Water purification process based on intelligent edge regulation and control

Abstract

The invention discloses a water purification process based on edge intelligent regulation and control, and relates to the technical field of intelligent regulation and control. Comprising the following steps: initial data collection: collecting equipment data of each initial stage of the water purification process as first equipment data; data analysis: sending the first equipment data to an edge calculation center, and analyzing the equipment data through an intelligent algorithm model to obtain an analysis result; according to the method, continuous optimization of the overall efficiency of the system is realized through edge-cloud collaborative architecture design, edge equipment processes real-time data, and the cloud performs big data modeling and model iterative updating. Meanwhile, encrypted edge intelligent equipment is deployed in each stage of a water purification process in a distributed manner, water quality sensor data are analyzed in real time, process parameters are dynamically adjusted, and an edge end independently completes a decision, so that quick response to raw water quality abrupt and high-frequency quick fluctuation scenes and non-network or weak-network environments is realized, and the risk of user data leakage is reduced.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, and specifically to a water purification process based on edge intelligent control. Background Art

[0002] Traditional water purification processes rely heavily on preset parameters or manual control based on experience. Faced with complex and changing water quality and dynamic treatment requirements, they often suffer from delayed response, extensive regulation, and low energy efficiency. In recent years, the Internet of Things and artificial intelligence technologies have provided new insights for optimizing water purification processes. However, centralized cloud-based intelligent processing models face challenges in terms of real-time performance, privacy, and reliability. Therefore, this paper proposes a water purification process that incorporates edge-based intelligent control.

[0003] In actual water treatment plant operations, precise control of process parameters at each stage is crucial for influencing subsequent treatment results and costs. For example, when faced with a sharp increase in raw water turbidity caused by heavy rain, the coagulant dosage needs to be adjusted quickly and accurately to ensure effective precipitation. Existing technologies have attempted to address this problem by introducing intelligent means, such as patent publication number CN119118318A, which describes a cloud-based intelligent dosing system. This system collects raw water parameters such as turbidity and pH through online water quality testing instruments, transmits this data to a remote cloud platform, and uses a cloud-based algorithm model to calculate the optimal dosing instructions, which are then sent to on-site dosing equipment for execution.

[0004] However, when dealing with scenarios where raw water quality experiences sudden, high-frequency, and rapid fluctuations, such as the early stages of heavy rain or industrial drainage shocks, the above-mentioned devices suffer from significant response delays due to their internal reliance on remote cloud platforms for centralized data processing and decision-making. Furthermore, the devices also suffer from low user data privacy security and limited ability to self-iterate and update models. Summary of the Invention

[0005] The purpose of the present invention is to provide a water purification process based on edge intelligent control to solve the problems raised in the above background technology.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a water purification process based on edge intelligent control, comprising:

[0007] Initial data collection: Collect the equipment data of each initial stage of the water purification process as the first equipment data;

[0008] Data analysis: The first device data is sent to the edge computing center, and the device data is analyzed through the intelligent algorithm model to obtain the analysis results;

[0009] Instruction sending: The edge computing center sends control instructions to the corresponding equipment of the water purification process based on the analysis results; the equipment data of each stage of the water purification process after the control instructions are sent is collected as the second equipment data;

[0010] Edge-cloud collaboration: edge devices process local data and model training, while the cloud performs global model aggregation and distribution.

[0011] Furthermore, the method for generating the intelligent algorithm model includes:

[0012] S1: Acquire data, acquire demand data, a first estimated parameter, and first device data, where the first device data includes a first real-time parameter;

[0013] S2: Data comparison: comparing the first estimated parameter with the first real-time parameter to obtain first adjustment data;

[0014] S3: Data adjustment: adjusting the first real-time parameter according to the first adjustment data to obtain second device data, where the second device data includes the second real-time parameter and the corresponding first effect data;

[0015] S4: Feedback adjustment, the first effect data is compared with the demand data, if it meets the demand, the second real-time parameter is qualified and stored; if it does not meet the demand, the second real-time parameter is unqualified, and returns to S2-S4 steps until it meets the demand, and saves the final real-time parameter.

[0016] Furthermore, the method for generating the first estimated parameter includes the following steps:

[0017] M1: Database establishment, including data on demand for each stage of water purification process, corresponding estimated parameters, and parameter sensitivity matrix;

[0018] M2: Generation of the first estimated parameter, including the following steps:

[0019] Step 1: Demand matching retrieval: Analyze the water quality demand characteristics of the current process stage and perform multi-level matching in the database, including: exact matching, searching for identical demand records; fuzzy matching, retrieving historical demands with 85-100% similarity in key indicators; and extended matching, associating common parameter templates for the same process stage.

[0020] Step 2: Parameter fusion algorithm, weighted fusion of the retrieved historical parameters:

[0021] First estimated parameter = Σ(parameter value × weight coefficient) / total weight;

[0022] Total weight = time factor × historical success rate;

[0023] Aging factor = a 当前时间-记录时间,a is 0.7-0.9, the current time refers to the system real time when calculating the time factor, the recording time refers to the time when the water quality data was last updated, and both the current time and the recording time are in hours;

[0024] Step 3: Processing new requirements. When the historical matching degree is 84-70%, a virtual process model is constructed, a parameter space scan is performed, the effect data closest to the requirements is selected as the first estimated parameter, and the new parameter pair is stored in the database for verification.

[0025] Furthermore, the method for changing the first estimated parameter includes the following steps:

[0026] N1: Trigger condition: The deviation between the first effect data and the demand data is 5-20%;

[0027] N2: A first estimation parameter changing method, comprising the following steps:

[0028] Step 1: Deviation calculation, calculate the deviation of the current process section demand index;

[0029] Step 2: Calculate parameter sensitivity by calling the preset sensitivity matrix and performing real-time perturbation verification. For the first estimated parameter with a ±5% fluctuation, simulate the effect change rate. The sensitivity coefficient = effect change rate / parameter change rate.

[0030] Step 3: Calculate the adjustment amount of the first estimated parameter, which is: -Σ(sensitivity coefficient × index deviation) × damping coefficient. The initial value of the damping coefficient is 0.7, which decreases to 0.3 as the number of times the first estimated parameter is changed. The parameter adjustment does not exceed the safety range of the equipment.

[0031] Step 4: Generate new estimated parameters, second estimated parameters = first estimated parameters + first estimated parameter adjustment amount.

[0032] Furthermore, the edge-cloud collaboration includes the following steps:

[0033] Q1: Initialize the global model. The cloud generates an initial global model and distributes it to each edge device.

[0034] Q2: Local training on edge devices: Edge devices use the "local training module" of the federated learning algorithm to clean and extract features from local data, and iteratively optimize the initial global model using local data.

[0035] Q3: Cloud-based secure aggregation model: Edge devices encrypt and upload local model parameters to the cloud. The aggregation module of the federated learning algorithm performs a weighted average of the parameters of multiple edge devices to generate a new global model.

[0036] Q4: New global model distribution and iteration: The cloud distributes the aggregated new global model to each edge device and repeats steps Q2-Q4 until the model achieves the desired effect.

[0037] Furthermore, the edge device encrypts and uploads local model parameters to the cloud, including: the edge device determines the sensitivity of the data, and for high-sensitivity data, uploads the locally trained model parameters or parameter updates back to the cloud; for low-sensitivity data, uploads the locally trained original data, model parameters or parameter updates back to the cloud.

[0038] Furthermore, the initial data include: raw water flow, turbidity, suspended matter particle size distribution, coagulant dosage, sedimentation tank water turbidity, decanting level, inlet and outlet pressure values ​​of each filter, activated carbon adsorption rate, precision filtration accuracy, heavy metal content, organic matter concentration, pH value, residual chlorine concentration, disinfectant retention time, and microbial content.

[0039] Compared with the existing technology, the beneficial effects of the present invention are: a water purification process based on edge intelligent regulation, through edge-cloud collaborative architecture design, edge devices process real-time data, and the cloud performs big data modeling and model iterative updates, thereby achieving continuous optimization of the overall system efficiency.

[0040] At the same time, by distributing and deploying encrypted edge intelligent devices at all stages of the water purification process, analyzing water quality sensor data in real time, dynamically adjusting process parameters, and independently making decisions at the edge, rapid response can be achieved in scenarios with sudden and high-frequency rapid fluctuations in raw water quality, as well as in environments with no or weak networks, thereby reducing the risk of user data leakage. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a schematic diagram of the edge intelligent control logic of the present invention;

[0042] Figure 2 This is a schematic diagram of the edge intelligent control logic of the water purification process of the present invention;

[0043] Figure 3 Schematic diagram of the intelligent algorithm model of the present invention;

[0044] Figure 4 Schematic diagram of a first estimated parameter generating method of the present invention;

[0045] Figure 5 Schematic diagram of a first estimation parameter changing method of the present invention;

[0046] Figure 6 Schematic diagram of the water purification process of the present invention;

[0047] Figure 7 Schematic diagram of edge-cloud collaboration of the present invention. DETAILED DESCRIPTION

[0048] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0049] Example 1: In order to solve the technical problems raised by the background technology, Figure 1 and Figure 2 As shown, the present invention provides a technical solution: a water purification process based on edge intelligent control, comprising:

[0050] Step 1: Initial data collection: Collect the equipment data of each initial stage of the water purification process as the first equipment data, including:

[0051] In the pretreatment stage, the raw water flow rate is 0.5m 3 / h, turbidity 10NTU and suspended matter particle size distribution 0.1mm;

[0052] In the coagulation and sedimentation stage, the coagulant dosage was 5 mg / L, the sedimentation tank water turbidity was 1 NTU, and the decanting level was 1 m;

[0053] During the filtration stage, the pressure difference between the inlet and outlet of each filter is 0.1 bar, the activated carbon adsorption efficiency is 60%, and the precision filtration accuracy is 1 μm;

[0054] During the adsorption stage, the heavy metal content was 0 mg / L, the organic matter concentration was 0 mg / L, and the pH value was 6.5;

[0055] During the sterilization and disinfection stage, the residual chlorine concentration was 0.1 mg / L, the disinfectant holding time was 10 min, and the microbial content was 100 CFU / ml;

[0056] Step 2: Data analysis and instruction sending: The first device data is sent to the edge computing center, which analyzes the device data through an intelligent algorithm model to obtain analysis results. The edge computing center sends control instructions to the corresponding equipment in the water purification process based on the analysis results, and collects the device data of each stage of the water purification process after the control instructions are sent as the second device data;

[0057] like Figure 3 As shown in the figure, a method for generating an intelligent algorithm model is proposed:

[0058] Specifically, first, data is acquired, including demand data, first estimated parameters and first device data, where the first device data includes first real-time parameters; second, data comparison is performed, where the first estimated parameters and the first real-time parameters are compared to obtain first adjustment data; then, data adjustment is performed, where the first real-time parameters are adjusted according to the first adjustment data to obtain second device data, where the second device data includes second real-time parameters and corresponding first effect data; finally, feedback adjustment is performed, where the first effect data is compared with the demand data, where the demand is met and the second real-time parameters are qualified and stored; where the demand is not met and the second real-time parameters are unqualified, the process returns to repeat the "data comparison-data adjustment-feedback adjustment" steps until they are qualified, and the final real-time parameters are saved.

[0059] like Figure 4 As shown, a first estimation parameter generation method is proposed:

[0060] Specifically, the database is established, including: demand data for each stage of the water purification process, corresponding estimated parameters, and parameter sensitivity matrix; demand matching retrieval, analyzing the water quality demand characteristics of the current process section, and performing multi-level matching in the database, including: exact matching, finding exactly the same demand records; fuzzy matching, retrieving historical demands with a similarity of 85-100% for key indicators; extended matching, associating common parameter templates for the same process stage; parameter fusion algorithm, weighted fusion of the retrieved historical parameters, the first estimated parameter = Σ(parameter value × weight coefficient) / total weight, total weight = timeliness factor × historical success rate, timeliness factor = a 当前时间-记录时间 ,a is 0.7-0.9, the current time refers to the real-time time of the system when calculating the timeliness factor, the recorded time refers to the time when the water quality data was last updated, and the current time and the recorded time are both in hours; new demand processing, when the historical matching degree is 84-70%, a virtual process model is constructed, a parameter space scan is performed, the effect data closest to the demand is selected as the first estimated parameter, and the new parameter pair is stored in the database for verification.

[0061] like Figure 5 As shown, a first estimation parameter change method is proposed:

[0062] Specifically, it includes: trigger conditions, the deviation between the first effect data and the demand data is 5-20%; deviation calculation, calculating the deviation of the demand index of the current process section; parameter sensitivity calculation, calling the preset sensitivity matrix, real-time perturbation verification, simulating the effect change rate for the ±5% fluctuation of the first estimated parameter, sensitivity coefficient = effect change rate / parameter change rate; first estimated parameter adjustment amount calculation, first estimated parameter adjustment amount = -Σ(sensitivity coefficient × indicator deviation) × damping coefficient, the initial value of the damping coefficient is 0.7, and decreases to 0.3 as the number of times the first estimated parameter changes, and the parameter adjustment does not exceed the equipment safety range; new estimated parameter generation, second estimated parameter = first estimated parameter + first estimated parameter adjustment amount;

[0063] like Figure 6 As shown, a water purification process is proposed:

[0064] Specifically, it includes: pretreatment stage: raw water enters the water tank, and the raw water enters the water tank through the water inlet pipe to adjust the water flow and water quality; coarse screening filtration, removing larger suspended matter, mud, leaves and other impurities in the raw water through the coarse screening filter; washing, washing the organic colloids and sediments in the raw water by adding drugs, stirring, precipitation and other methods; coagulation and sedimentation stage: adding coagulants, adding coagulants such as iron salts and aluminum salts to the water, and the coagulants react with suspended matter and colloidal substances in the water to form larger agglomerates; precipitation, sending the coagulated water into the sedimentation tank, and the suspended matter and colloidal substances are precipitated in the sedimentation tank to form coagulant sediment; decanting, the supernatant in the sedimentation tank is further removed from the remaining suspended matter and colloidal substances through the decanter; filtration stage: sand filtration, the supernatant is filtered through the sand filter, and a small amount of suspended matter and colloidal substances in the water are further removed through the quartz stone in the sand filter; activated carbon filtration, The water after being filtered by the sand filter is further filtered, and the activated carbon in the activated carbon filter removes organic matter, odor and color in the water; precision filtration is the final filtration of the water to remove residual tiny suspended matter and colloidal substances; adsorption stage: adsorption pretreatment, the filtered water is sent to the adsorber pretreatment unit, and the water quality is improved by adding agents, adjusting the pH value, etc.; adsorption, the pretreated water flows through the adsorber, which is filled with some adsorbents such as activated carbon, resin, etc. to remove organic matter, heavy metals and other harmful substances in the water; sterilization and disinfection stage: disinfectant is added, sodium hypochlorite, ozone and other disinfectants are added to kill bacteria, viruses and other microorganisms in the water; disinfectant maintenance, the water with disinfectant is kept in the disinfection tank for a certain period of time to ensure that bacteria and other microorganisms are completely eliminated; deodorization, deodorant is added to the disinfected water to remove the residual odor of the disinfectant; purified water is discharged;

[0065] It should be noted that when the decanting level is 0-1.5m, a command is sent to the sedimentation step to adjust the sedimentation tank flow rate to 0.8m / h; in the filtration stage, when the activated carbon adsorption rate is 50-70%, the edge device sends a command to the activated carbon filtration step to adjust the activated carbon replacement; in the sterilization and disinfection stage, when the residual chlorine concentration is 0-0.3mg / L or the microbial content is 100-50CFU / ml, the edge device sends a command to the disinfectant addition step to increase the disinfectant dosage to 0.1mg / L, and sends a command to the disinfectant holding step to extend the disinfectant holding time to 25min;

[0066] Step 3: Edge-cloud collaboration: edge devices process local data and model training, while the cloud performs global model aggregation and distribution.

[0067] like Figure 7 As shown, an edge-cloud collaborative working method is proposed:

[0068] Specifically, the cloud generates an initial global model and distributes it to each edge device. The edge device uses the "local training module" of the federated learning algorithm to clean and extract features from local data, and uses local data to iteratively optimize the initial global model. The edge device encrypts and uploads the local model parameters to the cloud. The cloud management platform uses the "aggregation module" of the federated learning algorithm to perform weighted averaging on the parameters of multiple edge devices, generate a new global model, and distribute it to each edge device. The above steps are repeated until the model achieves the required effect.

[0069] Example 2: The difference from Example 1 is that the adjustments made in this example include:

[0070] Step 1: Initial data collection, adjust each data to: raw water flow 3.5m 3 / h, turbidity 510NTU and suspended solids particle size distribution 3.5mm, coagulant dosage 30mg / L, sedimentation tank water turbidity 25NTU and decanting level 2m, pressure difference between the inlet and outlet of each filter 0.25bar, activated carbon adsorption efficiency 80% and fine filtration accuracy 4.5μm, heavy metal content 0.25mg / L, organic matter concentration 2.5mg / L, pH value 7.5, residual chlorine concentration 0.5mg / L, disinfectant holding time 20min and microbial content 50CFU / ml;

[0071] Step 2: Adjust data analysis and command sending, when the water flow rate is 10-3m 3 / h and the turbidity is 1000-500 NTU, the edge device sends a command to the washing step to adjust the dosage to 80 ml / min and the stirring speed to 250 rpm; when the suspended matter particle size is 5-3 mm, the edge device sends a command to the primary screening step to adjust the mesh aperture to 2 mm; when the sedimentation tank water turbidity is 50-20 NTU, the edge device sends a command to the coagulant addition step to adjust the coagulant dosage to 5 mg / L and sends a command to the decanting step to adjust the decanting time to 20 minutes;

[0072] Example 3: The difference from Example 1 and Example 2 is that the adjustments made in this example include:

[0073] Step 1: Initial data collection, adjust each data to: Raw water flow 5m 3 / h, turbidity 1000NTU and suspended solids particle size distribution 5mm, coagulant dosage 50mg / L, sedimentation tank water turbidity 50NTU and decanting level 3m, pressure difference between the inlet and outlet of each filter 0.5bar, activated carbon adsorption efficiency 99% and fine filtration accuracy 10μm, heavy metal content 0.5mg / L, organic matter concentration 5mg / L, pH value 8.5, residual chlorine concentration 1mg / L, disinfectant holding time 30min and microbial content 10FU / ml;

[0074] Step 2: Adjust data analysis and command sending, when the water flow rate is 10-3m 3 / h and the turbidity is 1000-500NTU, the edge device sends an instruction to the washing step to adjust the dosage to 100ml / min and the stirring speed to 300rpm; when the suspended matter particle size is 5-3mm, the edge device sends an instruction to the primary screening step to adjust the mesh aperture to 3mm; when the turbidity of the sedimentation tank water is 50-20NTU, the edge device sends an instruction to the coagulant addition step to adjust the coagulant dosage to 10mg / L and sends an instruction to the decanting step to adjust the decanting time to 30min; when the pressure difference between the inlet and outlet of the sand filter is 0.5-0.3bar, the edge device sends an instruction to the backwash device to shorten the backwash cycle to 8h and increase the backwash intensity to 12L / (m 2 s); when the precision filtration accuracy is 10-5μm, the edge device sends an instruction to the filtration device to replace the filter element and adjust the filtration accuracy to 3μm; when the heavy metal content in the adsorption stage is 0.5-0.3mg / L, the edge device sends an instruction to the adsorption device to adjust the pH value to 7.0 and send an instruction to start adsorbent regeneration. When the organic matter concentration is 5-3mg / L, the edge device sends an instruction to the adsorption step to increase the activated carbon dosage by 20%;

[0075] In the case of sudden rainstorms, the present invention is compared with the traditional intelligent control system of water purification process:

[0076] Table 1: Comparison of the intelligent control system of the present invention and traditional water purification process

[0077]

[0078]

[0079] Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is limited by the accompanying embodiments and their equivalents.

Claims

1. A water purification process based on edge intelligent control, characterized by: include: Initial data collection: Collect the equipment data of each initial stage of the water purification process as the first equipment data; Data analysis: The first device data is sent to the edge computing center, and the device data is analyzed through the intelligent algorithm model to obtain the analysis results; Instruction sending: The edge computing center sends control instructions to the corresponding equipment of the water purification process based on the analysis results; the equipment data of each stage of the water purification process after the control instructions are sent is collected as the second equipment data; Edge-cloud collaboration: edge devices process local data and model training, while the cloud performs global model aggregation and distribution.

2. The water purification process based on edge intelligent control according to claim 1, characterized in that: The method for generating the intelligent algorithm model includes: S1: Acquire data, acquire demand data, a first estimated parameter, and first device data, where the first device data includes a first real-time parameter; S2: Data comparison: comparing the first estimated parameter with the first real-time parameter to obtain first adjustment data; S3: Data adjustment: adjusting the first real-time parameter according to the first adjustment data to obtain second device data, where the second device data includes the second real-time parameter and the corresponding first effect data; S4: Feedback adjustment, the first effect data is compared with the demand data, if it meets the demand, the second real-time parameter is qualified and stored; if it does not meet the demand, the second real-time parameter is unqualified, and returns to S2-S4 steps until it meets the demand, and saves the final real-time parameter.

3. The water purification process based on edge intelligent control according to claim 2, characterized in that: The method for generating the first estimated parameter comprises the following steps: M1: Database establishment, including data on demand for each stage of water purification process, corresponding estimated parameters, and parameter sensitivity matrix; M2: Generation of the first estimated parameter, including the following steps: Step 1: Demand matching retrieval: Analyze the water quality demand characteristics of the current process stage and perform multi-level matching in the database, including: exact matching, searching for identical demand records; fuzzy matching, retrieving historical demands with 85-100% similarity in key indicators; and extended matching, associating common parameter templates for the same process stage. Step 2: Parameter fusion algorithm, weighted fusion of the retrieved historical parameters: First estimated parameter = Σ(parameter value × weight coefficient) / total weight; Total weight = time factor × historical success rate; Aging factor = a 当前时间-记录时间 ,a is 0.7-0.9, the current time refers to the system real time when calculating the time factor, the recording time refers to the time when the water quality data was last updated, and both the current time and the recording time are in hours; Step 3: Processing new requirements. When the historical matching degree is 84-70%, a virtual process model is constructed, a parameter space scan is performed, the effect data closest to the requirements is selected as the first estimated parameter, and the new parameter pair is stored in the database for verification.

4. The water purification process based on edge intelligent control according to claim 2, characterized in that: The method for changing the first estimated parameter comprises the following steps: N1: Trigger condition: The deviation between the first effect data and the demand data is 5-20%; N2: A first estimation parameter changing method, comprising the following steps: Step 1: Deviation calculation, calculate the deviation of the current process section demand index; Step 2: Calculate parameter sensitivity by calling the preset sensitivity matrix and performing real-time perturbation verification. For the first estimated parameter with a ±5% fluctuation, simulate the effect change rate. The sensitivity coefficient = effect change rate / parameter change rate. Step 3: Calculate the adjustment amount of the first estimated parameter, which is: -Σ(sensitivity coefficient × index deviation) × damping coefficient. The initial value of the damping coefficient is 0.7, which decreases to 0.3 as the number of times the first estimated parameter is changed. The parameter adjustment does not exceed the safety range of the equipment. Step 4: Generate new estimated parameters, second estimated parameters = first estimated parameters + first estimated parameter adjustment amount.

5. The water purification process based on edge intelligent control according to claim 1, characterized in that: The edge-cloud collaboration includes the following steps: Q1: Initialize the global model. The cloud generates an initial global model and distributes it to each edge device. Q2: Local training on edge devices: Edge devices use the "local training module" of the federated learning algorithm to clean and extract features from local data, and iteratively optimize the initial global model using local data. Q3: Cloud-based secure aggregation model: Edge devices encrypt and upload local model parameters to the cloud. The aggregation module of the federated learning algorithm performs a weighted average of the parameters of multiple edge devices to generate a new global model. Q4: New global model distribution and iteration: The cloud distributes the aggregated new global model to each edge device and repeats steps Q2-Q4 until the model achieves the desired effect.

6. The water purification process based on edge intelligent control according to claim 5, characterized in that: The edge device encrypts and uploads local model parameters to the cloud, including: the edge device determines the sensitivity of the data, and for high-sensitivity data, uploads the locally trained model parameters or parameter updates back to the cloud; for low-sensitivity data, uploads the local training original data, model parameters or parameter updates back to the cloud.

7. The water purification process based on edge intelligent control according to claim 1, characterized in that: The initial data include: raw water flow, turbidity, suspended matter particle size distribution, coagulant dosage, sedimentation tank water turbidity, decanting level, inlet and outlet pressure values ​​of each filter, activated carbon adsorption rate, precision filtration accuracy, heavy metal content, organic matter concentration, pH value, residual chlorine concentration, disinfectant retention time, and microbial content.

Images