Autonomous rainwater collection and distribution method and device

Abstract

The invention relates to a device (10) for collecting and distributing rainwater, the device comprising: - a collection surface (32) mounted on at least one movable element (22) of a frame provided with at least one actuator (29) configured to set each said movable element in motion between a "folded" configuration and a "deployed" configuration; - a system for storing the rainwater collected on the collection surface; - a means for distributing water from the storage system downstream of the device; and - a central control unit for each actuator of the frame; - a means for acquiring a prediction of at least one future amount collected by the device in a future time interval, the prediction being a function of measurable values; - a means for determining a prediction of at least one water distribution need downstream of the device in a future time interval, the prediction being a function of measurable values. The central unit is configured to calculate at least one amount of water to be distributed in a future time interval, as a function of each of the predictions, the central unit being configured to, in the absence of a contrary command from a user within a predetermined duration, control the distribution of each amount of water calculated in each of the time intervals.

Description

[0001] DESCRIPTION

[0002] TITLE OF THE INVENTION: AUTONOMOUS METHOD AND DEVICE FOR COLLECTING AND DISTRIBUTION OF RAINWATER

[0003] TECHNICAL FIELD OF THE INVENTION

[0004] The present invention relates to a self-contained method and device for collecting and distributing rainwater. The invention is applicable to filling water reservoirs (cisterns, basins, watering troughs, or any other water container) for applications such as irrigating crops or gardens, watering animals, filling swimming pools, air cooling by misting, passive air conditioning (adiabatic, evapotranspiration through raw earth cooling the air by humidity), natural plant cooling, vertical planting, vegetated screens, etc.

[0005] STATE OF THE ART

[0006] Rainwater harvesting is typically achieved by connecting a roof gutter to a tank equipped with a distribution tap. These characteristics limit rainwater collection to the immediate vicinity of a building with a roof. Furthermore, the amount of water collected is limited to the roof area. Additionally, it is not possible to filter the collected rainwater, for example, to avoid collecting water in the presence of air pollution or sandy soil. Finally, the systems used for this collection do not allow for automated water distribution.Document CN111802237A describes a rainwater harvesting irrigation system for arid lands, comprising a self-supporting, protected, inverted umbrella-shaped rainwater harvesting device, a removable impurity filtration system, a water collection tank, a control unit, a water pump, and a dual-mode irrigation device. Document CN115217183A describes a rainwater collector for irrigation comprising an inverted umbrella-shaped collection face, a flow channel, and a storage tank. Document DE102016015786A1 describes a rainwater harvesting screen comprising a potable water reservoir and a control unit. To extract water from the air (condensation), the device is equipped with a heat pump connected to a cooling fabric. The document CN107165223A describes a water collection device, which includes a collection bag, a support column and a rain shelter.A telescopic element controls the deployment of the collection bag. Document CN211143173U describes an automatic rainwater collection device comprising a base, a collection pipe, inverted umbrella ribs, an inverted umbrella cover, a drive device, and a control box.

[0007] DESCRIPTION OF THE INVENTION

[0008] The present invention aims to remedy all or part of the drawbacks of the prior art. To this end, according to a first aspect, the present invention relates to a rainwater collection and distribution device, which comprises: - a collection surface mounted on at least one movable element of a frame equipped with at least one actuator configured to move each said movable element between a so-called "folded" configuration, in which the collection surface is folded and has a first ground footprint, and a so-called "deployed" configuration, in which the collection surface is deployed and has a second ground footprint with a surface area greater than that of the first ground footprint.

[0009] - a rainwater storage system for water collected on the collection surface,

[0010] - a means of distributing water from the storage system to the downstream end of the device, and - a central control unit for each actuator of the armature;

[0011] - a means of acquiring a prediction of at least a future quantity of water collected by the device in a future time interval, said prediction being a function of measurable values,

[0012] - a means of determining a prediction of at least one water distribution need downstream of the device in a future time interval, said prediction being a function of measurable values;

[0013] in which the central unit is configured to calculate at least one quantity of water to be distributed in a future time interval, based on each of said predictions, the central unit being configured to, in the absence of contrary command from a user during a predetermined period, command the distribution of each calculated quantity of water in each said time interval.

[0014] Thanks to these provisions, two predictions are implemented to determine the quantities of water to be distributed in the future.

[0015] In some embodiments, at least one of these predictions is determined by a machine learning-trained model on data comprising at least some predicted values ​​of past rainfall, and at least some measured values ​​of collected water quantities.

[0016] The system thus uses a kind of micrometeorology, which locally corrects regional meteorological data in the predictions implemented.

[0017] In some embodiments, the model is trained by machine learning on data containing at least some predicted past sunshine values.

[0018] The predictions implemented thus take into account evaporation, or evapotranspiration, for at least one of the predictions, in particular for the one concerning future water distribution needs.

[0019] In embodiments, the device of the invention comprises at least one sensor of an environmental physical quantity, in which the model is trained by machine learning on data comprising at least data provided by at least one said sensor.

[0020] The device includes a data collection station enabling local correction of regional meteorological data in the implemented predictions. In embodiments, the central unit is configured to calculate at least one future evapotranspiration of plants receiving water distributed by the device and to calculate at least one quantity of water to be distributed in a future time interval, as a function of said future evapotranspiration.

[0021] The central unit thus optimizes the quantities of water to be distributed according to the plants being irrigated and, possibly, their development.

[0022] In embodiments, the device of the invention includes at least one sensor of a physical quantity representative of a quantity of water available downstream of the water distributor, the central unit being configured to calculate a quantity of water to be distributed as a function of the quantity of water available downstream of the distributor.

[0023] The central unit can thus take into account water reserves downstream, for example in the ground, and possibly the presence of other water sources besides rain and the device itself.

[0024] In some embodiments, the central unit is configured, in order to calculate the quantity of water to be distributed, to implement a machine learning trained model on representative measured data of quantities of water to be distributed calculated in the past and of distribution commands provided by a user in the past.

[0025] Learning is thus supervised.

[0026] In embodiments, the device of the invention further comprises a means for training a prediction model trained by machine learning, said prediction model being configured to associate at least one value of the quantity of future water collected by the device with a function of values ​​measured in the past representative of quantities of water collected by the device associated with dating values ​​of said measured values.

[0027] In embodiments, the device of the invention further comprises a means for training a prediction model trained by machine learning, said prediction model being configured to associate at least one value of future quantity of water collected by the device with a function of weather forecast values ​​acquired in the past associated with dating values ​​of said measured values.

[0028] In embodiments, the device of the invention further comprises a means for training a model for predicting future water needs trained by machine learning, said predictive model being configured to associate at least one value of future water need distributed by the device with a function of values ​​measured in the past representative of water needs associated with dating values ​​of said measured values.

[0029] In some embodiments, the model for predicting future water needs is configured to associate at least one value of future water needs distributed by the device with a function of predictions of water collections obtained in the past. In some embodiments, at least one of these predictions is determined by an expert system operating on data comprising at least predicted values ​​of past rainfall, and at least measured values ​​of collected water quantities.

[0030] The predictions are thus made according to rules provided by at least one expert. In some embodiments, at least once per time interval for which a quantity of water to be distributed has been determined by the central unit, the predictions and each quantity of water to be distributed in each future time interval are recalculated.

[0031] According to a second aspect, the present invention relates to a method for distributing rainwater collected on a collection surface mounted on at least one moving element of a frame equipped with at least one actuator configured to move each said moving element between a so-called "folded" configuration, in which the collection surface is folded and has a first footprint on the ground, and a so-called "deployed" configuration, in which the collection surface is deployed and has a second footprint on the ground with a surface area greater than that of the first footprint on the ground, the water thus collected being stored in a storage system equipped with a water distributor.

[0032] A process that includes:

[0033] - a control step for each actuator of the frame to move the collection surface from its folded configuration to its deployed configuration,

[0034] - a rainwater collection stage in a storage system, when the collection surface is in the deployed configuration,

[0035] - a step of acquiring a prediction of at least one future quantity collected by the device in a future time interval, said prediction being a function of measurable values,

[0036] - a step of determining a prediction of at least one water distribution need downstream of the device in a future time interval, said prediction being a function of measurable values,

[0037] - a step involving the calculation of at least one quantity of water to be distributed within a future time interval, based on each of the aforementioned predictions, and

[0038] - in the absence of a contrary order from a user during a predetermined period, a control step for the distribution of each quantity of water calculated in each said time interval.

[0039] The advantages, purposes and particular characteristics of this process being similar to those of the device which is the subject of the first aspect of the invention, they are not recalled here.

[0040] According to a third aspect, the present invention relates to a rainwater collection device, which comprises:

[0041] - a collection surface mounted on at least one moving element of a frame equipped with at least one actuator configured to move each said moving element between a so-called "folded" configuration, in which the collection surface is folded and has a first footprint on the ground, and a so-called "deployed" configuration, in which the collection surface is deployed and has a second footprint on the ground with a surface area greater than the surface area of ​​the first footprint,

[0042] - a rainwater storage system collected on the collection surface, and - a central control unit for each actuator of the frame;

[0043] in which at least one moving element is mounted on a pivot joint with a vertical axis of rotation.

[0044] Thanks to these features, the collection device can adapt its rainwater collection area to external conditions, such as the presence of rain, wind, rainwater quality, temperature, weather information, predictions of the future amount of water collected by the device and water needs, possibly supplemented by user instructions. Furthermore, the collection area can fold in a plane or extend around the vertical axis of rotation, which reduces the leverage from the ground of the wind force exerted on the collection area and therefore the risk of breakage or overturning of the device.

[0045] In some embodiments, the frame includes a vertical support, supporting at least two vertically offset pivot joints with vertical axis of rotation, each of which includes a rotary actuator, these rotary actuators being synchronized.

[0046] In some embodiments, the device of the invention includes a filter positioned between the collection surface and the storage system, in a position accessible by hand without dismantling the filter.

[0047] This makes unclogging and cleaning this filter easier.

[0048] In embodiments, the device of the invention includes at least one sensor of an environmental physical quantity such as the presence of rain, temperature and wind speed, in which the central unit is configured to move the collection surface from the folded configuration to the deployed configuration, or vice versa, according to at least one value captured by said sensor.

[0049] Thus, the device can avoid collecting snow or hail, deploying its collection surface when wind force might damage it, or deploying even when there is no rain. Conversely, the device can deploy its collection surface to protect crops, animals, people, or nearby storage areas from the sun during periods of intense heat.

[0050] In embodiments, the device of the invention includes at least one means of accessing meteorological data from a remote database, in which the central unit is configured to switch the collection surface from the folded configuration to the deployed configuration, or vice versa, according to at least one value received by this access means.

[0051] Thus, the device may have few, or no, local sensors and adapt its operation to meteorological data available online. In embodiments, the central unit is configured to determine if a water distribution instruction present in the storage system is present, determine if at least one inhibition factor applies to the requested water distribution and optimize the amount of water to be distributed according to the distribution instruction, each possible inhibition factor and a prediction of future amount of water collected by the device and future water needs downstream of the device.

[0052] Thus, the central unit can adjust the amount of water distributed to the immediate demand as well as to rainwater resources and future water needs.

[0053] In some embodiments, the central unit is configured, in order to determine the prediction of future quantities of water collected and future water needs, to implement a machine learning trained model on measured data including past rainfall, quantities of water distributed in the past and captured data representative of past physical quantities.

[0054] Thus, the central unit can refine predictions by benefiting from machine learning based on water resource data and past needs of at least one similar device and / or its own operation.

[0055] In some embodiments, the central unit is configured, in order to optimize the amount of water to be distributed, to implement a model trained by machine learning on measured data and decisions made by the user in the past.

[0056] Thus, the central unit can refine the optimization by benefiting from machine learning carried out on the basis of the real operation of at least one similar device and / or its own operation.

[0057] In some embodiments, the central unit is configured, in order to optimize the amount of water to be distributed, to implement parameterized or programmed rules.

[0058] Thus, the central unit can refine the optimization by benefiting from operating rules similar to those of an expert system.

[0059] In some embodiments, the central unit is configured to, in the absence of contrary commands from the user during a predetermined period, implement water distribution according to the optimization it has determined.

[0060] This avoids the risk that the user might, for example unintentionally, give no command, validation or invalidation of the optimized water distribution proposal.

[0061] According to a fourth aspect, the present invention relates to a method for collecting rainwater, characterized in that it comprises:

[0062] - a rainwater collection stage in a storage system, implementing a collection surface mounted on at least one moving element of a frame equipped with at least one actuator configured to move each said moving element between a so-called "folded" configuration, in which the collection surface is folded and has a first ground footprint, and a so-called "deployed" configuration, in which the collection surface is deployed and has a second ground footprint with a surface area greater than the surface area of ​​the first ground footprint,

[0063] - a step to determine if a water distribution instruction present in the storage system is present,

[0064] - a step to determine if at least one inhibiting factor applies to the requested water distribution,

[0065] - a step to determine an optimization of the quantity of water to be distributed based on the distribution instruction, each possible inhibiting factor, and a prediction of the future quantity of water collected by the device and future water needs, and

[0066] - a water distribution step according to the determined optimization, in the absence of a contrary order from the user during a predetermined period.

[0067] The advantages and purposes of this method being similar to those of the device which is the subject of the first or third aspect of the invention and of the method which is the subject of the second aspect of the invention, as briefly set out above, they are not recalled here.

[0068] BRIEF DESCRIPTION OF THE FIGURES

[0069] Other advantages, purposes and special features of the invention will become apparent from the following non-limiting description of at least one particular embodiment of the devices and methods that are the subject of the invention, with reference to the accompanying drawings, in which:

[0070] Figure 1 shows, in perspective, a first embodiment of the collection device that is the subject of the invention, in its folded configuration.

[0071] Figure 2 shows, in elevation view, the collection device illustrated in Figure 1, in its folded configuration.

[0072] Figure 3 shows, in elevation view, the collection device illustrated in Figure 1, in its deployed configuration.

[0073] Figure 4 shows, in perspective, a portion of the collection device of Figure 1. Figure 5 shows, in elevation, a second embodiment of the collection device of the invention, in its folded configuration.

[0074] Figure 6 shows, in perspective, the collection device illustrated in Figure 5, in its deployed configuration.

[0075] Figure 7 shows, in perspective, a third embodiment of the collection device that is the subject of the invention, in its folded configuration.

[0076] Figure 8 shows, in perspective, the collection device illustrated in Figure 7, in its deployed configuration.

[0077] Figure 9 shows, in perspective, a fourth embodiment of the collection device that is the subject of the invention, in its folded configuration.

[0078] Figure 10 shows, in perspective, the collection device illustrated in Figure 9, in its deployed configuration.

[0079] Figure 11 shows, in elevation, a control module for the deployment, retraction and water distribution of a device that is the subject of the invention. Figure 12 shows, in the form of a flowchart, a sequence of steps in a water collection process.

[0080] Figure 13 represents, in the form of a flowchart, a sequence of steps in a process for controlling the quality of stored water,

[0081] Figure 14 represents, in the form of a flowchart, a sequence of steps in a process for distributing stored water,

[0082] Figure 15 represents, in the form of temporal quantities, water collections, water reserves, distribution needs and an optimization of future distributions,

[0083] Figure 16 schematically represents a particular embodiment of a computer system implemented by certain embodiments of the device that is the subject of the invention, and Figure 17 schematically represents a particular embodiment of a computer system implemented by certain embodiments of the device that is the subject of the invention. DESCRIPTION OF EMBODIMENTS

[0084] The present description is given by way of non-limiting attribution, each feature of an embodiment being able to be advantageously combined with any other feature of any other embodiment.

[0085] It should be noted from the outset that figures 1 to 3, 5 to 10 are to scale, the other figures being schematic.

[0086] As can be understood from this description, various inventive concepts can be implemented by one or more of the processes or devices described below, several examples of which are provided herein. The actions or steps performed in implementing the process or device can be ordered in any appropriate manner. Consequently, it is possible to construct embodiments in which the actions or steps are performed in a different order than illustrated, which may include performing certain acts simultaneously, even if they are presented as sequential acts in the illustrated embodiments.

[0087] The expression "and / or," as used in this document, should be understood as meaning 'one or the other or both' of the elements thus joined, that is, elements that are present conjunctively in some cases and disjunctively in others. Multiple elements listed with "and / or" should be interpreted in the same way, that is, "one or more" of the elements thus joined. Other elements may also be present, other than those specifically identified by the "and / or" clause, whether or not they are related to those specifically identified elements.Thus, by way of non-limiting example, a reference to "A and / or B", when used in conjunction with an open language such as "including", may refer, in one embodiment, to A only (possibly including elements other than B); in another embodiment, to B only (possibly including elements other than A); in yet another embodiment, to A and B (possibly including other elements); etc.

[0088] As used herein, "or" is to be understood inclusively. As used herein, the expression "at least one," when referring to a list of one or more items, is to be understood as meaning at least one item chosen from one or more items in the list of items, but not necessarily including at least one of each item specifically listed in the list of items and not excluding any combination of items in the list of items. This definition also allows for the optional presence of items other than those specifically identified in the list of items to which the expression "at least one" refers, whether or not they are related to those specifically identified items.Thus, by way of non-limiting example, "at least one of A and B" (or, equivalently, "at least one of A or B", or, equivalently, "at least one of A and / or B") may refer, in one embodiment, to at least one, possibly including more than one, A, without B present (and possibly including elements other than B); in another embodiment, to at least one, possibly including more than one, B, without A present (and possibly including elements other than A); in yet another embodiment, to at least one, possibly including more than one, A, and at least one, possibly including more than one, B (and possibly including other elements); etc.

[0089] In this document, all transitional expressions such as "comprising", "including", "carrying", "having", "containing", "implying", "holding", "composed of", and others, should be understood as open, that is, as meaning including but not limited to. Only the transitional expressions "consisting of" and "consisting essentially of" should be understood as closed or semi-closed transitional expressions, respectively.

[0090] A "measured" value is a data point representing a measurement provided by at least one sensor of a physical quantity or calculated solely on the basis of such data and at least one calculation formula. A "measurable" value is a data point representing a measurement that can be provided by at least one sensor of a physical quantity or calculated solely on the basis of such data and at least one calculation formula.

[0091] A prediction of the future amount of water collected by the device is defined as at least one predicted value of the amount of water collected by the device in a future time interval. Preferably, this prediction includes at least two predicted values ​​for at least two future time intervals, these time intervals being different and preferably consecutive. The set of these intervals is called the water collection prediction period.

[0092] A prediction of water distribution needs downstream of a device is defined as at least one predicted value of the water needs to be distributed by the device downstream of the device within a future time interval. Preferably, this prediction includes at least two predicted values ​​for at least two future time intervals, these time intervals being different and preferably consecutive. The set of these intervals is called the water needs prediction period.

[0093] FIRST EMBODIMENT (FIGURES 1 TO 4) Figures 1 to 3 show a rainwater collection device 10, the subject of the invention, in a first embodiment. This device 10 comprises four feet 11, four upper peripheral cross members 12, four upper corners 13, two long upper central cross members 17, two short upper central cross members 19, four articulated arms 22, four lower peripheral cross members 25, four linear actuators 29, four linear actuator shaft support pieces 30, and four articulated arm support pieces 31. All these elements and the connectors that join them are preferably made of stainless steel.

[0094] The legs 11 are height-adjustable to maintain the vertical structure, as illustrated in Figures 1 to 3, by adapting their lengths to uneven or even sloping terrain. The upper peripheral cross members 12 are connected to each other and to the legs 11 by the upper corners 13, which have three right-angled connections. The two long central cross members 17 are connected at their ends to two upper peripheral cross members 12 by short connectors. The two short central cross members 19 are connected at their ends to the two long central cross members 17 by short connectors.

[0095] Each of the four articulated arms 22 is mounted on a pivoting T-joint, itself mounted either on a long upper central cross member 17 or on a short upper central cross member 19. The four lower peripheral cross members 25 are attached to the feet by hollow tube wedges. The linear actuators 29 are, for example, electric cylinders. Their lower end is mounted on a lower peripheral cross member 25 via a pivoting T-joint 30. Their upper end is mounted on an articulated arm 22 via a pivoting T-joint 30. All the pivoting T-joints provide pivot connections.

[0096] As illustrated in figure 3, a flexible, waterproof collection surface 32 is mounted on the arms 22.

[0097] The linear actuators 29 are configured to move these arms 22 between a folded configuration (illustrated in Figure 1) in which the collection surface 22 is folded and has a first ground footprint approximately equal to the area of ​​the square formed between the two short, upper central cross members 19, and an extended configuration (illustrated in Figure 3) in which the collection surface 22 is extended and has a second ground footprint larger than the first. In this extended configuration, the collection surface 22 forms a pyramidal funnel with its outlet at the bottom. Thus, rainwater can be collected over a large area.A system (not shown) for storing rainwater collected on the collection surface, for example a tank or cistern, is connected by a pipe 37 (see figure 4) to the outlet of a collector 35 located below the outlet of the collection surface 22.

[0098] In Figure 4, only some of the components of the device 10 are shown. For example, the feet 11 and the upper peripheral cross members 12 are not shown. Figure 4 shows, in the center of the collection surface 32, a casing 33, support beams 34 for the collector 35, and a filter 36 between the casing 33 and the collector 35. The casing 33 has no upper or lower wall. It allows the collected water to flow towards the filter 36. The support beams 34 create a space between the casing 33 and the filter 36, allowing the user to access and clean the filter 36. The filter 36, for example, consists of an upward-facing pyramidal grid. The collector 35 also has a pyramidal shape, but oriented downwards. The connection to the pipe 37 is located at the lower apex of the collector 35.Preferably, a second filtration stage (not shown), finer than the filtration performed by filter 36, is positioned between pipe 37 and the collected water storage system. Also preferably, a meter (not shown) measuring the quantity of collected water is positioned between pipe 37 and the storage system.

[0099] Thus, in the first specific embodiment illustrated in Figures 1 to 4, the rainwater collection device that is the subject of the invention comprises:

[0100] - a fixed tubular structure with four feet providing support for the device on the ground, - a mobile tubular structure composed of four arms articulated on this support and carrying a flexible, waterproof collection surface fixed to these arms and pierced in its center, ensuring the function of capturing rainwater,

[0101] - four linear actuators to configure the arms in deployed or folded position, and - a collector equipped with a filter in the central position of the fixed tubular structure to filter residues and convey the collected water to a collected water storage system.

[0102] Examples of technical manufacturing details for the first embodiment of the collection device are given below. The structure is made of galvanized steel tubes with a diameter of 42.4 millimeters. It consists of a lower fixed part and an upper movable part. The fixed part consists of four 200 cm long legs 11 anchored 50 cm into the ground. Eight 120 cm long crossbeams 12 and 25 form squares on two levels spaced 40 cm apart and connect the legs 11 using tube fittings. Two 120 cm long crossbeams 17 and two 30 cm long crossbeams 19 are fixed in the central part to form a 30x30 cm central square. The crossbeams and legs are interconnected by galvanized steel clamps.

[0103] Regarding the upper moving part, for arm mobility, four Type 125D adjustable knee joints need to be modified by replacing the joint pin with an M8 bolt and a washer with a nut and locknut to create a pivot joint. Four galvanized steel tubes, 42.4 mm in diameter and 300 cm long, form the arms 22 that hold the collection surface 32 in position. These four arms are drilled at their ends to allow the passage of retaining straps for the collection surface 32.

[0104] The power section consists of four linear motion actuators 29 with a maximum travel of 450 mm and a force of 1500 N. Each actuator 29 actuates an arm 22 of the moving part. The actuators 29 are fixed at their lower end to a lower peripheral cross member 25. The support for each actuator 29 is fixed to a single male terminal type 36D. A pin securing the actuator 29 to the support allows one degree of rotational freedom.

[0105] The attachment of an actuator 29 to the corresponding movable arm 22 is ensured by a part 31 which provides thrust directly above the movable arm 22. This part 31, or arm support, is in two parts and is rigidly held onto the arm 22 by means of two M8 bolts and two nuts. A pin connects the actuator to the arm support, also allowing a degree of rotational freedom.

[0106] The collection surface is made of PVC (polyvinyl chloride). In its deployed configuration, it takes the form of an inverted pyramid with four faces. It is therefore composed of four isosceles triangular sections with a base of 405 cm and a height of 217 cm. A 15 cm overlap between the triangular sections allows them to be welded together, leaving a sleeve for an arm 22 to pass between the two overlapping surfaces. The triangular sections are welded using a hot air gun at 250°C. The arm 22 is positioned during welding to precisely calibrate its degree of freedom. Once the four triangular sections are welded together, a 25 cm square cutout is made in the center of the collection surface 32. Tarpaulin eyelets are installed on each free end of an arm 22. The collection surface has an outlet for the collected water opposite the housing 33.The ground support is square in shape with four feet, designed to be securely anchored to the ground.

[0107] To facilitate the installation of the collection surface 32, it is positioned on the fixed part and at its center. Each arm 22 is then inserted into the sleeve formed on each edge of the pyramid that constitutes the collection surface. The arm 22 is then fixed to the fixed part of the device, and the arm support is then put in place. Once the four arms are mounted and the actuators are pinned, the bungee cords are attached to the eyelets of the collection surface and to the holes in the arms 22 at their free ends.

[0108] At the center of the device is a system that directs the rainwater collected on the collection surface towards pipe 37 and filters out coarse streams. This collector consists of three parts: the upper casing 33, the lower collector 35, and a support for the lower collector made up of longitudinal members 34. The lower collector 35 is itself in two parts and allows for the insertion of a textile, for example a net, between them, which acts as a filter.

[0109] SECOND METHOD OF IMPLEMENTATION (FIGURES 5 AND 6)

[0110] Figures 5 and 6 show a rainwater harvesting device 40, the subject of the invention, in a second embodiment. This device 40 comprises a vertical support 41, supporting two pivot joints 43 and 44, each of which includes a rotary actuator (not shown). These rotary actuators are synchronized. Each rotary actuator is, for example, composed of an electric motor mounted on a disk fixed to the vertical support 41, which drives a gear comprising a toothed ring. This gear is followed by reduction gears to drive five disks at angular speeds equal to the first five multiples of a predetermined angular speed. In another embodiment, five synchronized stepper motors drive the five moving disks relative to a fixed disk, with gears configured to drive these five moving disks at angular speeds equal to the first five multiples of a predetermined angular speed.The vertical support 41 is, for example, a mast, a longitudinal shaft or a post.

[0111] The lower pivot joint 43 supports seven arms 42 inclined so that their free end is higher than their end connected to the pivot joint 43. The first arm 42 is fixed relative to the vertical support 41. Each of the other six arms 42 is attached to one of the disks driven by an electric motor. The upper pivot joint 44 supports seven reinforcements 45 connected to the arms 42. The first reinforcement 45 is fixed relative to the vertical support 41 and supports the fixed arm 42. Each of the other reinforcements 45 is attached to one of the disks driven by an electric motor and to a movable arm 42 mounted on a disk that can rotate at the same angular speed.

[0112] In Figure 5, the collection device 40 is in its folded configuration, with the arms 42 pressed tightly together. In Figure 6, the collection device 40 is in its deployed configuration, with the arms 42 spread apart to form a hexagon. A collection surface 46 is supported by the arms 42. In the deployed configuration, the collection surface thus takes the form of an inverted pyramid with six faces, the apex of which is located in the vertical support 41.

[0113] Preferably, the first and last arms 42 have a smaller angle with the upward vertical than the other arms 42. Thus, the ends of the first and last arms 42 furthest from the vertical support 41 are higher than the ends of the other arms 42. Rainwater falling on the collection surface near these first and last arms 42 is therefore carried away from these arms by gravity. Alternatively, a gutter (not shown) is positioned below the junction between the first and last arms 42 and connects to the central collection box (not shown). Alternatively, the angles formed between successive arms 42 and the upward vertical are strictly increasing or decreasing, with such a gutter positioned below the arm with the largest angle (which is therefore either the first or the last arm 42).Alternatively, to connect one end of the collection surface 46 to the other and form a watertight seal on the edge of the pyramid located on the fixed arm, a magnetic connection can be provided, similarly to the magnetic seals of some shower cubicle doors.

[0114] As can be understood from reading the description above, the collection device 40 has similar advantages to those of the collection device 10.

[0115] THIRD METHOD OF REALIZATION (FIGURES 7 AND 8)

[0116] Figures 7 and 8 show a rainwater collection device 50, the subject of the invention, in a third embodiment. This device 50 comprises a shaft 51, supporting three pivot joints 53, 54 and 57 around the axis of the shaft 51, and two rotary actuators 59 symmetrical with respect to the axis of the shaft 51.

[0117] The lower pivot joint 53 supports six arms 52 angled so that their free end is higher than their end connected to the pivot joint 53. Two arms 52 are fixed relative to the shaft 51. Each of the other four arms 52 is mounted on a sliding joint that slides on a toroidal support of the lower pivot joint 53. The intermediate pivot joint 54 supports six reinforcements 55, each connected to an intermediate portion of one of the arms 52. Two reinforcements 55 are fixed relative to the shaft 51 and each support a fixed arm 52. Each of the other reinforcements 55 is mounted on a sliding joint that slides on a toroidal support of the intermediate pivot joint 54. The upper pivot joint 57 supports six reinforcements 58, each connected to a free end of one of the arms 52. Two reinforcements 58 are fixed relative to the shaft 51 and each support the free end of a fixed arm 52.Each of the other reinforcements 58 is mounted on a sliding link on a toroidal support of the upper pivot link 57.

[0118] The rotary actuators 59 are, in the device shown in figures 7 and 8, electric motors equipped with pulleys which pull on the movable arms 52 to deploy the collection surface 56 or fold it back.

[0119] In Figure 7, the collection device 50 is in its folded configuration, in which the arms 52 are pressed tightly together in two groups of three. In Figure 8, the collection device 50 is in its deployed configuration, in which the arms 52 are spread apart to form a hexagon. In this deployed configuration, the collection surface 59 thus takes the form of an inverted pyramid with six faces, the apex of which is located in the shaft 51.The variants described with regard to figures 5 and 6 apply equally to the collection device 50, in particular concerning the angles of the first and last arms 52 of each group of three arms, with regard to the angles of the intermediate arms, the gutters below the junction between a first arm 52 of one group and the last arm 52 of the other group, the helical shapes of the two half-collection surfaces in their deployed configuration and the magnetic links between the first and last arms 52 of the two groups of arms 52.

[0120] As can be understood from reading the description above, the collection device 50 has similar advantages to those of the collection device 40.

[0121] FOURTH METHOD OF IMPLEMENTATION (FIGURES 9 AND 10)

[0122] Figures 9 and 10 show a rainwater collection device 60, the subject of the invention, in a fourth embodiment. This device 60 comprises a vertical support 61, supporting a collector 69, three elbows 64 forming three pivot joints with horizontal axes of rotation, shown in dashed lines, and two rotary actuators 67 and 68.

[0123] Each elbow 64 supports a rotating telescopic arm 62 comprising a movable extension set in motion as described below. Each arm 62 carries an attachment 63 for a cable 65 set in motion by one of the rotary actuators 67 or 68, via a pulley 66 mounted on the vertical support 61.

[0124] A system of cables, pulleys, and stepper motors enables two movements. The first movement involves extending the telescopic arms 62 away from the axis of the vertical support 61. This is achieved by unwinding the cables 65, driven by the actuator 67 and guided by the upper pulleys 66. The weight of the arms 62 and the collection surface causes them to descend into their deployed position. Tension on the cables 65, exerted by the motor 67, causes the arms 62 to retract towards the axis of the vertical support 51. The second movement involves extending the telescopic arms 62. This is achieved by a system of springs (not shown) mounted in compression in the lower part of the arms 62 and bearing against the upper sliding part of these arms 62. Drive cables, driven by the actuator 68, allow these springs to be released or, conversely, compressed.The extension of arms 62 is achieved by the pressure of the internal springs within arms 62 when actuator 68 unwinds the three corresponding cables. The retraction of arms 62 occurs when actuator 68 rewinds these three cables. The internal springs within arms 62 are then compressed.

[0125] In Figure 9, the collection device 60 is in its folded configuration, in which the arms 62 are pressed against the vertical support 61 and their telescopic extensions are retracted. In Figure 10, the collection device 60 is in its deployed configuration, in which the arms 62 are spread apart to form a triangle and their telescopic extensions are extended. In this deployed configuration, the collection surface thus takes the form of an inverted pyramid with three faces, the apex of which is located in the vertical support 61, above the collector 69.

[0126] As can be understood from reading the description above, the collection device 50 has similar advantages to those of the collection device 10.

[0127] FIFTH METHOD OF IMPLEMENTATION

[0128] This embodiment is placed on an oya, a terracotta water reservoir with a neck for embedding the structure of the device. This water reservoir is preferably placed in the ground, the water slowly percolating through its porous wall to hydrate the soil near plants or trees. Alternatively, the oya is replaced by a water tank.

[0129] VARIANTS OF THE FIRST FIVE METHODS OF IMPLEMENTATION

[0130] Numerous variations of the embodiments described above can be implemented to achieve the same rainwater harvesting function by deploying a collection surface. In particular:

[0131] A / The deployed collection surface can be of any inverted pyramidal shape with a polygonal base, circular or elliptical shape, or even any shape provided that its central part is lower than its peripheral parts.

[0132] B / The mechanism for deploying and folding the collection surface can also be central (like that of a parasol), telescopic, central and telescopic, by unrolling the collection surface, by rotating deployment like a fan, or rotating and telescopic, for example.

[0133] C / The materials of the fixed structure and the moving part can also be rigid plastics, wood, or metals other than stainless or galvanized steel. D / The frame, which can be combined with the collection surface, can be inflatable and deployed or retracted by an actuator comprising a pressurized air source, for example, supplied by an air pump, and a pressurized air outlet, for example, a solenoid valve. Of course, an air pump operating in compression and suction can ensure the deployment and retraction of the frame and the collection surface. E / The flexible collection surface can also be made of coated canvas or sailcloth, for example. The collection surface may also be made of a rigid material, for example, hinged panels.

[0134] F / As described above, the collection surface may include at least one heating element to melt the snow on its upper surface.

[0135] Alternatively, a dew collection device, for example a net, is positioned above the rainwater collection surface to capture dew and / or frost.

[0136] CONTROL SENSORS FOR THE DEPLOYMENT AND FOLDING OF THE COLLECTION SURFACE

[0137] As shown in Figure 11, in the case of an isolated device, a deployment and retraction control module 70 for the collection device of the invention comprises at least one central unit 71 equipped with memory 72 for software operation, input and output ports 73, an electrical power source 74, and a power module 75 to supply the device's actuators. Preferably, the collection device of the invention is equipped with sensors. Alternatively or complementarily, the module 70 comprises a means 76 for accessing meteorological data from a remote database 38 updated by a meteorological service, watering or irrigation recommendations from an agronomic recommendation service, and / or a model 39 trained by machine learning on measured data.This model 39 provides predictions of future rainwater harvesting and / or future water needs distributed by the device 70, for example, a trained model as described opposite Figures 16 and 17. The meteorological data includes at least weather forecasts and, optionally, past measured environmental data and / or past provided weather forecasts. This data can be stored on computer servers, and these services can be implemented on remote servers that run expert systems and / or a machine learning-trained model on measured data.

[0138] Preferably, among the sensors, the device includes a rain sensor 77, an anemometer 78, and a temperature sensor 79, and optionally, a wind vane 88 indicating wind direction and / or a soil moisture sensor. These environmental physical quantity sensors are mounted on a mast 80 anchored to the ground, except for the moisture sensor, which penetrates the soil. Optionally, the device also includes a means of measuring electrical autonomy 81, a water volume meter 82 at the inlet of the collected water storage system 89, a flow meter 83 at the outlet 47 of the storage system 89, a dew point sensor 84, a hydrogen potential (pH) sensor 85 of the water being collected and / or of the water in the storage system 89, a turbidity sensor 86 (or turbidimeter) of the water in the storage system 89 and / or a sensor 87 of water activity and / or bacteriological activity of the water in the storage system 89.

[0139] The central processing unit 71, equipped with memory 72 and ports 73, can be a generic microcontroller board or computer. The electrical power source 74 includes a battery and a renewable energy source, for example, a photovoltaic panel or a wind turbine. The power module 75 includes electromechanical relays for distributing electrical power from a signal emitted by the central processing unit 71.

[0140] Method 76 for accessing meteorological data implements a remote communication protocol, such as a mobile phone protocol. It can also implement a communication protocol over a wireless network, enabling access to the web and / or the internet, such as the WiFi® protocol or a low-power radio protocol for the Internet of Things (IoT), such as LoRaWAN® (Long Range Wide Area Network). Depending on the protocol, an additional card is integrated into the system for its management.

[0141] Preferably, through such remote communication, the user can view the captured values ​​and / or statistics derived from their processing, and / or control the retraction or deployment of the rainwater harvesting system. The configuration and control of the device by the user may differ depending on the device model and the user type (professional or individual). Local control can be achieved via a control panel (not shown) or via a software application installed on a mobile device, such as a smartphone or tablet, or on a personal computer.

[0142] The electrical autonomy measuring device 81 is connected to the battery and, based on the voltage across its terminals and its discharge curve, provides an indication of the amount of energy available in the battery. Measuring the energy autonomy helps prevent power shortages and allows the collection device to be put into a safe mode, for example, in a folded configuration. The water volume meter 82, at the inlet of the storage system 89, can be positioned at any point along the path of the collected water between the collection surface and the storage system 89.

[0143] The bacteriological sensor 87 for the water in the storage system 89 is only necessary if the distributed water is intended for animal watering. The wind vane 88, indicating wind direction, is useful when the collection surface is asymmetrical or when the collection device presents different risks of tipping depending on the wind direction. The wind vane 88 can also be used to control the inclination of a symmetrical collection surface to optimize this inclination according to the wind, for example, to collect a greater quantity of rainwater. The storage system 89 can be a flexible tank, for example. The variety of sizes and capacities of this type of storage, its ease of implementation, and its low cost make it suitable for all types of applications, even for large professional, agricultural, or industrial water needs.The water storage system 89 is connected, at its outlet 47, to a standard pipe fitting 49 for distribution or irrigation leading to an activity area 48 (crop, watering trough, swimming pool ...) where the distributed water is used.

[0144] The 80 mm mast supports the meteorological sensors on an elevated mount, for example, at a height of 1.60 to 1.80 meters, preferably at the same height as the top of the rainwater collection device. At eye level, a waterproof enclosure (not shown) houses the electronic components. When using multiple collection devices on the same site, the control and measurement system can be shared by all the devices to reduce implementation and maintenance costs.

[0145] When using multiple devices on the same site, only one set of input sensors is required. The network of devices then operates in a master / slave configuration. The master is the device containing the input sensors. It transmits data to the other devices, which then function as slaves with this data. Note that this transmitted data can be limited to instructions to replicate the operation of the master device, provided the slave devices are identical to it.

[0146] OPERATION OF THE RAINWATER COLLECTION AND DISTRIBUTION DEVICE The process of the invention preferably comprises a method for controlling the collection of rainwater (see figure 12), a method for monitoring the quality of the stored water (see figure 13) and a method for distributing the stored water (see figure 14).

[0147] Figure 12 shows a flowchart 90 of the control steps for the rainwater harvesting device of the invention. The device being initially in the folded configuration 91, during a step 92, the central unit 71 determines whether it is raining, the temperature is above freezing, and the wind is calm or light. The wind gust speed measured by the anemometer 78 is then compared to a predetermined limit value above which there is a risk of the harvesting device tipping over or of damage to one of its components, particularly the collection surface. For example, this limit value is 50 km / h. If any of these three conditions is not met, the device remains in the folded configuration and the central unit 71 repeats step 92 periodically. If all three conditions are met, during a step 93, the device extends into its deployed configuration and collects rainwater.Alternatively, step 92 also determines whether the collected water storage system is already full and contains water of suitable quality. If both criteria are met, during step 93, the device remains in the folded configuration.

[0148] Then, during step 94, the central unit 71 determines whether at least one of the following conditions is met: no rain, zero or sub-zero temperature, strong wind (i.e., with gusts of speed equal to or greater than the predetermined limit described above), or a full water storage system. If one of these conditions is met, the central unit 71 returns the device to its folded configuration and repeats step 92 periodically. Otherwise, the device remains in its deployed configuration, and the central unit 71 repeats step 94 periodically. Alternatively, the device includes a heated collection surface for melting snow. In this case, if the temperature is zero or very slightly sub-zero (e.g., above -5 °C), the device either enters or remains in its deployed configuration during snowfall.

[0149] Alternatively, the device also assumes a folded configuration when the storage system is full or when the rain carries sand or hail, this information being obtained by consulting a database from a weather service provider. Alternatively, the device also assumes a folded configuration when the available electrical power in the battery is below a predetermined limit.

[0150] As an alternative to steps 92 and 94 described above, in a cultivation setting, the deployment of the collection surface can be controlled to prevent diseases related to humidity or the presence of water, such as fungal diseases (downy mildew, powdery mildew, gray mold, rust, and anthracnose), bacterial diseases (bacterial rot and soft rot), or to prevent risks related to excessive sunlight, such as leaf burn or sunscald, fruit sunburn, heat stress, and blossom-end rot. Alternatively, in a livestock setting, the shaded area created by the collection surface can serve as a refuge for animals in case of excessive sunlight. Alternatively, in an urban setting, the shaded area created by the collection surface can serve as a refuge for people in case of excessive sunlight.

[0151] The collection surface can thus be deployed to provide shade for crops, animals, or people, or to shade sensitive storage areas such as harvested fruits or vegetables. Combined with a scheduled calendar and / or a weather information service, this mode allows for fine-tuning deployment times based on sunlight conditions for effective sun exposure management. For example, a mobile application can be used to control the configuration of the collection surface and define the default operating mode: retracted, deployed, or fully open for a configurable number of hours on the sunniest days, according to seasonal or monthly time ranges.

[0152] Figure 13 represents a flowchart 100 of steps for monitoring the quality of the water stored in the storage system. This water being initially of suitable quality for its intended use (step 101), during step 102, the central unit 71 determines whether a decline in the quality of the stored water has been detected. For example, this detection is based on the values ​​captured by the pH sensor 85 of the water in the storage system 89, the turbidity sensor 86 (or turbidimeter) of the water in the storage system 89, or the bacteriological sensor 87 of the water in the storage system 89. Each of these measurements is then compared to a predetermined limit value beyond which the water is no longer of suitable quality for its intended use.For example, water intended for animal drinking must meet stricter requirements than water intended for a swimming pool, which itself meets stricter requirements than water intended for irrigation.

[0153] If none of the measurements exceed their corresponding limit value, the central unit 71 returns to step 101 before periodically repeating step 102. Conversely, if one of the measurements exceeds its corresponding limit value during step 103, the central unit 71 sends an alert to the user and, if automatic water treatment is planned for this case, the central unit 71 initiates this treatment. For example, a treatment similar to that applied to swimming pool water is applied to maintain a neutral pH. Then, during step 104, the central unit 71 determines whether the quality of the stored water has been restored. If so, the central unit 71 returns to step 101 before periodically repeating step 102. Otherwise, the central unit 71 returns to step 103 before periodically repeating step 104.

[0154] Figure 14 shows a flowchart 110 of steps for distributing water stored in the storage system. This water is initially stored in the storage system, step 111. During a step 112, the central unit 71 determines whether a water distribution instruction (including water discharge) is present. For example:

[0155] - Irreparable deterioration of water quality rendering it unfit for any use, - a warning of an imminent drought received from a meteorological information service,

[0156] - the arrival of a scheduled irrigation time or an irrigation recommendation for a crop received from a crop monitoring service or a soil moisture sensor,

[0157] - a drop in the water level of a swimming pool,

[0158] - a drop in water level in a drinking trough,

[0159] These are instructions for water distribution, respectively for discharge, irrigation, filling a swimming pool, and animal watering.

[0160] During step 113, the central unit 71 determines whether an inhibition factor applies to the requested water dispensing. For example, in the dispensing instruction examples described above, the following are inhibition factors for the requested water dispensing:

[0161] - a ban on the discharge of polluted water,

[0162] - an insufficient quantity of water in the storage system,

[0163] - Information indicating the presence of a disease that may develop if water is distributed, received from a crop monitoring sensor,

[0164] - a ban on filling swimming pools or stored water of unsuitable quality for filling a swimming pool and

[0165] - a quality of stored water unsuitable for animal watering.

[0166] During step 114, the central unit 71 obtains a prediction of at least one future quantity of water collected by the device and a prediction of at least one future water distribution requirement by the device. The prediction of the future quantity of water collected by the device is based on a forecast of future rainfall on the device's collection area. The quantities deduced from this forecast, by multiplying the rainfall by the collection area, can nevertheless be reduced to take into account:

[0167] periods of shrinkage of the collection surface, typically during periods of high wind, snow or frost,

[0168] periods during which the water storage system is predicted to be full and

[0169] from water evaporation on the collection surface and

[0170] past errors in regional forecasts for the site where the device is located. Thus, the prediction of the future quantity of water collected by the device makes it possible to determine a prediction of the quantity of water available in the future within the water storage system.

[0171] Similarly, predicting future needs may depend on predicting future rainfall over the device's water distribution area. For example, this could be the area of ​​a swimming pool, garden, or plot of land into which the device distributes water, possibly via distribution or irrigation channels or pipes.

[0172] To determine the prediction of future water collection volume by the device and future needs, the central unit 71 can implement a machine learning-trained model on measured data including past rainfall, quantities of water distributed during past water distributions, and captured data representative of past physical quantities. This data can be dated, possibly with time accuracy.

[0173] Similarly, to obtain at least one of these predictions, the central unit can implement a machine learning-trained model based on measured data, predictions provided by this model in the past, and / or water distribution commands given by the user in the past. This model preferentially continues its learning during each water distribution by the device or by a similar device preferably located near the device implementing this trained model, for example, on the same plot of land.

[0174] Figure 15 represents, as time-domain quantities, successively from top to bottom, water collection, water reserves, distribution needs, and an optimization of future distributions. Time is represented by days on the x-axis, and water quantities on the y-axis. The present is represented by a dashed vertical line. The data shown to the left of this line are water collection measurements taken by the device. The data shown to the right of this line are predictions or calculations, which can be performed locally by the central processing unit and / or remotely by a server configured to perform them and accessible, via a data network, by the device.

[0175] The first time-domain diagram, at the top of Figure 15, represents past meteorological data and data for a predetermined future forecast period (one week, in Figure 15). It shows the daily quantities of water that can be collected on the collection area, according to the meteorological data (i.e., rainfall multiplied by the collection area, with a maximum equal to the water retention capacity of the storage system), taking into account the periods of shrinkage of this area resulting from the implementation of process 90, assuming the temperature and wind speed are as predicted by the meteorological data. It should be noted that this meteorological data is preferably provided by a web-based meteorological service.Alternatively, this meteorological data is calculated by the device's central unit based on environmental data captured by the device's sensors (e.g., atmospheric pressure, wind speed, and sunlight sensors). The second time diagram represents the measured amounts of water actually collected in the past and the predicted amounts of water collected during the prediction period. In the simplest version of this prediction, no meteorological data is used. For example, the device's central unit has a climate model of daily rainfall throughout the year and adjusts this model based on the observed difference between the amount of water collected based on the model data and the amount actually collected by the device.

[0176] In a first embodiment, the prediction of at least a future quantity of water collected by the device is determined by implementing a machine learning model on measured or acquired data representative of past environmental data, including at least past rainfall values ​​and quantities of water collected in the past. This environmental data can be obtained locally through the implementation of environmental data sensors or a meteorological service.

[0177] Such an embodiment implements, for example, a time series prediction machine learning model, which may correspond to:

[0178] an ARIMA (AutoRegressive Integrated Moving Average) type model and / or variants (SARI MA, adapted to seasonal data, ARIMAX, adapted to the incorporation of exogenous data) of this model,

[0179] a model based on neural networks, of the RNN (Recurrent Neural Networks) type, and preferably of the LSTM (Long Short-Term Memory) type, and GRUs (Gated Recurrent Units),

[0180] an attention-based model, such as a Transformer adapted to time series (e.g., Temporal Fusion Transformer, or Informer, where

[0181] a hybrid model combining models briefly described above.

[0182] In a preferred embodiment, the quantities of water collected during the prediction period are predicted based on meteorological data, for example, in the form of the quantities shown in the first time diagram. Thus, preferably, past rainfall values ​​include future rainfall values ​​provided in the past by a meteorological service. The machine learning-trained model therefore accounts for past differences between the predicted quantities of rainwater collected from meteorological data (rainfall and, possibly, wind speed, temperature, humidity, and / or sunshine) provided by the meteorological service and the quantities actually collected by the device.

[0183] In an even more preferred embodiment, the device includes at least one sensor measuring an environmental physical quantity, and the environmental data used by the trained model includes data provided by at least one of these sensors. The machine learning-trained model thus enables the creation of a micrometeorological forecast for the site where the device is installed. This micrometeorological forecast takes into account past differences between meteorological data (rainfall and, possibly, wind speed, temperature, and / or sunshine) provided by the meteorological service and the environmental values ​​actually measured by the device's sensors. Then, based on this micrometeorological forecast, the central unit calculates the prediction of the quantities of water that will be collected daily during the forecast period.Alternatively, the trained model directly provides the prediction of the quantities of water that will be collected daily during the prediction period.

[0184] In other embodiments, the results provided by the trained model are supplemented or replaced by the results of an expert system execution. For example, this expert system implements rules for adjusting calculations of the quantities of water to be collected, taking into account the configuration of the site where the device is located, such as its altitude, slope, woodland cover, prevailing wind directions, distance from terrain features, watercourses, hills, mountains, wind corridors, local rainfall measurements, etc.For these embodiments, the central unit includes a means for executing an expert system configured to determine said prediction of at least one future quantity of water collected by the device based on stored data representative of past environmental data, including at least past rainfall values ​​and quantities of water collected in the past.

[0185] Such an expert system can implement rules defined by an expert user, via a graphical user interface, or by a third-party computer program, via an application programming interface. In some variations, such rules can be dynamically adjusted by the device based on captured values ​​or by a third-party user or computer program.

[0186] The third diagram in Figure 15 represents the maximum daily water quantities available in the water storage system; that is, the sum of the current water quantity available in the storage system and the predicted quantities of water collected during the forecast period. This third diagram also represents the maximum water retention capacity of this storage system.

[0187] The fourth diagram in Figure 15 represents predictions of daily water distribution needs downstream of the device. These needs are calculated, for example, as the difference between water consumption (evaporation from the surface of the water in a swimming pool, pond, or watering trough; evapotranspiration from plants; consumption by animals; diffusion into the soil from a crop, etc.) and the direct water input in the distribution area, primarily from rain (on the water surface, on the soil and plants, etc.) and, optionally, from a water source other than rain and the device (a well with a pump, an irrigation canal, etc.). This calculation of predicted daily needs can thus be performed without local measurements. However, ideally, the device includes at least one water availability sensor downstream of its water outlet. For a swimming pool or watering trough, this measurement could be the water level.For a garden or agricultural plot, this sensor can be a soil moisture level sensor.

[0188] Preferably, at least one environmental physical value sensor is also implemented. Regarding evapotranspiration, it is calculated, for example, according to the Penman-Monteith equation, based on data from sensors such as:

[0189] a capacitive soil moisture sensor measuring the soil's water content, a tensiometer measuring the water tension in the soil and indicating how easily roots can absorb water,

[0190] a TDR (Time Domain Reflectometry) or FD (Frequency Domain) sensor measuring the volumetric water content of the soil,

[0191] a water level sensor in a downstream installation,

[0192] a leaf water potential sensor measuring water tension in leaves, indicating plant water stress,

[0193] a dendrometer measuring variations in trunk or stem diameter, a leaf temperature sensor measuring leaf temperature, an electrical conductivity (EC) sensor measuring water salinity, a potential hydrogen (pH) sensor, and / or

[0194] a turbidity sensor measuring water clarity.

[0195] The result of past predictions of water distribution needs is shown, to the left of the line indicating the present, in the fourth diagram of Figure 15.

[0196] Of course, all processing (machine learning model training, expert system execution, reserve calculations, needs predictions and distribution optimization) can be updated regularly, preferably at least once a day.

[0197] Optionally, during successive stages, the weather predictions are assigned a label representing a confidence index in the predicted values.

[0198] As can be understood from the description in Figure 15, the present invention applies to different degrees of complexity:

[0199] - in simpler embodiments, the device only includes a flow meter to measure the quantity of water collected;

[0200] - in simple embodiments, the water collection prediction device or its server also accesses meteorological data;

[0201] - in more elaborate embodiments, the device also includes sensors for environmental physical quantities, such as rain detection, wind measurement, temperature and atmospheric humidity; - in servo-controlled embodiments, the device includes at least one sensor for distribution efficiency, for example, a water level in a swimming pool or drinking trough, or a soil moisture level in the area of ​​water distribution by the device.

[0202] More complex embodiments are described below, implementing a machine learning-trained model on at least some measured data.

[0203] AUTOMATIC LEARNING

[0204] The input data, preferably dated or even time-stamped, used by the machine learning-trained model includes all or part of the following data:

[0205] Fixed data representing the activity for which the water is distributed:

[0206] o Type of activity: cultivation, watering of animals, maintenance of swimming pool level, air cooling ...;

[0207] o Activity area: cultivated area, swimming pool area, watering troughs...

[0208] o For a crop: identification of cultivated plants) or gardening (with planting type), soil texture and / or retention capacity, plant development and / or health sensors, such as those described above for calculating evapotranspiration,

[0209] o For watering: identification of animals with the number of livestock for each animal raised;

[0210] Physical state of the activity for which water is distributed, before water distribution (and, optionally, between two water distributions):

[0211] o Volumetric moisture (capacity probe) which measures the moisture content or proportion of water in the soil,

[0212] Optionally, the state of plant development and health,

[0213] o Water level in drinking troughs or a swimming pool, respectively,

[0214] o Tensiometric probe, which measures the suction force that the plant must exert to extract water from the soil;

[0215] Weather data provided by a meteorological service:

[0216] o Precipitation,

[0217] o Sunshine,

[0218] Temperature,

[0219] o Humidity level in the air,

[0220] o Wind direction,

[0221] o Average and maximum wind speed,

[0222] o Atmospheric pressure;

[0223] Past environmental physical measurements from the device's sensors,

[0224] o Precipitation (rain gauge),

[0225] o Sunlight (pyranometer and / or UV probe),

[0226] o Temperature (thermometer), o Humidity level in the air (hygrometer)

[0227] o Wind direction (weather vane),

[0228] o Average and maximum wind speed (anemometer),

[0229] o Atmospheric pressure (barometer),

[0230] The measurements obtained by the device's operating sensors, o Quantity of water collected,

[0231] o Quantity of water stored,

[0232] o Quantity of water distributed,

[0233] o Measurement of electrical autonomy,

[0234] o Dew point measurement,

[0235] o Measurement of the pH of the stored water,

[0236] o Measurement of turbidity of stored water,

[0237] o Measurement of the activity of stored water,

[0238] o Measurement of the bacteriological activity of water;

[0239] Instructions regarding water distribution and the quantity of water to be distributed, received from the user or an expert, either directly or through a computer application or a crop irrigation recommendation service; and

[0240] Possibly, the availability of other water resources:

[0241] o Presence of an announcement of water distribution network power outages,

[0242] o Groundwater level or water retention basins.

[0243] Based on all or part of this data, updated at least during each water distribution, the machine learning-trained model determines a prediction of the future quantity of water collected by the device and / or a prediction of future water needs for the activity concerned.

[0244] The model training step is implemented, for example, by a learning method such as an electronic computing circuit associated with a database containing at least data on the quantities of water collected and the quantities of water distributed. In one embodiment, such a learning method is configured to perform the following steps:

[0245] - the loading of this data,

[0246] - scaling up this data to facilitate model learning,

[0247] - preparing a dataset for supervised learning,

[0248] - the creation of an LSTM model (acronym for Long Short Term Memory),

[0249] - the separation of the dataset into two subsets, a training dataset and a test dataset,

[0250] - learning the model, Tl

[0251] - the comparison between a prediction made from the training performed on the training dataset and the actual values ​​of the test dataset and

[0252] - a step to calculate performance metrics for the model.

[0253] Below, the LSTM model, Adam's parameter optimization algorithm and the associated recurring neural network (RNN) model are briefly presented.

[0254] The traditional neural network model goes from the input layer to the hidden layer to the output layer. The layers are fully connected, and the nodes between each layer are disconnected. Consequently, ordinary neural networks have certain limitations in solving time series. The recurrent neural network overcomes the shortcomings of traditional neural networks. Its operating formula is shown below, where X represents the value of the input layer, S represents the value of the hidden layer, U is the weighting matrix from the input layer to the hidden layer, Q represents the value of the output layer, V is the weighting matrix from the hidden layer to the output layer, and W is the last value of the hidden layer as an input at that time. The operating formula of the model is as follows:

[0255] [Formula 1]

[0256] {Q = ^(7S t ) S t = / ([7Xt + PVS t-1 )

[0257] Although the RNN can efficiently handle nonlinear time, this algorithm cannot handle time series with excessive delay due to gradient vanishing and gradient explosion. The LSTM is an enhanced cyclic neural network that solves the problems that the RNN cannot handle, namely long-range dependencies. In this LSTM model, ft, it, Ct, and Ot are the forget gate, the input gate, the output gate, and the time of the output gate, respectively, and W, b, and tanh are the corresponding weights, gaps, and excitation, respectively. The forget gate determines how much of the previous unit state Cti is maintained at the current time Ct; the input gate determines how much of the network's input Xt is recorded in the unit state Ct at the current time and how much of the output gate's control unit state is passed to the LSTM. The current output value is ht.Its calculation formula is indicated by the following equations:

[0258] - The equation for the forgetting gate ft is:

[0259] [Formula 2]

[0260]

[0261] - The equation for the output gate Ot is:

[0262] [Formula 3]

[0263]

[0264] The LSTM model can be divided into three parts: the input layer, the hidden layer, and the output layer. The input layer is primarily used for preprocessing and splitting the original dataset. The hidden layer is trained on the training dataset. Using the Adam optimizer, described above, the parameters are optimized, and the model is optimized with minimal loss as the guiding principle. The output layer predicts the data according to the model learned in the hidden layer and performs data restoration for scaling the previous data.

[0265] This learning stage takes place, for example, on a computer server that is remote from the rainwater collection device that performed the data collection.

[0266] Learning the model based on the data allows it to provide predictions of at least one probable future quantity of water collected and / or at least one probable future need for distributed water, predictions on the basis of which the central unit can calculate an optimization of water distribution in the future.

[0267] As an alternative or complement to the implementation of the trained model, obtaining at least one prediction can be done using parameterized or programmed rules, for example in the form of an expert prediction system.

[0268] FIRST EXAMPLE: WATER DISTRIBUTION TO MAINTAIN A PREDETERMINED WATER LEVEL IN A SWIMMING POOL.

[0269] The input data, preferably dated or even time-stamped, of the machine learning model trained on at least some measured data preferentially includes data from the following first group:

[0270] Physical state of the object of this activity:

[0271] o Water level in the pool;

[0272] Past and present weather data:

[0273] o Forecast rainfall,

[0274] o Sunshine,

[0275] o Air temperature,

[0276] o Wind speed,

[0277] o Maximum wind gust speed;

[0278] The measurements obtained by the device's operating sensors, o Quantity of water collected,

[0279] o Quantity of water stored,

[0280] o Quantity of water distributed,

[0281] o Measurement of turbidity of stored water,

[0282] o Measurement of the activity of stored water.

[0283] The input data may also include data from the following second group:

[0284] Past and present weather data:

[0285] o Air humidity level, o Wind direction;

[0286] Past environmental physical measurements from the device's sensors,

[0287] o Rain detector and rainfall measurement,

[0288] o Sunshine,

[0289] o Wind speeds measured by an anemometer,

[0290] o Air temperature;

[0291] The measurements obtained by the device's operating sensors, o pH measurement of the stored water;

[0292] The availability of other water resources:

[0293] o Presence of an announcement of water distribution network supply cuts.

[0294] Using data from the first group and, preferably, at least one data point from the second group, a prediction of the quantity of water collected in the future and a prediction of future water distribution needs are made, preferably by implementing a machine learning-trained model. Alternatively or complementaryly, these predictions implement an expert system and / or a model of water loss in a swimming pool (evaporation as a function of air temperature, sunlight, air humidity, and wind, possibly supplemented by a measurement or model of the water temperature in the pool, this temperature being used by the evaporation model).

[0295] Then, a prediction (or proposal) of the quantities of water to be distributed in the future is made. Preferably, this prediction of the quantities of water to be distributed takes into account at least one operating rule of the device, for example:

[0296] Provide, as a priority, a predictive alert indicating the impossibility of maintaining the water level in the pool above a predetermined low level (e.g., the lower limit for water intake from skimmers) over a predetermined period in the future (e.g., fifteen days) without the addition of water from another source (e.g., the public water supply network).

[0297] Maintain the water level in the pool at all times above a predetermined level (e.g., the lower suction limit of skimmers), and, under this condition, distribute any excess water collection relative to the capacity of the water storage system unless it causes the water level in the pool to exceed a predetermined high level (e.g., five centimeters below the edge of the pool) and maintain at all times a maximum quantity of water in the water storage system.

[0298] Finally, a prediction of the actual water level in the pool is made by implementing a machine learning trained model based on available measured data, the amount of rainwater that will be directly received by the surface of the pool, and / or models of water evaporation at the surface of the pool.

[0299] Note that the proposal made to the user may include a temporally programmed addition of water from another source (in particular the public water distribution network) to the pool if the prediction of the actual water level forecasts a drop in the water level in the pool below the predetermined low level at a given date.

[0300] SECOND EXAMPLE: DISTRIBUTION OF WATER TO WATER ANIMALS.

[0301] The predictions and trained models of the second example are similar to those of the first example, except that the water level controlled by the device of the invention is that of at least one drinking trough.

[0302] THIRD EXAMPLE: WATERING A GARDEN

[0303] It is assumed here that the garden has an irrigation system, drip irrigation and / or sprinkler irrigation, correctly calibrated to meet the respective water needs of the plants.

[0304] The input data, preferably dated or even time-stamped, of the machine learning model trained on at least some measured data preferentially includes data from the following first group:

[0305] Fixed data representing the activity for which the water is distributed:

[0306] o Planting area,

[0307] o For a crop: identification of cultivated plants) or gardening (with planting type), soil texture and / or retention capacity, plant development and / or health sensors, such as those described above for calculating evapotranspiration,

[0308] Past and present weather data:

[0309] o Forecast rainfall,

[0310] o Sunshine,

[0311] o Air temperature,

[0312] o Wind speed,

[0313] o Maximum wind gust speed;

[0314] The measurements obtained by the device's operating sensors, o Quantity of water collected,

[0315] o Quantity of water stored,

[0316] o Quantity of water distributed;

[0317] The input data may also include data from the following second group:

[0318] Physical condition of the garden:

[0319] o Soil moisture content,

[0320] o Optionally, the state of development and health of the plantations;

[0321] Weather data,

[0322] o Wind direction;

[0323] Past environmental physical measurements from the device's sensors,

[0324] Rainfall measurement,

[0325] o Measurement of sunshine,

[0326] o Wind measurements, o Air temperature measurements,

[0327] o Wind direction measurement;

[0328] The measurements obtained by the device's operating sensors, o Measurement of electrical autonomy;

[0329] Possibly, the availability of other water resources:

[0330] o Presence of an announcement of water distribution network power outages,

[0331] o Groundwater level or water retention basins.

[0332] Using data from the first group and, preferably, at least one data point from the second group, a prediction of the amount of water collected in the future and a prediction of future water distribution needs are made, preferably by implementing a machine learning model on measured data. Alternatively or complementarily, these predictions implement an expert system and / or a model of water consumption by garden plants (evapotranspiration as a function of air temperature, sunlight, air humidity, and wind).

[0333] Then, a prediction (or proposal) of the quantities of water to be distributed in the future is made. Preferably, this prediction of the quantities of water to be distributed takes into account at least one operating rule of the device, for example:

[0334] To supply the plantations with a predetermined quantity of water (for example, a quantity set by the user as normal), during each successive predetermined period over a predetermined future period (for example, for each period of two consecutive days without rain, over a period of ten future days), and, if this is impossible, to apply a multiplicative factor of less than one to the predetermined quantity of water supplied to the plantations,

[0335] Provide a predictive alert indicating that it will be impossible to maintain crop irrigation for a predetermined period in the future (e.g., fifteen days) without adding water from another source (e.g., the public water supply network).

[0336] Provide the crop with a predetermined minimum amount of water (e.g., half the amount set by the user as normal) for each successive predetermined period over a predetermined future period (e.g., for each period of two consecutive days without rain over a period of fifteen days to come), and, under this condition, distribute any excess water collected relative to the capacity of the water storage system and maintain a maximum amount of water permanently in the water storage system.

[0337] The following describes the use of rainfall predictions to suggest to the user whether or not to distribute water stored in the storage system and, if distribution is decided, the proposed quantity of water to be distributed or, proportionally, the duration of water distribution:

[0338] If the predicted rainfall within a predetermined timeframe (e.g., within the next twelve hours) exceeds a predetermined value (e.g., 10 mm), irrigation is suspended for one day. If the amount of stored water is less than a predetermined proportion (e.g., 20%) of the storage system's capacity, and the total predicted rainfall over the forecast period is less than a predetermined value that can be set according to the crop being irrigated (e.g., 10 mm), the device switches to or remains in "survival" mode, in which the amount of water distributed is equal to a predetermined proportion (e.g., 70%) of the predicted water requirement for the coming day. An alert indicating an urgent need to refill the storage system is sent to the user. As soon as the user has at least partially refilled the storage system, one of the following rules is implemented;

[0339] If the amount of water stored is within a predetermined range of values ​​(e.g., between 20% and 40%) of the storage system's capacity and the total rainfall predicted over the prediction period is less than a predetermined value that can be set according to the crop to be irrigated (e.g., 10 mm), the device switches to or remains in "economy" mode in which the amount of water distributed is equal to a predetermined proportion (e.g., 85%) of the water requirement predicted for the coming day;

[0340] If the amount of water stored is greater than a predetermined value (e.g., 40%) of the storage system's capacity, the device switches or remains in "normal" mode in which the amount of water distributed is equal to the predicted water requirement for the coming day;

[0341] Note that the proposal made to the user may include a distribution to the plantations, programmed in the future or immediately, of water from another source (in particular the public water distribution network) if the water distribution prediction provides, in a predetermined period to come (for example during the next fifteen days) an absence of stored water to distribute in the device after the distribution of the predetermined minimum quantity of water.

[0342] FOURTH EXAMPLE: IRRIGATION OF A CROP

[0343] The predictions and trained model of the fourth example are similar to those of the third example, except that a model of the water requirements of cultivated plants is implemented and the first data set also includes the soil moisture content in the plantation and measurements of the developmental and / or health status of the cultivated plants. For this purpose, a plant health sensor of a type commonly used in agriculture is added to the device. VARIANTS FOR ALL EXAMPLES

[0344] In some variations, the machine learning performed by the model is supervised by taking into account past distribution commands given by the user and / or an expert.

[0345] In some variations, learning is supervised by taking into account recommendations given by an expert, for example in agronomy.

[0346] During a step 115, the central unit 71 calculates an optimization of the dates (possibly supplemented by the times) and quantities of water to be distributed or rejected in at least one time interval of a future duration (for example, one week) as a function of the distribution instruction received in step 112, each possible inhibition factor and the prediction of at least one future quantity of water collected by the device and the prediction of at least one future need for water distribution by the device.

[0347] The central unit 71 is thus configured to calculate at least one quantity of water to be distributed within a future time interval, based on each of the aforementioned predictions. The central unit 71 is configured to, in the absence of a contrary command from a user within a predetermined period, control the distribution of each calculated quantity of water within each time interval. Preferably, at least once per time interval, the predictions and each quantity of water to be distributed within a future time interval are recalculated.

[0348] The bottommost diagram in Figure 15 represents, to the right of the vertical line representing the present, an optimization of water distribution by the device for the prediction period. This optimization follows rules such as, for example:

[0349] The central unit attempts, through water reserve prediction, to ensure future water distribution needs.

[0350] If ensuring future water distribution needs is not possible for at least one "shortage" day in the future, based on water reserve predictions, a quantity of water is chosen to be distributed which, supplemented with direct rainfall and after accounting for evapotranspiration, ensures the same daily quantity of water made available until the last day of shortage, provided that this quantity is greater than a predetermined minimum value corresponding to the cultivated plants (this configurable minimum value is, for example, obtained from a model of water consumption during the development of the cultivated plant and an analysis of the water retention capacity of the cultivated land), and if this quantity is less than the minimum value, the minimum value is distributed and an alert message is sent to the user, if it is predicted that the collected water storage system will be full in the future (for example, here, from the third day in the future),The surplus collected up to this event is distributed before it occurs, attempting to standardize the daily distribution described in the previous paragraph.

[0351] For example, regarding water distribution, optimization takes into account a first rule according to which animal watering takes priority over crop irrigation, which itself takes priority over filling a swimming pool. Furthermore, for a given quantity of water available for crop irrigation (or animal watering) during a given period before the next rainfall, a second optimization rule might consist of limiting the daily quantity of water to be distributed for this irrigation (or, respectively, this watering). A third optimization rule might consist of favoring water distribution at a predetermined time (for example, the time when evapotranspiration will be minimal, determined based on the forecast temperature, humidity, and wind speed received from a meteorological information service).

[0352] In variants, the method and device of the invention perform an economic optimization based on the cost of water from other available water resources (in particular the public water distribution network). Then, during a step 116, the central unit makes available to the user the water distribution proposal resulting from the optimization and, possibly, the distribution instruction, each possible inhibition factor, the quantity of available stored water as well as its quality, the prediction of future quantity of water collected by the device and future water needs and the information taken into account by the optimization.

[0353] During step 117, the user then validates or invalidates this proposal and, in the second case, orders the distribution or rejection, immediate or delayed, of stored water and the quantity of water thus distributed, or the non-distribution of water.

[0354] During step 118, the central unit 71 implements any distribution, non-distribution, or rejection command given by the user. The central unit 71 is configured to, in the absence of a contrary command from the user within a predetermined configurable period following step 116 (for example, 30 minutes), implement water distribution according to the optimization it determined during step 115.

[0355] Figure 16 shows a modular representation of a particular embodiment of a 300 system for training a model by machine learning on measured data and providing predictions of measurable data.

[0356] Such a modular representation includes:

[0357] Interface modules 305, including:

[0358] o a device for defining, for example by input or transfer from another computer system, fixed data representing the activity for which the water is distributed, operating rules and / or parameter values, for example corresponding to so-called "predetermined" values ​​in the preceding description 310, such as the duration of the prediction period, o optionally a graphical user interface or an API 320 and o optionally, at least one interface module 315 corresponding to an input or output device, as described in Figure 17, an optional set of modules for constructing a database 325 of measured or acquired physical quantity values, comprising:

[0359] o a sensor 330 for measuring the quantities of water collected for each time interval, captured by a sensor of the rainwater collection and distribution device that is the subject of the invention,

[0360] o a means 331 for acquiring past and / or present weather forecasts, for example by accessing a server of a meteorological service and / or by at least one sensor of the rainwater collection and distribution device,

[0361] o a sensor 332 for measuring environmental physical quantities of the rainwater collection and distribution system and / or o a sensor 333 for measuring a physical state of the activity for which water is distributed by the rainwater collection and distribution system, by at least one sensor connected to the rainwater collection and distribution system, an optional set of data processing modules, including:

[0362] o a data recovery device 335,

[0363] o a computing device 340 to produce predictions and, optionally, optimization, and / or

[0364] o a prediction model 350 to be trained, in a first phase, and made to provide predictions, in a second phase, it being noted that the training can continue in the second phase, and / or a water distribution control module 355, embodiments of which are described above, in particular with regard to figure 11,

[0365] at least two of said modules preferably being connected via a 301 network, as described in Figure 17.

[0366] Note that the same model 350, once trained, can provide either predictions of future collected water volumes or future downstream water distribution needs, or both. If two different trained models are implemented, the model providing the prediction of future water distribution needs can be trained on data containing the prediction of future collected water volumes.

[0367] The sensors 330, 332, 333, the acquisition means 331 and / or the database are configured to associate a timestamp (date and, possibly, time) of obtaining each data captured or acquired.

[0368] The weather forecast acquisition means 331 acquires these forecasts with their confidence index, which is stored and processed by the model being trained or trained.

[0369] Figure 17 shows a functional diagram illustrating an example of a computer system with which an embodiment can be implemented. In the example in Figure 17, a computer system 205 and instructions for implementing the disclosed technologies in the hardware, software, or a combination of hardware and software, are represented schematically, for example, in the form of boxes and circles, at the same level of detail commonly used by people of ordinary competence in the art to which this disclosure relates to communicate about computer architecture and computer system implementations.

[0370] The computer system 205 includes an input / output (I / O) subsystem 220, which may include a bus and / or one or more other communication mechanisms for communicating information and / or instructions between the components of the computer system 205 over electronic signal paths. The input / output subsystem 220 may include an input / output controller, a memory controller, and at least one input / output port. The electronic signal paths are represented schematically in the drawings, for example, as lines, unidirectional arrows, or bidirectional arrows.

[0371] At least one 210 processor is coupled to the 220 I / O subsystem for processing information and instructions. The 210 processor may include, for example, a general-purpose microprocessor or microcontroller and / or a special-purpose microprocessor such as an integrated system, a graphics processing unit (GPU), a digital signal processor, or an ARM processor. The 210 processor may include an integrated arithmetic logic unit (ALU) or may be coupled to a separate ALU.

[0372] The computer system 205 includes one or more memories 225, such as a main memory, which is coupled to the I / O subsystem 220 to electronically store data and instructions to be executed by the processor 210. The memory 225 may include volatile memory such as various forms of random access memory (RAM) or any other dynamic storage device. The memory 225 may also be used to store temporary variables or other intermediate information during the execution of instructions to be carried out by the processor 210. Such instructions, when stored in a non-transient, computer-readable storage medium accessible to the processor 210, can transform the computer system 205 into a special-purpose machine that is customized to perform the operations specified in the instructions.

[0373] The computer system 205 also includes non-volatile memory such as read-only memory (ROM) 230 or other static storage device coupled to the I / O subsystem 220 to store information and instructions for the processor 210. The ROM 230 may contain various forms of programmable ROM (PROM) such as erasable PROM (EPROM) or electrically erasable PROM (EEPROM). A persistent storage unit 215 may contain various forms of non-volatile random-access memory (NVRAM), such as FLASH memory, or solid-state storage, a magnetic disk, or an optical disk such as a CD-ROM or DVD-ROM, and may be coupled to the I / O subsystem 220 to store information and instructions.Memory 215 is an example of non-transient computer-readable media that can be used to store instructions and data which, when executed by processor 210, cause the execution of computer-implemented methods to carry out the techniques in this document.

[0374] Instructions in memory 225, ROM 230, or storage 215 can comprise one or more instruction sets that are organized into modules, methods, objects, functions, routines, or calls. Instructions can be organized as one or more computer programs, operating system services, or application programs, including mobile applications. Instructions can include an operating system and / or system software; one or more libraries to support multimedia, programming, or other functions; instructions or data protocol stacks to implement TCP / IP, HTTP, or other communication protocols; file format processing instructions to parse or render files encoded using HTML, XML, JPEG, MPEG, or PNG;user interface instructions to render or interpret commands for a graphical user interface (GUI, for "Graphics User Interface"), a command line interface, or a text-based user interface;Application software such as an office suite, Internet access applications, design and manufacturing applications, graphics applications, audio applications, software engineering applications, educational applications, games, or miscellaneous applications. The instructions may implement a web server, a web application server, or a web client. The instructions may be organized as a presentation layer, an application layer, and a data storage layer such as a relational database system using a structured query language (SQL) or no SQL, an object store, a graph database, a flat file system, or any other data storage.

[0375] The computer system 205 can be coupled via the I / O subsystem 220 to at least one output device 235. In one embodiment, the output device 235 is a digital computer display. Examples of displays that can be used in various embodiments include a touchscreen, a light-emitting diode (LED) display, a liquid crystal display (LCD), or an electronic paper display. The computer system 205 may include one or more other types of output devices 235, either in place of or in addition to a display device. Examples of other output devices 235 include printers, ticket printers, plotters, projectors, sound or video cards, loudspeakers, buzzers or piezoelectric or other audible devices, LED or LCD lamps or indicators, haptic devices, actuators, or servos.

[0376] At least one input device 240 is coupled to the I / O subsystem 220 to communicate signals, data, command selections, or gestures to the processor 210. Examples of input devices 240 include touch screens, microphones, digital still and video cameras, alphanumeric and other keys, keyboards, graphics tablets, image scanners, joysticks, clocks, switches, buttons, dials, sliders.

[0377] Another type of input device is a control device 245, which can perform cursor control or other automated control functions, such as navigating a graphical user interface on a display screen, either alternatively or in addition to input functions. The control device 245 can be a touchpad, mouse, trackball, or cursor direction keys to communicate direction information and control selections to the processor 210 and to control cursor movement on the screen 235. The input device can have at least two degrees of freedom in two axes, a first axis (e.g., x) and a second axis (e.g., y), which allows the device to specify positions in a plane.Another type of input device is a wired, wireless, or optical control device, such as a joystick, wand, console, steering wheel, pedal, gear shifter, or any other type of control device. A 240 input device can include a combination of several different input devices, such as a video camera and a depth sensor.

[0378] In another embodiment, the computer system 205 may include an Internet of Things (IoT) device in which one or more of the output device 235, input device 240, and control device 245 are omitted. Or, in such an embodiment, the input device 240 may include one or more cameras, motion detectors, thermometers, microphones, seismic detectors, other sensors or detectors, measuring devices, or encoders, and the output device 235 may include a special-purpose display such as a single-line LED or LCD display, one or more indicators, a display panel, a counter, a valve, a solenoid, an actuator, or a servomotor.

[0379] The output device 235 may include hardware, software, firmware, and interfaces for generating position report packets, notifications, pulse or heartbeat signals, or other recurring data transmissions that specify a position of the computer system 205, alone or in combination with other application-specific data, directed to the host 250 or server 255.

[0380] The computer system 205 can implement the techniques described herein using custom hardwired logic, at least one ASIC (Application-Specific Integrated Circuit) or FPGA (Field-Programmable Gate Array), firmware, and / or program instructions or logic that, when loaded and used or executed in combination with the computer system, cause or program the computer system to function as a purpose-specific machine. In one embodiment, the techniques described herein are executed by the computer system 205 in response to the processor 210, which executes at least one sequence of at least one instruction contained in main memory 225.These instructions can be read from main memory 225 from another storage medium, such as memory 215. Executing the instruction sequences contained in main memory 225 causes the processor 210 to execute the process steps described in this document. In other embodiments, hardwired circuits may be used instead of, or in combination with, software instructions.

[0381] The term "storage medium," as used in this document, refers to any non-transient medium that stores data and / or instructions enabling a machine to operate in a specific manner. Such storage media may include non-volatile and / or volatile media. Non-volatile media include, for example, optical or magnetic disks, such as memory-215. Volatile media include dynamic memory, such as memory-225. Common forms of storage media include, for example, a hard disk drive, a solid-state drive, a flash drive, a magnetic data storage medium, any optical or physical data storage medium, a memory chip, and so on.

[0382] Storage media are distinct from transmission media but can be used in conjunction with them. Transmission media facilitate the transfer of information between storage media. For example, transmission media include coaxial cables, copper wires, and optical fibers, including the wires that make up a bus in the I / O 220 subsystem. Transmission media can also take the form of acoustic or light waves, such as those generated during radio and infrared data communications.

[0383] Various forms of media can be involved in transporting at least one sequence of at least one instruction to the processor 210 for execution. For example, the instructions may initially be transported on a magnetic disk or solid-state drive of a remote computer. The remote computer may load the instructions into its dynamic memory and send the instructions over a communication link such as a fiber optic or coaxial cable or a telephone line using a modem. A modem or router local to the computer system 205 may receive the data over the communication link and convert the data into a format that can be read by the computer system 205.For example, a receiver such as a radio frequency antenna or an infrared detector can receive data carried in a wireless or optical signal, and a suitable circuit can provide the data to the I / O subsystem 220, for example, by placing the data on a bus. The I / O subsystem 220 carries the data to memory 225, from which the processor 210 retrieves and executes instructions. Instructions received by memory 225 may optionally be stored in memory 215 before or after execution by the processor 210.

[0384] The computer system 205 also includes a communication interface 260 coupled to a bus 220. The communication interface 260 provides bidirectional data communication coupling to the network link(s) 265 that are directly or indirectly connected to at least one communication network, such as a network 270 or a public or private cloud on the Internet. For example, the communication interface 260 can be an Ethernet® network interface, an Integrated Services Digital Network (ISDN) card, a cable modem, a satellite modem, or a modem to provide a data communication connection to a corresponding type of communication line, for example, an Ethernet® cable, a metallic cable of any type, a fiber optic line, or a telephone line.The 270 network broadly represents a local area network (LAN), a wide area network (WAN), a campus network, the Internet, or any combination thereof. The 260 communication interface may include a LAN card to provide a data communication connection to a compatible LAN, or a cellular radio interface that is wired to send or receive cellular data according to cellular radio wireless network standards, or a satellite radio interface that is wired to send or receive digital data according to satellite wireless network standards. In any such implementation, the 260 communication interface sends and receives electrical, electromagnetic, or optical signals over signal paths that carry digital data streams representing various types of information.

[0385] A network link 265 typically provides electrical, electromagnetic, or optical data communication directly or through at least one network to other data devices, using, for example, satellite, cellular, Wi-Fi®, or Bluetooth® technology. For example, a network link 265 can provide a connection across a network 270 to a host computer 250.

[0386] In addition, the network link 265 can provide a connection via the network 270 or to other computing devices through interconnect devices and / or computers operated by an Internet Service Provider (ISP) 275. The ISP 275 provides data communication services via a global packet-switched data communication network represented by the Internet 280. A server computer 255 can be coupled to the Internet 280. The server 255 broadly represents any computer, data center, virtual machine, or virtual computing instance with or without a hypervisor, or computer running a containerized program system such as Docker® or Kubernetes®. The server 255 can represent an electronic digital service that is implemented using more than one computer or instance and is accessed and used by transmitting web service requests,Uniform Resource Locator (URL) strings with parameters in HTTP payloads (Hypertext Transfer Protocol), API calls (Application Programming Interface), application service calls, or other service calls. The 205 computer system and the 255 server can form elements of a distributed computing system that includes other computers, a processing cluster, a server farm, or another organization of computers that cooperate to perform tasks or run applications or services. The 255 server can have one or more sets of instructions that are organized as modules, methods, objects, functions,of routines or calls. Instructions can be organized as one or more computer programs, operating system services, or application programs, including mobile applications. Instructions may include an operating system and / or system software; one or more libraries to support multimedia, programming, or other functions; instructions or data protocol stacks to implement TCP / IP (Transmission Control Protocol / Internet Protocol), HTTP, or other communication protocols; file format processing instructions to parse or render files encoded using HTML (Hypertext Markup Language), XML (Extensible Markup Language),JPEG (for "Joint Photography Experts Group"), MPEG (for "Moving Picture Experts Group"), or PNG (for "Portable Networks Graphie"); user interface instructions to render or interpret commands for a graphical user interface (GUI), a command-line interface, or a text-based user interface; application software such as an office suite, Internet access applications, design and manufacturing applications, graphics applications, audio applications, software engineering applications, educational applications, games, or miscellaneous applications. Server 255 may include a web application server that hosts a presentation layer,an application layer and a data storage layer such as a relational database system using a structured query language (known as "SQL", for "Structured Query Language") or no SQL, an object store, a graph database, a flat file system or any other data storage.

[0387] The computer system 205 can send messages and receive data and instructions, including program code, via the network(s), the network link 265, and the communication interface 260. In the Internet example, a server 255 can transmit requested code for an application program via the Internet 280, the ISP 275, the local network 270, and the communication interface 260. The received code can be executed by the processor 210 as it is received, and / or stored in memory 215, or in other non-volatile memory for later execution.

[0388] The execution of instructions as described in this section can implement a process in the form of an instance of a running computer program, consisting of program code and its current activity. Depending on the operating system (OS), a process may consist of multiple threads that execute instructions simultaneously. In this context, a computer program is a passive collection of instructions, while a process can be the actual execution of those instructions. Multiple processes can be associated with the same program; for example, opening multiple instances of the same program often means that more than one process is running. Multitasking can be implemented to allow multiple processes to share the CPU.Although each processor 210 or processor core executes only one task at a time, the computer system 205 can be programmed to implement multitasking to allow each processor to switch between running tasks without having to wait for each task to finish. In one embodiment, switching can occur when tasks perform input / output operations, when a task indicates it can be switched, or on hardware interrupts. Time-sharing can be implemented to allow for a rapid response to interactive user applications by quickly switching contexts to give the impression of multiple processes running concurrently.In one embodiment, for security and reliability reasons, an operating system may prevent direct communication between independent processes, by providing a strictly mediated and controlled interprocess communication functionality.

[0389] The technical characteristics of the embodiments of the process and device of the invention, as described above, are intended to be combined to form other embodiments of the process and device of the invention.

[0390] In conclusion, the rainwater harvesting system described in this invention operates autonomously. This system can function in open fields, away from buildings, electrical and / or telephone networks. It collects clean water for storage and use in various applications, such as agricultural irrigation (both residential and commercial), the creation of permanent, self-sustaining water reserves in sensitive forests to combat wildfires as quickly as possible, swimming pool filling, water collection for domestic or industrial use, air cooling through misting, evapotranspiration, or vegetated surfaces, and more generally, any system requiring the collection and use of clean water in an isolated area. This system is autonomous because it manages its moving parts and ensures its own safety.

Claims

43 DEMANDS 1. Rainwater collection and distribution system (10, 40, 50, 60), comprising: - a collection surface (32, 46, 56) mounted on at least one mobile element (22, 42, 52, 62) of a frame equipped with at least one actuator (29, 59, 67, 68) configured to move each said mobile element between a so-called "folded" configuration, in which the collection surface is folded and has a first footprint on the ground, and a so-called "deployed" configuration, in which the collection surface is deployed and has a second footprint on the ground with a surface area greater than the surface area of ​​the first footprint on the ground, - a storage system (89) for rainwater collected on the collection surface, - a means of distributing water from the storage system to the downstream side of the device, and - a central control unit (71) for each actuator of the frame; - a means of acquiring a prediction of at least a future quantity of water collected by the device in a future time interval, said prediction being a function of measurable values, - a means of determining a prediction of at least one water distribution need downstream of the device in a future time interval, said prediction being a function of measurable values, characterized in that the central unit (71) is configured to calculate at least one quantity of water to be distributed in a future time interval, according to each of said predictions, the central unit (71) being configured to, in the absence of contrary command from a user during a predetermined period, command the distribution of each quantity of water calculated in each said time interval.

2. Device (10, 40, 50, 60) according to claim 1, wherein at least one said prediction is determined by a machine learning trained model on data comprising at least past predicted rainfall values, and at least measured collected water quantity values.

3. Device (10, 40, 50, 60) according to claim 2, wherein the model is trained by machine learning on data including at least some predicted past sunshine values.

4. Device (10, 40, 50, 60) according to any one of claims 2 or 3, comprising at least one sensor for an environmental physical quantity, in which the model is trained by machine learning on data comprising at least some data provided by at least one of said sensor.44 5. Device (10, 40, 50, 60) according to any one of claims 1 to 4, wherein the central unit (71) is configured to calculate at least one future evapotranspiration of plants receiving water distributed by the device and to calculate at least one quantity of water to be distributed in a future time interval, as a function of said future evapotranspiration.

6. Device (10, 40, 50, 60) according to any one of claims 1 to 5, comprising at least one sensor of a physical quantity representative of a quantity of water available downstream of the water distributor, the central unit (71) being configured to calculate a quantity of water to be distributed as a function of the quantity of water available downstream of the distributor.

7. Device (10, 40, 50, 60) according to any one of claims 1 to 6, wherein the central unit (71) is configured, to calculate the quantity of water to be distributed, to implement a machine learning trained model on measured data representative of quantities of water to be distributed calculated in the past and of distribution commands provided by a user in the past.

8. Device according to any one of claims 1 to 7, further comprising a means for training a prediction model trained by machine learning, said prediction model being configured to associate at least one value of future quantity of water collected by the device with a function of values ​​measured in the past representative of quantities of water collected by the device associated with dating values ​​of said measured values.

9. Device according to any one of claims 1 to 8, further comprising a means for training a prediction model trained by machine learning, said prediction model being configured to associate at least one value of future quantity of water collected by the device with a function of weather forecast values ​​acquired in the past associated with dating values ​​of said measured values.

10. Device according to any one of claims 1 to 9, further comprising a means for training a predictive model of future water needs trained by machine learning, said predictive model being configured to associate at least one value of future water need distributed by the device with a function of values ​​measured in the past representative of water needs associated with dating values ​​of said measured values.

11. Device according to claim 10, wherein the model for predicting future water needs is configured to associate at least one value of future water need distributed by the device with a function of predicting water collections obtained in the past.45 12. Device (10, 40, 50, 60) according to any one of claims 1 to 10, wherein at least one said prediction is determined by an expert system operating on data comprising at least predicted values ​​of past rainfall, and at least measured values ​​of collected water quantities.

13. Device according to any one of claims 1 to 12, wherein, at least once per time interval for which a quantity of water to be distributed has been determined by the central unit, the predictions and each quantity of water to be distributed in each future time interval are recalculated.

14. Method (90, 100, 110) of distributing rainwater collected on a collection surface (32, 46, 56) mounted on at least one moving element (22, 42, 52, 62) of a frame equipped with at least one actuator (29, 59, 67, 68) configured to move each said moving element between a so-called "folded" configuration, in which the collection surface is folded and has a first footprint on the ground, and a so-called "deployed" configuration, in which the collection surface is deployed and has a second footprint on the ground with a surface area greater than that of the first footprint on the ground, the water thus collected being stored in a storage system (89) equipped with a water distributor, process characterized in that it comprises: - a control step for each actuator of the frame to move the collection surface from its folded configuration to its deployed configuration, - a rainwater collection stage in a storage system, when the collection surface is in the deployed configuration, - a step of acquiring a prediction of at least one future quantity collected by the device in a future time interval, said prediction being a function of measurable values, - a step of determining a prediction of at least one water distribution need downstream of the device in a future time interval, said prediction being a function of measurable values, - a step involving the calculation of at least one quantity of water to be distributed within a future time interval, based on each of the aforementioned predictions, and - in the absence of a contrary order from a user during a predetermined period, a control step for the distribution of each quantity of water calculated in each said time interval.

Images