METHOD FOR MANAGING ATMOSPHERIC WATER GENERATION DEVICES

Abstract

METHOD FOR MANAGING ATMOSPHERIC WATER GENERATION DEVICES The invention relates to a method for controlling at least one atmospheric water generation device. The method comprises generating (GEN1) at least one control indicator based on an estimate of meteorological conditions (EST1), an estimate of infrastructure water consumption needs (EST3), an estimate of electrical power availability (EST2), and an estimate of the performance (EST4) of the atmospheric water generation device. Figure for the abstract: Fig. 9

Description

Title of the invention: METHOD FOR MANAGING ATMOSPHERIC WATER GENERATION DEVICES Scope of the invention

[0001] The invention relates to a method for controlling the operation of an atmospheric water generation device or a network of devices intended to efficiently produce drinking water. State of the art

[0002] Water is a vital natural resource that is steadily dwindling on a planet considered to be blue. With the world's population expected to reach nearly 10 billion by 2050, the global water crisis is one of the greatest threats humanity will face in the coming decades. Currently, billions of people worldwide still lack adequate access to water, sanitation, and hygiene, according to a joint report by UNICEF and the World Health Organization (WHO). Despite its abundance on Earth, only 2.5% of the planet's available water is fresh, and of that, only 0.7% is accessible at the surface. Furthermore, climate change and global warming are altering the state of lakes, rivers, and streams.According to the current climate change scenario, nearly half the world's population will be living in areas experiencing high water stress by 2030. Meanwhile, several factors are further exacerbating water stress, including conflict, increased energy demand, demographic shifts, pollution, and urbanization. Among current solutions, seawater desalination leaves potentially toxic brine as a byproduct. For areas with access to water, the cost of drinking water includes the cost of public water, sanitation services, and fees and taxes used to offset pollution generated by water use. Therefore, the pricing required to cover the cost of all these steps is likely to increase and will continue to rise given water scarcity and frequent pollution of the resource.To solve the problems of water stress and rising water and energy prices, innovative solutions must be implemented that offer attractive investment opportunities while guaranteeing an economic and environmental strategy, a pillar of sustainable development.

[0003] Numerous implementations of atmospheric water generation systems are known; in other words, systems that extract water from the humidity contained in the ambient air. These atmospheric water generation systems have demonstrated, during the In recent decades, they have been considered emerging and promising solutions for overcoming global water scarcity, particularly in arid regions. Among these systems, two main methodologies are used: thermal cooling condensation systems (including air conditioning, refrigeration, thermoelectric, and thermoacoustic systems) and sorption systems. These thermal systems extract water vapor from ambient air by lowering its temperature below its dew point. The dew point, or dew-point temperature, is the temperature to which humid air must be cooled for water vapor to condense. For example, cooling condensation systems, such as vapor compression systems, primarily consist of a condenser, an expansion valve, an evaporator, and a compressor using a refrigerant.The refrigerant passes through the compressor, which transforms it into high-pressure, high-temperature vapor. It then condenses in the condenser. Due to the pressure drop across the expansion valve, the refrigerant undergoes a drastic temperature decrease, cooling the ambient air as it passes through the evaporator and causing the humidity to condense. This technology is considered energy-intensive due to the compressor's energy consumption. The second atmospheric water generation system is based on the sorption phenomenon, using desiccant materials. These desiccant materials can be solid or liquid adsorbents, such as zeolite, silica gel, lithium chloride, and lithium bromide. These desiccants are used to extract water vapor from the air through a hygroscopic process, and the water is then recovered through regeneration.This technology works exceptionally well when used in cooler temperatures or when a low dew point is required. It should be noted that such technology remains expensive. A combination of these two technologies provides a hybrid system that can also be used to improve humidity extraction.

[0004] This type of atmospheric water generation device has certain drawbacks. For example, the availability of water, and therefore the quantity of water that can be extracted through these devices, is closely linked to atmospheric conditions. This dependence is exacerbated by the variability of these conditions in space and time. Furthermore, since these systems consume electricity and some technologies remain expensive, the price per liter produced is not very competitive.

[0005] In order to resolve these drawbacks, processes for atmospheric water generation systems are known which manage the operation of such devices and allow their performance to be improved.

[0006] Moreover, processes for atmospheric water generation systems are known configured to control one or more characteristic ambient air parameters to manage the starting and / or stopping of the water generation device based on efficiency considerations.

[0007] US patent 20128321061 is known, describing a method for predictive planning of automated irrigation of various plants. This patent stipulates that water production is efficient when the air temperature and the dew point temperature are close. Estimates of the operating time or intervals of the devices are made according to this criterion.

[0008] The process disclosed in this document only provides an estimate of the watering time based on the temperature for irrigation, therefore it is not suitable for managing a drinking water network.

[0009] Also known is US patent 2021354080A1, describing a hybrid vapor compression and absorption system. The system includes a process that estimates weather conditions over the next few hours and estimates the water demand in order to connect or disconnect the vapor compression system. Operating hours planning is based on the assumption that the estimated ambient temperatures are close to the associated dew point temperatures.

[0010] It is clear that these types of processes offer some management of the operation of such atmospheric water generation devices; however, two major drawbacks emerge. Indeed, in the event of a succession of dry days, cloudy days, or seasonal climates with variable weather conditions, user satisfaction in terms of water production and energy efficiency will not be guaranteed.

[0011] Indeed, this type of process fails to anticipate the variability of atmospheric data (pressure, wind speed, solar irradiation, temperature, and humidity, for example). This type of process also does not solve the problem of the variability in users' water demand.

[0012] Furthermore, state-of-the-art solutions do not offer corrections to implemented forecasts and / or established commands.

[0013] Finally, it is known that, by 2030, 660 million people will still be without electricity. Prior art methods do not allow for atmospheric water production under conditions of limited or non-existent access to electrical energy. Summary of the invention

[0014] The invention aims to overcome the drawbacks of existing methods for controlling atmospheric water generation devices.

[0015] To this end, the invention relates to a method for controlling at least one device generation of atmospheric water which includes: • Initial acquisition of a first set of meteorological data from: - at least one source of meteorological data; and / or - a history of time-stamped weather data; and / or - a current date. • First estimation of a second set of meteorological data over a first predefined period using a calculator and based on the first set of meteorological data, said estimated meteorological data comprising: - A forecast of atmospheric temperature over a given period - A humidity forecast over a given period; • Second acquisition of an energy quantification data set from an energy data source of at least one power supply equipment; • Second estimation of a second set of energy quantification data, said set quantifying energy available over the first predefined period; • Third acquisition of an initial set of consumption data from: - at least one source of consumption data from a set of infrastructures and water consumption profiles; • Third estimate of a second set of consumption data over the first predefined period, the second set of consumption data including at least one consumption threshold; • Fourth acquisition of an initial set of performance data from: - At least one source of performance data from at least one atmospheric water generation device; • Fourth estimate of a second set of performance data over the first predefined period, the second set of performance data including at least one forecast of a performance indicator, said performance indicator defining at least a ratio between a quantity of water produced and a unit of electrical energy consumed and / or a water production flow rate; • Generation of an initial performance indicator for the first period of at least one atmospheric water generation device, based on the initial data. second third and fourth estimates, said control indicator defining at least one control data sent to the atmospheric water generating device, said control data defining at least one electrical supply duration and / or at least one electrical supply power of said at least one atmospheric water generating device; • Electrical control of said at least one atmospheric water generation device according to the first control indicator.

[0016] The control method according to the invention advantageously allows for the consideration of a variety of different factors for controlling at least one atmospheric water generation device. The system takes into account meteorological data, energy quantification data, user consumption data, and performance data of the atmospheric water generation devices. For each type of data, an estimate is made for a predefined period, which allows for the definition of a performance index that is subsequently used to control the atmospheric water generator(s). Such a control method for the generator(s) is particularly advantageous because it makes it possible to reach a necessary production threshold based on user needs while optimizing production, especially during periods when weather conditions are favorable for good atmospheric water production.

[0017] According to one embodiment, at least one meteorological data source is a source originating from: • An atmospheric pressure sensor; and / or • A humidity sensor; and / or • A temperature sensor; and / or • An anemometer; and / or • An initial monthly and / or daily meteorological database; and / or • A second hourly weather database; and / or • A third archived meteorological database.

[0018] According to one embodiment, the first control indicator generation step includes a step of estimating a first performance indicator over the first period of at least one atmospheric water generation device from the first and second estimates, said performance indicator defining a ratio between the quantity of water produced and a unit of electrical energy consumed, said performance indicator defining a pilot time period, the pilot step being carried out according to the pilot time period.

[0019] According to one embodiment, the available energy data source of at least Power supply equipment relates to: • A first energy source from an energy storage component; and / or • A second energy source from an electric power generator; and / or • A source of energy data available on an electrical network.

[0020] According to one embodiment, when a second energy source powers the atmospheric water generation device, the estimation of the amount of energy available from the second energy source is carried out from the estimates of the calculated meteorological data.

[0021] According to one embodiment, a second performance indicator is generated, said second performance indicator defining a ratio between the quantity of water produced and a unit of electrical energy consumed from the second energy source.

[0022] According to one embodiment, a third performance indicator is generated, said third performance indicator defining a ratio between the quantity of water produced and a unit of electrical energy consumed from an electrical network.

[0023] According to one embodiment, the process includes estimating a volume of water produced by a plurality of atmospheric water generation devices.

[0024] According to one embodiment, the method includes a step of measuring by at least one sensor of at least one atmospheric water generation device at least one water production flow rate and at least one electrical power consumed, the method further comprising a step of generating an alert indicator when the measured flow rate is less than a predefined flow rate threshold.

[0025] According to one embodiment, the method makes it possible to maximize the performance of the machine by controlling system components such as the degree of opening and / or closing of an expansion valve, the frequency variation of a compressor, the frequency variation of a fan, the activation and / or deactivation of a defrosting system according to estimates of meteorological and / or atmospheric data.

[0026] According to one embodiment, the method comprises the execution of a first machine learning model trained from a first set of training data comprising meteorological data, consumption data and energy quantity data and electrical regime data of at least one atmospheric water generation device, said machine learning model being configured to generate a performance indicator.

[0027] According to one embodiment, the method comprises the execution of a second machine learning model trained on a second training dataset comprising, on the one hand, estimates of meteorological data, estimates of consumption data and estimates of energy quantity data and on the other hand actual meteorological data, actual consumption data and actual energy quantity data, said machine learning model being configured to generate estimates from actual data.

[0028] According to one embodiment, the method comprises running a third machine learning model trained on a third training dataset comprising meteorological data and performance data for the atmospheric water generation device. The third dataset advantageously includes control parameters such as the degree of opening and / or closing of an electronic expansion valve of the device, the frequency of a fan of the device, and the frequency of a compressor of the device. Advantageously, the control parameters are associated with the meteorological and performance data. Advantageously, the third machine learning model is configured to generate a performance indicator.

[0029] According to one embodiment, machine learning models are configured to assign scores based on changes in the performance indicator. These machine learning models are advantageously trained based on the score corresponding to the change in the performance indicator. The use of the score advantageously enables supervised training of the machine learning models, allowing the total sum of scores to be maximized over time by selecting actions that improve system performance.

[0030] According to one embodiment, the measurement of the production flow rate and the electrical power consumed by the atmospheric water generating device or devices are included in the first set of training data.

[0031] According to one embodiment, the measurement of the production flow rate and the electrical power consumed by the atmospheric water generating device or devices is included in the third set of training data. Brief description of the figures

[0032] Other features and advantages of the invention will become apparent from the following detailed description, with reference to the accompanying figures, which illustrate:

[0033] [Fig.lA]: a synoptic diagram showing the architecture of an energy management system according to an embodiment of the invention;

[0034] [Fig.lB]: a synoptic diagram showing the architecture of an energy management system according to an embodiment of the invention;

[0035] [Fig.2]: an organizational chart illustrating the central process according to one embodiment of the invention;

[0036] [Fig.3]: a flowchart detailing a method for estimating meteorological data according to an embodiment of the invention;

[0037] [Fig.4]: an organizational chart detailing a method for estimating energy production on each of the sites according to an embodiment of the invention;

[0038] [Fig.5]: a flowchart detailing a method for estimating the performance of each device according to an embodiment of the invention;

[0039] [Fig.6]: a flowchart detailing a method for estimating water consumption needs according to an embodiment of the invention;

[0040] [Fig.7]: an organizational chart detailing a method for estimating operating hours according to an embodiment of the invention;

[0041] [Fig.8]: a flowchart detailing a method for starting the operation of the device according to an embodiment of the invention; and

[0042] [Fig.9]: a flowchart illustrating the main steps of the process according to one embodiment of the invention. Description of the invention

[0043] Now referring to figures 1 to 9, where identical numbers indicate the same elements on the figures, an electronic control method for managing the performance of atmospheric water generation systems is presented.

[0044] According to the preferred embodiment of the invention, but without limitation, the invention relates to a central server providing centralized management of a device or network of devices for generating atmospheric water. The invention also relates to a method for controlling at least one atmospheric water generation device (AWG). In this description, the control method may also be referred to as the central method 100. The central method 100 can be applied to any type of atmospheric water generation device (AWG) or air dehumidification system. By way of example, the atmospheric water generation device may include refrigeration or air conditioning systems. For example, the atmospheric water generation device may include an air cooler, a thermal cooler, a vapor compression system, or a sorption system.

[0045] According to one embodiment, the process 100 enables the control and command of one or more devices comprising air and water filtration components to produce water intended for human consumption meeting the guidelines and quality standards.

[0046] According to one embodiment, each of said water generation devices is equipped with an electronic control system. Each electronic control system is configured to control, via the central process, a water generation system located at a defined site.

[0047] According to one embodiment, the location of the site can be entered by the user.

[0048] According to one embodiment, the location of each of the devices can be collected automatically when the site is connected to the internet, as illustrated later.

[0049] According to one embodiment, the central process generates a map comprising devices with the same geographical and / or atmospheric characteristics and controls all the devices, forming the same area, on the basis of the same estimation of atmospheric data.

[0050] Advantageously, each water generation device can be equipped with appropriate sensors and / or modules to retrieve atmospheric and geographical data (for example, by sensors of temperature, humidity, wind speed and direction, pressure, solar irradiation, or weather station, GPS, WIFI, etc.).

[0051] According to one embodiment, each atmospheric water generation device is equipped with appropriate sensors to evaluate its operating state. By operating state, we mean the state of the components constituting the atmospheric water generation device (for example, control system for refrigeration, hydraulic and air circuits, on / off of components, energy performance, thermodynamic performance, etc.).

[0052] According to one embodiment, the central process allows one or more electronic control systems to be remotely controlled, and therefore one or more atmospheric water generation systems.

[0053] Advantageously, at least one database, derived from at least one water generation device, is collected, analyzed, and controlled by said central process. Such databases may include instantaneous databases and / or parameter histories and / or predictive databases comprising meteorological, atmospheric, and geographical conditions, user profiles and behaviors, and operating performance associated with each atmospheric water generation device.

[0054] According to one embodiment, said water generation devices can be interconnected with each other through centralized computer environments, where instructions are executed by remote processing units, which are connected by a communication network.

[0055] According to one embodiment, each water generation device can be operated separately and independently of each other.

[0056] According to one embodiment, the electronic control system 50 relating to each atmospheric water generation device comprises a power supply 51, a The system comprises a data acquisition and control unit 53, electrical components 59, measurement sensors / transducers 57, a control unit 58, and an HMI (Human-Machine Interface) unit 55 for user control and interaction. The power supply unit 51 provides energy to the various power stages required for the optimal operation of the different components of the device. For example, the power supply unit 51 provides 220 V AC at a frequency of 50 Hz. The output of this unit passes through a filtering and protection unit and then through two converters. The first is an AC / DC converter that delivers a 24 V DC voltage to power the control unit 58. This unit then passes through a DC / DC regulator, which outputs a 12 V voltage.

[0057] According to one embodiment, the acquisition and control block 53 allows the acquisition of one or more data collected through one or more sensors (digital or analog) 57. Said data is transmitted to a microcontroller 301 which allows the reading of the different data collected and which will then transfer them to a control card for processing.

[0058] According to one embodiment, the acquisition and control block 53 enables the control of the device, such that the control part comprises a microprocessor 303. For example, said microprocessor 303 comprises the following functions: • Retrieve the data from the sensors 57 acquired by the acquisition part, • Communicate with the various 305 connectivity modules, • Transfer this data to ensure machine control and monitoring real-time and remote. • Control the power supply of the various components (including, for example, compressor, fan(s), water pump(s), UV filter(s), electronic valve(s) via the acquisition board and relays). • Connect and configure a touchscreen.

[0059] According to one embodiment, the acquisition and control block 53 comprises: • 305 connectivity modules (i.e., 3G / 4G and WIFI modules, for example) to ensure the device's internet connectivity, which is necessary for the monitoring and remote control phase via web APIs, • A GPS module allowing the device's geographical location to be determined. • A BLE module that offers the possibility to control and monitor the status of the device through Bluetooth communication in case the device is installed in areas not accessible from the internet or in case of internet connection problems.

[0060] According to one embodiment, the acquisition and control block 53 comprises a display block 55 to facilitate interaction between the device and users.

[0061] Advantageously, the acquisition block 53, in addition to controlling and monitoring the device, allows for the verification of parameters and / or re-parameterization such as the choice of the connection method, and the verification of the lifespans of such components.

[0062] Thus, according to one embodiment, the acquisition block 53 comprises one or more sensors 57. According to one embodiment, at least one of the sensors 57 is located outside the atmospheric water generation device GEAb. The sensors 57 outside the water generation device are advantageously connected to the device by a link configured to transmit at least one data point measured by the sensor. Advantageously, this link comprises at least one cable configured to transmit at least one electrical signal. Advantageously, the link comprises at least one wireless transmitter, for example, a radio wave transmitter, configured to transmit a wireless signal. According to this example, the device comprises at least one wireless receiver configured to receive the wireless signal emitted by the transmitter. The receiver is configured to transmit the received signal to the acquisition block 53 and / or is understood by the acquisition block 53.These sensors 57 allow for the measurement of meteorological and atmospheric characteristics (e.g., temperature, humidity, pressure, wind direction and speed, solar irradiation). In one example, the sensors 57 include at least one atmospheric pressure sensor C1. In another example, the sensors 57 include at least one humidity sensor C2. In one embodiment, the sensors 57 include at least one temperature sensor C3. In one example, the sensors 57 include at least one anemometer C4. In yet another embodiment, the sensors 57 include at least one atmospheric pressure sensor.

[0063] According to one embodiment, the sensors 57 comprise at least one sensor for measuring geographic coordinates of the atmospheric water generation device(s) GEAi. In one example, the sensors 57 comprise at least one GPS device configured to generate at least one geographic coordinate.

[0064] According to one embodiment, the sensors 57 comprise at least one sensor measuring operational parameter data of the atmospheric water generation device GEAp. In one example, the device comprises at least one sensor configured to measure a quantity of water produced, for example, a water level sensor in a water tank of the device and / or a water flow sensor preferably arranged on a water outlet of the atmospheric water generation device GEAb.

[0065] In one example, the sensors 57 comprise at least one sensor configured to to measure the quality of the water produced. As an example, the 57 sensors include at least one conductivity sensor, and / or pH sensor, and / or turbidity sensor.

[0066] According to one embodiment, the sensors 57 include at least one sensor configured to measure energy produced and / or energy consumed by the GEAb water generation device. According to one example, the sensors 57 include at least one sensor for measuring the electrical power supplying the atmospheric water generation device.

[0067] In one embodiment, the sensors 57 comprise at least one sensor configured to measure at least one device efficiency parameter. In one embodiment, the sensors 57 comprise at least one sensor for determining the cooling capacity and / or at least one cooling temperature sensor. In one embodiment, at least one sensor 57 is configured to measure the electrical power supply to the atmospheric water generating device GEAp. In one embodiment, at least one of the sensors 57 is configured to measure the atmospheric water production flow rate. In one embodiment, the method includes a step for calculating a measured performance indicator INDPMi. Advantageously, the measured performance indicator comprises a ratio of the electrical power supply to the atmospheric water generating device GEAi and the water flow rate produced by said device.

[0068] Thus, the collection, analysis and correction of these parameters constitute a large database allowing the automated and efficient operation of the device to be optimized with respect to external constraints (availability of water, availability of energy).

[0069] According to one embodiment, the central process 100 is configured to optimize the cost per liter of water produced by each device by communicating with each of the associated electronic control systems. Without human intervention, control and command functions are implemented to execute the autonomous operation of each device and facilitate analysis and decision-making by implementing an energy efficiency strategy. The central process is characterized in that it comprises: • A procedure enabling weather forecasting 120. These forecasts are estimated and then corrected over an optimized time interval; • A procedure to estimate the amount of energy available 130 (in electricity production and storage, if available); • A procedure allowing the estimation of the performance of each device in terms of water production and energy consumption 140 or in performance ratio. These performance indicators are estimated, analyzed, and corrected, if necessary; • A procedure to estimate water consumption needs 150 taking into account atmospheric estimates as well as databases including user profiles and their consumption behaviors; • A procedure for estimating a threshold representing a minimum quantity that must be permanently available to the user; • A procedure allowing, in view of the collected databases (atmospheric conditions, user behavior, available energy) and the constraints to be met (user needs, device performance) to optimize the operating hours of each of the devices and the necessary operating period WPe of the water generation device.

[0070] The optimized time interval will sometimes be referred to in this application as the first predefined period. By first predefined period, we mean a period of time during which the atmospheric water generation device(s) (AWG) is / are controlled to produce atmospheric water.

[0071] In one example, the first predefined period lasts several hours, for example, eight. This time corresponds, in one example, to the duration of a night during which the atmospheric water generation devices (AWG) will be controlled to produce water. In another example, the first predefined period is at least 24 hours. In one embodiment, the predefined period is at least several days, for example, seven. This arrangement allows the atmospheric water generation device(s) to be controlled over a long period and thus spreads out electricity consumption, as well as enabling the selection of production times when weather conditions allow for optimal efficiency.

[0072] In one embodiment, the predefined time period is a variable time period. In one example, the predefined time period has a variable length that is advantageously scalable. In one embodiment, the first machine learning model and / or the second machine learning model are configured to generate correction data for the predefined time period. In one embodiment, the correction data advantageously allows for an adjustment of the duration of the predefined time period. Adjusting the predefined time period makes it possible to define a time period within which production can be optimized.

[0073] According to one embodiment, the method for controlling at least one atmospheric water generation device GEAi comprises at least one initial acquisition step ACQi of at least one initial meteorological dataset ENSp. In the present application, the ACQi acquisition step is sometimes referred to as the procedure 125 weather forecast retrieval data.

[0074] The ACQi acquisition of the first ENSi meteorological dataset is performed from at least one meteorological data source. For example, the meteorological data source is one of the 57 sensors described previously, such as a temperature sensor, an atmospheric pressure sensor, a solar radiation sensor, a relative humidity sensor, and / or an anemometer. For example, the meteorological data source includes at least one meteorological database from a national and / or regional meteorological service. For example, the meteorological data source includes at least one meteorological measuring station configured to measure one or more meteorological quantities. The weather station is advantageously connected to a communication network and configured to transmit the meteorological data over the communication network.

[0075] Advantageously, the initial ACQi acquisition step of the first ENSi meteorological dataset is performed from at least one current date. By current date, we mean the present date and / or time, or in other words, the date on which the piloting process is carried out.

[0076] Advantageously, the initial ACQi acquisition step of the first ENSi meteorological dataset is performed using at least one historical time-stamped meteorological data set. For example, the historical time-stamped meteorological data set comprises a set of measured meteorological data associated with a measurement date and / or time. Advantageously, the historical time-stamped meteorological data set includes at least one data point corresponding to the measurement of a meteorological quantity measured one year prior to the current date. For example, the historical time-stamped meteorological data set includes at least one data point corresponding to the measurement of a meteorological quantity measured on dates close to the date corresponding to one year prior to the current date; for example, data measured the week before and the week after the date corresponding to one year prior to the current date.For example, the historical time-stamped weather data includes at least one data point corresponding to the measurement of a meteorological quantity taken a few days before the current date. For example, the history includes the data corresponding to the measurement of the meteorological data each day of the week preceding the current date.

[0077] The piloting method includes a first estimation step ESTi of a second meteorological dataset ENSi' over a first predefined period Pi using a computer and based on the first meteorological dataset ENSi. This first estimation step ESTi is sometimes referred to in this request procedure for estimating meteorological data 120. The ESTi estimation step of meteorological data advantageously allows the generation of a set of ENS / which corresponds to a forecast of said meteorological data over the given period P,.

[0078] Advantageously, the meteorological data of the second ENS / set include at least one atmospheric temperature forecast over a given period. For example, the meteorological data of the second ENS / set include at least one ambient temperature, and / or at least one wet-bulb temperature, and / or at least one dew point temperature over the given period. The given period is, for example, a 24-hour period following the current date. For example, the given period is a 48-hour period following the current date. For example, the given period is a 7-day period following the current date.

[0079] Advantageously, the meteorological data of the second ENS / set include at least one forecast of absolute and / or relative ambient air humidity over a given period.

[0080] Advantageously, the piloting method includes a second acquisition step ACQ2 of an energy quantification data set ENS2 from an energy data source of at least one power supply equipment. In this application, the ACQ2 acquisition step is sometimes referred to as the procedure for estimating the amount of available energy 130. Power supply equipment means any device or infrastructure configured to store and / or distribute energy. Advantageously, power supply equipment is an electric accumulator and / or a battery of accumulators. In one embodiment, power supply equipment is an electric power generator. Electric power generator means any device configured to generate electrical power, such as a generator set, a power plant, and / or a hydroelectric dam.In one embodiment, the electrical power generator includes at least one renewable energy source. For example, the electrical power generator is one or more solar panels. For example, the electrical power generator is one or more wind turbines. For example, the electrical power supply equipment is an electrical grid.

[0081] According to one embodiment, the second energy quantification dataset ENS2 comprises at least one energy production history from at least one power supply equipment. In one example, the second energy quantification dataset ENS2 comprises several energy production histories from several power supply equipment. For example, each production history includes at least one quantified data point for electricity production, such as power output and / or energy produced. In another example, the quantified production data is associated with at least one temporal data point, such as a production date and / or production duration.

[0082] The control method includes a second estimation step EST2 of a second dataset ENS2'. Advantageously, the second dataset ENS2' quantifies the available energy over the first defined period. In one embodiment, the second dataset ENS2' includes at least one data point defining the energy availability of at least one power supply unit.

[0083] In this application, the second estimation step EST2 is sometimes referred to as the calculation of electrical energy estimates 132.

[0084] In one embodiment, the second estimation step EST2 includes a copy of the historical data from the energy quantification dataset ENS2 into the second energy quantification dataset ENS2'. Advantageously, the historical data includes production data from an anniversary of the current date. In one example, the data inserted into the second energy quantification dataset ENS2' includes the energy production data of the equipment whose energy production is estimated on the anniversary of the current date, for example, the production data one year prior to the current date. In one embodiment, the anniversary data corresponds to the equipment's production data on a date one day and / or one week prior to the current date.According to one embodiment, the estimation of energy availability is carried out by calculating a moving average of the production of said equipment over a few days preceding the current date, for example, two days, three days, four days, five days, six days or seven days.

[0085] In one embodiment, the historical data includes weather conditions associated with the energy availability data. In one example, the second estimation step includes a comparison between at least one weather data point estimated in the first estimation step and at least one weather data point associated with the energy availability data. In this way, the second estimation step is performed by searching for energy availability data corresponding to a date with weather conditions similar to the estimated weather conditions. In another example, the second estimation is performed by comparing the temperature and / or humidity forecast for the given period with the temperature and / or humidity data from the historical data.According to this example, the date on which the weather conditions are most similar is selected, and the availability forecast corresponds to the . Availability value of the selected date in the history.

[0086] In one embodiment, the historical data includes weather conditions associated with the energy efficiency data. In one example, the fourth estimation step includes a comparison between at least one weather data point estimated in the first estimation step and at least one weather data point associated with the energy efficiency data. In this way, the fourth estimation step is performed by searching for energy efficiency data corresponding to a date with weather conditions similar to the estimated weather conditions. In another example, the fourth estimation is performed by comparing the temperature and / or humidity forecast for the given period with the temperature and / or humidity data from the historical data.In this example, the date with the most similar weather conditions is selected, and the energy efficiency forecast corresponds to the energy efficiency value of the selected date in the history.

[0087] According to one embodiment, the energy quantification dataset ENS2 includes availability data provided by an energy distribution network manager and / or an energy production network manager.

[0088] According to one embodiment, the second energy quantification dataset ENS2' includes at least one availability data item from the first energy quantification dataset ENS2. This arrangement allows for the direct inclusion of data retrieved from the energy distribution and / or production network operator.

[0089] In one embodiment, the second estimation step EST2 of the second energy quantification dataset includes the inclusion of meteorological data estimates calculated in the first energy quantification dataset ENS2. In one example, the second estimation EST2 is performed by calculating one or more energy production data points from the second estimated meteorological dataset ENSp. In another embodiment, the second estimated meteorological dataset ENSi includes at least one data point related to solar irradiance. In this example, at least one power supply unit includes at least one solar panel. A calculator calculates an estimate of the second energy quantification dataset ENS2' from the solar irradiance forecast. The details of such a calculation are described below.

[0090] According to one embodiment, the second set of estimated meteorological data ENSi includes at least one data point relating to wind intensity and / or geographical wind direction. In this example, at least one power supply unit includes at least one wind turbine. The calculator calculates a Estimation of the second data set for quantifying ENS2' energy, i.e., at least the electrical power supplied by the wind turbine and / or the quantity of electrical energy supplied by the wind turbine over the first predefined period based on the wind conditions forecast. The methods for such a calculation are described below.

[0091] The control method includes a third acquisition step AQC3 of a first set of consumption data. ENS3. The third acquisition step ACQ3 is carried out using at least one consumption data source BDCi from a set of infrastructure and water consumption profiles. Infrastructure is defined as any building, equipment, or structure intended to accommodate users.

[0092] In one embodiment, the first BDCi consumption data set includes at least a history of water consumption quantities at the infrastructure or infrastructure set level. In one embodiment, the history of water consumption quantities is a daily and / or weekly and / or monthly and / or annual history.

[0093] The piloting method includes a third estimation step of a second consumption dataset ENS3'. The second consumption dataset includes at least one consumption threshold. Advantageously, the consumption threshold corresponds to a water consumption forecast at the infrastructure level. The third acquisition step ACQ3 is sometimes referred to as the user requirements estimation procedure 150 in this application.

[0094] According to one embodiment, the consumption threshold is calculated by the calculator from the first set of consumption data BDCi.

[0095] In one embodiment, the third estimation step EST3 includes a copy of the historical data from the first consumption dataset BDC i into the second consumption dataset ENS3'. Advantageously, the historical data includes water consumption data for the infrastructure from an anniversary of the current date. In one example, the data inserted into the second consumption dataset ENS3' includes the consumption data for the infrastructure whose water consumption is estimated on the anniversary of the current date, for example, consumption data one year prior to the current date. In one embodiment, the anniversary data corresponds to the infrastructure consumption data from a date one day and / or one week prior to the current date.According to one embodiment, the estimation of water consumption is carried out by calculating a moving average of the water consumption of said infrastructure over a few days preceding the current date, for example, two days, three days, four days, five days, six days or seven days.

[0096] In one embodiment, the historical data includes weather conditions associated with the consumption data. In one example, the third estimation step includes a comparison between at least one weather data point estimated in the first estimation step and at least one weather data point associated with the consumption data. In this way, the third estimation step is performed by searching for consumption data corresponding to a date with weather conditions similar to the estimated weather conditions. In another example, the third estimation is performed by comparing the temperature and / or humidity forecast for the given period with the temperature and / or humidity data from the historical data.In this example, the date with the most similar weather conditions is selected, and the consumption forecast corresponds to the consumption value of the selected date in the history.

[0097] The piloting method includes a fourth acquisition step ACQ4 of a first set of performance data ENS4. The first set of performance data ENS4 is obtained from at least one performance data source of at least one atmospheric water generation device GEAi. Advantageously, the performance data includes historical water production data from the atmospheric water generation device(s) GEAi. In one embodiment, the historical data includes at least one energy consumption data point and / or at least one water production flow rate data point for the device. Advantageously, the historical data points are time-stamped. Advantageously, the historical data points include at least one data point for the quantity of water produced.

[0098] The piloting method includes a fourth estimation step EST4 of a second performance dataset ENS4'. This step is sometimes referred to as the estimation of energy efficiency parameters 141 in this application. The second performance dataset comprises predictive data on the water production performance of the atmospheric water generation device(s) GEAi over the first predefined period. The second performance dataset ENS4' includes at least one performance indicator. The performance indicator defines at least one ratio between a quantity of water produced and a unit of electrical energy consumed over the first predefined period, or vice versa.

[0099] According to one embodiment, the fourth estimation step EST4 includes a copy of the historical data from the first performance dataset ENS4 into the second performance dataset ENS4'. Advantageously, the historical data includes performance data from the atmospheric water generation device on an anniversary of the current date. In one example, the The data included in the second performance dataset ENS4' comprises the performance data of the GEAi generation device, the performance of which is estimated on the anniversary of the current date, for example, the performance data one year prior to the current date. In one embodiment, the anniversary data corresponds to the device's performance data on a date one day and / or one week prior to the current date. In another embodiment, the performance estimation is performed by calculating a rolling average of the device's performance over a few days preceding the current date, for example, two, three, four, five, six, or seven days.

[0100] The method includes a step of generating GENi a first pilot indicator PILi over the first period Pi of at least one atmospheric water generation device GEAh. This GENi generation is performed from the first, second, third, and fourth estimates. The pilot indicator defines at least one power supply duration of the atmospheric water generation device GEAi. The indicator defines at least one power supply output of the atmospheric water generation device GEAi.

[0101] According to one embodiment, the control data is generated based on the previously determined consumption threshold. The control data is generated based on the estimated consumption threshold for water production to reach the threshold.

[0102] The method includes an electrical control step of at least one atmospheric water generation device.

[0103] Advantageously, the electrical control comprises the control by the device's computer of at least one electrical power generator and / or at least one electrical power supply to said device. In one embodiment, the control comprises a control electrical current or power and a supply duration.

[0104] According to one embodiment, a comparison, at least on an hourly basis, of the estimated water production with the actual water level produced, allows the operating program developed to be updated and corrected continuously.

[0105] According to one embodiment, said comparison is made every minute or fifteen minutes to obtain more precise estimates.

[0106] According to one example, the scheduling, intended to be implemented over a period WPe 200 equal to 54 hours for the production of an estimated quantity of water of 67 liters, will be reorganized over a shorter revised hourly period (WPe - 1) when the estimated value is less than the value actually produced, and this will continue until the user's need is met. Otherwise, the process stops the device during off-peak hours (i.e., hours that do not coincide with peak hours). functioning).

[0107] In one embodiment, the analyses and decision-making processes use at least one artificial intelligence model based on machine learning algorithms. For example, this model uses supervised reinforcement learning. Reinforcement learning is understood to be autonomous learning. This learning process involves the agent learning which actions to take in order to optimize a quantitative reward over time, through iterated experience. Thus, the sum of rewards over time is maximized. This method allows for self-correction, minimizes error margins, and accelerates corrective actions.

[0108] According to one embodiment, the central process optimally executes the interconnection of the different procedures, which can be successive, simultaneous or convergent.

[0109] According to one embodiment, the central process allows for dynamic and evolving decision-making to execute the various predictive procedures, analyze them and revise them.

[0110] Moreover, the objectives of this process include economic (e.g., optimization of the cost per liter produced), energy (e.g., rationalization of energy consumed while ensuring user satisfaction) and environmental (reduction of greenhouse gas emissions) purposes, making it possible to provide pure, safe and potable water in a sustainable manner.

[0111] Another advantage of this solution is that water is produced by striking a balance between user needs, the performance of the atmospheric water generation system, water availability, and the availability of energy sources, relative to each site. Each site can be subject to an optimized energy source of renewable (wind, solar, etc.), non-renewable, or hybrid type.

[0112] To do this, according to one embodiment, the central process considers different scenarios in order to automate the operation of the water generation device effectively and efficiently depending on whether the site connectivity is established or not and whether the site is connected to the electrical network or not.

[0113] According to one embodiment, the geolocation system relating to each device allows the geographical coordinates of the specific site to be automatically returned when it is connected to the internet.

[0114] According to one embodiment, the user defines the geographical coordinates of his site when said site is not connected to the internet.

[0115] According to one embodiment, the processing of meteorological data relating to each site is different depending on whether the region is connected to the internet network or not, as illustrated later.

[0116] According to one embodiment, when the connectivity test is successful, the electronic control system associated with the given atmospheric water generation device is connected to the Internet (e.g., via SIM card, or via Wi-Fi, etc.). Geographic coordinates are thus established. The availability of the API allows for the collection of meteorological and / or atmospheric estimates 125 from the estimates established through the central process.

[0117] According to one embodiment, each device may be equipped with one or more sensors (for example, digital or analog) for at least one characteristic of ambient air, enabling the measurement of at least one characteristic parameter of atmospheric conditions (for example, air temperature, humidity, dew point, pressure, sunshine, wind direction and speed, etc.). The data received and / or stored constitute a good database of historical weather conditions.

[0118] According to one embodiment, the central process uses historical weather and / or atmospheric conditions to estimate future weather databases using artificial intelligence through learning functions.

[0119] According to one embodiment, in order to quickly and reliably automate the operation of each atmospheric water generation device, it is necessary to estimate the meteorological and / or atmospheric data 123 with greater accuracy relative to the geographic database. In one example, the method compares the estimates of atmospheric data received by the APIs with those derived from historical data 127 in order to correct said estimates of meteorological and / or atmospheric data.

[0120] Indeed, in order to improve the estimates of operational parameters (such as, for example, predictions in water production and user needs), the process uses at least two meteorological and / or atmospheric databases.

[0121] The first database provides monthly and / or daily historical data over a given period, allowing for the compilation of statistics on trends in water availability, in other words, on water production potential relative to that period. For example, the statistics are based on averages corresponding to previous years (for example, the last year or the last 3, 5, or 10 years).

[0122] According to one embodiment of the invention, the process generates a typical meteorological year derived from a multi-year time series or collects such data from suitable sites. The process uses said multi-year time series for statistical purposes in water production potential to be used to estimate this potential in the medium and long term.

[0123] This first database presents daily or monthly averages of at least one of the characteristic air parameters allowing analysis of the trends in daily or monthly water production potential during an optimized period.

[0124] According to an example, this database could be: • Collected from sensors implanted in the atmospheric water generation system; • Collected from databases provided by neighboring atmospheric stations; • Received by weather forecasting platforms; • Stored in the system's memory card.

[0125] According to one embodiment, each electronic control system 50 can collect and / or store the characteristics of the study site (i.e., geographical coordinates and historical, instantaneous and forecast data of atmospheric conditions).

[0126] In one embodiment, a second database provides more detailed hourly weather forecasts. This database includes the history of recent databases in order to predict upcoming weather conditions over an optimized period. For example, this database estimates the weather conditions for the next few hours (48 h, 54 h, 72 h, 120 h, or even 168 h) based on the history of the last two weeks.

[0127] According to one embodiment, the weather forecasts informed from the APIs are compared with the forecasts from the hourly database 127, and then corrected 129, if necessary.

[0128] According to one embodiment, when the area is not connected to the Internet, the user can enter the site's geographic coordinates. The local (slave) process uses the historical data for at least one characteristic air parameter 121 of the specified site (such as ambient temperature, pressure, air humidity, dew point, etc.) to estimate the associated atmospheric forecasts 123. The historical meteorological data can be measured by one or more sensors previously installed in the atmospheric water generation device or from databases stored in the control system's memory card. The control system generates the meteorological data forecasts 123 using artificial intelligence, for example, by means of machine learning functions based on supervised machine learning.Similarly, the local process can use statistics compiled from daily and / or monthly historical data to identify trends in water production potential over the coming days and / or months.

[0129] According to one embodiment of the invention, the devices can be installed on sites equipped with an electrical network (connected to the national grid, cogeneration system, for example) or located on isolated sites (non-electrified but equipped with sources) renewable energies).

[0130] Devices installed on sites connected to the local electrical network can also be powered by hybrid energy sources integrating renewable energy sources (for example, solar energy, geothermal energy, etc.).

[0131] According to one embodiment, for sites connected to the electricity grid, no limit on the energy supplied is envisaged. Water production is then ensured while reducing energy consumption by automating the operation of each atmospheric water generation device. The autonomy of each of these devices is ensured by adjusting their operation to the optimal operating hours that allow for a competitive price per liter produced. Consequently, for this type of site, the only constraint will be water availability, since electricity is always available.

[0132] According to one embodiment, isolated sites require atmospheric water generation devices using autonomous energy systems such as renewable energy sources (solar, wind, etc.), non-renewable energy sources (generators, etc.), or hybrid sources. However, the intermittent nature of renewable energy sources generally requires additional storage systems to meet electrical load demands.

[0133] In one embodiment, the estimation of meteorological data (temperature, solar irradiance, wind speed and direction, for example) will be used, among other things, to determine the characteristic parameters of the electrical energy required. For example, for an isolated site, the renewable energy source used is solar energy via a photovoltaic system. The photovoltaic system comprises one or more photovoltaic modules, storage accumulators, i.e., batteries, an energy regulator, and a DC / AC conversion unit, if necessary.

[0134] The central process receives at least one meteorological parameter (solar irradiance data, for example) from at least one atmospheric water generation device and estimates the electrical energy 132 produced and stored, when batteries are used. In this case, the characteristics include estimates of solar irradiance, energy produced, and energy stored.

[0135] According to one embodiment, a hybrid solution can be proposed for said off-grid regions, comprising a photovoltaic solar installation, a storage unit, i.e., battery and / or a generator set, as a backup power source.

[0136] According to one embodiment, as output variables, the process calculates estimates of electrical energy produced 132, according to the nature of the site and the source of installed energy, based on at least one database of meteorological data estimates 120 associated with said site. The hourly estimate relating to said site, in particular, solar irradiation, uses either meteorological data via weather forecasting platforms 125, or estimates 123 using the historical data 121 from databases retrieved by each of the devices.

[0137] According to one embodiment, the weather estimates can be of the instantaneous, hourly or daily type.

[0138] According to one embodiment, the method calculates the energy estimates 132 by the photovoltaic modules, the energy stored by the storage units and / or the energy required as a backup by the use of a generator set, for example.

[0139] According to one embodiment, the calculation of the solar energy produced by the photovoltaic installation can be done, for example, using the following formula [Math 1]:

[0140] EEP = Pc * K * Irr

[0141] Where:

[0142] Pc: Installed peak power [kWp / kW.m2]

[0143] EEP: Electrical energy produced [kWh / day]

[0144] Irr: Solar irradiance [kWh / m2.d]

[0145] K: Conversion factor

[0146] According to one embodiment, the calculation of the solar energy stored by the batteries is done using the following formula [Math 2]:

[0147] Cbat = Ej*J / Kb

[0148] With:

[0149] Cbat: Battery capacity (kWh)

[0150] Ej: daily electrical energy requirements (kWh / day)

[0151] J: number of days of autonomy (days)

[0152] Kb: loss factor

[0153] According to one embodiment of the invention, the method is configured for estimating the performance 130 of each of the atmospheric water generation devices.

[0154] The performance of said system is understood to be the ratio between water production and associated electrical consumption, or vice versa. In what follows, the first ratio, in litres produced per kilowatt-hour consumed, will be used.

[0155] The process collects at least one database of the history 121 related to water production coupled with energy consumption and / or directly to the ratio of these two parameters of at least one atmospheric water generation device and delivers estimates 131 of at least one of these characteristics over an optimized period.

[0156] According to one embodiment, the process performs an estimation of the databases 131 of the water production / electricity consumption pair from the estimates in meteorological data established 120.

[0157] As illustrated previously, according to one embodiment, estimates of water production and user needs of each atmospheric water generation device are developed in the short, medium and long term.

[0158] According to one embodiment, the method allows the predicted system performance database to be compared to a reference model. The reference model can be a model derived from an experimental characterization of the (water production, electricity consumption) pair carried out for a wide range of meteorological conditions. For example, the characterization is performed for a wide range of air characteristics, for instance, for ambient air temperatures and relative humidity ranging, by way of example, from T (-10°C to 70°C) and RH (0% to 100%), respectively.

[0159] In general, such characterization is carried out at the time of development of the atmospheric water generation device.

[0160] An example of a reference model is represented by tables [Table 1] and [Table 2] for the pair (water production “p” and electrical consumption “q”), respectively, according to the weather conditions in (T, HR).

[0161] With:

[0162] HR; translates a relative humidity belonging to the range of ] 10% - 100% [, with an increasing step of 5%, for example.

[0163] Tj translates the ambient temperature belonging to the interval of ]-10 °C - 70 °C [ with an increasing step of 5 °C, for example.

[0164] For example, let T = 27 °C and RH = 76% • Tj = 25 °C < T < T2 = 30 °C and RH! = 75% < HR < HR2 = 80%

[0165] With pu = p(Tb HRJ and qn = q(Tb HRJ, etc.

[0166] [Table 1]: Experimental characterization of the water produced T (°C) RH(%) -10 Tj 70 0 Pu Pu Pl,n 10 P2,l P2j P2,n HR, Pi,l PiJ Pi,n 100 Pk,l Pkj Pk,n

[0167] [Table 2]: Experimental characterization of energy consumed T(°C) -10 Tj 70 HR(%) 0 qi.i qu qi,n 10 q2,i q2,j q2,n HR, qu qu n 100 qk,i qkj qk,n

[0168] According to one embodiment, the determination of the couple (water production, energy consumption) of the associated device, over the entire temperature / relative humidity range of the air, can be extended by bilinear interpolation according to the following formulas [Math 3] and [Math 4].

[0169] Calculation of water production

[0170] [Math 3]: . \TT, / \HR-HR, / \TT; HR-HR, p(T,HR) = (p21-p]J.^ + (p12-piJ.w^ + (pil + p22-p21-p12).Trn.ÏS?HSI+ pn

[0171] Calculation of electrical energy consumption

[0172] [Math 4]: q(T, HR) - (q21 -i / T;,-Tt + ^ii)'HR2-HR] + ^22-^-^12 / ^-^^2-^1 +

[0173] According to one embodiment, the reference model representing the performance of the system can be obtained by considering numerical simulations on the mass and enthalpy balances through the refrigeration and aerodynamic circuits of such devices.

[0174] In order to ensure greater reliability of the process, the predictions of the performance of the associated device, for each site, are compared to the data provided by the reference model with the same weather forecast data.

[0175] According to one embodiment, a calculation of error 135 is provided using the following equation [Math 5]:

[0176] [Math 5] : error=lxref-xmodeil / xref

[0177] Where:

[0178] xmodei and xref are related to the output parameters (water produced, electrical energy consumed) corresponding to the prediction model and the reference value, respectively.

[0179] The output parameters of the prediction model and the reference model are compared to verify the operating status of each device. The error calculation is crucial for judging the device's performance. If the error value exceeds a certain threshold, the process alerts users 137 to the possibility of failures or malfunctions of the device in question. Otherwise, the process validates the value obtained 139 of the device's performance estimate. According to a For example, the aforementioned threshold corresponds to a 50% error. In another example, the aforementioned threshold corresponds to an estimated 20% error. In yet another example, the aforementioned threshold corresponds to an 80% error.

[0180] According to one embodiment, the process compares the performance of several devices distributed in the same territory or neighboring territories, and deduces, according to learning functions, the state of operation of each of the devices.

[0181] According to one embodiment, the process is configured to predict user needs 150.

[0182] The process makes it possible to estimate water consumption 157 by gathering a database of the history of water consumption behavior of users 121 in relation to the associated device.

[0183] According to one embodiment, the predictions of user needs relate to a database focused on consumer profiles of each device.

[0184] According to one embodiment, the predictions take into account user consumption trends and profiles associated with each of the devices (for example, number of users, user activity, frequency of use, ambient temperature, relative humidity of the air).

[0185] Since consumption habits vary, weather forecasts, and therefore long-term forecasts of water production potential, allow the process to adapt to the variability of weather conditions, and therefore, to the variability in water production.

[0186] In this case, this embodiment is characterized in that it is configured to address potential water shortages for sites with variable and / or seasonal weather characteristics.

[0187] The coarse meteorological database (monthly and / or daily) provides monthly and / or daily statistics and trends. These statistics allow for medium- and long-term forecasting of user needs and therefore of the quantity of water to be produced.

[0188] The hourly meteorological database makes it possible to estimate the potential for water produced in the short term.

[0189] Depending on the operating state of each device and the associated meteorological estimates, each forecast water production corresponds to a quantity of energy consumed.

[0190] The process is configured to generate statistics and to estimate the missing quantities of water consumption in the medium and long term. These missing quantities of water relate to months in which the average monthly potential water production is below the estimated value of user needs. Therefore, the process can anticipate future water needs and plan accordingly. water storage when energy efficiency is guaranteed.

[0191] According to one embodiment, the process is configured to plan the mode of estimating required water consumption according to at least two scenarios: the first mode prioritizes the learning functions to estimate the water needs of users 157, otherwise to switch to a mode associated with a setpoint value defined by each user 153.

[0192] According to one embodiment, the process is configured to plan the mode of estimating required water consumption according to at least two scenarios: the first mode prioritizes the learning functions to estimate the water needs of users 157 (economic mode), otherwise to focus on maximizing water production 153.

[0193] In one embodiment, the method includes a step of comparing the performance of the atmospheric water generation device under the two scenarios. The method then includes a step of notifying a user of a performance value for each scenario. In one embodiment, the notification includes an indication of the scenario with the highest performance.

[0194] According to one embodiment, the process will alert the user to the non-satisfaction of the need in the short, medium or long term if the setpoint defined 153 cannot be reached over the desired period.

[0195] According to one embodiment, the process is configured to automate the operation of each of the atmospheric water generation devices autonomously. As mentioned previously, the start-ups and shutdowns of each of these devices are governed by procedures including estimations of meteorological and / or atmospheric parameters 120, estimation of parameters related to the performance of the device 130, estimation of user needs 150 and estimation of the availability of energy sources.

[0196] According to one embodiment, in order to meet user needs effectively and efficiently, the process gathers the various necessary databases 161, identifies the optimal period 163 to satisfy user needs, sorts the minimum operating hours with decreasing energy efficiency ratio 165, and finally, plans the operating program of each water generation device 167.

[0197] The operating hours of each of the atmospheric water generation devices can be continuous or discontinuous in time.

[0198] According to one embodiment, the method will check whether the current time coincides with a scheduled operating time or not 182. If this time turns out to be an operating time, the procedure will check the availability of the different energy sources, for an isolated site, taking into account the successive availabilities of solar energy 183, battery storage state 185, or backup by a generator set 186.

[0199] The method is configured to start each of the devices 184 according to the availability of energy sources, otherwise, to stop the device if the current time corresponds to a shutdown time 195.

[0200] According to one embodiment, the method will continuously verify and adjust the established instructions, that is to say, the different estimates leading to the estimates in the automated operating program of each device.

[0201] According to one embodiment, the process will reiterate the forecast parameters 200 to execute a revised operating program if the actual water level produced in the reservoir is lower than the estimated water production level. The process will stop the operation of the associated device based on the scheduled off-peak hours when the actual water level produced is greater than or equal to the estimated water volume.

[0202] According to one embodiment, for sites without network connectivity, each electronic control system can itself control the associated atmospheric water generation device.

[0203] According to one embodiment, each electronic control system uses edge computing to reduce latency and provide protection against internet outages. The edge computing provided by this control system allows data to be stored and processed closer to the users and the applications that consume it.

[0204] Thus, in order to control the associated device via a dynamic and evolving analysis using learning functions, each process (slave) also includes the estimation procedures for: • atmospheric and / or meteorological data 120; • data on energy production 130; • performance of the atmospheric water generation device 140; • water consumption 150; • the planning of operating hours 160; • availability of energy source 180.

[0205] Therefore, the invention relates to a dynamic control and command method enabling efficient and economical satisfaction of atmospheric water supply, optimization of the cost per liter produced, and correction of associated operational parameters based on the supervised artificial learning method. Nomenclature:

[0206] GEAi: atmospheric water generation device

[0207] ACQi: Acquisition of a first set of meteorological data

[0208] ACQ2: Acquisition of a data set for the quantification of energy

[0209] ACQ3: Acquisition of a first set of consumption data

[0210] ACQ4: Acquisition of a first set of performance data

[0211] ESTi: estimation of a second meteorological dataset

[0212] EST2: estimation of a second dataset quantifying an energy available

[0213] EST3: estimation of a second set of consumption data

[0214] EST4: Estimation of a second performance data set

[0215] GENi: Generation of a first performance indicator

[0216] PILi: electrical control of the atmospheric water generation device

[0217] ENSi: first meteorological dataset

[0218] ENSi': second meteorological dataset

[0219] ENS2: energy quantification data set

[0220] ENS2': second dataset quantifying available energy

[0221] ENS3: first set of consumption data

[0222] ENS3': second set of consumption data

[0223] ENS4: first performance data set

[0224] ENS'4: second performance data set

[0225] BDCi: source of consumption data for a set of infrastructures

[0226] Pi: first predefined period

[0227] INDpi: first performance indicator

[0228] INDP2: Second Performance Indicator

[0229] INDP3: Third Performance Indicator

[0230] INDpm1: Measured performance indicator

[0231] Ci: atmospheric pressure sensor

[0232] C2: humidity sensor

[0233] C3: temperature sensor

[0234] C4: anemometer

[0235] BD1: first monthly and / or daily meteorological database

[0236] BD2: Second hourly meteorological database

[0237] BATI: first energy source from an energy storage component

[0238] SOL1: second energy source from an electrical energy generator

[0239] RES1: source of energy data available on an electrical network

[0240] MLI: first machine learning model

[0241] DATAi: first training dataset

[0242] DATA2: second training dataset

[0243] DATA3: third training dataset

[0244] 130 Quantity of available energy

[0245] 50: Electronic control system

[0246] 51: Power supply

[0247] 53: Control and acquisition unit

[0248] 55: Display unit

[0249] 57: Sensors

[0250] 58: Control level

[0251] 59: Electrical components

[0252] 100 Central process

[0253] 120 Meteorological data estimation procedure

[0254] 121 Parameter history

[0255] 123 Estimation of meteorological data from historical data

[0256] 125 Retrieval of atmospheric forecasts

[0257] 127 Comparison of atmospheric data

[0258] 129 Correction of atmospheric data

[0259] 130 Procedure for estimating electrical energy produced

[0260] 132 Calculation of electrical energy estimates

[0261] 140 Device performance calculation procedure

[0262] 141 Estimation of energy efficiency parameters

[0263] 143 Comparison between estimated values ​​and reference values ​​

[0264] 145 Error testing

[0265] 147 Alertuser

[0266] 149 validation device performance estimation

[0267] 150 user needs estimation procedure

[0268] 151 Adjust user needs

[0269] 153 Define user needs

[0270] 155 user profile databases

[0271] 157 user needs estimation

[0272] 160 operating hours estimation procedure

[0273] 161 retrieval of estimates

[0274] 163 optimization of the operating period

[0275] 165 Sorting of operating hours

[0276] 167 development of the operating program

[0277] 180 start-up procedure according to energy availability

[0278] 182 operating hours test

[0279] 183 Solar energy availability test

[0280] 184 Device start-up

[0281] 185 Stored Energy Availability Test

[0282] 186: availability of backup power source

[0283] 187: increment test time

[0284] 190: test procedure

[0285] 195: Device shutdown

[0286] 200: upgrade of the operating period

[0287] 301: microcontroller

[0288] 303: microprocessor

[0289] 305: connectivity modules

[0290] API: Application Programming Interface

[0291] BLE: Bluetooth

[0292] EEC: Electrical energy consumed

[0293] EEP: Electrical energy produced

[0294] EEf: energy efficiency ratio

[0295] Irr: Solar irradiation

[0296] HRe: relative humidity

[0297] HUC: User consumption history

[0298] GPS: Global Positioning System

[0299] Tre: temperature

[0300] Pre: pressure

[0301] WPr: water production

[0302] WPE: optimal period

[0303] Wdi: wind direction

[0304] WSp: wind speed

Claims

1. Demands A method for controlling at least one atmospheric water generation device (AWG) characterized in that it comprises: • First acquisition (ACQi) of a first meteorological dataset (ENSi) from: • at least one meteorological data source and / or • a history of time-stamped and / or meteorological data • a current date; • First estimate (ESTi) of a second meteorological dataset (ENSi') over a first predefined period (PJ) using a calculator and from the first meteorological dataset (ENSi), said estimated meteorological data including: • A forecast of atmospheric temperature over a given period; • A humidity forecast over a given period; • Second acquisition (ACQ2) of an energy quantification dataset (ENS2) from an energy data source of at least one power supply equipment; • Second estimation (EST2) of a second energy quantification dataset (ENS2'), said dataset (ENS2') quantifying an available energy over the first predefined period (Pi); • Third acquisition (ACQ3) of a first set of consumption data (ENS3) from: • at least one source of consumption data (BDCi) from a set of water infrastructure and consumption profiles; • Third estimate (EST3) of a second set of consumption data (ENS'3) over the first predefined period, the second set of consumption data including at least one consumption threshold; • Fourth acquisition (ACQ4) of an initial set of

2. performance data (ENS4) from: • At least one performance data source from at least one atmospheric water generation (AWG) device; • Fourth estimate (EST4) of a second set of performance data (ENS'4) over the first predefined period, the second set of performance data including at least one forecast of a performance indicator, said performance indicator defining at least a ratio between a quantity of water produced and a unit of electrical energy consumed and / or a water production flow rate; • Generation (GENi) of a first control indicator (PILi) on the first period (Pi) of at least one atmospheric water generation device (GEAi) from the first, second, third and fourth estimates (ESTb EST2, EST3, EST4), said control indicator (PILi) defining at least one control data sent to the atmospheric water generation device (GEAi), said control data defining at least one electrical supply duration and / or at least one electrical supply power of said at least one atmospheric water generation device (GEAi); • Electrical control (PILi) of said at least one atmospheric water generation device (GEAi) as a function of the first control indicator (PILi). A method according to claim 1, characterized in that at least one source of meteorological data is a source originating from: • From an atmospheric pressure sensor (Ci) and / or • A humidity sensor (C2); and / or • A temperature sensor (C3); and / or • An anemometer (C4); and / or • A first monthly and / or daily meteorological database (BDi); and / or • A second hourly meteorological database (BD2); and / or • A third archived meteorological database (BD3).

3. A method according to claim 1 characterized in that the available energy data source of at least one power supply equipment relates to: • A first energy source (BATi) from an energy storage component; and / or • A second energy source (SOLO) from an electrical power generator; and / or • an available energy data source on an electrical network (RESi).

4. Method according to claim 1 characterized in that when a second energy source (SOLi) powers the atmospheric water generation device, the estimation of the amount of energy available from the second energy source is carried out from the estimates of the calculated meteorological data (ESTi).

5. The method according to claim 3 characterized in that a second performance indicator (INDP2) is generated, said second performance indicator (INDP2) defining a ratio between the quantity of water produced and one unit of electrical energy consumed from the second energy source (SOLi).

6. A method according to claim 1 characterized in that a third performance indicator (INDP3) is generated, said third performance indicator (INDP3) defining a ratio between the quantity of water produced and a unit of electrical energy consumed from an electrical network (RESÛ.

7. Method according to claim 1 characterized in that it comprises estimating a volume of water produced by a plurality of atmospheric water generation devices ({AWG).

8. A method according to any one of the preceding claims comprising a step of measuring, by at least one sensor of at least one atmospheric water generation device (AWG) of at least one water production flow rate and at least one electrical power consumption, the method further comprising a step of generating an alert indicator when the measured flow rate is below a threshold of predefined flow rate.

9. A method according to claim 1 characterized in that it comprises executing at least one first machine learning model (MLi) trained from a first training data set (DATAi) comprising meteorological data, consumption data, energy quantity data, and electrical regime data from at least one atmospheric water generation device (GEAi), said machine learning model (MLi) being configured to generate a performance indicator (INDPi)

10. )■ Method according to claims 8 and 9 wherein the measurement of the production flow rate and the electrical energy consumed by the atmospheric water generating device (GEAi) are included in the first set of drive data (DATAi).

11. A method according to claim 1 characterized in that it comprises the execution of at least one second machine learning model (ML 2) trained from a second training data set (DATA2) comprising on the one hand estimates of meteorological data, estimates of consumption data and estimates of energy quantity data and on the other hand real meteorological data, real consumption data and real energy quantity data, said machine learning model (ML2) being configured to generate estimates from real data.

12. The method according to claim 1 characterized in that it comprises the execution of at least one third machine learning model (ML3) trained from a third training data set (DATA3) comprising meteorological data and performance data of the atmospheric water generation device (GEAi), the third data set comprising control parameters, such as a degree of opening and / or closing of an electronic expansion valve of the water generation device (GEAi), a frequency of a fan of the water generation device GEAb, a frequency of a compressor of the water generation device (GEA1), the control parameters being associated with the meteorological data and the performance data, the third machine learning model being configured to generate a performance indicator.

Images