Water consumption prediction system

Abstract

Provided is a water consumption prediction system in which a water meter includes: a flow rate measurement unit that is provided in a flow path through which water flows and measures the flow rate of the water flowing through the flow path; and a communication unit that communicates with a center server. The center server includes: a central communication unit that communicates with the communication unit; a first calculation unit that calculates an estimated daily water consumption for the current day or a subsequent day for each consumer house on the basis of measurement results from the flow rate measurement unit; a pattern information acquisition unit that acquires usage pattern information indicating hourly water usage tendencies for each of the consumer houses on the basis of the measurement results from the flow rate measurement unit; and a second calculation unit that calculates an estimated hourly water consumption for each of the consumer houses on the basis of the estimated consumption and the usage pattern information.

Description

Water demand forecasting system

[0001] The present disclosure relates to a water demand prediction system that predicts water demand at a consumer's home.

[0002] A water management plan support device has been known that includes a water management plan creation unit that creates a water management plan for a water facility, and a re-planning determination unit that calculates the amount of water surplus or shortage in a distributing reservoir where a constraint related to the upper and lower storage limits is violated based on the predicted value of the water storage volume obtained by a water storage volume prediction simulation in each distributing reservoir and predetermined upper and lower limits (Patent Document 1). This water management plan support device further includes a distributing reservoir selection unit that calculates the water exchange capacity for each distributing reservoir other than the above distributing reservoir by adjusting the amount of water received.

[0003] JP 2023-070580 A

[0004] The above-mentioned conventional technology involves transferring water between multiple reservoirs when a water storage constraint is violated due to a water surplus or shortage. However, the challenge is to avoid water shortages without transferring water between multiple reservoirs.

[0005] Therefore, an object of the present disclosure is to provide a water demand prediction system that can improve the reliability of predictions of water demand at consumer homes.

[0006] The water demand prediction system disclosed herein is a water demand prediction system that includes a water meter installed at each consumer's home and a center server that can communicate with the water meter, wherein the water meter has a flow measurement unit that is installed in a flow path through which water flows and measures the flow rate of water flowing through the flow path, and a communication unit that communicates with the center server, and the center server has a general communication unit that communicates with the communication unit, a first calculation unit that calculates an estimated daily water demand for the current day or for the days beyond the current day for each consumer's home based on the measurement results by the flow meter unit, a pattern information acquisition unit that acquires usage pattern information for each consumer's home that indicates hourly water usage trends based on the measurement results by the flow meter unit, and a second calculation unit that calculates an estimated hourly water demand for each consumer's home based on the estimated demand and the usage pattern information.

[0007] According to the present disclosure, the second calculation unit calculates the estimated hourly water demand for each consumer home based not only on the estimated daily water demand for the current day or beyond the current day calculated by the first calculation unit, but also on usage pattern information indicating hourly water usage trends acquired by the pattern information acquisition unit. This allows for highly accurate prediction of the water demand that will be required at each consumer home. This improves the reliability of predictions of water demand at each consumer home. In particular, it allows for highly accurate prediction of the water demand that will be required at each consumer home in the next few hours. This allows a water purification facility, which is a supply source of water to each consumer home, to accurately determine the amount of water that will be required in a short period of time. Therefore, it is possible to minimize the excess or deficiency of the amount of purified water compared to the required supply amount at the water purification facility, which also leads to reduced electricity costs for water purification.

[0008] According to the present disclosure, it is possible to provide a water demand prediction system that can improve the reliability of predictions of water demand at consumer homes.

[0009] Fig. 1 is a block diagram showing a water demand prediction system according to one embodiment. Fig. 2 is a block diagram showing the detailed configuration of a customer's home and a center server in the water demand prediction system of Fig. 1. Fig. 2(a) is a diagram for explaining usage pattern information, and Fig. 2(b) is a diagram for explaining an estimated water demand per hour based on the usage pattern information of Fig. 1.

[0010] A water demand prediction system according to an embodiment of the present disclosure will be described below with reference to the drawings. The water demand prediction system described below is merely one embodiment of the present disclosure. Therefore, the present disclosure is not limited to the following embodiment, and additions, deletions, and modifications are possible within the scope of the present disclosure.

[0011] Fig. 1 is a block diagram showing a water demand prediction system 100 according to one embodiment. Fig. 2 is a block diagram showing the detailed configuration of a center server 10 and a customer's home 20 in the water demand prediction system 100 of Fig. 1. Fig. 3(a) is a diagram for explaining usage pattern information, and Fig. 3(b) is a diagram for explaining the estimated water demand per hour based on the usage pattern information of Fig. 3(a).

[0012] As shown in Figure 1, the water demand prediction system 100 includes a center server 10 and a plurality of consumer homes 20. Although Figure 1 illustrates four consumer homes 20, the water demand prediction system 100 may include three or fewer consumer homes 20, or five or more consumer homes 20. The water flow paths leading to the consumer homes 20 connect to a water purification plant, a reservoir, or the like. Therefore, the number of consumer homes 20 is generally a value required when predicting the total water demand.

[0013] As shown in Figure 2, each consumer home 20 is provided with a water meter 13. The water meter 13 has a flow meter unit 11 and a communication unit 12. The water meter 13 may also have a storage unit (not shown) that is made up of various types of memory or a hard disk. Examples of consumer homes 20 include, but are not limited to, ordinary homes, hospitals, schools, municipal facilities, nursing homes, single-person homes, multi-generational homes, elderly housing, university dormitories, and commercial facilities, and may be any building that can use water.

[0014] Examples of methods for acquiring data on the flow rate used from the water meter 13 are given below. As a first example, as shown in FIG. 2 , data is acquired from a water meter 13 having a flow meter side unit 11 and a communication unit 12 via the communication unit 12. In this case, a message related to the flow rate value (indicator value) measured by the flow meter side unit 11 is transmitted to the center server 10 via the communication unit 12. As a second example, data is acquired from a water meter 13 that has a flow meter side unit 11 but does not have a communication unit 12. In this case, the communication unit 12 is provided independently outside the water meter 13. Note that the flow meter side unit 11 in the first and second examples is, for example, an impeller type or an ultrasonic type.

[0015] A third example involves acquiring data from a water meter 13 that has a flow meter unit 11 capable of pulse output, a communication unit 12, and no battery. In this case, the communication unit 12 is provided external to the water meter 13, converts the pulse output value from the flow meter unit 11 into a pointer value, etc., and transmits it to the center server 10. A fourth example involves attaching an attachment to a water meter 13 that does not have a communication unit 12 or a battery. In this case, the pointer value displayed on the water meter 13 is photographed by a camera provided on the attachment, and the captured image is digitized using the OCR (Optical Character Reader) function of the attachment. This digitized data is then transmitted to the center server 10 via the communication function of the attachment. The flow meter unit 11 in the third and fourth examples is, for example, an impeller type.

[0016] In this embodiment, the flowmeter unit 11 is, for example, an ultrasonic flowmeter. The flowmeter unit 11 is provided in a flow path 11a through which water flows and measures the flow rate of water flowing through the flow path 11a. Specifically, the flowmeter unit 11 measures the flow rate of water based on a propagation time calculated based on an ultrasonic signal transmitted from one of two ultrasonic transmitter-receivers and received by the other ultrasonic transmitter-receiver.

[0017] The communication unit 12 has a function of wirelessly communicating with the central communication unit 5 (described later) in the center server 10. A communication network such as the Internet, LAN, LPWA (Low Power Wide Area), or WI-SUN can be used as a wireless communication method between the communication unit 12 and the central communication unit 5 of the center server 10. The communication unit 12 periodically (e.g., once a day) transmits information related to guideline values, which are information on the amount of water used (consumption) at the consumer's home 20, to the central communication unit 5 along with date information on whether water is used or not (i.e., zero usage), an ID identifying the consumer's home 20, and location information for the consumer's home 20. The information related to the guideline values ​​transmitted from the communication unit 12 to the central communication unit 5 includes, for example, 24 pieces of information. Each of these 24 pieces of information represents the amount of water used per hour in a day. The location information may be provided in the center server 10.

[0018] The center server 10 has a first calculation unit 1, a pattern information acquisition unit 2, a second calculation unit 3, a classification unit 4, and an integrated communication unit 5. Of the above components of the center server 10, the first calculation unit 1, the pattern information acquisition unit 2, the second calculation unit 3, and the classification unit 4 are functionally realized by a microcontroller including a CPU (Central Processing Unit) and memory (ROM (Read Only Memory) and RAM (Random Access Memory)) that stores a program, or an ASIC (Application Specific Integrated Circuit), etc. The center server 10 may also have a storage unit (not shown) that is made up of various types of memory, a hard disk, etc. Furthermore, the center server 10 may be configured as a cloud system.

[0019] The central communication unit 5 communicates with the communication unit 12 of each consumer home 20. The communication method of the central communication unit 5 is the same as the above-mentioned communication method of the communication unit 12. The central communication unit 5 periodically (for example, once a day) receives information related to guideline values, which are information on the amount of water used (consumption) in the consumer home 20, from the communication unit 12.

[0020] The first calculation unit 1 calculates the estimated daily water demand for the current day or for the future days for each consumer home 20 based on the measurement results by the flowmeter unit 11. Specifically, the first calculation unit 1 linearly calculates the estimated daily water demand for the current day or for the future days based on information about daily usage measured by the flowmeter unit 11 for the past week. Alternatively, the first calculation unit 1 may calculate the average water usage for the past week as the estimated water demand. Alternatively, the first calculation unit 1 may calculate the average water usage for each day of the week based on information about daily usage for the past month, and calculate the average water usage for the day of the week corresponding to the day of the week at the time of prediction as the estimated water demand. Alternatively, the first calculation unit 1 may calculate the maximum water usage for the past week as the estimated water demand. Furthermore, the first calculation unit 1 may perform each of the above calculations for each attribute of the consumer's residence 20 (for example, an ordinary household, a hospital, a school, a municipal facility, a nursing home, a single-family home, a multi-generational home, a senior housing unit, a university dormitory, a commercial facility, etc.). Note that each of the calculation functions performed by the first calculation unit 1 described above may be performed by an artificial intelligence function.

[0021] The pattern information acquisition unit 2 acquires usage pattern information indicating hourly water usage trends for each consumer home 20 based on the measurement results from the flowmeter unit 11. Specifically, the pattern information acquisition unit 2 calculates the average usage amount for each day of the week and for each hour based on, for example, information about daily usage amounts for each hour over the past week. The pattern information acquisition unit 2 also calculates the average daily usage amount for each day of the week based on the information about the past week. Then, the pattern information acquisition unit 2 acquires, as usage pattern information, the ratio of the average hourly usage amount to the average daily usage amount for the same day of the week, as shown in FIG. 3( a). In this way, the usage pattern information is data for each day of the week and for each hour.

[0022] Alternatively, the pattern information acquisition unit 2 may calculate the average usage amount for each day of the week and for each hour based on, for example, information about daily hourly usage amounts for the past month. In this case, as in the above case, the pattern information acquisition unit 2 calculates the average daily usage amount for each day of the week based on the information for the past month. Then, the pattern information acquisition unit 2 acquires, as usage pattern information, the ratio of the average hourly usage amount to the average daily usage amount for the same day of the week. Note that it is not necessary to calculate the average usage amount for each day of the week. Alternatively, the average daily usage amount for each hour of the week may be calculated based on information for one month for the same period in the past year. In this case, as in the above case, the pattern information acquisition unit 2 calculates the average daily usage amount for each day of the week based on information for one month for the same period in the past year. Then, the pattern information acquisition unit 2 acquires, as a usage pattern, the ratio of the average hourly usage amount to the average daily usage amount for the same day of the week. It should be noted that even when calculating the average usage amount based on information for one month for the same period one year in the past, it is not essential to calculate it for each day of the week.

[0023] The pattern information acquisition unit 2 may have an artificial intelligence 2a. In this case, the pattern information acquisition unit 2 receives the measurement results from the flowmeter unit 11 as input and acquires the above-mentioned usage pattern information by the function of the artificial intelligence 2a.

[0024] The classification unit 4 classifies the consumer homes 20 into groups based on the attributes of the consumer homes 20. The attributes of the consumer homes 20 are as described above. In an aspect in which such a classification unit 4 is provided, the pattern information acquisition unit 2 acquires, as usage pattern information, information uniformly indicating the hourly water usage trends of each of multiple consumer homes 20 classified into the same group based on the attributes. In this case, the pattern information acquisition unit 2 calculates, for example, average values ​​of usage by day of the week and by hour for each consumer home 20 belonging to the same group based on information about daily hourly usage for the past week, the past month, or the same period of the past year, and calculates an average value of the calculated average values ​​(hereinafter referred to as the overall average value for each hour). Furthermore, the pattern information acquisition unit 2 calculates an average value of daily usage for each consumer home 20 belonging to the same group based on information about daily hourly usage for the past week, the past month, or the same period of the past year, and calculates an average value of the calculated average values ​​(hereinafter referred to as the overall average value for each day). Then, the pattern information acquiring unit 2 acquires, as the usage pattern information, the ratio of the overall average value for each hour to the overall average value for one day, in the same manner as described above with reference to FIG. 3(a).

[0025] The second calculation unit 3 calculates the estimated hourly water demand for each consumer home 20 based on the estimated demand obtained by the first calculation unit 1 and the usage pattern information obtained by the pattern information acquisition unit 2. Specifically, as shown in FIG. 3( b), the second calculation unit 3 calculates the product of the ratio of the average hourly usage amount to the average daily usage amount and the estimated water demand Q obtained by the first calculation unit 1. The estimated water demand per hour is obtained by such calculation by the second calculation unit 3.

[0026] The second calculation unit 3 may add a predetermined margin to the estimated hourly water demand and output the result. In this case, the second calculation unit 3 may add, for example, 20% of the estimated hourly water demand to the estimated water demand. Note that the amount added is not limited to 20% and can be changed appropriately taking into account empirical rules.

[0027] Alternatively, in an embodiment in which the classification unit 4 is provided, the first calculation unit 1 and the second calculation unit 3 may perform the following processing using the usage pattern information acquired by the pattern information acquisition unit 2. The first calculation unit 1 calculates the sum of the estimated water demand per day for the current day or for the days beyond the current day for each of multiple consumer homes 20 classified in the same group based on the uniform usage pattern information. In this case, a common estimated demand is calculated for each consumer home 20 belonging to the same group, and the calculated estimated demand is multiplied by the number of consumer homes 20 belonging to the group to calculate the sum. The second calculation unit 3 then calculates the estimated hourly water demand for each of multiple consumer homes 20 belonging to the same group based on the sum of the estimated demands and the uniform usage pattern information. Therefore, in this case, the estimated hourly water demand is not calculated for each consumer home 20.

[0028] As described above, according to the water demand prediction system 100 of this embodiment, the second calculation unit 3 calculates the estimated hourly water demand for each consumer home 20 based not only on the estimated daily water demand for the current day or beyond the current day calculated by the first calculation unit 1 but also on the usage pattern information indicating the hourly water usage trend acquired by the pattern information acquisition unit 2. This allows the water demand likely to be required at each consumer home 20 to be predicted with high accuracy. This improves the reliability of the water demand prediction for each consumer home. In particular, it allows the water demand likely to be required at each consumer home 20 in several hours to be predicted with high accuracy. This allows the water purification facility, which is the supply source of water to each consumer home 20, to determine the amount of water that will need to be supplied in a short period of time. Therefore, it is possible to minimize the excess or deficiency of the amount of purified water compared to the required supply amount at the water purification facility, which also leads to a reduction in the electricity bill for water purification.

[0029] The present disclosure is not limited to the above-described embodiments, and various modifications are possible without departing from the spirit of the present disclosure. For example, the following modifications are possible.

[0030] In the above embodiment, the first calculation unit 1, the pattern information acquisition unit 2, and the second calculation unit 3 are provided in the center server 10, but this is not limitative. The first calculation unit 1, the pattern information acquisition unit 2, and the second calculation unit 3 may also be provided in the water meter 13.

[0031] In the above embodiment, the attributes of the consumer homes 20 are classified by the buildings that can use water, but this is not limiting. The attributes of the consumer homes 20 may be classified, for example, by water usage patterns. In this case, the usage patterns may be classified based on a morning and evening usage pattern, a night usage pattern, a daytime usage pattern, a daily usage pattern, a daytime usage pattern, a maximum usage flow rate, etc.

[0032] In addition, in the above embodiment, the estimated hourly water demand in the water supply network that supplies water from the water supply facility to each consumer home 20 is the sum of the estimated hourly water demand at each consumer home 20 in the water supply network.

[0033] In the above embodiment, the estimated hourly water demand in a housing complex such as an apartment building is the sum of the estimated hourly water demands of each consumer's home 20 in the housing complex.

[0034] Furthermore, in the above embodiment, the pattern information acquisition unit 2 is provided with the artificial intelligence 2a, but the pattern information acquisition unit 2 only needs to obtain usage pattern information, and the artificial intelligence 2a is not an essential component.

[0035] (Additional Notes) The above description of the embodiments discloses the following techniques.

[0036] (Technology 1) A water demand prediction system comprising a water meter installed at each consumer's home and a center server capable of communicating with the water meter, wherein the water meter has a flow measurement unit installed in a flow path through which water flows and measures the flow rate of water flowing through the flow path, and a communication unit that communicates with the center server, and the center server has a general communication unit that communicates with the communication unit, a first calculation unit that calculates an estimated daily water demand for the current day or for the days beyond the current day for each consumer's home based on the measurement results by the flow meter unit, a pattern information acquisition unit that acquires usage pattern information indicating hourly water usage trends for each consumer's home based on the measurement results by the flow meter unit, and a second calculation unit that calculates an estimated hourly water demand for each consumer's home based on the estimated demand and the usage pattern information.

[0037] With this configuration, the second calculation unit calculates the estimated hourly water demand for each consumer home based not only on the estimated daily water demand for the current day or beyond the current day calculated by the first calculation unit, but also on usage pattern information indicating hourly water usage trends acquired by the pattern information acquisition unit. This allows for highly accurate prediction of the water demand likely to be required at each consumer home. This improves the reliability of predictions of water demand at each consumer home. In particular, it allows for highly accurate prediction of the water demand likely to be required at each consumer home in the next few hours. This allows water purification facilities, which are the supply sources for water to each consumer home, to accurately determine the amount of water that will be required in a short period of time. Therefore, it is possible to minimize the excess or deficiency of the amount of purified water compared to the required supply amount at the water purification facility, which also leads to reduced electricity costs for water purification.

[0038] (Technology 2) The pattern information acquisition unit has an artificial intelligence function and acquires the usage pattern information using the artificial intelligence function with the measurement results from the flow meter side unit as input. This is a water demand prediction system described in Technology 1.

[0039] With this configuration, reliable usage pattern information can be obtained by continuing machine learning or deep learning using artificial intelligence.

[0040] (Technology 3) A water demand prediction system described in Technology 1 or 2, in which the pattern information acquisition unit uses measurement results from the flow meter unit over a predetermined period in the past to acquire average water usage amounts for each day of the week and each hour as the usage pattern information.

[0041] With this configuration, average values ​​for each day of the week and each hour are acquired as usage pattern information, which can further improve the reliability of predictions of water demand.

[0042] (Technology 4) A water demand prediction system described in any one of technologies 1 to 3, further comprising a classification unit that classifies the consumer homes into groups based on the attributes of the consumer homes, and the pattern information acquisition unit acquires, as the usage pattern information, the average amount of water usage per hour at multiple consumer homes classified into the same group based on the attributes.

[0043] This configuration allows consumer homes with similar water usage patterns to be classified into one group. The average hourly water usage for multiple consumer homes classified into the same group is then obtained as usage pattern information. This allows the usage pattern information to be an average amount of water usage, excluding irregular water usage, making it less likely that the water demand forecast will be biased.

[0044] (Technology 5) A water demand prediction system as described in Technology 4, wherein the first calculation unit calculates the sum of the estimated daily water demand for the current day or for the days beyond the current day for each of the multiple consumer homes classified into the same group based on the uniform usage pattern information, and the second calculation unit calculates the estimated hourly water demand for each of the multiple consumer homes classified into the same group based on the sum of the estimated demand and the usage pattern information.

[0045] With this configuration, the calculation process by the first calculation unit and the calculation process by the second calculation unit only need to be performed for each group, and these calculation processes do not need to be performed for each consumer home. This reduces the number of calculation processes, thereby reducing the processing load on the center server. Therefore, an inexpensive center server with low specifications can be used.

[0046] (Technology 6) A water demand prediction system according to any one of technologies 1 to 5, wherein the second calculation unit adds a predetermined margin to the estimated water demand per hour and outputs the result.

[0047] This configuration further reduces the possibility of water shortages occurring in water supply facilities such as water purification facilities.

[0048] REFERENCE SIGNS LIST 1 First calculation unit 2 Pattern information acquisition unit 3 Second calculation unit 4 Classification unit 5 General communication unit 10 Center server 11 Flow meter side unit 11a Flow path 12 Communication unit 13 Water meter 20 Customer's house 100 Water demand prediction system

Claims

1. A water demand prediction system comprising a water meter installed at each consumer's home and a center server capable of communicating with the water meter, wherein the water meter has: a flow measurement unit installed in a flow path through which water flows and measuring the flow rate of water flowing through the flow path; and a communication unit for communicating with the center server, and the center server has: a general communication unit for communicating with the communication unit; a first calculation unit for calculating an estimated daily water demand for the current day or for the days beyond the current day for each consumer's home based on the measurement results by the flow meter unit; a pattern information acquisition unit for acquiring usage pattern information indicating hourly water usage trends for each consumer's home based on the measurement results by the flow meter unit; and a second calculation unit for calculating an estimated hourly water demand for each consumer's home based on the estimated demand and the usage pattern information.

2. A water demand prediction system as described in claim 1, wherein the pattern information acquisition unit has an artificial intelligence function and acquires the usage pattern information using the artificial intelligence function with the measurement results from the flow meter unit as input.

3. The water demand prediction system of claim 1, wherein the pattern information acquisition unit acquires the average water usage amount for each day of the week and each hour as the usage pattern information using measurement results from the flow meter unit over a predetermined period of time in the past.

4. The water demand prediction system of claim 1, further comprising a classification unit that classifies the consumer homes into groups based on the attributes of the consumer homes, and the pattern information acquisition unit acquires as the usage pattern information information that uniformly indicates the hourly water usage trends of each of the multiple consumer homes classified into the same group based on the attributes.

5. The water demand prediction system described in claim 4, wherein the first calculation unit calculates the sum of the estimated daily water demand for the current day or beyond the current day for each of the multiple consumer homes classified into the same group based on the uniform usage pattern information, and the second calculation unit calculates the estimated hourly water demand for each of the multiple consumer homes classified into the same group based on the sum of the estimated demand and the usage pattern information.

6. The water demand prediction system according to claim 1, wherein the second calculation unit adds a predetermined margin to the estimated water demand per hour and outputs the result.

Images