determines whether the water facility is in an unsteady state different from a normal operating state based on at least one or more of the amount of water distributed to consumers, raw water turbidity, damage information on the water facility, and a natural disaster forecast; a second processing unit that sets an operation target and an operation constraint of the water facility corresponding to the unsteady state when the first processing unit determines that the water facility is in the unsteady state; a third processing unit that derives an operation planning problem corresponding to the unsteady state based on the operation target and the operation constraint set by the second processing unit, and a demand forecast calculated based on a water distribution amount record; and A fourth processing unit that solves the derived operation planning problem and determines the water volume of the water facility; having The first processing unit determines that the water supply is in an unsteady state based on a change in water volume due to at least one or more of the following: a water supply volume to a consumer, raw water turbidity, damage information on water facilities, and a natural disaster forecast; The second processing unit derives a water volume increase (maintaining the storage volume), a water volume increase (increasing the storage volume), and a water volume decrease including a stop based on the water volume change judgment corresponding to the unsteady state by the first processing unit, and derives the operation constraints and the operation targets for each. Water operations planning system.

3. The second processing unit introduces the deviation between the water level and the water level target value as one term of an objective function of the operation planning problem as the operation target for determining whether to increase the water volume (maintain the storage volume) or increase the water volume (increase the storage volume), The water operation planning system according to claim 2.

4. The third processing unit, when the rate of change in the water volume satisfies a predetermined condition and a correlation coefficient calculated from the measured water distribution volume value and the predicted water distribution volume value becomes equal to or less than a predetermined value, sets the operation plan creation cycle and the operation plan creation period to values ​​smaller than normal values ​​when the operation plan creation cycle and the operation plan creation period do not correspond to non-stationary conditions, and uses a short-term prediction shorter than 24 hours as a demand forecast amount. The water operation planning system of claim 1.

5. The third processing unit performs short-term demand forecasting using a Gaussian process regression model, and sets an operation plan formulation period so that a forecast error of the model falls within a predetermined range. The water operation planning system according to claim 4.

6. The storage unit is The driving goal and the driving constraint set by the second processing unit; The operation plan creation period and the operation plan creation period derived by the third processing unit; A management table for managing the demand forecast amount is stored, The fourth processing unit solves the operation planning problem by referring to the operation planning management table. The water operation planning system of claim 1.

7. The storage unit is a normal time management table for managing each data item, such as an objective function, a constraint condition, a planning cycle, a time interval for planning, a planning period, and a demand forecast amount, which are used in formulating an operation planning problem during normal times; A constraint condition conversion table for managing the operation target and the operation constraint related to the water volume increase (maintaining the storage volume), the water volume increase (increasing the storage volume), and the water volume reduction including the stoppage, which are used in the non-steady state, obtained based on the data managed in the normal state management table; The water operation planning system of claim 2, further comprising:

8. A water operation planning method for a water facility, which is executed by a computer having a processor that executes a program and a memory unit that stores data, the computer comprising: a first processing step of determining whether or not the water facility is in an unsteady state different from a normal operating state based on at least one or more of the amount of water distributed to consumers, raw water turbidity, damage information on the water facility, and a natural disaster forecast; a second processing step of setting an operational target and an operational constraint of the water facility corresponding to the unsteady state when the first processing step determines that the water facility is in an unsteady state; a third processing step of deriving an operation planning problem corresponding to the unsteady state based on the operation target and the operation constraint set by the second processing step, and a demand forecast amount calculated based on a water distribution amount record; A fourth processing step of solving the derived operational planning problem to determine the water volume of the water facility; The third processing step is to calculate the change in the water distribution amount based on a water distribution amount change rate calculated from the water distribution amount and a correlation coefficient calculated from the measured water distribution amount value and the predicted water distribution amount value. A first mode for forming a water distribution amount during normal times; - The second mode is when a single pipeline accident causes a burst leak, causing a temporary surge in water supply. - The third mode is when an earthquake causes multiple burst leaks in the pipeline network, causing a sudden increase in water distribution volume, resulting in a water distribution volume that differs from the original amount. The operation plan is classified into three operation modes, and the operation plan period, operation plan period, and demand forecast amount are determined for each operation mode. The fourth processing step solves the operation planning problem by using the determined operation planning period, the operation planning period, and the demand forecast amount. Water operation planning methods.

9. A water operation planning method for a water facility, which is executed by a computer having a processor that executes a program and a memory unit that stores data, the computer comprising: a first processing step of determining whether or not the water facility is in an unsteady state different from a normal operating state based on at least one or more of the amount of water distributed to consumers, raw water turbidity, damage information on the water facility, and a natural disaster forecast; a second processing step of setting an operational target and an operational constraint of the water facility corresponding to the unsteady state when the first processing step determines that the water facility is in an unsteady state; a third processing step of deriving an operation planning problem corresponding to the unsteady state based on the operation target and the operation constraint set by the second processing step, and a demand forecast amount calculated based on a water distribution amount record; A fourth processing step of solving the derived operational planning problem to determine the water volume of the water facility; The first processing step determines that the water supply system is in an unsteady state based on a change in water volume due to at least one or more of the following: the amount of water distributed to consumers, the turbidity of raw water, information on damage to water facilities, and a natural disaster forecast; The second processing step derives a water volume increase (maintaining the storage volume), a water volume increase (increasing the storage volume), and a water volume decrease including a stop based on the water volume change judgment corresponding to the non-steady state by the first processing step, and derives the operation constraints and the operation targets for each. Water operation planning methods.

Images