AU2019222779A1 - Systems and methods for forecasting a water level in a water storage receptacle - Google Patents

Abstract

Abstract A system and method is provided for forecasting a water level in a water storage receptacle, the water level transmitted to at least one computing device connected to a communications network, the system including at least one processor in operable 5 connection with a memory configured by a database, the processor configured to: receive an initial water level in the water storage receptacle; receive a water storage receptacle location; determine from the water storage receptacle location, an associated rainfall catchment area; retrieve from a database an actual rainfall recorded for the location; determine a change in water level in the water storage 10 receptacle based on the catchment area and the actual rainfall; retrieve an estimated water consumption rate for the water storage receptacle; and calculate a current water level based on the initial water level, the change in water level and the estimated water consumption rate; and display the current water level to a user.

Description

Systems and Methods for Forecasting a Water Level in a Water Storage Receptacle [0001] This application claims priority from Australian Application No. 2018903173 5 filed on 29 August 2018, the contents of which are to be taken as incorporated herein by this reference.

Technical Field [0002] The present invention relates generally to systems and methods for forecasting the water level in a water storage receptacle, and more particularly the 10 use of computers to facilitate real-time delivery of modelling and forecasting of the water level to a user.

Background of Invention [0003] Although water is innately a renewable resource, providing reliable access to water can be a challenge, particularly in regions that are not connected to a mains 15 water supply. That is, if rainfall is not occurring in the locations where it is needed, then water storages in those areas are not being replenished. Moreover, due to heightened awareness of the prevalence of drought in Australia and elsewhere, even those households having access to a mains water supply are installing water storage receptacles or tanks in increasing numbers to capture and store water for use inside 20 or outside the home, i.e. for flushing toilets or irrigating the garden, in addition to those who have always relied on their own water storage to supply household, commercial, industrial and agricultural purposes.

[0004] Thus it will be understood that it is useful to monitor the water level within water storage tanks. In some cases, monitoring is necessary since an alternative 25 source of water must be obtained if water levels are running low, or in other cases, for interest.

[0005] Various methods and apparatus for measuring the water level within a storage tank are known, including manual means such as using a stick, or knocking on the side of the tank to assess acoustically where the approximate water level 30 resides, or sensor based systems including pressure sensors, depth sensors, float

2019222779 26 Aug 2019 systems and the like. However, the manual means exemplified require regular input from a user which may be inconvenient. Moreover, continuous monitoring of sensor based systems at a remote location has the disadvantage of requiring power to be provided to the sensors at all times in what may be a remote location. Additionally, sensor based monitoring systems cause power and data to be constantly consumed resulting in significant costs.

[0006] Furthermore, knowing the current water level in the water storage receptacle does not alone assist a user in understanding whether the water level is above or below a critical level, nor does it indicate when it is necessary to arrange for 10 a top up of the water storage levels by a water carter. Generally, the decision to order water is made based on a “gut feel”.

[0007] Therefore it would be desirable to provide means for monitoring, with a reasonable degree of accuracy, the water level in a water storage tank over time, without having to make continuous measurements and additionally to reliably forecast 15 when the water level will fall below a critical level.

[0008] A reference herein to a patent document or other matter which is given as prior art is not to be taken as an admission that that document or matter was known or that the information it contains was part of the common general knowledge as at the priority date of any of the claims.

Summary of Invention [0009] According to an aspect of the present invention, there is provided a system for forecasting a water level in a water storage receptacle, the water level transmitted to at least one computing device connected to a communications network, the system including at least one processor in operable connection with a memory configured by 25 a database, the processor configured to receive an initial water level in the water storage receptacle; receive a water storage receptacle location; determine from the water storage receptacle location, an associated rainfall catchment area; retrieve from a database an actual rainfall recorded for the location; determine a change in water level in the water storage receptacle based on the catchment area and the actual rainfall; retrieve an estimated water consumption rate for the water storage receptacle; and calculate a current water level based on the initial water level, the

2019222779 26 Aug 2019 change in water level and the estimated water consumption rate; and display the current water level to a user.

[0010] The processor is preferably further configured to forecast a period of time remaining before the water level drops below a predetermined threshold based on the 5 current water level and the water consumption estimate.

[0011] In some embodiments, the system includes a water storage receptacle location input component which receives an input from the user in the form of an address.

[0012] The system may further include a global positioning system enabled device positioned in the vicinity of the water storage and a water storage receptacle location input component which receives coordinates or a latitude or longitude from the global positioning system enabled device.

[0013] The system may provide access to a database storing overhead imagery wherein determining the associated rainfall catchment area includes identifying, based on the water storage receptacle location, a nearby rainfall catchment area from the overhead imagery and estimating a size of the rainfall catchment area. The rainfall catchment area may comprise a roof.

[0014] In some embodiments, the system includes a water storage receptacle volume input component for receiving a receptacle volume from the user.

[0015] The change in water level may be calculated at intervals based on a volume of water that should have flowed into the tank based on the size of the rainfall catchment area and the actual rainfall recorded at the location since a previous calculation of the change in water level. The intervals at which the change in water level is calculated may be regular intervals.

[0016] In certain embodiments, the estimated water consumption rate is derived from an estimated volume of water consumed per person per interval and the number of residents known to reside at the water storage receptacle location.

[0017] According to some embodiments, the estimated water consumption rate for the water storage receptacle is determined by retrieving at least two water levels

2019222779 26 Aug 2019 received for the water storage receptacle and determining the consumption rate over a period of time based on the change in water level from a first water level to a second water level together with any volume of water that should have flowed into the tank based on the size of the rainfall catchment area and the actual rainfall recorded 5 at the location during the period of time.

[0018] Preferably, more than two water levels are received for the water storage receptacle and more than one consumption rate is calculated, each consumption rate being determined for a different period of time and the estimated consumption rate is determined by calculating an average of the more than one consumption rates.

[0019] The estimated water consumption rate may be determined via application of machine learning techniques. The machine learning techniques may include Random Forecast, Neural Networks, ARI MA or regression methods.

[0020] In some embodiments, forecasting a period of time remaining before the water level drops below a predetermined threshold involves retrieving a most recent 15 water level, based on the estimated consumption rate determining a period of time in which the water level will reach zero and checking whether rainfall is forecast during the period of time, and if no rain is forecast, displaying the period of time in which the water level will reach zero to the user and if rain is forecast during the period of time in which the water level will reach zero, forecasting the volume of water that should 20 flow into the tank based on the size of the rainfall catchment area and the rainfall forecast for the location and recalculating the period of time in which the water level will reach zero.

[0021] According to another aspect of the present invention, there is provided a method for forecasting a water level in a water storage receptacle, the method 25 including the following steps: receiving an initial water level in the water storage receptacle; receiving a water storage receptacle location; determining from the water storage receptacle location, an associated rainfall catchment area; retrieving from a database an actual rainfall determined for the location; determining a change in water level in the water storage receptacle based on the catchment area and the actual 30 rainfall; retrieving an estimated water consumption rate for the water storage receptacle; and calculating a current water level based on the initial water level, the

2019222779 26 Aug 2019 change in water level and the estimated water consumption rate; and displaying the current water level to a user.

[0022] The method may further include the step of forecasting a period of time remaining before the water level drops below a predetermined threshold based on the 5 current water level and the water consumption estimate.

[0023] In some embodiments, the step of receiving a water storage receptacle location involves receiving an input from the user in the form of an address. In other embodiments, the step of receiving a water storage receptacle location involves receiving coordinates or a latitude or longitude from a GPS enabled device positioned 10 in the vicinity of the water storage receptacle location.

[0024] The step of determining the associated rainfall catchment area may include identifying, based on the water storage receptacle location, a nearby rainfall catchment area from overhead imagery and estimating a size of the rainfall catchment area. The rainfall catchment area may comprise a roof.

[0025] In some embodiments, the method further includes the step of receiving an input of a water storage receptacle volume.

[0026] The step of calculating the change in water level may be executed at intervals based on a volume of water that should have flowed into the tank based on the size of the rainfall catchment area and the actual rainfall recorded at the location 20 since a previous calculation of the change in water level. The intervals at which the step of calculating the change in water level is executed may be regular intervals.

[0027] In certain embodiments, the step of calculating the estimated water consumption rate involves deriving an estimated water consumption rate from an estimated volume of water consumed per person per interval and the number of 25 residents known to reside at the water storage receptacle location.

[0028] The step of calculating the estimated water consumption rate may involve retrieving at least a first water level received at a first time and a second water level received at a second time and determining a consumption rate over a period of time elapsed between the first time and the second time based on the change in water 30 level together with any volume of water that should have flowed into the tank based on the size of the rainfall catchment area and the actual rainfall recorded at the location during the period of time.

[0029] Preferably, more than two water levels are received for the water storage receptacle and more than one consumption rate is calculated, each consumption rate being determined for a different period of time and the estimated consumption rate is determined by calculating an average of the more than one consumption rates.

[0030] The estimated water consumption rate may be determined via application of machine learning techniques. The machine learning techniques may include Random Forecast, Neural Networks, ARI MA or regression methods.

[0031] In some embodiments, the step of forecasting a period of time remaining before the water level drops below a predetermined threshold involves retrieving a most recent water level and based on the estimated consumption rate determining a period of time within which the water level will reach zero, checking whether rainfall is forecast during the period of time, and if no rain is forecast during the period of time displaying the period of time in which the water level will reach zero to the user, and if rain is forecast during the period of time in which the water level will reach zero, forecasting the volume of water that should flow into the tank based on the size of the rainfall catchment area and the rainfall forecast for the location and recalculating the period of time in which the water level will reach zero.

[0032] According to another aspect of the present invention, there is provided a non-transitory computer readable media embodied with software for forecasting a water level in a water storage receptacle, the software when executed configured to receive an initial water level in the water storage receptacle; receive a water storage receptacle location; determine from the water storage receptacle location, an associated rainfall catchment area; retrieve from a database an actual rainfall determined for the location; determine a change in water level in the water storage receptacle based on the catchment area and the actual rainfall; retrieve an estimated water consumption rate for the water storage receptacle; and calculate a current water level based on the initial water level, the change in water level and the estimated water consumption rate; and display the current water level to a user.

Brief Description of Drawings [0033] Embodiments of the invention will now be described with reference to the accompanying drawings. It is to be understood that the embodiments are given by way of illustration only and the invention is not limited by this illustration. In the drawings:

[0034] Figure 1 shows a system diagram of an exemplary embodiment of a system for monitoring a water level in a water storage receptacle.

[0035] Figure 2 shows a schematic diagram showing the processes that interact according to an embodiment of the present invention.

[0036] Figure 3 shows a process flow diagram of an exemplary method for calculating a current water level in the water storage receptacle.

[0037] Figure 4 shows a process flow diagram of an exemplary method for determining the water consumption rate for the water storage receptacle.

[0038] Figure 5 shows a process flow diagram of an exemplary method for forecasting the expected water level in a water storage receptacle.

[0039] Figure 6 shows an exemplary graphical use interface for the displaying the water level forecast to a user.

[0040] Figure 7 shows an exemplary computing device in a computer network which may be used to implement the present invention.

Detailed Description [0041] Referring to Figure 1, there is shown a system 100 for implementing the method for forecasting the water level in a water storage receptacle 110. The water storage receptacle 110 is associated with and located generally nearby to a structure or building 120, which could take the form of a private residence, business or government owned building. By associated with, it will be understood that the water storage receptacle 110 will store water that is utilised by the residents or other inhabitants of the building or structure 120. Typically, the building or structure 120 will include a roof structure 130 which may serve as the only, or one of several rainfall catchment areas associated with the water storage receptacle 110. That is, the rainfall catchment area 130 provides a catchment area from which rainfall flows into water storage receptacle 110.

[0042] A communications network includes one or more servers 170 including one communicatively coupled to a wireless communications network 160. The server 170 including at least one processor in operable connection with a memory configured by a database 180. The communications network could take the form of a packet switching network, such as the Internet, a network WLAN or a cloud computing environment. The cloud computing environment includes a set of network services that are capable of being used remotely over a network, and the method described herein may be implemented as a set of instructions stored in a memory and executed by a cloud computing platform.

[0043] Throughout the description there are references to computers as example devices where information is moved to and from. In other embodiments, the implemented devices are smartphones, tablets, laptop computers, desktop computers, server computers, among other forms of computer systems.

[0044] User 150 may include an owner, resident or manager of building or structure 120 having an interest in the water level in the water storage receptacle 110. That is, user 150 wishes to know whether the water level in the water storage receptacle 110 has or will drop below a predetermined threshold and is approaching critical levels. This provides user 150 with an opportunity to take measures to replenish the water level using means other than relying on rainfall. For instance, by having water carted to the water storage receptacle 110. The user 150 is provided with a user computer 140 which may include any one or more of a mobile communication device, tablet, desktop computer or the like.

[0045] In some embodiments, the water storage receptacle 110 includes a sensor (not shown) for sensing the water level, either on demand or at predetermined intervals. Such a water level sensor may be a pressure sensor, ultrasonic sensor, float system, or the like. In the alternative, it will be understood that there are other means for determining the current water level in the water storage receptacle including banging on the wall of the tank to sense the approximate water level

2019222779 26 Aug 2019 acoustically, manually reading a measuring device placed inside of the water tank which can be used to manually ascertain a current water level that can be input into the system by the user 150 using the mobile communications device. Regardless of the means for determining the water level, it will be understood that the current, or at 5 least a recent water level, is a necessary input into the forecasting model.

[0046] In some embodiments, the water storage receptacle 110 is provided with a GPS receiver (not shown) or the mobile communications device 140 associated with the user 150 is GPS enabled in order that a location of the water storage receptacle can be accurately determined. That is, in the case of the mobile communications 10 device being GPS enabled, determining the location of the water storage receptacle

110 can be determined by the user 150 positioning the mobile communications device in proximity to the water storage receptacle to ascertain the GPS location. Other means of determining the water storage receptacle location include prompting the user 150 to input an address for a location associated with the water storage receptacle on registration for the water level forecasting service. The requirement for the water storage receptacle location as an input to the forecasting model will become apparent below. Other data input by the user 150 during registration for the water level forecasting service includes a count of the number of residents or people associated with the building or structure. The number of people occupying the building is relevant to determination of water consumption which is described in more detail below.

[0047] A further input into the forecast model is the rainfall catchment area, i.e. dimensions thereof. It will be understood that catchment area data may be obtained, for example, from overhead imagery that could be supplied by satellite 190 and stored 25 in database 195. Rainfall catchment areas such as rooves are readily discernible from overhead imagery provided, for example, by Google Earth and Google Maps, Landsat and ESRI and the like, and the roof area can be readily estimated from the overhead imagery and extrapolated into real world dimensions to provide a relatively accurate estimate of the rainfall catchment area. In turn, knowing the rainfall catchment area 30 can be used to estimate the rainfall that would flow into the associated water storage receptacle for different rates of rainfall.

2019222779 26 Aug 2019 [0048] Referring now to Figure 2 there is shown a schematic diagram 200 showing the various processes that interact to provide water level forecasts to a user 150 (see Figure 1). Sensed and calculated data concerning water storage receptacle 110 (see Figure 1) is stored in one or more databases 210 which are accessible via communications network 160, also described with reference to Figure 1. This includes sensor readings taken at the water storage receptacle 110 to determine the current water level.

[0049] Using other inputs such as the rainfall catchment area, which is approximated in the manner previously described, together with local rainfall data and water consumption rates, the determination of which is detailed below, a future water level of the water storage receptacle is forecast at 230. This model can be executed at predetermined intervals or on a regular schedule, say every 12 to 24 hours to provide a current, up-to-date estimate of current water levels in the water storage receptacle.

[0050] In an embodiment, the present invention may be provided in a number of locations over a geographical area and the water storage receptacle systems may communicate with one another. For example, tens, hundreds or thousands of units (which may be battery powered) may be present and in-use at any given point in time over a particular geographic location. Data from a water storage receptacle in a particular geographic location may be provided to another water storage receptacle system - to improve accuracy and avoiding the need for duplication in sending data from water storage receptacle systems which are in the same geographic location with similar weather patterns.

[0051] The sensor associated with the water storage receptacle system may be both low power and low frequency avoiding complexity in terms of electrical and RF interference and, in turn, increasing the ease of which data may be reliably captured and transmitted wirelessly. For example sensors utilising a low-power narrowband are low cost and provide long battery life.

[0052] The sensors may be provided across a geographical area in a distributed wireless sensor system via a low-density mesh of base stations - avoiding the cost and complexity of high-density mesh systems. A messaging protocol system based on IEEE 802.15.4, such as ZigBee, may be used with the sensor in order to simplify the messaging structures and network mesh controls.

[0053] Advantageously, the physical implementation of the sensors allows for the data to be provided in a reliable and low cost manner.

[0054] At 240, a water consumption rate is estimated for the water storage receptacle. The water consumption rate is based on the historical actual measured or sensed water levels for the water storage receptacle, together with actual rainfall, i.e. the rate at which the water levels are being replenished by rainfall. Determination of water consumption rate involves a machine learning element. For example, the system is trained over time to determine a typical water consumption rate for a particular household or business, based on regular readings of the actual water level in the water storage receptacle, whilst correlating that usage data with actual rainfall data. The rainfall data and water consumption rates will be understood to have a seasonal element, for example, water consumption rates will typically increase over the warmer summer months and decrease over the cooler winter months. Moreover, typically higher levels of rainfall will be recorded, during the cooler, wetter season, than during the hotter, drier season. The water consumption model can be executed on a regular schedule, say every 7 to 10 days to provide a current, up-to-date water consumption rate.

[0055] At 250, a forecasting model is executed to predict how long the current water level in the water storage receptacle will last based the water consumption rate and forecast rainfall. More detail regarding the respective models is provided with reference to Figures 3 to 5 below.

[0056] A user 150 associated with mobile communications device 150 makes a request to forecast the water level for a water receptacle at a particular location. In response the system 100 examines data associated with one or more data sources which may be stored on databases 180 and 195 which are accessible via a telecommunications network. Specifically, the processor is configured to forecast a period of time remaining before the water level drops below a predetermined threshold based on the current water level, the water consumption rate determined for the water storage receptacle and the forecast rainfall.

[0057] Referring now to Figure 3, there is shown a process flow diagram for the method for calculating the current water level in the water storage receptacle. This process flow is an expansion of process 230 referenced in Figure 2. The water level is sensed or measured at predetermined intervals employing means described with reference to Figure 2. Each time a current water level is sensed at step 310, at step 320 there is a decision point to query whether any rainfall has occurred since the immediately previous water level reading. If the answer to the query at step 320 is no, then at step 330 a current water level is determined having regard to the current volume of water residing in the water storage receptacle, the water consumption rate per person, per day, the number of residents/people associated with the water storage receptacle and the period of time elapsed since the immediately previous water level reading, i.e. :

Result_1 = Max (A - B * C * D, 0) wherein:

A = Current water volume

B = Volume consumed per person per day

C = Number of residents

D = Time since last water level reading [0058] On the other hand, in the event that the answer to the query at step 320 is yes, then at step 340 the amount of water assumed to have flowed into the water storage receptacle since the immediately previous water level reading is calculated. This involves determining, via weather data, the amount of rainfall recorded at the location associated with the water storage receptacle, in the period since the immediately previous water level reading, together with the estimated rainfall catchment area and the time elapsed since the previous water level reading, i.e. :

Result_2 = (A * B) * C wherein:

A = Rainfall in location, over period

B = Estimated rainfall catchment area

C = Time since last reading

2019222779 26 Aug 2019 [0059] At step 350 the the new water level (Result_1) and the amount of water that would have flowed into the water storage receptacle since the immediately previous water level reading (Result_2) are stored in a database 180 where they can be retrieved as inputs into related models.

[0060] Referring now to Figure 4, there is shown a process flow diagram for calculating a water consumption rate for the water storage receptacle. This process flow is an expansion of process 240 referenced in Figure 2. As described with reference to Figure 2, determining the water consumption rate involves the application of machine learning and is preferably calculated over a number of water storage receptacles, to provide an average water consumption rate per person, which may be seasonally adjusted as required. The use of readings from multiple water storage receptacles will be understood to enhance the machine learning aspects by providing a larger data store for training the system, ultimately resulting in more accurate determination of typical water consumption rates.

[0061] The machine learning component uses previously calculated water consumption rates and divides those patterns into features and utilises those features to apply various machine learning techniques to see which technique provides an appropriate level of accuracy, i.e. best fit. That machine learning technique is then selected to calculate the water consumption rate for the next periods. For each water consumption rate that is calculated, the accuracy is subsequently checked against actual measured water level data for the period. If the accuracy of the applied machine learning technique drops below a predetermined threshold, then alternative machine learning techniques are applied to the existing water consumption rate data to select the technique providing optimal accuracy. Examples of suitable machine learning techniques include Random Forecast, Neural Networks with various configurations, ARIMA and various regression methods.

[0062] Firstly, at step 410 the process starts and at step 420 a previous water level reading is retrieved for preferably more than one water storage receptacle from the database. For example, in the case where a water level is being sensed or measured at 12 hour intervals for a group of water storage receptacles, the following readings are retrieved for each of the water storage receptacles:

i) immediate last reading (i.e. 12 hours ago);

ii) preceding day reading (i.e. 24 hours ago);

iii) one week ago reading; and iv) one month ago reading.

[0063] At step 430, for each time period, confirm that are at least two readings, i.e. associated with two different water storage receptacles are available to ensure sufficient data is available for training of machine learning techniques at decision point labelled as step 440. Subsequently, at step 450, for each water storage receptacle, and for each pair of readings, a change in water volume over the time period is determined, together with the rainfall recorded at the location over the time period, and the change in time, i.e. = the time period, to calculate the water consumption rate for that particular water storage receptacle over the period, i.e.:

Result = (A + B) / C wherein:

A = Change in volume over the period

B = Rainfall over the period

C = Change in time

The calculated value is then stored into a data store buffer. Stored values are subsequently sent to a neural network or similar for learning and testing recognition.

[0064] At step 460, the water consumption rate is determined for a particular period by calculating an average consumption rate from the values stored in the data store buffer. At step 470 the recently calculated water consumption rate is stored for later retrieval.

[0065] Referring now to Figure 5, there is shown a process flow for forecasting a future water level for the storage receptacle based on the current water level calculated according to the process described with reference to Figure 3 and the water consumption rate calculated according to the process described with reference to Figure 4, together with a rainfall forecast for the location obtained from weather data. This process flow is an expansion of process 250 referenced in Figure 2.

[0066] Firstly, at step 510 the process starts and at step 520 a previous water level reading is retrieved for the water storage receptacle queried by the user. At step 530, the number of days X until the water level will drop below a predetermined threshold, i.e. for example, when the tank will be empty, is determined based on the current water level retrieved at step 520 and the recently calculated water consumption rate stored in the data store buffer. The number of days X to empty is stored in another temporary data storage buffer together with the most recent water level sensed or measured for the water storage receptacle.

[0067] At step 540, if the water level in the buffer indicates that the water storage receptacle is currently full, then the process is terminated at step 545 and the output displayed to the user will be the number of days X until the water level will drop below the predetermined threshold. If the tank is not full, at step 550 the weather data is queried to determine whether any rainfall is forecast in the following period.

[0068] In the event that rainfall is forecast, at step 555 an amount of water that will flow into the water storage receptacle. This is calculated based on the rainfall forecast for the location over a period, the rainfall catchment area associated with the water storage receptacle; and the time elapsed since the last water level reading, i.e.:

Result_B = (A * B) * C wherein:

A = Rainfall at location, over period

B = Rainfall catchment area

C = Time elapsed since last water level reading [0069] In the event that no rainfall is forecast, at step 560 a new water level is determined based on the current water level, the water consumption rate per person, the number of people residing at the location and the time elapsed since the last water level reading, i.e.:

Result_A = Max (A - B * C * D, 0) wherein:

A = Last water level reading

B = Water consumption rate

C = Number of residents at location

D = Time elapsed since last water level reading [0070] At step 570, the new water level, i.e. Result_A and the amount of water predicted to flow into the water storage receptacle as a result of the forecast rainfall,

i.e. Result_B, are combined and added to the data storage buffer. If the water level in the buffer indicates that the tank is full at step 540, then the process is terminated at step 545 and the output displayed to the user will be the number of days X until the water level will drop below a predetermined threshold. If the tank is not full, at step 550 the weather data is queried as to whether there is any rainfall forecast and the process flow continues.

[0071] Referring now to Figure 6, there is shown a graphical user interface (GUI) on a mobile communications device for presenting the water level forecast to a user in an easy to interpret format. The GUI includes a water level indicator 620 as a percentage of full, e.g. showing 85% full, together with a count of days 630 remaining until the water level drops below a predetermined threshold. Optionally, the GUI may include an order water icon, which links directly to an online ordering facility or optionally a telephone number for a local water carter.

[0072] Referring now to Figure 7, there is shown schematically, an illustrative computer 700 on which any aspect of the present disclosure may be implemented. In the embodiment shown in Figure 7, the computer 700 includes a processing unit 705 having one or more processors and a non-transitory computer-readable storage medium 710 that may include, for example, volatile 715 and/or non-volatile 720 memory. The memory 710 may store one or more instructions to program the processing unit 705 to perform any of the functions described herein. The computer 700 may also include other types of non-transitory computer-readable mediums, such as storage 725 (e.g., one or more disk drives) in addition to the system memory 710. The storage 725 may also store one or more application programs and/or external components used by application programs (e.g., software libraries), which may be loaded into the memory 730.

2019222779 26 Aug 2019 [0073] The computer 700 may have one or more input devices and/or output devices, such as devices 735 and 740 illustrated in Figure 7. These devices can be used, among other things, to present a user interface. Examples of output devices that can be used to provide a user interface include printers or display screens for visual presentation of output and speakers or other sound generating devices for audible presentation of output. Examples of input devices that can be used for a user interface include keyboards and pointing devices, such as mice, touch pads, and digitizing tablets. As another example, the input devices 740 may include a microphone for capturing audio signals, and the output devices 735 may include a display screen for visually rendering, and/or a speaker for audibly rendering, recognized text.

[0074] As shown in Figure 7, the computer 700 may also comprise one or more network interfaces (e.g., the network interface 745) to enable communication via various networks (e.g., the network 750). Examples of networks include a local area 15 network or a wide area network, such as an enterprise network or the Internet. Such networks may be based on any suitable technology and may operate according to any suitable protocol and may include wireless networks, wired networks or fiber optic networks.

[0075] It will be understood that the method and system of the present invention 20 solves a problem for users reliant on water storage who find it difficult to determine how long the current water storage levels will last. Conventionally, the appropriate time to order water to replenish water storage levels is based on “gut feel” or in some cases, when the water runs out, which leads to inconvenience, and could result in unnecessary costs and ineffective use of natural resources. The methods described 25 herein take the guesswork out of determining how long current water storage levels will last and provide means for reliably forecasting when the water storage level will drop below a critical predetermined threshold. This is achieved by using a combination of data sources, for example to monitor the actual water level in the water storage receptacle, track actual and forecast rainfall for the location, determine 30 the rainfall catchment area, and machine learning to determine water consumption rates.

2019222779 26 Aug 2019 [0076] Where the terms comprise, comprises, comprised or comprising are used in this specification (including the claims) they are to be interpreted as specifying the presence of the stated features, integers, steps or components, but not precluding the presence of one or more other features, integers, steps or components, or group 5 thereof.

[0077] While the invention has been described in conjunction with a limited number of embodiments, it will be appreciated by those skilled in the art that many alternative, modifications and variations in light of the foregoing description are possible. Accordingly, the present invention is intended to embrace all such alternative, modifications and variations as may fall within the spirit and scope of the invention as disclosed.

Claims (31)

1. The claims defining the invention are as follows:

   1. A system for forecasting a water level in a water storage receptacle, the water level transmitted to at least one computing device connected to a communications network, the system including at least one processor in operable connection with a memory configured by a database, the processor configured to:

   a. receive an initial water level in the water storage receptacle;

   b. receive a water storage receptacle location;

   c. determine from the water storage receptacle location, an associated rainfall catchment area;

   d. retrieve from a database an actual rainfall recorded for the location;

   e. determine a change in water level in the water storage receptacle based on the catchment area and the actual rainfall;

   f. retrieve an estimated water consumption rate for the water storage receptacle; and

   g. calculate a current water level based on the initial water level, the change in water level and the estimated water consumption rate; and

   h. display the current water level to a user.
2. 2. The system for forecasting a water level in a water storage receptacle according to claim 1, wherein the processor is further configured to forecast a period of time remaining before the water level drops below a predetermined threshold based on the current water level and the water consumption estimate.
3. 3. The system for forecasting a water level in a water storage receptacle according to claim 1 or 2, further including a water storage receptacle location input component which receives an input from the user in the form of an address.
4. 4. The system for forecasting a water level in a water storage receptacle according to claim 1 or 2, the system further including a global positioning system enabled device positioned in the vicinity of the water storage and a water storage receptacle location input component which receives coordinates or a latitude or longitude from the global positioning system enabled device.
5. 5. The system for forecasting a water level in a water storage receptacle according to claim 3 or 4, further including access to a database storing overhead imagery wherein determining the associated rainfall catchment area includes identifying, based on the water storage receptacle location, a nearby rainfall catchment area from the overhead imagery and estimating a size of the rainfall catchment area.
6. 6. The system for forecasting a water level in a water storage receptacle according to claim 5, wherein the rainfall catchment area comprises a roof.
7. 7. The system for forecasting a water level in a water storage receptacle according to any one of claims 1 to 6, further including a water storage receptacle volume input component for receiving a receptacle volume from the user.
8. 8. The system for forecasting a water level in a water storage receptacle according to claim 7, wherein the change in water level is calculated at intervals based on a volume of water that should have flowed into the tank based on the size of the rainfall catchment area and the actual rainfall recorded at the location since a previous calculation of the change in water level.
9. 9. The system for forecasting a water level in a water storage receptacle according to claim 8, wherein the intervals at which the change in water level is calculated are regular intervals.
10. 10. The system for forecasting a water level in a water storage receptacle according to any one of claims 1 to 9, wherein the estimated water consumption rate is derived from an estimated volume of water consumed per person per interval and the number of residents known to reside at the water storage receptacle location.
11. 11. The system for forecasting a water level in a water storage receptacle according to any one of claims 1 to 10, wherein the estimated water consumption rate for the water storage receptacle is determined by retrieving at least two water levels received for the water storage receptacle and determining the consumption rate over a period of time based on the change in water level from a first water level to a second water level together with any volume of water that should have flowed into the tank based on the size of the rainfall catchment area and the actual rainfall recorded at the location during the period of time.
12. 12. The system for forecasting a water level in a water storage receptacle according to any one of claims 1 to 11, wherein more than two water levels are received for the water storage receptacle and more than one consumption rate is calculated, each consumption rate being determined for a different period of time and the estimated consumption rate is determined by calculating an average of the more than one consumption rates.
13. 13. The system for forecasting a water level in a water storage receptacle according to any one of claims 10 to 12, wherein the estimated water consumption rate is determined via application of machine learning techniques.
14. 14. The system for forecasting a water level in a water storage receptacle according to claim 13, wherein the machine learning techniques include Random Forecast, Neural Networks, ARIMA or regression methods.
15. 15. The system for forecasting a water level in a water storage receptacle according to any one of claims 1 to 14, wherein forecasting a period of time remaining before the water level drops below a predetermined threshold involves retrieving a most recent water level, based on the estimated consumption rate determining a period of time in which the water level will reach zero and checking whether rainfall is forecast during the period of time, and if no rain is forecast displaying the period of time in which the water level will reach zero to the user and if rain is forecast during the period of time in which the water level will reach zero, forecasting the volume of water that should flow into the tank based on the size of the rainfall catchment area and the rainfall forecast for the location and recalculating the period of time in which the water level will reach zero.
16. 16. A method for forecasting a water level in a water storage receptacle, the method including the following steps:

    a. receiving an initial water level in the water storage receptacle;

    2019222779 26 Aug 2019

    b. receiving a water storage receptacle location;

    c. determining from the water storage receptacle location, an associated rainfall catchment area;

    d. retrieving from a database an actual rainfall determined for the location;

    5 e. determining a change in water level in the water storage receptacle based on the catchment area and the actual rainfall;

    f. retrieving an estimated water consumption rate for the water storage receptacle; and

    g. calculating a current water level based on the initial water level, the

    10 change in water level and the estimated water consumption rate; and

    h. displaying the current water level to a user.
17. 17. The method for forecasting a water level in a water storage receptacle according to claim 16, wherein the method further includes the step of forecasting a period of time remaining before the water level drops below a predetermined

    15 threshold based on the current water level and the water consumption estimate.
18. 18. The method for forecasting a water level in a water storage receptacle according to claim 16 or 17, wherein the step of receiving a water storage receptacle location involves receiving an input from the user in the form of an address.
19. 19. The method for forecasting a water level in a water storage receptacle
20. 20 according to claim 16 or 17, wherein the step of receiving a water storage receptacle location involves receiving coordinates or a latitude or longitude from a GPS enabled device positioned in the vicinity of the water storage receptacle location.

    20. The method for forecasting a water level in a water storage receptacle according to claim 18 or 19, wherein the step of determining the associated rainfall

    25 catchment area includes identifying, based on the water storage receptacle location, a nearby rainfall catchment area from overhead imagery and estimating a size of the rainfall catchment area.

    2019222779 26 Aug 2019
21. 21. The method for forecasting a water level in a water storage receptacle according to claim 20 wherein the rainfall catchment area comprises a roof.
22. 22. The method for forecasting a water level in a water storage receptacle according to any one of claims 16 to 21, further including the step of receiving a

    5 receptacle volume for the water storage receptacle.
23. 23. The method for forecasting a water level in a water storage receptacle according to claim 21, wherein the step of calculating the change in water level is executed at intervals based on a volume of water that should have flowed into the tank based on the size of the rainfall catchment area and the actual rainfall recorded

    10 at the location since a previous calculation of the change in water level.
24. 24. The method for forecasting a water level in a water storage receptacle according to claim 23, wherein the intervals at which the step of calculating the change in water level is executed are regular intervals.
25. 25. The method for forecasting a water level in a water storage receptacle

    15 according to any one of claims 16 to 24, wherein the step of calculating the estimated water consumption rate involves deriving an estimated water consumption rate from an estimated volume of water consumed per person per interval and the number of residents known to reside at the water storage receptacle location.
26. 26. The method for forecasting a water level in a water storage receptacle

    20 according to any one of claims 16 to 25, wherein the step of calculating the estimated water consumption rate involves retrieving at least a first water level received at a first time and a second water level received at a second time and determining a consumption rate over a period of time elapsed between the first time and the second time based on the change in water level together with any volume of water that should

    25 have flowed into the tank based on the size of the rainfall catchment area and the actual rainfall recorded at the location during the period of time.
27. 27. The method for forecasting a water level in a water storage receptacle according to any one of claims 16 to 23, wherein more than two water levels are received for the water storage receptacle and more than one consumption rate is

    30 calculated, each consumption rate being determined for a different period of time and the estimated consumption rate is determined by calculating an average of the more than one consumption rates.
28. 28. The method for forecasting a water level in a water storage receptacle according to any one of claims 25 to 27, wherein the estimated water consumption rate is determined via application of machine learning techniques.
29. 29. The method for forecasting a water level in a water storage receptacle according to claim 28, wherein the machine learning techniques include Random Forecast, Neural Networks, ARI MA or regression methods.
30. 30. The method for forecasting a water level in a water storage receptacle according to claim 17 or any one of claims 18 to 29, when dependent on claim 17, wherein the step of forecasting a period of time remaining before the water level drops below a predetermined threshold involves retrieving a most recent water level and based on the estimated consumption rate determining a period of time within which the water level will reach zero, checking whether rainfall is forecast during the period of time, and if no rain is forecast during the period of time displaying the period of time in which the water level will reach zero to the user, and if rain is forecast during the period of time in which the water level will reach zero, forecasting the volume of water that should flow into the tank based on the size of the rainfall catchment area and the rainfall forecast for the location and recalculating the period of time in which the water level will reach zero.
31. 31. Non-transitory computer readable media embodied with software for forecasting a water level in a water storage receptacle, the software when executed configured to:

    a. receive an initial water level in the water storage receptacle;

    b. receive a water storage receptacle location;

    c. determine from the water storage receptacle location, an associated rainfall catchment area;

    d. retrieve from a database an actual rainfall determined for the location;

    2019222779 26 Aug 2019

    e. determine a change in water level in the water storage receptacle based on the catchment area and the actual rainfall;

    f. retrieve an estimated water consumption rate for the water storage receptacle; and

    5 g. calculate a current water level based on the initial water level, the change in water level and the estimated water consumption rate; and

    h. display the current water level to a user.