Water purifier filter core life intelligent prediction method based on finite element analysis

Abstract

The application discloses a water purifier filter core life intelligent prediction method based on finite element analysis, relates to the technical field of water treatment life prediction, and comprises the following steps: collecting a filter core model and water inlet quality parameters of a target water purifier, and constructing a filter core three-dimensional finite element model according to the filter core model. The water inlet quality parameters are input into the model as boundary conditions to perform multi-physical field coupling simulation, and a simulation result of the time-space distribution of pollutants in the filter core is generated, according to which the current cumulative interception capacity of the filter core is calculated. The current cumulative interception capacity is taken as an initial state, and the preset maximum interception capacity of the filter core is combined to complete the prediction of the remaining service life of the filter core. The method can truly simulate the dynamic distribution law of pollutants in the filter core, is in line with the actual water purification working condition loss characteristics, gets rid of the limitations of traditional experience estimation, and realizes the accurate and intelligent prediction of the service life of the water purifier filter core.

Description

Technical Field

[0001] This invention belongs to the field of water treatment life prediction technology, specifically a method for intelligent prediction of water purifier filter life based on finite element analysis. Background Technology

[0002] Currently, the lifespan of water purifier filter cartridges is mostly determined by fixed usage cycle statistics, manual experience estimation, and simple comparison of basic water quality parameters. These methods rely solely on external usage conditions to set replacement time points, without in-depth analysis considering the filter cartridge's structure and internal fluid flow. This traditional approach can only roughly assess filter cartridge wear at a macroscopic level, failing to delve into the actual state of pollutant deposition and retention within the filter cartridge. Its applicability is limited to standardized, general water purification scenarios.

[0003] Traditional methods for determining filter cartridge lifespan do not incorporate 3D structural modeling and multiphysics simulation analysis. They cannot depict the dynamic changes of contaminants within the filter cartridge's internal space and over time. The cumulative retention capacity can only be roughly estimated using external data, resulting in a significant discrepancy with the actual retention load of the filter cartridge. Conventional prediction logic does not consider the dynamic influencing factors of the filter cartridge's structural parameters and the quality of the influent water. This makes it unsuitable for varying retention conditions caused by differences in filter cartridge models and regional influent water quality. Consequently, the lifespan prediction results have a low degree of consistency with the actual wear and tear of the filter cartridge, making it difficult to meet the refined operation and maintenance needs of diverse household and commercial water purification equipment. Summary of the Invention

[0004] The present invention aims to at least solve the technical problems of inaccurate estimation of cumulative retention and inaccurate prediction of filter life in the prior art.

[0005] Therefore, this invention proposes an intelligent prediction method for the lifespan of water purifier filter cartridges based on finite element analysis, including: Obtain the filter cartridge model and inlet water quality parameters of the target water purifier; A three-dimensional finite element model of the filter element is established based on the filter element model. The influent water quality parameters are used as boundary conditions to input the three-dimensional finite element model of the filter element, and multi-physics field coupling simulation is performed to generate the spatiotemporal distribution simulation results of pollutants inside the filter element. The current cumulative retention of the filter element is calculated based on the spatiotemporal distribution simulation results. Using the current cumulative retention as the initial state, and combined with the preset maximum retention capacity of the filter element, the remaining service life of the filter element is predicted.

[0006] Furthermore, the step of establishing a three-dimensional finite element model of the filter element based on the filter element model specifically includes: Obtain the filter element structure parameters corresponding to the filter element model of the target water purifier. The filter element structure parameters include filter element diameter, filter element length, filter element wall thickness, filter pore diameter, and filter pore density. Construct a geometric model of the filter element in 3D modeling software based on the filter element structure parameters; The filter element geometric model is meshed to generate a three-dimensional finite element model of the filter element composed of multiple finite element elements; The material properties of each finite element in the three-dimensional finite element model of the filter element are set, and the material properties include the porosity, permeability, density and elastic modulus of the filter layer material; Physical field constraints are applied to the three-dimensional finite element model of the filter element. These physical field constraints include the inlet velocity boundary of the flow field, the outlet pressure boundary of the flow field, and the pressure field continuity boundary.

[0007] Furthermore, the step of inputting the influent water quality parameters as boundary conditions into the three-dimensional finite element model of the filter element, performing multiphysics coupling simulation, and generating simulation results of the spatiotemporal distribution of pollutants inside the filter element specifically includes: The influent flow rate, influent turbidity, influent particulate matter size distribution and influent organic matter concentration among the influent water quality parameters are used as time-related boundary conditions and assigned to the flow field inlet boundary of the three-dimensional finite element model of the filter element, respectively. The flow field distribution described by the Navier-Stokes equations, the pollutant transport process described by the convection-diffusion equations, and the seepage process inside the porous media of the filter element described by Darcy's law are simultaneously coupled and solved on the three-dimensional finite element model of the filter element. At the end of each simulation time step, the cumulative concentration and mass of pollutants at each finite element location in the three-dimensional finite element model of the filter element are recorded; The cumulative concentration and mass of pollutants recorded at all simulation time steps are organized according to time sequence and spatial location to form the simulation results of the spatiotemporal distribution of pollutants inside the filter element.

[0008] Furthermore, the simulation results of the spatiotemporal distribution of pollutants inside the filter element include the pollutant concentration field, the pollutant deposition thickness field, and the filter element permeability change field.

[0009] Furthermore, the step of calculating the current cumulative retention of the filter element based on the spatiotemporal distribution simulation results specifically includes: Extract the cumulative mass of pollutants at each finite element unit location at the current simulation moment from the simulation results of the spatiotemporal distribution of pollutants inside the filter element; The total mass of contaminants trapped inside the filter element is obtained by summing the cumulative mass of contaminants in all finite element elements of the three-dimensional finite element model of the filter element. Obtain the initial filter cartridge mass corresponding to the filter cartridge model of the target water purifier, and add the initial filter cartridge mass to the total mass of the intercepted contaminants to obtain the current total mass of the filter cartridge; The difference between the current total mass and the initial filter element mass is determined as the current cumulative retention of the filter element.

[0010] Furthermore, the step of predicting the remaining service life of the filter element, using the current cumulative retention as the initial state and combining it with the preset maximum retention capacity of the filter element, specifically includes: Obtain the maximum filter cartridge capacity corresponding to the filter cartridge model of the target water purifier. The maximum filter cartridge capacity is determined by the saturated adsorption capacity of the filter cartridge material. Calculate the difference between the maximum retention capacity of the filter element and the current cumulative retention amount to obtain the remaining retention capacity of the filter element; The average inlet flow rate and average inlet turbidity of the target water purifier are obtained, and the filter cartridge retention rate per unit time is calculated based on the average inlet flow rate and the average inlet turbidity. Divide the remaining retention capacity by the filter cartridge retention rate per unit time to obtain the theoretical remaining service life of the filter cartridge. The predicted remaining service life of the filter element is obtained by subtracting the preset safety redundancy period from the theoretical remaining service life.

[0011] Furthermore, the step of obtaining the maximum filter cartridge capacity corresponding to the filter cartridge model of the target water purifier specifically includes: Retrieve the filter material type corresponding to the filter model from the filter database. The filter material type includes activated carbon, polypropylene meltblown, ceramic filter, or ultrafiltration membrane. The saturated adsorption capacity per unit mass of filter material is obtained from the material performance library based on the filter material type. The total mass of the filter element is obtained by summing the masses of all finite element elements in the three-dimensional finite element model of the filter element. Multiply the total mass of the filter element by the saturated adsorption capacity per unit mass of filter element material to obtain the maximum retention capacity of the filter element.

[0012] Furthermore, the step of assigning the influent flow rate, influent turbidity, influent particulate matter size distribution, and influent organic matter concentration from the influent water quality parameters as time-dependent boundary conditions to the flow field inlet boundary of the three-dimensional finite element model of the filter element specifically includes: The influent water quality parameters are arranged in chronological order to form an influent water quality time series. Interpolate the influent flow rate in the influent water quality time series to generate a flow rate boundary function over continuous time; Piecewise linear fitting is performed on the influent turbidity in the influent water quality time series to generate a turbidity boundary function over continuous time. The particle size distribution of the influent particulate matter is divided into multiple particle size groups according to the particle size range, and a concentration boundary function that varies with time is established for each particle size group. The concentration of organic matter in the influent is divided into multiple groups according to the type of organic matter, and a concentration boundary function that changes with time is established for each group. All generated flow boundary functions, turbidity boundary functions, concentration boundary functions for each particle size group, and concentration boundary functions for each type group are collectively bound to the flow field inlet boundary of the three-dimensional finite element model of the filter element.

[0013] Furthermore, the steps of simultaneously coupling and solving the flow field distribution described by the Navier-Stokes equations, the pollutant transport process described by the convection-diffusion equations, and the seepage process inside the porous media of the filter element described by Darcy's law on the three-dimensional finite element model of the filter element specifically include: The three-dimensional finite element model of the filter element is divided into a fluid domain and a porous media domain, wherein the fluid domain corresponds to the water inlet channel and water outlet channel of the filter element, and the porous media domain corresponds to the filter layer of the filter element. Solving the Navier-Stokes equations within the fluid domain yields the velocity and pressure field distributions for each finite element element within the fluid domain. The velocity field distribution and pressure field distribution at the interface between the fluid domain and the porous medium domain are passed to the porous medium domain as boundary conditions. Darcy's law is solved within the porous medium domain to obtain the seepage velocity field distribution and pore pressure distribution of each finite element element within the porous medium domain. Simultaneously solve the convection-diffusion equation within the porous medium domain to obtain the concentration distribution of particulate matter in each particle size group and the concentration distribution of organic matter in each type group within the porous medium domain. At the end of each simulation time step, the flux boundary conditions at the interface between the fluid domain and the porous medium domain are updated to achieve bidirectional coupling between the flow field and the concentration field.

[0014] Furthermore, the steps for obtaining the filter cartridge model and inlet water quality parameters of the target water purifier specifically include: Receive the water purifier device serial number entered by the user through a mobile terminal, and query the corresponding filter model from the cloud device database based on the water purifier device serial number; Receive the water purifier installation location information input by the user through a mobile terminal, and obtain historical and real-time water quality data for that location from the water quality monitoring platform based on the water purifier installation location information; The historical water quality data and the real-time water quality data are integrated into influent water quality parameters, which include influent flow rate, influent turbidity, influent particulate matter size distribution and influent organic matter concentration. The filter cartridge model and the influent water quality parameters are stored as input data for the prediction model.

[0015] Compared with the prior art, the beneficial effects of the present invention are: Existing technologies do not take into account the spatial and temporal distribution of water flow inside the filter element. This invention solves the pain points of "inaccurate estimation of filter element lifespan and incompatibility with water quality".

[0016] Based on the actual filter cartridge model, a dedicated 3D finite element model of the filter cartridge is constructed. The influent water quality parameters are set as boundary conditions and integrated into the model for multiphysics coupling simulation. This can completely replicate the flow trajectory of water inside the filter cartridge and the entire process of pollutant migration and deposition with the water flow, fully presenting the distribution and changes of pollutants at different locations and time periods inside the filter cartridge. It breaks away from the inherent mode of evaluating the filter cartridge status solely based on external water quality data and usage time, and fully restores the coupled operating state of the fluid field and pollutant diffusion field inside the filter cartridge, closely matching the physical changes inside the filter cartridge under real water purification working conditions.

[0017] Based on simulation results of the spatiotemporal distribution of contaminants inside the filter element, the real-time cumulative retention capacity is calculated. Using the real-time calculated retention capacity as an initial benchmark, and combined with the predetermined maximum retention capacity of the filter element, lifespan is extrapolated, abandoning the fixed-cycle, one-size-fits-all lifespan determination method. It follows the rhythm of retention load changes caused by fluctuations in influent water quality and differences in filter element structure, closely matching the dynamic process of continuous contaminant accumulation in the filter element during actual use. It also matches the wear and tear rhythm of water purifier filter elements under different application environments, adapting to the wear and tear characteristics of various specifications of water purifiers and different water quality environments. It covers more than 95% of household filter elements and is suitable for complex and ever-changing actual water purification usage scenarios. Attached Figure Description

[0018] Figure 1 The flowchart is a process for the intelligent prediction method of water purifier filter cartridge life based on finite element analysis as described in this invention. Figure 2 A flowchart for generating spatiotemporal distribution results in multiphysics coupling simulation; Figure 3 A flowchart for calculating the predicted remaining service life of the filter element and obtaining the maximum filter element retention capacity. Detailed Implementation

[0019] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0020] See Figure 1 This invention provides an intelligent prediction method for the lifespan of water purifier filter cartridges based on finite element analysis. The method includes: obtaining the filter cartridge model and inlet water quality parameters of the target water purifier; establishing a three-dimensional finite element model of the filter cartridge based on the filter cartridge model; inputting the inlet water quality parameters as boundary conditions into the three-dimensional finite element model of the filter cartridge; performing multi-physics coupling simulation to generate simulation results of the spatiotemporal distribution of pollutants inside the filter cartridge; calculating the current cumulative retention of the filter cartridge based on the spatiotemporal distribution simulation results; using the current cumulative retention as the initial state and combining it with the preset maximum retention capacity of the filter cartridge, predicting the remaining lifespan of the filter cartridge.

[0021] In one embodiment of the present invention, the system receives the serial number of a water purifier device input by a user via a mobile terminal, and queries the corresponding filter model from a cloud-based device database based on the serial number. It also receives the installation location information of the water purifier input by the user via the mobile terminal, and obtains historical and real-time water quality data for that location from a water quality monitoring platform based on the installation location information. The historical and real-time water quality data are integrated into influent water quality parameters, including influent flow rate, influent turbidity, influent particulate matter size distribution, and influent organic matter concentration. The filter model and influent water quality parameters are stored as input data for a prediction model to obtain the target water purifier. The filter element model corresponds to the filter element structure parameters, which include filter element diameter, filter element length, filter element wall thickness, filter pore diameter, and filter pore density. Based on the filter element structure parameters, a filter element geometric model is constructed in 3D modeling software. The filter element geometric model is meshed to generate a 3D finite element model of the filter element composed of multiple finite element elements. The material properties of each finite element element in the 3D finite element model of the filter element are set. The material properties include the porosity, permeability, density, and elastic modulus of the filter layer material. Physical field constraints are applied to the 3D finite element model of the filter element. The physical field constraints include the inlet velocity boundary of the flow field, the outlet pressure boundary of the flow field, and the pressure field continuity boundary.

[0022] In practice, a household reverse osmosis water purifier was installed in a residential community in a certain city. The user entered the water purifier's serial number SN: RO-2024-001258 through a water purifier management application installed on a mobile terminal. The mobile terminal sent the water purifier's serial number to a cloud device database. The cloud device database stores the correspondence between the serial numbers of registered water purifiers and the filter cartridge models. Based on the received water purifier serial number SN: RO-2024-001258, the cloud device database retrieved the corresponding filter cartridge model as PPC-12 composite filter cartridge. The PPC-12 composite filter cartridge contains a polypropylene melt-blown layer and an activated carbon fiber layer.

[0023] In some embodiments, the user inputs the water purifier installation location information as "Room 102, Building 3, Sunshine Garden Community, XX District, XX City, XX Province" via a mobile terminal. The mobile terminal sends this installation location information to the water quality monitoring platform. The water quality monitoring platform is a subordinate platform of the National Water Quality Monitoring Data Center, which stores historical and real-time water quality data for various geographical regions. Based on the received installation location information "Room 102, Building 3, Sunshine Garden Community, XX District, XX City, XX Province", the water quality monitoring platform retrieves the water supply network node to which the location belongs, extracts historical water quality data for the past 30 days from the historical water quality database of that node, and simultaneously obtains the current real-time water quality data from the real-time water quality monitoring terminal of that node. The historical water quality data includes the daily average influent turbidity, daily average influent flow rate, statistical values ​​of daily influent particulate matter size distribution, and the daily average influent organic matter concentration. The real-time water quality data includes the current influent turbidity, current influent flow rate, instantaneous values ​​of current influent particulate matter size distribution, and instantaneous values ​​of current influent organic matter concentration.

[0024] In practical implementation, historical and real-time water quality data are integrated into influent water quality parameters. Specifically, the following methods are used: The daily average influent turbidity from the past 30 days in historical water quality data and the current influent turbidity from real-time water quality data are arranged in chronological order of collection time to form an influent turbidity time series; the daily average influent flow rate from the past 30 days in historical water quality data and the current influent flow rate from real-time water quality data are arranged in chronological order of collection time to form an influent flow rate time series; and the statistical values ​​of the daily influent particulate matter size distribution from the past 30 days in historical water quality data (including particle size ranges of 0.1-1 micrometers, 1-5 micrometers, etc.) are used to form influent particulate matter size distribution parameters. The volume percentages of the four particle size ranges (5-10 micrometers, 10-50 micrometers) and the instantaneous values ​​of the current influent particulate matter size distribution in the real-time water quality data are arranged in chronological order of collection time to form the concentration time series for each particle size range; the average daily influent organic matter concentration over the past 30 days in the historical water quality data and the instantaneous values ​​of the current influent organic matter concentration in the real-time water quality data are arranged in chronological order of collection time to form the influent organic matter concentration time series; the above influent turbidity time series, influent flow rate time series, concentration time series for each particle size range, and influent organic matter concentration time series are used together as the influent water quality parameters.

[0025] In some embodiments, the filter cartridge model PPC-12 and influent water quality parameters are stored as input data for the prediction model. The prediction model runs on a cloud server. After receiving the filter cartridge model PPC-12, the cloud server retrieves the corresponding filter cartridge structure parameters from the filter cartridge structure database. The filter cartridge structure parameters include a filter cartridge diameter of 65 mm, a filter cartridge length of 250 mm, a filter cartridge wall thickness of 12 mm, a pore diameter of 0.5 micrometers, and a pore density of 1200 pores per square centimeter. Based on the filter cartridge diameter of 65 mm, ... The filter element has a length of 250 mm and a wall thickness of 12 mm. The geometric model of the filter element is constructed in the 3D modeling software SolidWorks. Specifically, a cylinder with a diameter of 65 mm and a length of 250 mm is created as the filter element base. A through channel with a diameter of 31 mm is created in the center of the cylinder as the water outlet channel. Based on the parameters of a filter pore diameter of 0.5 micrometers and a filter pore density of 1200 per square centimeter, a uniformly distributed filter pore feature is generated on the outer wall surface of the cylinder. The filter pore feature is arranged in a hexagonal honeycomb array between the outer wall surface and the inner wall surface of the cylinder.

[0026] In the specific implementation, the filter element geometric model is meshed, and tetrahedral elements are used to discretize the filter element geometric model. The global element size is set to 0.5 mm, and the local mesh refinement size is set to 0.05 mm in the area near the filter pores. Finally, a three-dimensional finite element model of the filter element composed of 2,847,392 finite element elements is generated. Each finite element element has a unique element number and spatial coordinates.

[0027] Optionally, the material properties of each finite element in the three-dimensional finite element model of the filter element can be set. For the finite element located in the polypropylene meltblown layer region, its material properties are set to include porosity of 0.78, permeability of 2.5e-12 m², density of 910 kg / m³, and elastic modulus of 1.2 GPa. For the finite element located in the activated carbon fiber layer region, its material properties are set to include porosity of 0.65, permeability of 1.8e-13 m², density of 450 kg / m³, and elastic modulus of 0.8 GPa. For the finite element located in the filter pore channel, its material properties are set to fluid domain properties, with porosity and permeability both set to 1, density of 998 kg / m³, and elastic modulus not applicable.

[0028] It is understandable that physical field constraints are applied to the three-dimensional finite element model of the filter element. These physical field constraints include the inlet velocity boundary, the outlet pressure boundary, and the pressure field continuity boundary. Specifically, an inlet velocity boundary is applied at the filter hole inlet on the outer wall of the three-dimensional finite element model of the filter element. The value of this inlet velocity boundary is calculated based on the inlet flow rate time series and the total area of ​​the filter holes, resulting in an initial inlet velocity of 0.05 m / s. An outlet pressure boundary is applied at the outlet end face of the central outlet channel of the three-dimensional finite element model of the filter element. This outlet pressure boundary is set to a standard atmospheric pressure of 101325 Pa.

[0029] To more realistically simulate the actual operating conditions of a household water supply network and overcome the distortion of simulation results under ideal boundary conditions, dynamic boundary modeling is introduced when applying physical field constraints. For the inlet boundary of the flow field, instead of a constant inlet velocity, a transient flow boundary function is constructed. By performing spectral analysis on the inlet flow time series and superimposing a step pulse signal generated by the mechanical switch of the faucet, the impact process of water flow from rest to rated flow and the water hammer effect when the flow field is closed are simulated. For the outlet boundary of the flow field, a variable resistance pressure boundary model is established. Frictional losses along the pipeline are calculated using the Darcy-Weisbach formula, and local resistance losses are calculated in conjunction with the characteristics of pipe fittings such as elbows and valves. The outlet pressure boundary is set as the pressure at the end of the pipeline minus the total pressure drop, thus making the outlet pressure dynamically change with the flow rate, rather than being constant at standard atmospheric pressure. A pressure field continuity boundary is applied at the interface between the polypropylene meltblown layer and the activated carbon fiber layer, requiring equal pressure values ​​and a continuous pressure gradient on both sides of the interface. Optionally, after completing the establishment of the above three-dimensional finite element model of the filter element, the model is stored in the finite element model library of the cloud server and an index is established with the filter element model PPC-12. The three-dimensional finite element model of the filter element is used for subsequent multiphysics coupling simulation calculations.

[0030] In one embodiment of the present invention, see [reference] Figure 2The influent flow rate, turbidity, particle size distribution, and organic matter concentration are used as time-dependent boundary conditions and assigned to the flow field inlet boundary of the three-dimensional finite element model of the filter element. The flow field distribution described by the Navier-Stokes equation, the pollutant transport process described by the convection-diffusion equation, and the seepage process inside the porous media of the filter element described by Darcy's law are simultaneously coupled and solved on the three-dimensional finite element model of the filter element. At the end of each simulation time step, the cumulative concentration and mass of pollutants at each finite element unit in the three-dimensional finite element model of the filter element are recorded. The cumulative concentration and mass of pollutants recorded for all simulation time steps are organized according to time sequence and spatial location to form the spatiotemporal distribution simulation results of pollutants inside the filter element. The spatiotemporal distribution simulation results of pollutants inside the filter element include the pollutant concentration field, the pollutant deposition thickness field, and the filter element permeability change field.

[0031] In practical implementation, for the constructed three-dimensional finite element model of the PPC-12 composite filter element, the influent flow rate, influent turbidity, influent particulate matter size distribution, and influent organic matter concentration are used as time-dependent boundary conditions and assigned to the flow field inlet boundary of the three-dimensional finite element model of the filter element. The water quality time series obtained by integrating the influent water quality parameters specifically includes: dividing the flow rate value at each time point in the influent flow rate time series by the total area of ​​the filter pores on the outer wall of the three-dimensional finite element model of the filter element (3.14 square millimeters) to calculate the inlet velocity sequence that changes with time, and loading this inlet velocity sequence onto the flow field inlet boundary. The inlet particulate matter concentration boundary was obtained by converting the turbidity values ​​(in NTU) in the influent turbidity time series into particulate matter mass concentrations (in milligrams per liter) using a standard turbidity-particulate matter conversion factor of 1.2. The concentration time series for each particle size range (0.1-1 μm, 1-5 μm, 5-10 μm, 10-50 μm) were then loaded onto the flow field inlet boundary as inlet concentration boundaries for different particle size groups. The total organic carbon value (in milligrams per liter) in the influent organic matter concentration time series was directly loaded onto the flow field inlet boundary as the organic matter inlet concentration boundary. Considering the significant impact of water temperature on fluid viscosity and filtration efficiency, a temperature field was introduced as an independent variable input before coupling the solution of the Navier-Stokes equations. Influent water temperature time series were obtained from a water quality monitoring platform, and temperature correction functions for fluid dynamic viscosity and density were established. During the simulation, fluid properties were updated in real time according to the current water temperature at each time step to avoid flow field distribution deviations caused by ignoring temperature changes.

[0032] In some embodiments, the flow field distribution described by the Navier-Stokes equations, the pollutant transport process described by the convection-diffusion equations, and the seepage process inside the porous media of the filter element described by Darcy's law are simultaneously solved on a three-dimensional finite element model of the filter element. The solution process uses the multiphysics coupling module of the finite element analysis software COMSOL Multiphysics, with a time step of 0.5 seconds and a total simulation time of 3600 seconds. The expression of the Navier-Stokes equations is as follows: Where: ρ represents fluid density, t represents time, u represents velocity vector, p represents pressure, I represents unit tensor, and μ represents fluid dynamic viscosity; the convection-diffusion equation is solved independently for each particle size group and organic species group, and Darcy's law is used to solve the relationship between seepage velocity and pressure gradient in the porous media domain of the filter element.

[0033] Optionally, at the end of each simulation time step of 0.5 seconds, the cumulative concentration and mass of pollutants at each finite element location in the three-dimensional finite element model of the filter element are recorded. Specifically, the post-processing function of the finite element analysis software is used to traverse all 2,847,392 finite element elements. For each finite element element, the total concentration of particulate matter and the total concentration of organic matter at the center point coordinates of the element are extracted and multiplied by the volume of the element to obtain the cumulative mass of pollutants in the element. The cumulative mass of particulate matter and the cumulative mass of organic matter are recorded respectively. At the same time, the porosity change value and the permeability change value of the element are also recorded.

[0034] In the specific implementation, the cumulative concentration and mass of pollutants recorded at all simulation time steps are organized according to time sequence and spatial location to form the simulation results of the spatiotemporal distribution of pollutants inside the filter element. The specific organization method is as follows: For each simulation time step (t=0.5 seconds, 1.0 seconds, 1.5 seconds, ..., 3600 seconds), a data snapshot corresponding to that time step is generated. Each data snapshot contains a matrix with a dimension of (number of grid nodes, 4), where the number of grid nodes is the number of discrete nodes in the three-dimensional finite element model, and the 4 columns store the node coordinates (X, Y, Z) and the cumulative concentration of pollutants at that node, respectively. At the same time, another matrix with a dimension of (number of elements, 5) is generated, where the number of elements is 2,847,392, and the 5 columns store the element number, element volume, cumulative mass of pollutants at the element center point, real-time value of element porosity, and real-time value of element permeability, respectively. The data snapshots of all time steps are stored in the simulation result database in the cloud according to the time index.

[0035] In some embodiments, the simulation results of the spatiotemporal distribution of pollutants inside the filter element include a pollutant concentration field, a pollutant deposition thickness field, and a filter element permeability change field. The pollutant concentration field is generated as follows: for each time step, the cumulative pollutant concentration (the sum of particulate matter and organic matter concentrations) of each finite element unit is spatially interpolated according to the coordinates of the unit's center point to obtain a continuous concentration distribution cloud map in three-dimensional space. The pollutant deposition thickness field is generated as follows: the equivalent deposition thickness (unit: micrometer) is obtained by dividing the cumulative mass of pollutants in each finite element unit by the surface area of ​​the filter material in that unit (for the polypropylene meltblown layer region, the surface area is calculated using pore diameter and porosity; for the activated carbon fiber layer region, the surface area is calculated using fiber diameter and porosity). The equivalent deposition thickness of each unit is then organized into a deposition thickness distribution field according to spatial coordinates. The filter element permeability change field is generated as follows: using the Kozeny-Carman equation, the permeability of each finite element unit at the current moment is calculated based on the ratio of the current porosity to the initial porosity. The permeability change field is expressed as the ratio field of the current permeability to the initial permeability of each unit.

[0036] It is understandable that after completing the calculation of the above spatiotemporal distribution simulation results, the results are stored in a cloud server and directly used to calculate the current cumulative retention of the filter element. Optionally, after all simulation time steps are completed, a summary report of the spatiotemporal distribution simulation results is automatically generated. This summary report includes the coordinates of the maximum value of the pollutant concentration field, the maximum value of the pollutant deposition thickness field, and the coordinates of the minimum value of the filter element permeability change field at the last time step (3600 seconds).

[0037] To address the maintenance conditions of filter cartridges and the operational characteristics of specific membrane material filter cartridges, the multi-physics coupling mechanism is further improved. When the water purifier switches to the backwashing stage, the boundary conditions are reversed through the logic control unit: the original flow field outlet is set as the high-pressure inlet, and the original flow field inlet is set as the low-pressure outlet. At this time, within the porous media domain of the three-dimensional finite element model of the filter cartridge, the shearing and peeling kinetic equation of the filter cake layer is solved to simulate the shedding and discharge process of pollutants under the action of reverse flow, thereby accurately evaluating the performance recovery rate of the filter cartridge after backwashing. For ultrafiltration or reverse osmosis membrane type filter cartridges, the concentration polarization transport equation is added to the wall boundary layer of the porous media domain. By calculating the solute accumulation within the mass transfer boundary layer, the phenomenon of membrane surface osmotic pressure increase is quantified, thereby correcting the effective filtration driving force (transmembrane pressure difference) and ensuring the accuracy of membrane cartridge life prediction. It can be understood that the update frequency of the spatiotemporal distribution simulation results is consistent with the time resolution of the influent water quality parameters. When the influent water quality parameters are updated, the coupling solution process is re-executed to obtain the updated spatiotemporal distribution simulation results.

[0038] In one embodiment of the present invention, the cumulative mass of contaminants at each finite element unit location at the current simulation moment is extracted from the spatiotemporal distribution simulation results of contaminants inside the filter element. The cumulative mass of contaminants in all finite element units of the three-dimensional finite element model of the filter element is summed to obtain the total mass of contaminants retained inside the filter element. To accurately characterize the actual retention load of the filter element, the current cumulative retention amount of the filter element is calculated based on the law of conservation of mass. The cumulative mass of contaminants at each finite element unit location at the current simulation moment is extracted from the spatiotemporal distribution simulation results of contaminants inside the filter element, and the cumulative mass of contaminants in all finite element units is summed to obtain the total mass of contaminants retained inside the filter element. (Unit: milligrams), this value serves as an intermediate verification parameter. Perform the corrected cumulative retention calculation steps: obtain the data from the start of the simulation to the current time. The time integration interval is used to read the time series of influent pollutant mass flow rates at the inlet boundary of the flow field within that time period. The flow rate is determined by the product of the influent flow rate and the influent turbidity / organic matter concentration. Simultaneously, at the outlet end face of the central effluent channel in the three-dimensional finite element model of the filter element, the time series of effluent pollutant mass flow rates at each simulation time step is extracted. The flow rate is determined by the product of the outlet velocity and the pollutant concentration at the outlet. The difference between the two In the time interval Numerical integration is performed internally to obtain the current cumulative retention capacity of the filter element. The calculation formula is as follows: in, As the integral variable, it can be understood that the above calculation logic ensures the physical nature of the cumulative retention, that is, the mass of contaminants retained by the filter cartridge equals the total amount of contaminants entering the system minus the total amount of contaminants that permeate the filter cartridge and are discharged with the water. In a specific numerical example, after performing the above integral operation, we obtain... The value is 1256.7 mg. Optionally, this value can be compared with the aforementioned intermediate calibration parameters. If the error between the two is less than a preset threshold (e.g., 1%), the simulation is considered to have converged; otherwise, the mesh or time step is adjusted and the simulation is recalculated.

[0039] In the specific implementation, the simulation results of the three-dimensional finite element model of the PPC-12 composite filter element and its generated spatiotemporal distribution are completed. The cumulative mass of pollutants at each finite element unit position at the current simulation moment is extracted from the spatiotemporal distribution simulation results of pollutants inside the filter element. The current simulation moment is selected as the end time of the total simulation duration of 3600 seconds, i.e., the 3600th second. The spatiotemporal distribution simulation results store the data snapshot corresponding to this moment. The data snapshot contains a matrix with dimensions (2,847,392,5). The five columns of the matrix store the finite element unit number, finite element unit volume, cumulative mass of pollutants at the center point of the finite element unit, real-time value of finite element unit porosity, and real-time value of finite element unit permeability, respectively. Traversing all 2,847,392 rows of data in the matrix, the value of the third column is read from each row to obtain the cumulative mass of pollutants at each finite element unit position. For finite element units with a pollutant cumulative mass value of zero, it is also recorded as zero value.

[0040] In some embodiments, the cumulative mass of contaminants in all finite element elements of the three-dimensional finite element model of the filter element is summed to obtain the total mass of contaminants retained inside the filter element. The summation operation adopts an accumulation method, adding the cumulative mass of contaminants of 2,847,392 finite element elements one by one. The expression for the summation operation is as follows: in: This indicates the total mass of contaminants trapped inside the filter element. This indicates the total number of finite element elements in the three-dimensional finite element model of the filter element. The value is 2,847,392. Indicates the first The cumulative mass of pollutants in each finite element element is obtained after performing the above summation operation. The value is 1256.7 mg.

[0041] Optionally, obtain the initial filter cartridge mass corresponding to the filter cartridge model of the target water purifier. The filter cartridge model of the target water purifier is PPC-12 composite filter cartridge. Retrieve the factory quality parameters of PPC-12 composite filter cartridge from the filter cartridge quality database. These factory quality parameters are provided by the filter cartridge manufacturer and stored in the cloud database. The retrieved initial filter cartridge mass of PPC-12 composite filter cartridge is 385.4 grams, which is equivalent to 385,400 milligrams.

[0042] In practice, the initial filter element mass is added to the total mass of retained contaminants to obtain the current total mass of the filter element. An addition operation is performed: 385400 mg plus 1256.7 mg, resulting in a current total mass of 386656.7 mg. This current total mass represents the total mass of the filter element and the retained contaminants after 3600 seconds of filtration operation. It can be understood that the difference between the current total mass and the initial filter element mass is determined as the current cumulative retention amount of the filter element. The difference between the current total mass (386656.7 mg) and the initial filter element mass (385400 mg) is 1256.7 mg, which is the current cumulative retention amount of the filter element, expressed in milligrams.

[0043] In some embodiments, the current cumulative retention amount of 1256.7 mg is equal to the previously summed total retained contaminant mass of 1256.7 mg, verifying the consistency of the calculation process. The current cumulative retention amount is stored in the filter cartridge status database on the cloud server and associated with the filter cartridge model PPC-12 and the water purifier device serial number SN: RO-2024-001258. Optionally, after calculating the current cumulative retention amount, this current cumulative retention amount is used as the initial state input for predicting the remaining service life of the filter cartridge, and simultaneously pushed to the water purifier management application interface on the user's mobile terminal for display.

[0044] In one embodiment of the present invention, see [reference] Figure 3 The filter material type corresponding to the filter model is retrieved from the filter material database. Filter material types include activated carbon, polypropylene melt-blown, ceramic filter, or ultrafiltration membrane. Based on the filter material type, the saturated adsorption capacity per unit mass of filter material is obtained from the material performance library. The total mass of all finite element elements in the three-dimensional finite element model of the filter is obtained to get the total mass of the filter. The total mass of the filter is multiplied by the saturated adsorption capacity per unit mass of filter material to get the maximum retention capacity of the filter. The difference between the maximum retention capacity and the current cumulative retention is calculated to get the remaining retention capacity of the filter. The average influent flow rate and average influent turbidity of the target water purifier are obtained, and the filter retention growth rate per unit time is calculated based on the average influent flow rate and average influent turbidity. The remaining retention capacity is divided by the filter retention growth rate per unit time to get the theoretical remaining service life of the filter. The preset safety redundancy time is subtracted from the theoretical remaining service life to get the predicted remaining service life of the filter.

[0045] In practice, given the current cumulative retention of the filter cartridge is 1256.7 mg and the target water purifier's filter cartridge model is PPC-12, the filter cartridge material type corresponding to filter cartridge model PPC-12 is retrieved from the filter cartridge database. The filter cartridge database is stored on a cloud server and contains a mapping table between filter cartridge models and filter cartridge material types. The retrieved filter cartridge model PPC-12 is found to be a composite structure of activated carbon and polypropylene melt-blown type, with activated carbon accounting for 65% of the total mass of the filter cartridge and polypropylene melt-blown type accounting for 35% of the total mass of the filter cartridge.

[0046] In some embodiments, the saturated adsorption capacity per unit mass of filter material is obtained from a material performance library according to the filter material type. The material performance library stores saturated adsorption parameters for different filter material types under standard test conditions. For activated carbon, the saturated adsorption capacity per unit mass of filter material is 320 mg of organic matter per gram of activated carbon; for polypropylene meltblown, the saturated adsorption capacity per unit mass of filter material is 85 mg of particulate matter per gram of polypropylene meltblown material. The above saturated adsorption capacity data are provided by the filter material supplier and entered into the material performance library after being verified by laboratory dynamic adsorption tests.

[0047] Optionally, the total mass of the filter element can be obtained by summing the masses of all finite element units in the three-dimensional finite element model of the filter element. The constructed three-dimensional finite element model of the PPC-12 composite filter element contains 2,847,392 finite element units. The mass of each finite element unit is calculated by multiplying its material property (density) by the unit volume. Traversing all finite element units, the density of the finite element units located in the polypropylene meltblown layer region is taken as 910 kg / m³, the density of the finite element units located in the activated carbon fiber layer region is taken as 450 kg / m³, and the density of the finite element units located in the filter pore channel is taken as 998 kg / m³. However, the mass of this part of the unit is not included in the calculation of the mass of the filter element body. The total mass of all finite element units in the polypropylene meltblown layer region is 135.2 g, and the total mass of all finite element units in the activated carbon fiber layer region is 250.2 g. The sum of the two is the total mass of the filter element, which is 385.4 g.

[0048] In practice, the maximum retention capacity of the filter element is obtained by multiplying its total mass by the saturated adsorption capacity per unit mass of filter element material. The specific multiplication method is as follows: the mass of the activated carbon portion (250.2 grams) multiplied by 320 mg / g yields 80064 mg; the mass of the polypropylene meltblown portion (135.2 grams) multiplied by 85 mg / g yields 11492 mg. Adding these two parts together gives the maximum retention capacity of the filter element as 91556 mg. This can be understood as calculating the difference between the maximum retention capacity and the current cumulative retention capacity to obtain the remaining retention capacity of the filter element. The current cumulative retention capacity is calculated to be 1256.7 mg. Subtracting 1256.7 mg from 91556 mg gives the remaining retention capacity of the filter element as 90299.3 mg.

[0049] In some embodiments, the average influent flow rate and average influent turbidity of the target water purifier are obtained. The influent flow rate time series and influent turbidity time series of the past 30 days are read from the integrated influent water quality parameters. The arithmetic mean of the 30 daily average influent flow rate values ​​in the influent flow rate time series is calculated to obtain an average influent flow rate of 0.18 liters per minute. The arithmetic mean of the 30 daily average influent turbidity values ​​in the influent turbidity time series is calculated to obtain an average influent turbidity of 2.3 NTU.

[0050] In practical implementation, the growth rate of filter cartridge retention per unit time is calculated based on the average influent flow rate and average influent turbidity. The growth rate of retention is calculated using the following formula: in: This indicates the rate of increase in the amount of filter cartridge residue per unit time. This represents the average influent flow rate, calculated as 0.18 liters per minute, which is equivalent to 0.003 liters per second. This represents the average influent turbidity, calculated as 2.3 NTU multiplied by a conversion factor of 1.2, resulting in 2.76 mg / L. This indicates the overall retention efficiency of the filter element; for the PPC-12 composite filter element, the value is 0.98. The conversion factor between turbidity particulate matter and the intercepted mass is taken as 1.0 mg / L corresponding to mg / L. The calculation yields... Milligrams per second.

[0051] Optionally, the remaining retention capacity is divided by the filter cartridge retention rate per unit time to obtain the theoretical remaining service life of the filter cartridge. Performing this division, 90299.3 mg / s² = 0.0081144 mg / s², yields 11,128,752 seconds. Converting this to days, 11,128,752 divided by 86,400 seconds per day equals 128.8 days. Alternatively, subtracting the preset safety margin from the theoretical remaining service life yields the predicted remaining service life of the filter cartridge. The preset safety margin is set to 7 days according to the filter cartridge manufacturer's recommendation. Subtracting 7 days from 128.8 days yields a predicted remaining service life of 121.8 days. Optionally, the predicted remaining lifespan of 121.8 days is rounded down to 121 days and stored in the filter status database on the cloud server, associated with the filter model PPC-12 and the water purifier serial number SN: RO-2024-001258. Simultaneously, this information is pushed to the user's mobile terminal's water purifier management application interface, displaying "Filter lifespan approximately 121 days". In some embodiments, when the average influent flow rate and average influent turbidity change, the steps for obtaining the average influent flow rate and average influent turbidity, as well as the calculation steps for the filter cartridge retention rate increase per unit time, are re-executed, and the updated predicted remaining lifespan replaces the previous stored value.

[0052] In one embodiment of the present invention, the influent water quality parameters are arranged in chronological order to form an influent water quality time series. The influent flow rate in the influent water quality time series is interpolated to generate a continuous-time flow boundary function. The influent turbidity in the influent water quality time series is piecewise linearly fitted to generate a continuous-time turbidity boundary function. The influent particulate matter particle size distribution is divided into multiple particle size groups according to particle size intervals, and a time-varying concentration boundary function is established for each particle size group. The influent organic matter concentration is divided into multiple type groups according to organic matter type, and a time-varying concentration boundary function is established for each type group. All generated flow rate boundary functions, turbidity boundary functions, concentration boundary functions for each particle size group, and concentration boundary functions for each type group are collectively bound to the flow field inlet boundary of the three-dimensional finite element model of the filter element. The three-dimensional finite element model of the filter element is divided into a fluid domain and a porous media domain, wherein the fluid domain corresponds to the influent flow field of the filter element. The water channel and outlet channel, the porous media domain corresponding to the filter layer of the filter element, solve the Navier-Stokes equations in the fluid domain to obtain the velocity and pressure field distributions of each finite element element in the fluid domain. The velocity and pressure field distributions at the interface between the fluid domain and the porous media domain are then passed to the porous media domain as boundary conditions. Darcy's law is solved in the porous media domain to obtain the seepage velocity field distribution and pore pressure distribution of each finite element element in the porous media domain. Simultaneously, the convection-diffusion equations are solved in the porous media domain to obtain the concentration distributions of particulate matter of each particle size group and the concentration distributions of organic matter of each type group in the porous media domain. At the end of each simulation time step, the flux boundary conditions at the interface between the fluid domain and the porous media domain are updated to achieve bidirectional coupling of the flow field and the concentration field.

[0053] In the specific implementation, the integrated influent water quality parameters are arranged in chronological order to form an influent water quality time series. The influent water quality parameters include historical water quality data and real-time water quality data from the past 30 days from the water quality monitoring platform, totaling 31 time points. Each time point corresponds to a collection time. The data are arranged in order from the earliest (30 days ago) to the latest (current time) collection time to form a time series containing 31 data points. Each data point includes influent flow rate, influent turbidity, influent particulate matter size distribution vector (including the volume percentage of four particle size ranges: 0.1-1 micrometer, 1-5 micrometer, 5-10 micrometer, and 10-50 micrometer), and influent organic matter concentration.

[0054] In some embodiments, the influent flow rate in the influent water quality time series is interpolated to generate a continuous-time flow boundary function. The influent flow rate values ​​at 31 time points in the influent water quality time series are 0.17, 0.18, 0.19, 0.18, 0.17, 0.16, 0.17, 0.18, 0.19, 0.20, 0.19, 0.18, 0.17, 0.16, 0.15, 0.16, 0.17, 0.18, 0.19, 0.18, 0.17, 0.16, 0.15, 0.14, 0.15, 0.16, 0.17, 0.18, 0.19, 0.18, 0.17 (unit: liters per minute). Using cubic spline interpolation with time t as the independent variable and influent flow rate Q as the dependent variable, a piecewise cubic polynomial function is constructed. This ensures that the function value and its first and second derivatives are continuous within each time interval, ultimately yielding the flow boundary function defined over the continuous time interval [0, 30 days]. .

[0055] In practice, piecewise linear fitting is performed on the influent turbidity in the influent water quality time series to generate a turbidity boundary function over continuous time. The influent turbidity values ​​are 1.2, 1.3, 1.5, 1.6, 1.8, 2.0, 2.2, 2.4, 2.6, 2.8, 3.0, 3.2, 3.4, 3.5, 3.6, 3.5, 3.4, 3.2, 3.0, 2.8, 2.6, 2.4, 2.2, 2.0, 1.8, 1.6, 1.5, 1.4, 1.3, 1.2, and 1.1 (unit: NTU). Using a piecewise linear fitting method, the turbidity values ​​between adjacent time points are connected by straight line segments, forming a turbidity boundary function over continuous time consisting of 30 straight line segments. The function is continuous at every time point, but its first derivative is discontinuous.

[0056] Optionally, the particle size distribution of the influent particulate matter is divided into multiple particle size groups according to the particle size range. These groups include: Group A (0.1-1 μm), Group B (1-5 μm), Group C (5-10 μm), and Group D (10-50 μm), for a total of four groups. A time-varying concentration boundary function is established for each group. For Group A, the volume percentage of the 0.1-1 μm particle size range at each time point is extracted from the influent water quality time series. This percentage is multiplied by the particulate matter mass concentration converted from turbidity at that time point to obtain the mass concentration time series for Group A. A cubic spline interpolation method is then used to generate the concentration boundary function over continuous time. For groups B, C, and D, concentration boundary functions are established in the same manner. , , .

[0057] In some embodiments, the influent organic matter concentration is divided into multiple groups according to the type of organic matter. These groups include group E (corresponding to humic acids), group F (corresponding to fulvic acids), and group G (corresponding to other natural organic matter). The influent organic matter concentration data obtained from the water quality monitoring platform includes the total organic carbon value and the proportion of each type of organic matter. The concentration of humic acids at each time point is calculated as total organic carbon multiplied by 0.45, the concentration of fulvic acids is calculated as total organic carbon multiplied by 0.35, and the concentration of other natural organic matter is calculated as total organic carbon multiplied by 0.20. A concentration boundary function over time is established for each group, and a piecewise linear fitting method is used to generate the boundary function. , , .

[0058] In practical implementation, all generated flow boundary functions Turbidity boundary function Concentration boundary function for each particle size group to and the concentration boundary function for each species group. to They are all bound to the flow field inlet boundary of the three-dimensional finite element model of the filter element. The binding operation is achieved through the boundary condition setting interface of the finite element analysis software COMSOL Multiphysics, which associates each boundary function with the corresponding physical quantity input port of the flow field inlet boundary.

[0059] It can be understood that the three-dimensional finite element model of the filter element is divided into a fluid domain and a porous media domain. The fluid domain corresponds to the water inlet channel and water outlet channel of the filter element, and the porous media domain corresponds to the filter layer of the filter element. In the constructed three-dimensional finite element model of the PPC-12 composite filter element, the area between the outer wall of the filter element and the central water outlet channel is the filter layer of the filter element, and this area is set as the porous media domain. The water inlet channel area outside the outer wall of the filter element and the cavity area inside the central water outlet channel are set as the fluid domain. The two domains are connected by an interface.

[0060] Optionally, the Navier-Stokes equations are solved within the fluid domain to obtain the velocity and pressure field distributions for each finite element element. The Navier-Stokes equations are solved in a steady-state form, with the fluid density set to 998 kg / m³ and the dynamic viscosity to 0.001 Pascals-second. The inlet boundary conditions use the aforementioned bound flow boundary function. The ratio of the inlet area to the outlet area is converted into a velocity boundary condition. The outlet boundary condition is set as a pressure outlet with an outlet pressure of 0 Pascals (relative pressure). After the solution is completed, the velocity vector (in meters per second) and pressure value (in Pascals) at the centroid coordinates of each finite element in the fluid domain are extracted.

[0061] In practice, the velocity and pressure field distributions at the interface between the fluid domain and the porous medium domain are passed to the porous medium domain as boundary conditions. There are 45,892 finite element nodes at the interface, and each node stores the velocity components and pressure values ​​obtained from the fluid domain. These data are applied as Dirichlet boundary conditions to the interface nodes of the porous medium domain, requiring that the velocity and pressure at the interface of the porous medium domain be consistent with the calculation results of the fluid domain.

[0062] In some embodiments, Darcy's law is solved within the porous medium domain to obtain the seepage velocity field distribution and pore pressure distribution of each finite element element within the porous medium domain. The expression for Darcy's law is: in: This represents the Darcy velocity vector. The permeability tensor represents the permeability tensor of a porous medium. For isotropic porous media, Simplified to scalar multiplied by unit tensor. Indicates the dynamic viscosity of a fluid. The pore pressure gradient is represented by the equation above. For each finite element in the porous medium domain, the pore pressure distribution and seepage velocity distribution of each element are obtained by solving the above equation based on the element material properties (the permeability of the polypropylene meltblown layer is taken as 2.5e-12 m², and the permeability of the activated carbon fiber layer is taken as 1.8e-13 m²) and the pressure boundary conditions transmitted from the interface.

[0063] Optionally, the convection-diffusion equations can be solved simultaneously within the porous media domain to obtain the concentration distributions of particulate matter for each particle size group and the concentration distributions of organic matter for each species group within the porous media domain. The general form of the convection-diffusion equations is as follows: in: Indicates the concentration of a certain pollutant (particulate matter or organic matter). Indicates time, This represents the Darcy velocity vector. This represents the diffusion coefficient, with a value of 1e-11 for particulate matter and 5e-10 for organic matter. This indicates the reaction term; for particulate matter grouping, The value is zero (ignoring chemical reactions); for grouping organic compounds, The value is taken as the adsorption rate term. The Langmuir adsorption kinetic model is used for calculation. The convection-diffusion equation is solved independently for the four particle size groups and the three organic species groups to obtain the concentration value of each group at each finite element in the porous medium domain.

[0064] In practice, at the end of each simulation time step, the flux boundary conditions at the interface between the fluid domain and the porous medium domain are updated to achieve bidirectional coupling between the flow field and the concentration field. Specifically, the update method is as follows: from the pollutant concentration distribution obtained from the porous medium domain, the pollutant flux (in milligrams per square meter per second) on the porous medium domain side at the interface is calculated, and this flux value is used as the mass flux boundary condition of the fluid domain at the interface and applied to the solution of the fluid domain in the next time step; at the same time, the permeability of each element in the porous medium domain is updated in real time according to the cumulative pollutant concentration of each finite element element, and the porosity-permeability correlation model is used to substitute the updated permeability value into the Darcy's law solution in the next time step.

[0065] It is understandable that at the end of each simulation time step, the above-mentioned operations of solving the Navier-Stokes equations in the fluid domain, transferring interface boundary conditions, solving Darcy's law in the porous media domain, solving the convection-diffusion equations in the porous media domain, and updating interface flux boundary conditions are repeated until the preset total simulation time of 3600 seconds is reached. Optionally, after completing the coupled solution for all time steps, the pollutant concentration distribution of all finite element elements in the porous media domain at each time step is stored as part of the simulation results of the spatiotemporal distribution of pollutants inside the filter element.

[0066] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A method for intelligent prediction of water purifier filter cartridge life based on finite element analysis, characterized in that, The method includes: Obtain the filter cartridge model and inlet water quality parameters of the target water purifier; A three-dimensional finite element model of the filter element is established based on the filter element model. The influent water quality parameters are used as boundary conditions to input the three-dimensional finite element model of the filter element, and multi-physics field coupling simulation is performed to generate the spatiotemporal distribution simulation results of pollutants inside the filter element. The current cumulative retention of the filter element is calculated based on the spatiotemporal distribution simulation results. Using the current cumulative retention as the initial state, and combined with the preset maximum retention capacity of the filter element, the remaining service life of the filter element is predicted.

2. The intelligent prediction method for water purifier filter cartridge life based on finite element analysis according to claim 1, characterized in that, The steps for establishing a three-dimensional finite element model of the filter element based on the filter element model specifically include: Obtain the filter element structure parameters corresponding to the filter element model of the target water purifier. The filter element structure parameters include filter element diameter, filter element length, filter element wall thickness, filter pore diameter, and filter pore density. Construct a geometric model of the filter element in 3D modeling software based on the filter element structure parameters; The filter element geometric model is meshed to generate a three-dimensional finite element model of the filter element composed of multiple finite element elements; The material properties of each finite element in the three-dimensional finite element model of the filter element are set, and the material properties include the porosity, permeability, density and elastic modulus of the filter layer material; Physical field constraints are applied to the three-dimensional finite element model of the filter element. These physical field constraints include the inlet velocity boundary of the flow field, the outlet pressure boundary of the flow field, and the pressure field continuity boundary.

3. The intelligent prediction method for water purifier filter cartridge life based on finite element analysis according to claim 2, characterized in that, The steps of inputting the influent water quality parameters as boundary conditions into the three-dimensional finite element model of the filter element, performing multiphysics coupling simulation, and generating simulation results of the spatiotemporal distribution of pollutants inside the filter element specifically include: The influent flow rate, influent turbidity, influent particulate matter size distribution and influent organic matter concentration among the influent water quality parameters are used as time-related boundary conditions and assigned to the flow field inlet boundary of the three-dimensional finite element model of the filter element, respectively. The flow field distribution described by the Navier-Stokes equations, the pollutant transport process described by the convection-diffusion equations, and the seepage process inside the porous media of the filter element described by Darcy's law are simultaneously coupled and solved on the three-dimensional finite element model of the filter element. At the end of each simulation time step, the cumulative concentration and mass of pollutants at each finite element location in the three-dimensional finite element model of the filter element are recorded; The cumulative concentration and mass of pollutants recorded at all simulation time steps are organized according to time sequence and spatial location to form the simulation results of the spatiotemporal distribution of pollutants inside the filter element.

4. The intelligent prediction method for water purifier filter cartridge life based on finite element analysis according to claim 3, characterized in that, The simulation results of the spatiotemporal distribution of pollutants inside the filter element include the pollutant concentration field, the pollutant deposition thickness field, and the filter element permeability change field.

5. The intelligent prediction method for water purifier filter cartridge life based on finite element analysis according to claim 3, characterized in that, The steps for calculating the current cumulative retention of the filter element based on the spatiotemporal distribution simulation results specifically include: Extract the cumulative mass of pollutants at each finite element unit location at the current simulation moment from the simulation results of the spatiotemporal distribution of pollutants inside the filter element; The total mass of contaminants trapped inside the filter element is obtained by summing the cumulative mass of contaminants in all finite element elements of the three-dimensional finite element model of the filter element. Obtain the initial filter cartridge mass corresponding to the filter cartridge model of the target water purifier, and add the initial filter cartridge mass to the total mass of the intercepted contaminants to obtain the current total mass of the filter cartridge; The difference between the current total mass and the initial filter element mass is determined as the current cumulative retention of the filter element.

6. The intelligent prediction method for water purifier filter cartridge life based on finite element analysis according to claim 5, characterized in that, The steps for predicting the remaining service life of the filter element, based on the current cumulative retention amount as the initial state and combined with the preset maximum retention capacity of the filter element, specifically include: Obtain the maximum filter cartridge capacity corresponding to the filter cartridge model of the target water purifier. The maximum filter cartridge capacity is determined by the saturated adsorption capacity of the filter cartridge material. Calculate the difference between the maximum retention capacity of the filter element and the current cumulative retention amount to obtain the remaining retention capacity of the filter element; The average inlet flow rate and average inlet turbidity of the target water purifier are obtained, and the filter cartridge retention rate per unit time is calculated based on the average inlet flow rate and the average inlet turbidity. Divide the remaining retention capacity by the filter cartridge retention rate per unit time to obtain the theoretical remaining service life of the filter cartridge. The predicted remaining service life of the filter element is obtained by subtracting the preset safety redundancy period from the theoretical remaining service life.

7. The intelligent prediction method for water purifier filter cartridge life based on finite element analysis according to claim 6, characterized in that, The steps for obtaining the maximum filter cartridge capacity corresponding to the filter cartridge model of the target water purifier specifically include: Retrieve the filter material type corresponding to the filter model from the filter database. The filter material type includes activated carbon type, polypropylene melt-blown type, ceramic filter type or ultrafiltration membrane type. The saturated adsorption capacity per unit mass of filter material is obtained from the material performance library based on the filter material type. The total mass of the filter element is obtained by summing the masses of all finite element elements in the three-dimensional finite element model of the filter element. Multiply the total mass of the filter element by the saturated adsorption capacity per unit mass of filter element material to obtain the maximum retention capacity of the filter element.

8. The intelligent prediction method for water purifier filter cartridge life based on finite element analysis according to claim 3, characterized in that, The step of assigning the influent flow rate, influent turbidity, influent particulate matter size distribution, and influent organic matter concentration from the influent water quality parameters to the flow field inlet boundary of the three-dimensional finite element model of the filter element as time-dependent boundary conditions specifically includes: The influent water quality parameters are arranged in chronological order to form an influent water quality time series. Interpolate the influent flow rate in the influent water quality time series to generate a flow rate boundary function over continuous time; Piecewise linear fitting is performed on the influent turbidity in the influent water quality time series to generate a turbidity boundary function over continuous time. The particle size distribution of the influent particulate matter is divided into multiple particle size groups according to the particle size range, and a concentration boundary function that varies with time is established for each particle size group. The concentration of organic matter in the influent is divided into multiple groups according to the type of organic matter, and a concentration boundary function that changes with time is established for each group. All generated flow boundary functions, turbidity boundary functions, concentration boundary functions for each particle size group, and concentration boundary functions for each type group are collectively bound to the flow field inlet boundary of the three-dimensional finite element model of the filter element.

9. The intelligent prediction method for water purifier filter cartridge life based on finite element analysis according to claim 8, characterized in that, The steps of simultaneously coupling and solving the flow field distribution described by the Navier-Stokes equations, the pollutant transport process described by the convection-diffusion equations, and the seepage process inside the porous media of the filter element described by Darcy's law on the three-dimensional finite element model of the filter element specifically include: The three-dimensional finite element model of the filter element is divided into a fluid domain and a porous media domain, wherein the fluid domain corresponds to the water inlet channel and water outlet channel of the filter element, and the porous media domain corresponds to the filter layer of the filter element. Solving the Navier-Stokes equations within the fluid domain yields the velocity and pressure field distributions for each finite element element within the fluid domain. The velocity field distribution and pressure field distribution at the interface between the fluid domain and the porous medium domain are passed to the porous medium domain as boundary conditions. Darcy's law is solved within the porous medium domain to obtain the seepage velocity field distribution and pore pressure distribution of each finite element element within the porous medium domain. Simultaneously solve the convection-diffusion equation within the porous medium domain to obtain the concentration distribution of particulate matter in each particle size group and the concentration distribution of organic matter in each type group within the porous medium domain. At the end of each simulation time step, the flux boundary conditions at the interface between the fluid domain and the porous medium domain are updated to achieve bidirectional coupling between the flow field and the concentration field.

10. The intelligent prediction method for water purifier filter cartridge life based on finite element analysis according to claim 1, characterized in that, The steps to obtain the filter cartridge model and inlet water quality parameters of the target water purifier specifically include: Receive the water purifier device serial number entered by the user through a mobile terminal, and query the corresponding filter model from the cloud device database based on the water purifier device serial number; Receive the water purifier installation location information input by the user through a mobile terminal, and obtain historical and real-time water quality data for that location from the water quality monitoring platform based on the water purifier installation location information; The historical water quality data and the real-time water quality data are integrated into influent water quality parameters, which include influent flow rate, influent turbidity, influent particulate matter size distribution and influent organic matter concentration. The filter cartridge model and the influent water quality parameters are stored as input data for the prediction model.

Images