CN116813028A

CN116813028A - Method for intelligently detecting service life of filter element of water purifier

Info

Publication number: CN116813028A
Application number: CN202310803334.1A
Authority: CN
Inventors: 赵凯; 韦承佐; 胡第平; 张建芳; 刘锶鸿
Original assignee: Shenzhen Angel Drinking Water Equipment Co Ltd
Current assignee: Shenzhen Angel Drinking Water Equipment Co Ltd
Priority date: 2023-06-30
Filing date: 2023-06-30
Publication date: 2023-09-29

Abstract

The application provides a method for intelligently detecting the service life of a filter element of a water purifier. The water purifier comprises a booster pump and a filter element. The method comprises the following steps: collecting a current value of the booster pump; collecting a TDS value of a water purifying end of the filter element; and predicting the service life of the filter element by using the current value and the TDS value through a trained filter element service life prediction model. The method can save the service life cost of the detection filter element and ensure the accuracy.

Description

Method for intelligently detecting service life of filter element of water purifier

Technical Field

The application relates to an intelligent detection method, in particular to a method for intelligently detecting the service life of a filter element of a water purifier.

Background

Along with the improvement of living standard in the new era, the requirements of people on drinking water are higher and higher, and the intelligent water purifier also gradually goes into the vision of people. However, the detection of the life of the filter element of the intelligent water purifier is complicated. Factors such as ambient temperature, water quality in different regions, water pressure and the like can greatly influence the prediction of the service life of the filter element.

The life detection mode of the filter element in the current market mainly comprises the following steps: (1) determining the service life of the filter element according to the empirical parameters by measuring the working time of the water pump; (2) determining the service life of the filter element by detecting the TDS value of purified water; (3) detecting the front-back pressure change of the RO filter element through a pressure sensor to determine the service life of the filter element; (4) the life of the cartridge is determined by a flow sensor. Although the service life of the filter element can be detected by all the four detection methods, the calculation mode of the first detection method is too fuzzy, so that the detection results of different water quality areas are greatly different, and the conclusion is not accurate enough; the second detection method only considers the TDS value, but does not consider insoluble substances in water, and meanwhile, the TDS error is very large, and the calculated life error is also very large; the third detection method has higher accuracy of detection results than the first detection method, but has too high detection cost, and is not suitable for wide popularization and use; the fourth detection method has high cost and also has the problem that the detection of the flow sensor is inaccurate when the flow rate of the purified water is lower than 200 ml/min. In summary, the existing filter element life detection method has the problems of inaccurate detection or high detection cost.

Disclosure of Invention

The application aims to provide a method for detecting the service life of a filter element of a water purifier, which can realize more accurate prediction of the residual service life of the filter element of the water purifier.

According to one aspect of the application, a method for intelligently detecting the service life of a filter element of a water purifier is provided, the water purifier comprises a booster pump and a filter element, the method comprises the steps of collecting a current value of the booster pump and a TDS value of a water purifying end of the filter element, and then predicting the service life of the filter element through a trained filter element service life prediction model by utilizing the current value and the TDS value.

According to some embodiments, the method includes collecting TDS values of the water quality, and selecting a corresponding cartridge life prediction model to predict cartridge life based on a range of the obtained TDS values.

According to some embodiments, the filter cartridge life prediction model comprises a multiple linear regression model.

According to some embodiments, the method includes pre-collecting information during use of different filter cartridges and training the filter cartridge life prediction model using the collected information.

According to some embodiments, the method includes turning on the booster pump after receiving a tap discharge command, and then collecting a current value of the water pump.

According to some embodiments, the method includes collecting a TDS value of the cartridge water purification end, and collecting the TDS value of the cartridge water purification end with a TDS sensor.

According to some embodiments, the method includes obtaining a filter element cumulative usage time and inputting the cumulative operation time with the current value and the TDS value into the filter element life prediction model to predict filter element life.

According to some embodiments, the method includes uploading a filter element life prediction result to a cloud end through a WIFI module, and then prompting the filter element life on a mobile terminal application program.

According to some embodiments, the method includes alerting the filter element after determining that the filter element has reached its useful life.

According to some embodiments, the water purifier further comprises: booster pump: pressurizing the water to ensure the purified water yield; a filter element: filtering tap water into purified water; and (3) an electric control board: the current value and the TDS value of the booster pump are collected, data are input into the model, the service life of the filter element is predicted, the data are uploaded to the cloud end through WIFI, and a user can know the service condition of the filter element in real time through APP or a small program.

According to the method for detecting the service life of the filter element of the water purifier, the service life of the filter element is predicted through the filter element service life prediction model by utilizing the current value of the booster pump and the TDS value of the TDS sensor, so that the residual service life of the filter element can be more accurately determined while the cost is kept low. Further, according to some embodiments, model selection may be made based on local water quality, so that cartridge life predictions may be made more accurately.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application as claimed.

Drawings

Other objects, features and advantages of the present application will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings. The drawings described below are only some of the embodiments of the present application and are not intended to limit the present application.

FIG. 1 illustrates a flow chart of a method of predicting filter cartridge life in accordance with an exemplary embodiment of the application.

FIG. 2 illustrates a flow chart of a method of predicting filter cartridge life in accordance with another exemplary embodiment of the application.

FIG. 3 shows a flowchart of a method of training a multiple linear regression model according to an exemplary embodiment of the application.

Fig. 4A illustrates a block diagram of a water purifier system according to an exemplary embodiment of the present application.

Fig. 4B illustrates a waterway diagram according to an example embodiment of the present application.

Fig. 5 shows a block diagram of an electronic control board system according to an example embodiment of the application.

Detailed Description

Example embodiments will now be described more fully with reference to the accompanying drawings. However, the exemplary embodiments can be embodied in many forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of the example embodiments to those skilled in the art. The same reference numerals in the drawings denote the same or similar parts, and thus a repetitive description thereof will be omitted.

The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the disclosure. One skilled in the relevant art will recognize, however, that the disclosed aspects may be practiced without one or more of the specific details, or with other methods, components, materials, devices, or the like. In these instances, well-known structures, methods, devices, implementations, materials, or operations are not shown or described in detail.

The flow diagrams depicted in the figures are exemplary only, and do not necessarily include all of the elements and operations/steps, nor must they be performed in the order described. For example, some operations/steps may be decomposed, and some operations/steps may be combined or partially combined, so that the order of actual execution may be changed according to actual situations.

The terms first, second and the like in the description and in the claims and in the above-described figures are used for distinguishing between different objects and not necessarily for describing a sequential or chronological order. Furthermore, the terms "comprise" and "have," as well as any variations thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those listed steps or elements but may include other steps or elements not listed or inherent to such process, method, article, or apparatus.

Those skilled in the art will appreciate that the drawings are schematic representations of example embodiments and that the modules or flows in the drawings are not necessarily required to practice the application and therefore should not be taken to limit the scope of the application.

In the existing prediction method of the service life of the filter element of the water purifier, the problems of high cost or insufficient accuracy exist. Therefore, the application provides a novel intelligent water purifier filter element service life detection method, which is used for jointly predicting the water purifier filter element service life by collecting the booster pump current value and the water TDS value, so that the cost can be reduced, and the detection precision can be improved. Further, according to some embodiments, a predictive model may be selected based on local water quality ranges to improve accuracy of cartridge life estimates.

The technical scheme according to the embodiment of the present application will be described in detail below with reference to the accompanying drawings.

Referring to fig. 1, a method flow for intelligently detecting a life of a water purifier cartridge according to an exemplary embodiment of the present application includes: collecting a TDS value S101 of a water purifying end of the filter element; collecting a booster pump current value S103; and predicting filter cartridge life S105.

Referring to fig. 1, according to an example embodiment, at S101, a current value of the booster pump is acquired.

The booster pump is generally used for pressurizing water in the water purifier, utilizes the low air pressure of the large-area piston to generate high hydraulic pressure of the small-area piston, and realizes the operation of liquid in a cyclic reciprocating mode.

According to some embodiments, after receiving a tap water discharge instruction, a booster pump is turned on, and then a current value of the booster pump can be collected.

According to some embodiments, the current value of the booster pump may be acquired by a current detection circuit. In the prior art, the service life of the filter element is generally predicted only through the TDS value, the booster pump current is not utilized, and the accuracy is not high enough. In the application, the current value of the booster pump is used as an input index of the filter element life model, so that the defect that the filter element life is incorrect by simply using the TDS value can be overcome, and the prediction accuracy can be improved.

And S103, collecting the TDS value of the water purifying end of the filter element.

TDS (Total Dissolved Solids), i.e. total dissolved solids, refers to the concentration of total dissolved substances in water in milligrams per liter. The TDS value mainly reflects the relationship between the concentration of calcium, magnesium and potassium ions in water and the hardness and conductivity of water.

According to some embodiments, the TDS value may reflect the filtering effect of the body of water, such that the life of the filter element may be predicted, but with insufficient accuracy. In the present application, the prediction accuracy is improved by the common use of the TDS value and the booster pump current value.

In S105, the filter element life is predicted by a trained filter element life prediction model using the current value and the TDS value. For example, the cartridge life prediction model may be a multiple linear regression model.

According to some embodiments, the TDS value of the incoming water may also be collected, and then a corresponding filter element life prediction model may be selected to predict filter element life based on a range of TDS values of the incoming water, as will be described later.

According to some embodiments, information during use of different cartridges may also be pre-collected and used to train the cartridge life prediction model, as will be described in detail later. The pre-trained model can be stored in a chip of the water purifier when the water purifier leaves the factory, and can be downloaded or updated through a cloud network.

According to some embodiments, a filter element cumulative usage time may also be obtained and then input into the filter element life prediction model along with the current value and the TDS value to predict filter element life.

In addition, according to the method of the embodiment, the filter element service life prediction result can be uploaded to the cloud end through the WIFI module, the service life of the filter element is prompted on the mobile terminal application program, and an alarm prompt is performed after the filter element is judged to reach the service life, and the filter element service life prediction method is not repeated herein.

According to an example embodiment, the filter cartridge life prediction model is a multiple linear regression model. The estimated residual service life of the filter element can be obtained by inputting the current of the booster pump and the TDS value into a multiple linear regression model. When the residual service life of the filter element is close to or lower than 0, the filter element can be considered to be required to be replaced, and a filter element replacement prompt is sent to a user.

Compared with other schemes which only use TDS values and/or accumulated working time as reference data, the technical scheme of the embodiment adds current value detection, so that the data of a prediction model can be more perfect, and the result of predicting the service life of the filter element is more accurate.

Referring to fig. 2, a method flow of predicting filter cartridge life according to an example embodiment of the application includes: tap water S201; the first TDS sensor acquisition value is brought into the test model S203; the booster pump boosts and collects the current S205 of the water pump; the second TDS sensor collects TDS data S207 of the water purifying end; the master control chip collects the system running time S209; the data is stored for subsequent use S211.

According to some embodiments, after the faucet is turned off, the information is transmitted to the main control chip, and the main control chip starts to send a next instruction to the filter life prediction.

The master control chip may be an MCU (Microcontroller Unit; MCU). The MCU is also called a single chip microcomputer (Single Chip Microcomputer) or a single chip microcomputer, which properly reduces the frequency and specification of a central processing unit (Central Process Unit; CPU), and integrates peripheral interfaces such as a memory (memory), a counter (Timer), USB, A/D conversion, UART, PLC, DMA and the like, and even LCD driving circuits on a single chip to form a chip-level computer for different application occasions to perform different combination control.

At S203, the first TDS sensor acquires a TDS value.

According to an example embodiment, after measuring the TDS value of the water flow, the first TDS sensor transmits data to the main control chip, and the main control chip selects a model corresponding to the TDS value according to a pre-stored TDS range comparison table.

At S205, the booster pump receives the instruction, and starts the boosting operation on the water flow.

According to an example embodiment, after the booster pump increases the water flow pressure to a required value, the water flow is continuously flowed forward, and meanwhile, the water pump current is measured according to the current circuit and is transmitted to the main control chip to store data.

At S207, the second TDS sensor collects a clean water side TDS value.

According to an example embodiment, the TDS sensor transmits this value to the host chip for storage as model input data.

In S209, the MCU host chip collects the system run time.

According to an example embodiment, after system run time is collected, it is stored in memory for subsequent model-based cartridge life estimation.

At S211, the data is stored by the main control chip for subsequent use.

According to an embodiment, when the predicted usage time is about to reach the upper limit, a prompt signal is displayed and sent out through a display screen.

Thus, according to the exemplary embodiment, the value measured by the first TDS sensor may also reflect the local water quality during the use of the water purifier, so that a more suitable prediction model may be selected, and the prediction result may be more accurate. The data prediction model can be different according to different water quality conditions of multiple areas, and the service life of the filter element is calculated and predicted by selecting a proper data model according to the local water quality conditions, so that the service life of the filter element of the water purifier under the condition of different water quality conditions of different areas is more accurate.

Referring to fig. 3, a multiple linear regression model derivation flow according to an exemplary embodiment of the present application includes: collecting data S301; data preprocessing S303; defining a loss function S305; the partial derivatives are assembled into a system of equations and solved S307 and a multiple linear regression model is determined S309.

Referring to fig. 3, in S301, a certain amount of data is collected, including the water pump current level, TDS value, filter cartridge usage time, etc.

In S303, data processing is performed to divide the data into a training set and a test set according to a certain ratio, and standardized processing is performed on the data.

For example, we can use 80% of the data as training set and 20% as test set. For the pump current magnitude and TDS values, we can scale their values to between 0 and 1 using a normalization process.

At S304, a model is built.

According to an example embodiment, the model training is to train a multiple linear regression model using a training set to learn the relationship between the water pump current magnitude, TDS value, etc. and the filter cartridge life. Examples of multiple linear regression models can be found in the following formulas.

y＝w0+w1x1+w2x2+...+wnxn+ε

Wherein:

w0 is the intercept;

w1, w2,..wn is the weight of the independent variables x1, x2, …, xn;

y is the predicted value of the dependent variable (cartridge life);

epsilon is the error term.

For the present example embodiment, the independent variable may be the pump current magnitude and TDS value variation, and the dependent variable is the cartridge life. Assume that the training model has m training samples: ((x 1, x1, 2), y 1), ((x 2,1, x2, 2), y 2), ((xm, 1, xm, 2), ym), where xi,1 represents the pump current level of the ith training sample, xi,2 represents the TDS value change of the ith training sample, and y represents the corresponding cartridge life.

In S305, a loss function is defined.

According to an example embodiment, the loss function is one defined during training, such as a Mean Square Error (MSE), and the least squares algorithm minimizes the loss function. The model training aims to learn the relation between the current of the water pump, the TDS value and the like and the service life of the filter element, and obtain the optimal weight and intercept. The specific algorithm steps are as follows:

according to an example embodiment, the model training aims to learn the relationship between the water pump current magnitude, the TDS value and the filter element life, resulting in optimal weights and intercepts.

The specific algorithm steps are as follows:

the independent and dependent variables of n samples can be expressed in vector form as:

X＝[X1,X2,...,Xn] ^T and y= [ Y1, Y2, ], yn] ^T

The sum of squares of the errors is minimized by solving the estimates of the regression coefficients using the least squares method. The sum of squares of the errors can be expressed as:

to minimize the sum of squares of the errors, the regression coefficients need to be partially derivative and equal to 0.

Namely:

where j=0, 1,2,..n.

At S307, the partial derivatives are assembled into a system of equations and solved.

The partial derivatives are combined into an equation set, and the estimated value of the regression coefficient can be obtained:

w＝(X ^T X) ^-1 X ^T Y

wherein X is ^T Is a transpose of the argument matrix, X and Y are vectors of the arguments and the dependent variables, respectively, (X) ^T X) ^-1 Is an independent variable matrix X ^T An inverse matrix of X.

By solving this equation, we can get an estimate of the regression coefficients.

At S309, a multiple linear regression model is determined.

And determining a multiple linear regression model according to the estimated value of the regression coefficient.

The following describes an embodiment of an apparatus of the present application which may be used to perform the aforementioned method according to an embodiment of the present application. For details not disclosed in the embodiments of the device according to the application, reference is made to the embodiments of the method according to the application.

Referring to fig. 4A, a water purifier system waterway according to an example embodiment of the present application includes an electronic control board 401; a water inlet solenoid valve 403; a first TDS sensor 405; a booster pump 407; RO cartridge 409; a second TDS sensor 411; a faucet 413 and a waste water solenoid valve 415.

According to an exemplary embodiment, when the main control chip 401 detects the faucet 413 opening signal, the electronic control board 401 sequentially opens the water inlet solenoid valve 403 and the wastewater solenoid valve 415, so as to prepare for the operation of the RO filter element 409.

According to an exemplary embodiment, after the water inlet solenoid valve 403 is opened, the first TDS sensor 405 transmits the acquired TDS value to the electronic control board 401. The electronic control board 401 can judge the water quality according to different TDS value ranges, and accordingly decides to use a corresponding filter element life prediction model.

According to an exemplary embodiment, when the electronic control board 401 detects the tap 413 on signal, the booster pump 407 is started while the current value of the booster pump 407 is collected. For example, the booster pump 407 receives an instruction to start a boosting operation on the water flow. When the pressure is increased to the required pressure, the water flow continues to flow, and the water pump current is measured according to the current circuit and is transmitted to the electronic control board 401 for data storage.

According to some embodiments, after the water flow is pressurized by the booster pump 407, the water flow enters the RO filter cartridge 409 to start filtering after being pushed by the water pressure.

RO is the abbreviation of English Reverse Osmosis, chinese means Reverse Osmosis, under a certain pressure, water molecules can pass through an RO membrane filter core, and impurities such as inorganic salts, heavy metal ions, organic matters, colloid, bacteria, viruses and the like in source water can not pass through the RO membrane, so that permeable pure water and impermeable concentrated water can be strictly distinguished.

According to an exemplary embodiment, the filtered wastewater is diverted to wastewater solenoid valve 415 for effluent treatment.

The filtered clean water passes through a second TDS sensor 411, and the second TDS sensor 411 transmits TDS data to the electronic control board 401 for predictive use.

The electric control board 401 collects the current value of the booster pump 407 and receives the TDS value from the second TDS sensor, and inputs the current value and the TDS value into the trained filter element life prediction model. According to some embodiments, the electronic control board 401 may be further configured to obtain a system running time and accumulate the system running time to an accumulated service time of the filter element, so that data analysis such as a TDS value, a current, and a time may be used to predict a service life of the filter element.

Referring to fig. 4B, the waterway according to the example embodiment of the present application includes: a first TDS sensor 421; a water inlet solenoid valve 423; a first filter element 425; a booster pump 427; a second cartridge 429; a flush valve 431; a second TDS sensor 433; a tap 435.

According to an exemplary embodiment, water flows from the water inlet through the first TDS sensor 421, the first TDS sensor 421 collects TDS values, and then the water flows to the water inlet solenoid valve 423.

According to an exemplary embodiment, the water inlet solenoid valve 423 is opened to allow water to pass through, flow to the first filter element 425 for filtering, and the filtered water flows to the booster pump 427.

According to an exemplary embodiment, the booster pump 427 boosts the flow of water, which enters the second filter insert 429 to begin filtering, and the filtered wastewater flows to the rinse valve 431 and then out the wastewater port.

According to an exemplary embodiment, the purified water filtered through the second filter cartridge 429 flows to the second TDS sensor 433, the second TDS sensor 433 collects TDS values, and finally the water flows out of the faucet 435.

Referring to fig. 5, an electronic control board system according to an exemplary embodiment of the present application includes: a WIFI module interface 501; a display screen interface 503; a waste water solenoid valve interface 505; a water inlet solenoid valve interface 507; tap interface 509; a power adapter 511; a current detection circuit 513; a booster pump interface 515; TDS sensor interface 517; MCU master control chip 519.

According to an exemplary embodiment, after the tap is turned on, a 5V voltage is output to the MCU main control chip 519 by the hall sensor, and when the MCU main control chip 519 receives a tap on signal, the MCU main control chip 519 first turns on the waste water solenoid valve and the water inlet solenoid valve. The wastewater electromagnetic valve can control the outflow of filtered wastewater, and the water inlet electromagnetic valve can control the inflow of water before filtration. The electromagnetic valve is a switch which can be controlled by current, not only can be controlled remotely, but also can save manpower.

According to an example embodiment, after the solenoid valve is activated, the MCU main control chip 519 activates the booster pump, and the current detection circuit 513 starts to collect the current of the booster pump and returns the current value of the booster pump to the MCU main control chip 519 along with the TDS value detected by the second TDS sensor and the running time.

According to an example embodiment, the booster pump current value, the second TDS value, and the run time are brought into a trained model system to estimate the remaining useful life of the filter element.

According to an example embodiment, the estimation result may be uploaded to the cloud end through the WIFI module. In addition, the main control chip can transmit the estimation result to the display screen through the serial port to display the service life condition of the filter element.

From the above description, it is readily understood that technical solutions according to embodiments of the present application may have one or more of the following advantages.

According to an example embodiment, the current value, the water outlet TDS value, the water pump running time and the like of the water pump are acquired during self-learning, and the learned information data is stored in the internal FLASH of the MCU.

According to the example embodiment, the stored model data is directly used when the filter element is normally replaced, so that the filter element can be quickly identified, and the quick response and reliability of the system for predicting the service life of the filter element can be realized.

According to the example embodiment, the current good TDS value of the water pump is detected to be brought into the memory AI model when water is taken in normal operation, so that the service life of the filter element can be rapidly predicted.

It should be clearly understood that the present application describes how to make and use particular examples, but the present application is not limited to any details of these examples. Rather, these principles can be applied to many other embodiments based on the teachings of the present disclosure.

Furthermore, it should be noted that the above-described figures are merely illustrative of the processes involved in the method according to the exemplary embodiment of the present application, and are not intended to be limiting. It will be readily appreciated that the processes shown in the above figures do not indicate or limit the temporal order of these processes. In addition, it is also readily understood that these processes may be performed synchronously or asynchronously among the plurality of modules.

The exemplary embodiments of the present application have been particularly shown and described above. It is to be understood that this application is not limited to the precise arrangements, instrumentalities and instrumentalities described herein; on the contrary, the application is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.

Claims

1. A method of intelligently detecting the life of a water purifier cartridge, the water purifier comprising a booster pump and a cartridge, the method comprising:

collecting a current value of the booster pump;

collecting a TDS value of a water purifying end of the filter element;

and predicting the service life of the filter element by using the current value and the TDS value through a trained filter element service life prediction model.

2. The method as recited in claim 1, further comprising:

collecting a TDS value of inlet water;

and selecting a corresponding filter element life prediction model according to the range of the TDS value of the inlet water so as to predict the filter element life.

3. The method of claim 2, wherein the water purifier further comprises:

the first TDS sensor is arranged at the upstream of the filter element and is used for detecting the TDS value of the inlet water;

and the second TDS sensor is positioned at the downstream of the filter element and is used for detecting the TDS value of the water purifying end of the filter element.

4. The method of claim 1, wherein the cartridge life prediction model comprises a multiple linear regression model.

5. The method as recited in claim 1, further comprising:

information in the use process of different filter elements is collected in advance;

and training the filter element life prediction model by utilizing the acquired information.

6. The method of claim 1, wherein collecting the current value of the booster pump comprises:

after receiving a tap water discharge instruction, opening a booster pump;

and collecting the current value of the booster pump.

7. The method as recited in claim 1, further comprising:

acquiring accumulated service time of the filter element;

the accumulated use time is input to the filter element life prediction model together with the current value and the TDS value to predict the filter element life.

8. The method as recited in claim 1, further comprising:

uploading a filter element service life prediction result to the cloud end through the WIFI module; and/or

And prompting the service life of the filter element on the mobile terminal application program.

9. The method as recited in claim 1, further comprising:

and alarming after judging that the filter element reaches the service life.

10. The method of claim 1, wherein the water purifier further comprises:

a booster pump for boosting the intake water;

and the electric control board is used for collecting the current value and the TDS value of the booster pump, inputting data into a model, and inputting the current value and the TDS value into a trained filter element life prediction model.