Method and device for quality control of a lightly loaded liquid in a system comprising membrane filtration

Abstract

The invention relates to a method for monitoring a liquid filtration system, the filtration system comprising at least one membrane filter and a sensor configured to measure a value of a quantitative performance parameter of the interaction between the membrane filter and the liquid. After obtaining a plurality of sensor values, values ​​of the plurality of values ​​characterizing the quantitative performance of the interaction between the filter and the liquid at different times, and obtaining a signature based on the plurality of values ​​obtained, a qualitative state of the liquid is obtained as a function of the signature obtained. (See Figure 2 for an abbreviation.)

Description

Title of the invention: Method and device for quality control of a lightly loaded liquid in a system comprising membrane filtration TECHNICAL FIELD OF THE INVENTION

[0001] The field of the invention relates to the control of the quality of a lightly loaded liquid in a system comprising membrane filtration, for example for controlling the filtration of water in a basin such as a swimming pool comprising a membrane filter. STATE OF THE ART

[0002] Depending on the nature and use of a pool, its water may be filtered and, if necessary, treated to meet its needs. Thus, for example, to ensure the comfort and safety of users, the water in a swimming pool is generally filtered and treated, in particular to prevent the proliferation of bacteria and viruses and to ensure the chemical neutrality of the water. Such treatment essentially consists of eliminating foreign bodies, for example residues from the decomposition of leaves, and diffusing chemicals into the water.

[0003] The addition of chemicals can be carried out using diffusers comprising these chemicals. For example, these can take the form of pebbles which decompose over time and release the chemicals. The number of pebbles placed in the diffusers is adjusted, in particular, according to the size of the pebbles, their formulation, the volume of water to be treated and conditions such as the temperature of the water. There are also chemical diffusion systems which calculate the quantity of chemicals to be released depending on the water, by measuring, for example, its chemical content, its pH and its temperature.

[0004] Filtration is typically performed by circulating water through a filtration system using a pump. The pump is often started manually by a user or automatically according to predetermined time slots. There are different types of filtration.

[0005] Frequently, filtering is carried out using a sand filter comprising a multifunction valve and a tank containing sand, often in several layers of different granulometries, to filter impurities from the water. Although it is very simple to operate and maintain, such a filter has disadvantages, particularly with regard to the size of the filtered particles (the filtration fineness is approximately 40 microns), the filtration quality which is difficult to control and the quantity of water rejected during cleaning of the filter. During this operation, called backwashing (or backwash in Anglo-Saxon terminology), the pollutant load is partially evacuated by reversing the direction of the water in the filter and the dirty water is discharged into the sewer.

[0006] Sand can be replaced by glass, which reduces the formation of biofilm (a substance created from impurities that tends to clog the sand). The use of glass reduces the frequency of backwashing and improves the filtration fineness (around 15 microns). Zeolites can also be used, on the same principle, to further improve the filtration fineness (to around 5 microns). However, this material is not compatible with certain treatments.

[0007] There are also diatomaceous earth filters (grains obtained from the crushing of fossil rock). For this type of filtering, the filter comprises a cylindrical tank containing a membrane or a filter frame on which the suspended diatoms stick when filtration is in progress. Due to a high risk of clogging, the filter must be washed frequently. During backwashing, the diatomaceous earth grains are sent to the sewer. It is therefore necessary to refill them. Furthermore, although they improve the filtration fineness (to around 3 to 5 microns), diatomaceous earth grains are not compatible with certain treatments.

[0008] There are also cartridge filters. For this type of filtering, the filter comprises a cylindrical tank containing one or more removable cartridges. Each cartridge is made of a synthetic or vegetable membrane folded in a star shape around a rigid perforated tube. The cartridges can be removed for cleaning to avoid clogging. Such cartridges offer a filtration fineness of approximately 20 microns. They are not compatible with certain treatments.

[0009] Finally, there has recently been membrane filtration, in which the water to be filtered is pumped under pressure through a semi-permeable membrane. This filtration, which is still poorly understood and not very widespread, offers a filtration fineness of around 5 microns. It allows for consistent filtration quality, but its performance in terms of the volume of water filtered is variable.

[0010] Although these solutions have proven their effectiveness, they are not entirely satisfactory, particularly in terms of filtration quality, water discharge and / or filter material, filter management, etc. There is therefore a need to improve and better control the filtration of basins, particularly swimming pools. Statement of the invention

[0011] The invention thus relates, in a first aspect, to a method for monitoring a liquid filtration system, the filtration system comprising at least one membrane filter and at least one sensor configured to measure a value of a quantitative performance parameter of an interaction of said membrane filter and of said liquid, the method comprising the following steps: • acquisition of a plurality of values ​​of said at least one sensor, values ​​of said plurality of values ​​characterizing a quantitative performance of the interaction of said filter and said liquid at different times; • obtaining a signature based on said plurality of values ​​obtained; and • estimating a qualitative state of said liquid based on said signature obtained.

[0012] The invention makes it possible in particular to detect and characterize a clogging of the membrane filter, depending on the quality of the filtered liquid, in order, for example, to make it possible to carry out an appropriate action in order to ensure a filtration flow rate adapted to the needs of the system comprising the filter.

[0013] Preferred, simple, convenient and economical characteristics of the device according to the invention are presented below.

[0014] For example, the method further comprises a step of estimating an action to be performed.

[0015] According to embodiments, the method further comprises a step of validating or correcting said estimated qualitative state.

[0016] Still according to particular embodiments, the method further comprises a step of identifying one or more reference signatures, said qualitative state being estimated as a function of an identified reference signature.

[0017] Said qualitative state is for example estimated according to at least one rule of an expert system. Said signature is for example obtained using at least one artificial intelligence engine.

[0018] The invention also relates, in a second aspect, to a system for estimating the quality of a liquid, the system comprising at least one membrane filter and at least one sensor configured to measure a value of a quantitative performance parameter of an interaction of said membrane filter and said liquid, the system further comprising at least one data processing unit configured to receive a plurality of values ​​of said at least one sensor, values ​​of said plurality of values ​​characterizing a quantitative performance of an interaction of said filter and said liquid at different times, to obtain a signature based on said plurality of values ​​obtained and to estimate a qualitative state of said liquid as a function of said signature obtained.

[0019] The invention makes it possible in particular to detect and characterize a clogging of the membrane filter, depending on the quality of the filtered liquid, in order, for example, to make it possible to carry out an appropriate action in order to ensure a filtration flow rate adapted to the needs of the system comprising the filter.

[0020] Preferred, simple, convenient and economical features of the system according to the invention are presented below.

[0021] For example, said at least one sensor comprises at least one flow meter, one pressure sensor, one turbidimeter or one spectrometer.

[0022] According to particular embodiments, the system comprises a pump for circulating the liquid in said membrane filter. Still according to particular embodiments, said at least one sensor comprises at least one rotation speed sensor of said pump or a power sensor consumed by said pump. BRIEF DESCRIPTION OF THE FIGURES

[0023] Other advantages, aims and particular characteristics of the present invention will emerge from the following non-limiting description of at least one particular embodiment of the devices, system and methods which are the subject of the present invention, with reference to the appended drawings, in which: • [Fig.l] is a schematic view of an environment comprising a basin, allowing the implementation of the invention according to particular embodiments; • [Fig.2] illustrates an example of a filtration system implementing certain embodiments of the invention, which can be used to filter the water of the pond shown in [Fig.l] and monitor the quality of the water; • [Fig.3] illustrates a first example of steps in a process for determining and monitoring the evolution of the quality of the filtered water in real time as well as for determining the evolution of the performance of the filtration system; • [Fig.4] illustrates a second example of steps of a method for determining and monitoring the evolution of the quality of the filtered water in real time as well as for determining the evolution of the performance of the filtration system; and • [Fig.5] illustrates an example of a calculator capable of implementing a method for determining and monitoring the evolution of the quality of the filtered water in real time as well as for determining the evolution of the performance of the filtration system according to particular embodiments. DETAILED DESCRIPTION OF THE INVENTION

[0024] The objective of a filtration system is to retain elements suspended in the liquid to be filtered. Concerning a swimming pool, the objective of the filtration system is to retain the pollution which arrives in the pool in order to maintain the quality of the bathing water, it being noted that the origin of the pollution is multiple (pollution linked to the presence of bathers, rainwater, the contribution of dust and dead leaves by the wind, etc.).

[0025] When a membrane filter is used in a filtration system, the liquid to filtering passes through pores in the membrane, which allows for a constant quality of the filtered liquid regardless of the incoming pollutant load. During filtration, the pollution retained by the membrane creates a layer on the surface of the membrane, which partially clogs it, then completely if it is not cleaned.

[0026] The inventors have observed that characteristics of this clogging and unclogging, for example a clogging index and a clogging / unclogging speed or curve, are representative of the nature of the pollution and / or its source.

[0027] Thus, depending on the characteristics of the clogging, for example its evolution over time, it is possible to determine the characteristics of the pollution and to identify actions to be taken to combat the pollution and / or optimize the use of the filtration system.

[0028] For these purposes, data representative of the clogging of the membrane, for example data representing a variation in flow rate and pressure, over time, between the inlet and the outlet of the filtration system, can be used to identify a type of pollution and, where appropriate, measures to be taken to limit the pollution and / or clean the filter. According to embodiments, this data is processed to calculate a signature which can be compared to signatures in a database to which a type of pollution, a list of one or more actions, etc. are associated. It is thus possible, in a swimming pool, to detect the installation of a robot, the opening or closing of a shutter, the supply of rainwater or even the lack of oxidant in the water, from data characterizing the behavior of the membrane filter.

[0029] Thus, according to embodiments, the invention relates to a real-time hybrid sensor based on the use of a membrane filtration system and one or more physical measurement sensors representative of a behavior of the membrane filtration system. This hybrid sensor makes it possible to determine and monitor in real time the evolution of the quality of the filtered liquid as well as the impact of this on the evolution of the performance of the filtration system. Still according to particular embodiments, the physical measurements measured are used to calculate a signature which can be compared to signatures of known events to make it possible to determine a type of pollution causing a loss of performance of the filtration system (clogging), the nature of the degradation of the quality of the filtered water, to adapt the filtration operation, etc.

[0030] The physical measurements representative of a behavior of the membrane filtration system are for example values ​​of a quantitative performance parameter of an interaction of the membrane filter and the filtered liquid, which can represent a quantity (or a ratio) of liquid passing through the filter, a quantity of particles, bacteria, etc., retained by the membrane filter, etc.

[0031] The invention thus offers numerous advantages, including: - real-time adaptation of the membrane process management based on variations in the quality of the liquid to be filtered, - avoidance of irreversible clogging of the membrane during a sudden increase in the pollutant load of the liquid to be filtered, requiring washing with washing agents (which generates toxic discharges and water consumption), - an improvement in filtration by avoiding (or slowing down) the reduction in its performance due to membrane clogging, - detection of the non-functioning, in the filtration environment, of certain equipment having an effect on the quality of the liquid to be filtered, - a reduction in the volume to be filtered in the case of batch applications, for example in a swimming pool (the invention allows dynamic management of the filtration, depending on a deviation from expected performance, unlike a filtration duration generally determined solely according to the temperature of the water to be filtered) and - the provision of information enabling phenomena to be identified by recognizing their signature.

[0032] By way of illustration, the following description essentially relates to the filtration of water in a basin such as a swimming pool. However, this example is not limiting. The invention finds applications in many other fields, for example in the food, pharmaceutical, cosmetic and chemical industries.

[0033] [Fig. 1] is an example of an environment comprising a swimming pool, allowing the implementation of the invention according to embodiments.

[0034] As illustrated, the swimming pool and its surroundings, referenced 100, comprise a pool 102 surrounded by a border 104 here comprising coping stones. The pool is accessible by a staircase 106. The perimeter of the pool comprises a space 108 arranged to allow the circulation of people enjoying the pool and / or ensuring maintenance. The pool, when not in use, may be covered by a shutter 110 or a safety cover. By way of illustration, this shutter or safety cover is here shown partially open. In addition to its role in terms of safety, the shutter or cover has a thermal and protective role to limit pollution of the pool water.

[0035] The pool water is maintained by a treatment and filtration system (not shown). Such a system is generally placed in a pool shelter 112, located a few meters or tens of meters from the pool and connected to it via pipes 114. This system may also include means for heating the pool water, for example a heat pump. This system may be connected to a server remote 122 via a communication network 124, for example to allow monitoring of certain elements or parameters of the swimming pool, for example the filtration performance, and / or remote control of certain parameters of the swimming pool such as its temperature, its level of chemical agents (eg, chlorine, salts, etc.), the activation ranges of the filtration and the cleaning of the filter, etc.

[0036] [Fig. 2] illustrates an example of a filtration system 200 implementing certain embodiments of the invention, which can be used to filter the water of the pond 102 shown in [Fig. 1] and monitor the quality of the water.

[0037] As illustrated, a pump 202 makes it possible to circulate the water from the pool through a membrane filter 204, by sucking it in through one or more skimmers and discharging it through one or more discharge drains (not shown). The membrane of the filter used here is a multi-channel inorganic microfiltration membrane. The membrane filter 204 is for example a filter using multi-layer membranes of recrystallized silicon carbide (R-SiC) carried by a unique honeycomb monolith geometry. This is for example the Crystar filter from the company Saint-Gobain (Crystar and Saint-Gobain are trademarks).

[0038] The filtration system 200 further comprises a data processing device 206 enabling the acquisition and transmission of data. According to certain embodiments, the data processing device 206 further enables the processing of acquired data, for example to transform them into one or more signatures and / or to analyze them. The data processing device 206 can receive data from one or more sensors of the same type or of different types, for example sensors 208, 210 and 212. These sensors are for example sensors of rotation speed of the pump 202, amperage consumed by the pump 202, flow meters, pressure sensors, turbidimeters, spectrometers, etc. They enable data representative of the behavior of the membrane filter to be obtained, directly (e.g., flow rate or pressure) or indirectly (e.g., rotation speed of the pump or power consumed).

[0039] The sensors implemented are connected to the data processing device 206, wired and / or wirelessly, for example using a local communication protocol of the WiFi or Bluetooth type (WiFi and Bluetooth are trademarks).

[0040] According to the example illustrated in [Fig. 2], the acquired data, processed or not, are transmitted to the remote server 214, via the communication network 216, to be analyzed, for example compared to previously acquired data with which a diagnosis, one or more actions, etc., stored here in the database 218, are associated. In other embodiments, the data are analyzed locally in the data processing device 206 or in another device, for example the computer 220. The acquired data can in particular be processed to generate one or more signatures which can be compared to previously stored signatures and to which particular characteristics such as a diagnosis and / or particular actions have been associated.

[0041] The data are for example compared in the form of signatures. Examples of signatures are described below. The diagnosis(es) can then be displayed on a terminal, for example the computer 220 or the tablet 222.

[0042] If one or more particular actions are associated with a signature close to a signature characterizing a behavior of the membrane filter, these actions can be triggered. This may be, for example, backwashing of the membrane, adding a chemical agent to the pool water, closing or opening a shutter or a protective cover, regulating a heating system and / or triggering a pool filling / draining system. The triggering of actions may be automatic, manual (for example from the tablet 222) or semi-automatic (the action(s) are performed after validation by a user).

[0043] Data exchanges between the different devices can be carried out using wired or wireless communication means, for example wireless communication means of the GPRS type (acronym for global packet radio service in English terminology), 3G, 4G, SigFox or LoRa (SigFox and LoRa are trademarks).

[0044] [Fig.3] illustrates an example of steps of a method for determining and monitoring the evolution of the quality of the filtered water in real time as well as for determining the evolution of the performance of the filtration system.

[0045] As illustrated, a first step here has as its object the acquisition of data representative of the behavior of the membrane filter used (step 300). These data are, for example, a water flow rate, a water pressure, a pressure drop in the membrane filter, a rotation speed of the pump used, an instantaneous consumption (for example electrical) of the pump used, etc. These data can be acquired at an interval of between 10 milliseconds (ms) and 60 seconds (s) or more. The frequency of data acquisition can vary according to the nature of the data measured. For example, a water flow rate or a pressure drop can be measured according to an interval of between 10 milliseconds and 1 second while the temperature or the pH can be measured, in the membrane filter, according to an interval of between 1 and 10 minutes. It is observed that the frequency of data acquisition can also vary according to the circumstances.For example, a water flow rate can be measured at a frequency of between 10 and 60 measurements per minute during a filtration phase and at a frequency of between 10 and 100 measurements per second during a membrane filter unclogging phase.

[0046] In a next step, the acquired data is processed to calculate (or train) a signature (step 305). According to a particularly simple embodiment, a signature is a simple concatenation of the parameters measured over time, according to a sliding window. Such a signature can be represented in the form of a matrix comprising as many rows as there are measured parameters and as many columns as there are measurements to be taken into account (for example 60 columns representing a one-minute observation window for data measured at a frequency of one measurement per second). According to other embodiments, certain measurements are averaged over time or weights are assigned to the measurements to give more importance to recent measurements than to older measurements. Still according to particular embodiments, weights can be assigned to the different measured parameters.

[0047] By way of illustration, a signature can be formed with a pressure measured at the inlet of the membrane filter, a pressure measured at the outlet of the membrane filter and an electrical power consumed by the pump. A weight equal to one can be assigned to the measured flow rates and a weight equal to one half can be assigned to the electrical power consumed. Furthermore, a weight equal to one can be assigned to all the measurements made during the last ten seconds, a weight equal to one half can be assigned to all the measurements made during the previous ten seconds, a weight equal to one third can be assigned to all the measurements made during the previous ten seconds, and so on, for a measurement window equal, for example, to one minute.

[0048] According to a particular embodiment, the signature is determined using an artificial intelligence-based system, for example a neural network, used to identify the most relevant parameters and measurement times.

[0049] Still according to certain particular embodiments, a signature can be corrected to take into account certain elements, for example characteristics of the environment. By way of illustration, a signature can be corrected as a function of the temperature of the liquid to be filtered to correct a drift which would be linked to the viscosity of this liquid.

[0050] In a following step, the calculated signature is compared with a set of reference signatures previously calculated and stored in a database 315 in order to identify the closest signature(s) (step 310). The database 315 may be a local or remote database. It may be specific to the basin considered, common to basins of the same type or generic. It may also be a generic database or common to basins of the same type, enriched with signatures specific to the basin considered.

[0051] The reference signatures are, for example, signatures corresponding to previously identified situations and for which a diagnosis could be established and / or for which actions to be taken could be identified. These diagnoses and / or actions to be taken are stored in association with the corresponding signatures, in the same database or in a separate database. A confidence index, for example a value between zero and one, can be associated with each diagnosis and / or action to be taken.

[0052] The comparison between the calculated signature and the reference signatures is for example based on a distance calculation, for example using the Euclidean distance.

[0053] The diagnosis and / or the actions to be taken associated with the signature closest to the calculated signature or with the signatures closest to the calculated signature are then obtained (step 320). The diagnosis can be presented to a user or an operator, for example on a display screen such as a tablet screen, who can thus take note of it. The confidence level, if it exists, is preferably also indicated. Optionally, the user or the operator can further validate it or, on the contrary, indicate an error. If an error is detected, the user or the operator can enter their diagnosis. The database can then be updated, for example by adjusting the confidence level (e.g., by incrementing it if it is validated or by decrementing it if it is incorrect) or by adding the calculated signature with the entered diagnosis.Similarly, an action to be taken associated with a signature can be presented, validated or corrected.

[0054] In a following step, one or more actions associated with one or more reference signatures considered to be closest to the calculated signature are executed (step 330), for example by controlling actuators. This may involve, for example, backwashing the membrane or adding a chemical agent to the pool water. As described previously, the triggering of such actions may be automatic, manual or semi-automatic (for example after validation by a user or an operator).

[0055] [Fig.4] illustrates a second example of steps of a method for determining and monitoring the evolution of the quality of the filtered water in real time as well as for determining the evolution of the performance of the filtration system.

[0056] As illustrated, the raw data from the sensors implemented can be preprocessed (step 400). This optional step can have the purpose of normalizing the measurement values, for example to code them in a particular format, for example on 8 bits, and / or to filter values ​​which would be outside predetermined ranges.

[0057] In a following step (step 405), the raw data or some of the raw data (preprocessed or not) are transmitted to one or more artificial intelligence engines to identify signatures, preferably predetermined. As illustration, a first set of raw data (preprocessed or not), for example a change in water pressure at the inlet and outlet of the membrane filter, is sent to a first artificial intelligence engine to determine a first signature, a second set of raw data (preprocessed or not), for example a change in the rotation speed of the motor and its electrical consumption, which may also include results from the first artificial intelligence engine, is sent to a second artificial intelligence engine to obtain a second signature and a third set of raw data (preprocessed or not), for example flow rate and turbidity data, is sent to a third artificial intelligence engine to obtain a third signature. Other implementations are possible.

[0058] It is observed here that the artificial intelligence engine(s) implemented are capable of using past usage to predict future usage. These are predictive systems. It is also observed that the learning of the artificial intelligence engine(s) can be completed during their use, making it possible to improve the quality of the learning.

[0059] The results obtained from the artificial intelligence engine(s) are transmitted to an expert system (step 410), alone or with raw data (preprocessed or not), to determine and monitor the evolution of the quality of the filtered water in real time, to determine the evolution of the performance of the filtration system and / or to determine actions to be carried out. For these purposes, the expert system uses predetermined rules stored in a rule base 415. The rules are for example determined by an expert according to his knowledge or automatically, for example by an artificial intelligence engine, from a data history. By way of illustration, the expert system used may be of the KBES type (acronym for knowledge based expert System in English terminology), the inference engine being able, for example, to be a Bayesian system.

[0060] The artificial intelligence engine(s) and the expert system may be implemented in a local device, for example the computer 220 in [Fig. 2], or in a remote server, for example the server 214 in [Fig. 2].

[0061] The results obtained from the expert system make it possible to control one or more actuators (step 420) to perform actions such as backwashing the membrane, adding a chemical agent to the pool water, closing or opening a shutter or protective cover, regulating a heating system and / or triggering a pool filling / draining system. An alarm may also be triggered (step 425) to draw the attention of a user or operator to a particular risk.

[0062] The results of the artificial intelligence engine(s) and the expert system may be stored in a local and / or remote database (e.g. database 430) constituting a history of water quality and filtration system performance.

[0063] [Fig. 5] illustrates an example of a computer 500 capable of implementing a method for determining and monitoring the evolution of the quality of the filtered water in real time as well as for determining the evolution of the performance of the filtration system according to particular embodiments. The computer 500 is for example a PC type computer (acronym for personal computer in English terminology).

[0064] As illustrated, the computer 500 comprises a power supply 525 providing the electrical energy necessary for the components of the computer 500. It further comprises one or more communication buses, shared or not, to which are connected:

[0065] - a central processing unit or microprocessor 505 (CPU, acronym for central Processing unit in Anglo-Saxon terminology) further comprising, preferably, one or more coprocessors, for example of the GPU type (acronym for graphical processing unit in Anglo-Saxon terminology), making it possible to accelerate calculations (for example by parallelizing them), in particular calculations of an artificial intelligence engine;

[0066] - a random access memory or cache memory 510 (RAM, acronym for random access memory in Anglo-Saxon terminology) comprising registers adapted to record variables and parameters created and modified during the execution of programs implementing the steps described previously;

[0067] - a 515 read-only memory (ROM, acronym for read only memory in terminology Anglo-Saxon) which may include an operating system and programs implementing the steps described above;

[0068] - a storage medium 520, fixed or removable, which may in particular comprise rules used by the expert system and / or which can be used to store results from the artificial intelligence engine(s) and / or the expert system;

[0069] - an input interface 530 for selectively receiving data from sensors; and

[0070] - an output interface 535 for controlling one or more actuators. The interfaces Input and output may or may not be separate. For example, it may be a standard input / output interface.

[0071] The computer 500 further comprises, preferably, a communication interface 540 connected to a communication network, for example a wireless communication network and / or a local communication network, the interface being capable of transmitting and receiving data, in particular to or from another of the servers, computers, tablets and / or smartphones.

[0072] Optionally, the computer 500 can also have a human- machine 545, comprising for example a display or a touch display allowing a user to interact with programs implemented by the computer 500, and input means such as a keyboard and / or a mouse allowing a user to interact with programs implemented by the computer 500.

[0073] The communication bus allows communication and interoperability between the different elements included in the computer 500 or connected to it. The representation of the bus is not limiting and, in particular, the central processing unit is capable of communicating instructions to any element of the computer 500 directly or via another element of the computer 500.

[0074] The executable code of the programs allowing the computer 500 to implement, in whole or in part, the method according to the invention, can be stored, for example, in the read-only memory 515. According to a variant, the executable code of the programs can be received via the communication network, via the interface 540, to be stored in a manner identical to that described previously. More generally, the program(s) can be loaded into one of the storage means of the computer 500 before being executed.

[0075] The central processing unit 505 will control and direct the execution of the instructions or portions of software code of the program(s) according to the invention, instructions which are stored, for example, in the read-only memory 515 or in the other aforementioned storage elements. When the power is switched on, the program(s) which are stored in a non-volatile memory, for example the read-only memory 515, are transferred into the random access memory 510 which then contains the executable code of the program(s), as well as registers for storing the variables and parameters necessary for implementing the method according to the invention.

[0076] The calculator 500 can be installed near a swimming pool or remotely, for example in a third-party installation responsible for maintaining the swimming pool. It can also be split into several elements, some of which can be installed near the swimming pool and others remotely. Similarly, part of the processing and / or calculations can be carried out in or near the sensors. By way of illustration, recognition of people can be carried out in a camera, the information being transmitted in addition to or instead of images to a swimming pool control system. Still by way of illustration, the artificial intelligence module can be implemented remotely and the expert system can be implemented locally.

[0077] Of course, the present invention is not limited to the embodiments described above as examples. It extends to other variants.

[0078] Depending on the embodiment chosen, certain acts, actions, events or functions of each of the methods described herein may be performed or occur in a different order than that in which they were described, or may be added, merged, or may not be performed or occur, as the case may be. Furthermore, in certain embodiments, certain acts, actions, or events are performed or occur concurrently and not successively.

[0079] Although described through a number of detailed exemplary embodiments, the proposed device, system, and method include various variations, modifications, and improvements that will be apparent to those skilled in the art, it being understood that these various variations, modifications, and improvements are within the scope of the invention, as defined by the claims that follow. In addition, various aspects and features described above may be implemented together, or separately, or substituted for each other, and all various combinations and subcombinations of the aspects and features are within the scope of the invention.Further, some systems and equipment described above may not incorporate all of the modules and functions described for the preferred embodiments.

Claims

Claims

1. A method for monitoring a liquid filtration system, the filtration system comprising at least one membrane filter and at least one sensor configured to measure a value of a quantitative performance parameter of an interaction of said membrane filter and said liquid, the method comprising the following steps; acquiring a plurality of values ​​of said at least one sensor, values ​​of said plurality of values ​​characterizing a quantitative performance of the interaction of said filter and said liquid at different times; obtaining a signature based on said plurality of values ​​obtained; and estimating a qualitative state of said liquid as a function of said signature obtained.

2. The method of claim 1, further comprising a step of estimating an action to be performed, the action to be performed being appropriate to ensure a filtration flow rate adapted to the needs of the filtration system.

3. A method according to claim 1 or claim 2, further comprising a step of validating or correcting said estimated qualitative state.

4. Method according to any one of claims 1 to 3, further comprising a step of identifying one or more reference signatures, said qualitative state being estimated as a function of an identified reference signature.

5. Method according to any one of claims 1 to 3, according to which said qualitative state is estimated according to at least one rule of an expert system.

6. Method according to any one of claims 1 to 5, according to which said signature is obtained using at least one artificial intelligence engine.

7. System (200) for estimating the quality of a liquid, the system comprising at least one membrane filter (204) and at least one sensor (208, 210, 212) configured to measure a value of a quantitative performance parameter of an interaction of said membrane filter and said liquid, the system further comprising at least one data processing unit (206, 220, 214) configured to receive a plurality of values ​​of said at least one sensor, values ​​of said plurality of values ​​characterizing a quantitative performance of an interaction of said filter and said liquid at different times, to obtain a signature based on said plurality of values ​​obtained and to estimate a qualitative state of said liquid as a function of said signature obtained.

8. The system of claim 7, wherein said at least one sensor comprises at least one flow meter, a pressure sensor, a turbidimeter or a spectrometer.

9. A system according to claim 7 or claim 8, further comprising a pump (202) for circulating liquid through said membrane filter.

10. System according to claim 9, according to which said at least one sensor comprises at least one sensor of rotation speed of said pump or a sensor of power consumed by said pump (202).

Images