Reverse osmosis system

Abstract

And a reverse osmosis system. A reverse osmosis system (100) comprising: a first conductivity sensor (1) for measuring the conductivity (L1) of water (2) supplied to the reverse osmosis system (100); a second conductivity sensor (3) for measuring the conductivity of a permeate (4; 4 ') produced by the reverse osmosis system (100); 23) of the electrical conductivity (L2); and an AI unit (5) designed to use a statistical model as a basis to measure the electrical conductivity (L1) of the water (2) supplied to the reverse osmosis system (100) and to measure the permeate (4; 23) produced by the reverse osmosis system (100) to be recirculated is calculated and set accordingly, by means of the electrical conductivity (L2) of the concentrate (6; 20), which has been trained by means of training data.

Description

[0001] DE 4239867 A1 discloses a method for treating a liquid by means of an apparatus comprising a membrane module including a part of the recycled concentrate according to the principle of reverse osmosis, the method comprising continuously measuring the salt concentration of the permeate flowing out of the membrane module and achieving an increase in production by setting the recycle ratio of the concentrate at such a level that reaches but does not exceed a specified allowable limit of the salt concentration in the permeate.

[0002] The object of the present invention is to provide a reverse osmosis system that automatically sets an optimal production based on a target conductivity value of the permeate.

[0003] The reverse osmosis system comprises: a first conductivity sensor for measuring the conductivity of the water supplied to the reverse osmosis system or to a filter of the reverse osmosis system, the filter comprising a membrane; a second conductivity sensor for measuring the conductivity of the permeate produced by the reverse osmosis system; and an artificial intelligence (AI) unit. The AI unit is designed to calculate and accordingly set the ratio or production of the concentrate produced by the reverse osmosis system to be recycled based on the measured conductivity of the water supplied to the reverse osmosis system and based on the measured conductivity of the permeate produced by the reverse osmosis system using a statistical model as a basis, the statistical model having been trained with training data.

[0004] The training data may include, for example, a large number of training data sets, where a corresponding training data set contains at least the conductivity of the water supplied to the reverse osmosis system, the conductivity of the permeate produced by the reverse osmosis system, and the associated optimal production. The training data sets differ in terms of the conductivity of the water supplied to the reverse osmosis system and / or in terms of the conductivity of the permeate produced by the reverse osmosis system. The training data can be determined / generated empirically and / or by a model.

[0005] The AI unit may be further designed to calculate and accordingly set the ratio or production of the concentrate produced by the reverse osmosis system to be recycled based on the measured conductivity of the water supplied to the reverse osmosis system, based on the measured conductivity of the permeate produced by the reverse osmosis system, and also based on the target conductivity value of the permeate produced by the reverse osmosis system using a statistical model as a basis.

[0006] In one embodiment, the reverse osmosis system further comprises: a temperature sensor for measuring the temperature, in particular the temperature of the permeate. The AI unit is further designed to calculate and accordingly set the ratio of the concentrate produced by the reverse osmosis system to be recycled based on the measured conductivity of the water supplied to the reverse osmosis system, based on the measured conductivity of the permeate produced by the reverse osmosis system, and based on the measured temperature using a statistical model as a basis.

[0007] In one embodiment, the AI unit is further designed to use a statistical model as a basis to calculate and accordingly set the proportion of the concentrate produced by the reverse osmosis system to be recycled based on the reverse osmosis system parameters of the reverse osmosis system.

[0008] In one embodiment, the reverse osmosis system parameters are selected from a set of reverse osmosis system parameters, i.e., they include at least one reverse osmosis system parameter from the following set: (one or more) overflow factors; the opening interval and / or the degree of opening of the discharge valve (i.e., the valve that controls the volumetric flow rate of the drained water); the pump speed of the pump of the reverse osmosis system, especially the pump speed that affects the volumetric flow rate of water through the reverse osmosis system; the power consumption of the pump of the reverse osmosis system; the volumetric flow rate of the permeate; the pressure of the water supplied to the filter containing the membrane; the retention capacity of the membrane; and the conductivity of the water upstream of the membrane.

[0009] In one embodiment, the AI unit is designed to compare the measured conductivity of the permeate produced by the reverse osmosis system with a target conductivity value and update the statistical model so as to minimize the difference between the measured conductivity of the permeate produced by the reverse osmosis system and the target conductivity value.

[0010] The present invention allows for determining or recommending the optimal water production in a reverse osmosis system through machine learning. The water production here indicates the amount of drained water produced per product / permeate. A typical optimal water production can be, for example, between 50% and 95%.

[0011] Regarding the term "production" used in this application, there are other terms in use, such as WCF (Water Conversion Factor), recovery rate, system production, etc.

[0012] The operation of a reverse osmosis system typically requires specifying the production. The production here describes the ratio of the permeate (product) to the consumed water or drained water. The production can be specified by a formula, for example, as follows:

[0013]

[0014] where VP is the volumetric flow rate of the permeate, and VF is the so-called volumetric flow rate of the feed water. The feed water refers to the water supplied to the (filtering) membrane. The volumetric flow rate of the feed water here corresponds to the sum of the volumetric flow rate of the permeate and the volumetric flow rate of the drained water.

[0015] The production must be below 100% because the reverse osmosis process retains ions from the feed water, and without partially discharging the ion-rich drained water, these ions would precipitate on the membrane, which is called fouling. Over time, fouling leads to a decline in the performance of the reverse osmosis system or a reduction in the quality of the permeate.

[0016] The manufacturer of the system has specified a range of typical production settings. However, the specific production settings are left to the operator of the reverse osmosis system. This is especially because the supplied water or feed water (soft water) may differ in terms of conductivity (i.e., ion concentration). Therefore, more drain water must be discharged at locations with high ion concentration compared to locations with low ion concentration.

[0017] Since the reverse osmosis system is subject to various influences, such as temperature, ion concentration, permeate production, hydraulic structure, customer-specific parameter settings, etc., it is difficult to make generally valid statements about the production to be set. The present invention solves this problem by an automatic AI-based production setting according to the relevant operating parameters of the reverse osmosis system. The conductivity of the supplied water or soft water can be, for example, between 100 - 2000 μS / cm, while the typical conductivity of the permeate can be, for example, between 1 - 30 μS / cm.

[0018] The present invention will be described in detail below with reference to the accompanying drawings. Here:

[0019] Figure 1 A reverse osmosis system according to a first embodiment is shown in a highly schematic form,

[0020] Figure 2 A reverse osmosis system according to a second embodiment is shown in a highly schematic form, and

[0021] Figure 3 A reverse osmosis system according to a third embodiment is shown in a highly schematic form.

[0022] Figure 1 The reverse osmosis system 100 according to the first embodiment is shown. Soft water 2 flows into the storage tank 13 via the solenoid valve 12. At the same time, the soft water 2 is continuously measured by the first conductivity sensor 1 for the conductivity L1 of the soft water 2 without ion specificity. The pipeline shown above the storage tank 13 conveys the unconsumed permeate back to the storage tank 13.

[0023] Via the pump 9, the tank water and a certain proportion of the concentrate 6 are introduced as feed water 25 together into the filter 15 including at least one (filter) membrane 11. The produced permeate 4 is measured by the second conductivity sensor 3 for the conductivity L2 of the produced permeate 4.

[0024] In the shown embodiment, the filter 15 includes the (filter) membrane 11. It is to be understood that the filter 15 may include more than one (filter) membrane 11. In addition, multiple filters 15 may be connected in series or in parallel.

[0025] The concentrate 6 generated during the filtration process is recycled via the pump 10 and thus becomes part of the feed water 25, or flows out of the reverse osmosis system 100 as the drain 14 via a solenoid valve or a drain valve 8. The amount of the drain 14 depends on the set production rate. Here, the production rate can be defined as the ratio between the volumetric flow rate (numerator) of the permeate 4 and the volumetric flow rate (denominator) of the feed water 25.

[0026] Regarding the above features, reference is also made to the relevant technical literature.

[0027] According to the present invention, based on the conductivity or ion concentration L1 and L2 measured by the conductivity sensors 1 and 3, the AI unit 5 automatically sets or classifies the production rate or the proportion RA of the concentrate 6 generated by the reverse osmosis system 100 to be recycled, based on a statistical model, through machine learning.

[0028] Here, the production rate or the proportion RA to be recycled is automatically set by the AI unit 5 by appropriately controlling the pump 10 and the drain valve 8, such that the conductivity L2 of the permeate 4 is maintained at an adjustable level, for example, between 1 - 30 μS / cm. The achievement of the correct setting is checked in the permeate 4 via the conductivity sensor 3.

[0029] If the classification of the production rate or the proportion RA to be recycled does not result in the desired conductivity L2 of the permeate 4, the AI unit 5 gradually changes the production rate or the proportion RA to be recycled until the desired conductivity L2 is reached. Then, the newly created data points are provided to the classification algorithm in order to update the statistical model accordingly.

[0030] In addition to the two conductivities L1 and L2, other variables can also be evaluated by the AI unit 5 for automatically setting the production rate or the proportion RA to be recycled. One example thereof is the temperature T of the permeate 4, which is measured, for example, by the temperature sensor 7. The temperature T has a direct influence on the conductivity L2 of the permeate 4 and thus also on the production rate or the proportion RA to be recycled. It is known that increasing the temperature T results in lower ion retention, which in turn results in a higher conductivity L2 of the permeate 4.

[0031] Furthermore, the system and the classification may be affected by the parameters of the reverse osmosis system. Examples thereof are the (one or more) overflow factors, the opening intervals of the drain valve 8, or other measured values, such as the speeds of the pumps 9 and 10, the power consumption of the pumps 9 and 10, the power consumption of the entire system, etc.

[0032] The acceptable constant level of the conductivity L2 of the permeate 4, i.e., the target conductivity value, is adjustable by the user. The target conductivity value results in the adaptation of the classification or statistical model. A higher target conductivity value shifts the production rate to a higher value.

[0033] In addition to the solenoid valve 8, other types of valves that produce a continuous and adjustable drain flow can also be used, such as motor control valves and the like.

[0034] Figure 2 A two-stage reverse osmosis system 100' according to the second embodiment is shown in a highly schematic form. Here, the first stage basically corresponds to Figure 1 the reverse osmosis system 100 shown in. Downstream of the first stage is the second stage, which further processes the permeate 4 of the first stage and outputs an additional permeate 23. The second stage includes: pumps 16 and 17 and a filter 18 including a (filtering) membrane 24. The drain 20 from the second filter 18 flows back to the storage tank 13 via the valve 19 and / or is supplied to the second filter 18 via the pump 17.

[0035] Here, the production rate or the proportion RA of the first stage to be recycled and the proportion of the second stage to be recycled are automatically set by the AI unit 5, with or without interacting with a conventional controller for controlling the pump(s) and valve(s), by appropriately controlling the pump 10 and the discharge valve 8 and by appropriately controlling the pump 17 and the valve 19, such that the conductivity L2 of the permeate 23 is maintained at an adjustable level.

[0036] Figure 3 A two-stage reverse osmosis system 100'' according to the third embodiment is shown in a highly schematic form. Here, the first stage basically corresponds to Figure 1 the reverse osmosis system 100 shown in. Downstream of the first stage is the second stage, which further processes the permeate 4 of the first stage and outputs an additional permeate 23. The second stage includes: pumps 16 and 17 and a filter 18 including a (filtering) membrane 24. The drain 22 from the second filter 18 is discharged via the discharge valve 21.

[0037] Here, the production rate or the proportion of the two stages to be recycled is automatically set by the AI unit 5 by appropriately controlling the pumps 10 and 17 and the discharge valves 8 and 21, such that the conductivity L2 of the permeate 23 is maintained at an adjustable level.

Claims

1. A reverse osmosis system (100), comprising: - a first conductivity sensor (1) for measuring the conductivity (L1) of the water (2) supplied to the reverse osmosis system (100), - a second conductivity sensor (3) for measuring the conductivity (L2) of the permeate (4; 23) produced by the reverse osmosis system (100), and - an AI unit (5) designed to calculate and accordingly set the proportion (RA) of the concentrate (6; 20) produced by the reverse osmosis system (100) to be recycled, based on the measured conductivity (L1) of the water (2) supplied to the reverse osmosis system (100) and based on the measured conductivity (L2) of the permeate (4; 23) produced by the reverse osmosis system (100), using a statistical model as a basis, the statistical model having been trained by means of training data.

2. The reverse osmosis system (100) according to claim 1, characterized in that: The reverse osmosis system (100) further comprises: a temperature sensor (7) for measuring the temperature (T), in particular the temperature of the permeate (4; 23), - wherein the AI ​​unit (5) is further designed to use a statistical model as a basis for calculating and accordingly setting the proportion (RA) of the concentrate (6; 20) produced by the reverse osmosis system (100) to be recycled based on the measured conductivity (L1) of the water (2) supplied to the reverse osmosis system (100), based on the measured conductivity (L2) of the permeate (4; 23) produced by the reverse osmosis system (100) and based on the measured temperature (T).

3. The reverse osmosis system (100) according to any one of the preceding claims, characterized in that The AI ​​unit (5) is further designed to use the statistical model as a basis to calculate and accordingly set the proportion (RA) of the concentrate (6; 20) produced by the reverse osmosis system (100) to be recycled, depending on the reverse osmosis system parameters of the reverse osmosis system (100).

4. The reverse osmosis system (100) according to claim 3, characterized in that - The reverse osmosis system parameter is selected from the set of reverse osmosis system parameters: -(one or more) overflow factors, - the opening interval and / or the degree of opening of the discharge valve (8; 21), - the pumping speed of the pumps (9, 10) of the reverse osmosis system (100), - the power consumption of the pumps (9, 10) of the reverse osmosis system (100), - volume flow rate of the permeate (4; 23), - the pressure of the water (2) supplied to the filter (15) comprising the membrane (11), - the retention capacity of the membrane (11), and - The electrical conductivity of the water upstream of the membrane (11).

5. The reverse osmosis system (100) according to any one of the preceding claims, characterized in that - the AI ​​unit (5) is designed to compare the measured conductivity (L2) of the permeate (4; 23) produced by the reverse osmosis system (100) with a target conductivity value and to update the statistical model in order to minimize the difference between the measured conductivity (L2) of the permeate (4; 23) produced by the reverse osmosis system (100) and the target conductivity value.

Images