Control system and control method for AI dynamic regulation and control of low-cost magnetic coagulating sedimentation

Abstract

The invention discloses a control system and a control method for AI dynamic regulation and control of low-cost magnetic coagulating sedimentation, and relates to the technical field of sewage treatment, the system comprises: a dynamic water quality monitoring module for performing dynamic water quality monitoring on a target water body through a predetermined probe; the precipitate adding amount information composition module is used for reading the to-be-added category and forming precipitate adding amount information; the real-time precipitation control strategy obtaining module is used for dynamically adjusting the precipitation control standard; and the magnetic coagulating sedimentation control treatment module is used for performing magnetic coagulating sedimentation control treatment on the target water body according to the real-time sedimentation control strategy. The technical problems that in the prior art, the micro-morphology of magnetic flocs cannot be sensed in real time, flocculation conditions are difficult to dynamically optimize according to water quality changes, a high-precision sensor and a controller are relied on, and then the magnetic coagulation sedimentation treatment cost is high, and flexibility and stability are insufficient are solved. The technical effects of reducing the magnetic coagulation sedimentation treatment cost and improving the water quality treatment flexibility and stability are achieved.

Description

Technical Field

[0001] This application relates to the technical field of sewage treatment, and specifically relates to a control system and control method for AI dynamic regulation of low-cost magnetic coagulation precipitation. Background Art

[0002] In the field of sewage treatment, magnetic coagulation technology, as an important means to improve the efficiency and quality of water treatment, has been widely used in recent years. However, most of the sedimentation control methods of traditional magnetic coagulation equipment adopt fixed dosing strategies, that is, fixed parameters are used to control the dosing amounts of coagulants and magnetic powder, and the dynamic changes of water quality cannot be sensed in real time. In the actual sewage environment, the water quality fluctuates frequently due to various factors, such as intermittent drainage in industrial production, diurnal changes in the amount of urban domestic sewage, and the mixing of rainwater. The fixed-parameter dosing method cannot adapt to the dynamic changes of water quality in time. When the water quality deteriorates, insufficient dosing leads to poor treatment effects and the effluent water quality is difficult to meet the standards stably; while when the water quality is better, excessive dosing causes high drug consumption, greatly increasing the sewage treatment cost; and currently, magnetic coagulation equipment lacks the ability to sense the microscopic morphology of magnetic flocs in real time, and thus it is difficult to dynamically optimize the flocculation conditions according to the actual situation, making it difficult for the entire magnetic coagulation process to achieve the best effect. In addition, the upgrade scheme of existing magnetic coagulation equipment requires the configuration of high-precision sensors and special controllers to improve the equipment performance. For small and medium-sized sewage treatment plants, the transformation cost is high and unaffordable, and due to the dependence on cloud computing, it affects the timely response to water quality changes and also brings data privacy risks.

[0003] Therefore, in the current related technologies, there are technical problems such as the inability to sense the microscopic morphology of magnetic flocs in real time, the difficulty in dynamically optimizing the flocculation conditions according to water quality changes, the dependence on high-precision sensors and controllers, which in turn lead to high costs of magnetic coagulation precipitation treatment and insufficient flexibility and stability. Summary of the Invention

[0004] By providing a control system and control method for AI dynamic regulation of low-cost magnetic coagulation precipitation, this application solves the technical problems in the prior art, such as the inability to sense the microscopic morphology of magnetic flocs in real time, the difficulty in dynamically optimizing the flocculation conditions according to water quality changes, the dependence on high-precision sensors and controllers, which in turn lead to high costs of magnetic coagulation precipitation treatment and insufficient flexibility and stability, and achieves the technical effects of reducing the cost of magnetic coagulation precipitation treatment and improving the flexibility and stability of water treatment.

[0005] This application provides a control system for AI dynamic regulation of low-cost magnetic coagulation precipitation. The system includes: a dynamic water quality monitoring module for dynamically monitoring the water quality of the target water body through a predetermined probe to obtain real-time water quality information; a precipitation dosing amount information composition module for reading the categories to be dosed and analyzing to obtain the first precipitation dosing amount of the first category in the categories to be dosed, forming precipitation dosing amount information with the real-time water quality information as a constraint; a real-time precipitation control strategy obtaining module for forming dynamic variable constraints based on the real-time water quality information and the precipitation dosing amount information, and dynamically adjusting the precipitation control benchmark under the dynamic variable constraints to obtain a real-time precipitation control strategy; and a magnetic coagulation precipitation control processing module for performing magnetic coagulation precipitation control processing on the target water body according to the real-time precipitation control strategy.

[0006] In a possible implementation, the control system for AI dynamic regulation of low-cost magnetic coagulation precipitation further performs the following processing: The real-time water quality information at least includes real-time turbidity, real-time pH value, and real-time temperature.

[0007] In a possible implementation, the control system for AI dynamic regulation of low-cost magnetic coagulation precipitation further performs the following processing: The categories to be dosed include flocculants and magnetic powder.

[0008] In a possible implementation, the control system for AI dynamic regulation of low-cost magnetic coagulation precipitation further performs the following processing: Extract the predetermined water quality information in the precipitation control benchmark; compare the real-time water quality information with the predetermined water quality information to obtain a real-time water quality deviation coefficient; extract the predetermined dosing amount information in the precipitation control benchmark; compare the precipitation dosing amount information with the predetermined dosing amount information to obtain a dosing amount deviation coefficient; perform weighted calculation on the real-time water quality deviation coefficient and the dosing amount deviation coefficient to obtain a dynamic variable factor; and adjust the precipitation control benchmark based on the dynamic variable factor to obtain the real-time precipitation control strategy.

[0009] In a possible implementation, the control system for AI dynamic regulation of low-cost magnetic coagulation precipitation further performs the following processing: Obtain any water quality index; respectively traverse and match in the real-time water quality information and the predetermined water quality information to obtain the real-time water quality parameter and the predetermined water quality parameter corresponding to the any water quality index; introduce a deviation evaluation function to evaluate and analyze the real-time water quality parameter and the predetermined water quality parameter to obtain the real-time water quality deviation coefficient; where the expression of the deviation evaluation function is: ; refers to the real-time water quality deviation coefficient, refers to the value of the refers to the the value of a predetermined water quality parameter, refers to the weight coefficient of the th parameter, and is an integer greater than or equal to 3.

[0010] In a possible implementation, the control system for AI dynamic regulation of low-cost magnetic coagulation sedimentation further performs the following processing: extracting the mixing control information in the sedimentation control reference, where the mixing control information includes a predetermined stirring intensity and a predetermined reaction duration; weighted adjustment of the predetermined stirring intensity and the predetermined reaction duration using the dynamic variable factor as the weight to obtain a real-time stirring intensity and a real-time reaction duration respectively; forming the real-time sedimentation control strategy based on the real-time stirring intensity and the real-time reaction duration.

[0011] In a possible implementation, the control system for AI dynamic regulation of low-cost magnetic coagulation sedimentation further performs the following processing: dynamically collecting a real-time microscopic image of the target magnetic flocs in the target water body; performing feature evaluation and analysis on the real-time microscopic image according to the microscopic state evaluation mechanism to obtain a real-time microscopic state; obtaining an arbitrary information value coefficient of the real-time microscopic state for any water quality index; combining the arbitrary information value coefficient to perform predictive analysis on the real-time water quality parameter to obtain a predicted real-time water quality parameter; calibrating the real-time water quality parameter with the predicted real-time water quality parameter.

[0012] In a possible implementation, the control system for AI dynamic regulation of low-cost magnetic coagulation sedimentation further performs the following processing: collecting real-time image feature information of the real-time microscopic image; performing normalized weighting on the real-time image color parameter and the real-time image structure parameter in the real-time image feature information to obtain a real-time image feature value; combining the microscopic state coefficient stored in the microscopic state evaluation mechanism to perform standardized correction on the real-time image feature value to obtain the real-time microscopic state.

[0013] In a possible implementation, the control system for AI dynamic regulation of low-cost magnetic coagulation sedimentation further performs the following processing: extracting a first historical data set from the magnetic coagulation sedimentation database; performing mutual information analysis on the first historical microscopic state of the first magnetic flocs and the first historical index parameter corresponding to the first historical any water quality index in the first historical data set to obtain a first historical mutual information coefficient; using the first historical mutual information coefficient as the arbitrary information value coefficient.

[0014] The present application also provides a control method for AI dynamic regulation of low-cost magnetic coagulation precipitation, including: dynamically monitoring the water quality of the target water body through a predetermined probe to obtain real-time water quality information; reading the types of substances to be added, and analyzing to obtain the first precipitation dosage of the first type in the types of substances to be added based on the real-time water quality information to form precipitation dosage information; forming dynamic variable constraints based on the real-time water quality information and the precipitation dosage information, and dynamically adjusting the precipitation control benchmark under the dynamic variable constraints to obtain a real-time precipitation control strategy; and performing magnetic coagulation precipitation control treatment on the target water body according to the real-time precipitation control strategy.

[0015] It is intended to solve the technical problems existing in the prior art, such as the inability to perceive the microscopic morphology of magnetic flocs in real time, the difficulty in dynamically optimizing the flocculation conditions according to water quality changes, and the dependence on high-precision sensors and controllers, which lead to relatively high costs of magnetic coagulation precipitation treatment and insufficient flexibility and stability, through the control system and control method for AI dynamic regulation of low-cost magnetic coagulation precipitation proposed in the present application. A dynamic water quality monitoring module is used to dynamically monitor the water quality of the target water body through a predetermined probe; a precipitation dosage information composition module is used to read the types of substances to be added and form precipitation dosage information; a real-time precipitation control strategy obtaining module is used to dynamically adjust the precipitation control benchmark to obtain a real-time precipitation control strategy; and a magnetic coagulation precipitation control treatment module is used to perform magnetic coagulation precipitation control treatment on the target water body according to the real-time precipitation control strategy. The technical effect of reducing the cost of magnetic coagulation precipitation treatment and improving the flexibility and stability of water quality treatment is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments of the present disclosure will be briefly introduced below. Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the operations before or below do not necessarily need to be executed precisely in sequence. On the contrary, according to the need, they can be executed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or several operations can be removed from these processes.

[0017] Figure 1 It is a schematic structural diagram of a control system for AI dynamic regulation of low-cost magnetic coagulation precipitation provided by an embodiment of the present application.

[0018] Figure 2 It is a schematic flowchart of a control method for AI dynamic regulation of low-cost magnetic coagulation precipitation provided by an embodiment of the present application.

[0019] Description of reference numerals: Dynamic water quality monitoring module 10, precipitation dosage information composition module 20, real-time precipitation control strategy obtaining module 30, magnetic coagulation precipitation control treatment module 40. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] The above description is only an overview of the technical solution of the present application. In order to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the specific embodiments of the present application are hereinafter specifically exemplified.

[0021] In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations of the present application. All other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0022] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments. However, it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict. The terms "first\second" involved are only used to distinguish similar objects and do not represent a specific order for the objects. The terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, system, product or server comprising a series of steps or units need not be limited to those steps or units clearly listed, but may include other steps or modules not clearly listed or inherent to these processes, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0023] The embodiment of the present application provides a control system for AI dynamic regulation of low-cost magnetic coagulation sedimentation, as Figure 1 shown. The system includes: A dynamic water quality monitoring module 10, configured to perform dynamic water quality monitoring on a target water body through a predetermined probe to obtain real-time water quality information.

[0024] Further, the specific configuration of the dynamic water quality monitoring module 10 further includes that the real-time water quality information at least includes real-time turbidity, real-time pH value and real-time temperature.

[0025] Preferably, a predetermined probe device (multi-parameter water quality probe) is used to continuously and automatically detect and collect data from the target water body to obtain real-time water quality information. The multi-parameter water quality probe integrates various sensors such as a turbidity sensor, a pH sensor, and a temperature sensor, and can measure multiple water quality parameters simultaneously, including the turbidity, pH value, and temperature of the water body. Moreover, the multi-parameter water quality probe can replace single-parameter sensors, reducing the number of probes to lower costs. Specifically, dynamic water quality monitoring means that the predetermined probe detects the target water body at regular time intervals or continuously (for example, collecting data every 5 minutes) to promptly capture the changes in water quality over time and reflect the dynamic characteristics of the water quality. Through the measurement of the multi-parameter water quality probe, the instant data of the target water body is obtained as real-time water quality information, including real-time turbidity, real-time pH value, and real-time temperature. Among them, turbidity is used to reflect the content of suspended particles in the water body. By measuring the degree of light scattering or absorption by the water body, the turbidity sensor in the multi-parameter water quality probe can obtain the turbidity information of the water body in real time, thereby understanding the amount of suspended particles in the water body. The pH value is used to measure the acidity and alkalinity of the water body. The pH sensor in the multi-parameter water quality probe reacts with hydrogen ions in the water body to generate corresponding electrical signals, thereby measuring the pH value of the water body in real time. In sewage treatment, different treatment processes need to be within a specific pH value range to achieve the best effect. The temperature sensor in the multi-parameter water quality probe usually uses components such as thermistors or thermocouples to obtain the temperature information of the water body in real time by measuring the temperature change of the water body. The change in water temperature will affect the solubility of chemical substances in the water body, the activity of microorganisms, and the reaction rate of chemical reactions, etc. Through the dynamic water quality monitoring of the target water body by the multi-parameter water quality probe, accurate data support is provided for the control of magnetic coagulation precipitation, so as to adjust the treatment strategy in a timely manner according to the changes in water quality and improve the treatment effect and stability.

[0026] The precipitation dosage information composition module 20 is configured to read the category to be added and, constrained by the real-time water quality information, analyze and obtain the first precipitation dosage of the first category in the category to be added, and form precipitation dosage information.

[0027] Furthermore, the specific configuration of the precipitation dosage information composition module 20 further includes that the category to be added includes a flocculant and magnetic powder.

[0028] Preferably, during the magnetic coagulation sedimentation treatment process, specific substances are added to the target water body to promote the sedimentation effect. The substances to be added mainly involve flocculants and magnetic powder. Among them, flocculants can cause suspended particles in water to aggregate into larger flocs, facilitating subsequent sedimentation separation; magnetic powder can increase the specific gravity of the flocs, accelerate the sedimentation speed, and improve the treatment efficiency. Specifically, under different water quality conditions, the dosage of the chemicals required to achieve the ideal sedimentation effect is different. Constrained by the real-time water quality information obtained by a predetermined probe, the dosage is determined. For example, a high turbidity indicates a large number of suspended particles in the water, and more flocculants and magnetic powder are required to promote particle aggregation and sedimentation. If the turbidity is low, the required dosage can be reduced; the pH value affects the hydrolysis and ionization processes of flocculants, and different flocculants have different flocculation effects within different pH value ranges. For example, some flocculants have good effects in a weakly acidic environment, while some are more effective in an alkaline environment; water temperature affects the rate of chemical reactions and the hydrolysis rate of flocculants. When the temperature is low, the chemical reaction slows down, and the hydrolysis of the flocculant is incomplete. It may be necessary to increase the dosage to ensure the flocculation effect. When the temperature is high, the reaction rate accelerates, and the dosage may be appropriately reduced.

[0029] Preferably, based on the real-time water quality information and referring to historical data, analyze and calculate the first sedimentation dosage corresponding to the first category in the categories to be added, that is, the dosage required to achieve the best sedimentation effect under the current water quality conditions. Among them, the first category is one of the categories to be added (flocculant or magnetic powder). Specifically, collect a large amount of historical water quality data and corresponding dosage data, perform preprocessing operations such as data cleaning and normalization on the data, and divide it into a training set and a test set. Then select a suitable machine learning algorithm, such as artificial neural network (ANN), support vector machine (SVM), etc. Use historical water quality parameters (turbidity, pH value, temperature, etc.) as input features and the dosage of the first category as the output label to train the model. During the training process, continuously adjust the parameters of the model so that the model can accurately learn the mapping relationship between water quality parameters and dosage, and obtain a dosage prediction model. Then input the real-time obtained water quality information into the trained dosage prediction model, and then output the first sedimentation dosage of the first category. For example, after training the artificial neural network model, when the real-time turbidity, real-time pH value, and real-time temperature are input, the output dosage of the flocculant is 6 mg / L, which is the first sedimentation dosage. Similarly, calculate the sedimentation dosage corresponding to the magnetic powder, and form sedimentation dosage information, which may include the numerical value of the dosage, the corresponding water quality parameters, the calculation time, etc., to guide the chemical addition operation during the magnetic coagulation sedimentation process.

[0030] The real-time sedimentation control strategy acquisition module 30 is used to form dynamic variable constraints based on the real-time water quality information and the sedimentation dosage information, and dynamically adjust the sedimentation control benchmark under the dynamic variable constraints to obtain a real-time sedimentation control strategy.

[0031] Preferably, the real-time water quality information (such as real-time turbidity, real-time pH value, and real-time temperature) and the precipitation dosage information (the precipitation dosage for each category) together constitute dynamic variable constraint conditions to reflect the actual condition of the current water body and the required chemical dosage. For example, a higher real-time turbidity indicates that the dosage of the flocculant needs to be increased to achieve a better precipitation effect; while the real-time pH value and temperature will affect the reaction activity of the flocculant, thus also imposing certain restrictions and requirements on the dosage; the precipitation dosage information directly stipulates the currently recommended chemical dosage, which cannot be randomly exceeded or below this range, otherwise it may lead to poor precipitation effect or chemical waste. Then, the precipitation control benchmark is dynamically adjusted according to the dynamic variable constraints, where the precipitation control benchmark is the control parameters and target values set under ideal working conditions (such as the standard dosage of flocculant and magnetic powder, precipitation time, stirring speed, etc. under conventional water quality conditions), and is used to guide the operation of the magnetic coagulation precipitation process.

[0032] Preferably, the precipitation control benchmark is adjusted based on the dynamic variable constraints. For example, if the real-time water quality information indicates an increase in the turbidity of the water body, according to the relationship between turbidity and dosage in the dynamic variable constraints, the dosage of the flocculant needs to be increased from the standard value, and the increased part is the dosage adjustment value calculated according to the change in turbidity. At the same time, other control parameters such as precipitation time and stirring speed may also need to be adjusted according to the real-time water quality information and precipitation dosage information, such as appropriately extending the precipitation time or adjusting the stirring speed to enable the increased flocculant to fully mix and react with the water body; and then a real-time precipitation control strategy is obtained, including adjusted chemical dosage, precipitation time, stirring speed, equipment operation parameters and other information, as well as response measures and adjustment methods under different water quality changes. For example, when the turbidity is between 50 - 80 NTU, the dosage of the flocculant is adjusted to 6 - 8 mg / L, the precipitation time is extended to 30 - 40 minutes, and the stirring speed is increased to 150 - 200 r / min in the initial stage and then gradually decreased to 50 - 80 r / min, etc. The real-time precipitation control strategy can be dynamically adjusted according to the continuously changing real-time water quality information and precipitation dosage information, thereby realizing a more efficient and stable magnetic coagulation precipitation process, improving the effluent water quality, and reducing chemical consumption and operating costs.

[0033] Further, the specific configuration of the real-time sedimentation control strategy acquisition module 30 further includes extracting predetermined water quality information from the sedimentation control benchmark; comparing the real-time water quality information with the predetermined water quality information to obtain a real-time water quality deviation coefficient; extracting predetermined dosage information from the sedimentation control benchmark; comparing the sedimentation dosage information with the predetermined dosage information to obtain a dosage deviation coefficient; performing weighted calculation on the real-time water quality deviation coefficient and the dosage deviation coefficient to obtain a dynamic variable factor; and adjusting the sedimentation control benchmark based on the dynamic variable factor to obtain the real-time sedimentation control strategy.

[0034] Preferably, extracting predetermined water quality information from the sedimentation control benchmark includes predetermined water quality indicators, such as predetermined turbidity, predetermined pH value, predetermined temperature, etc., representing the water quality state expected to be achieved under standard conditions. For example, in the sedimentation control benchmark, the predetermined turbidity may be set to 50 NTU, the predetermined pH value is 7.5, and the predetermined temperature is 22 °C; then comparing the real-time water quality information with the predetermined water quality information to calculate the deviation degree of each water quality indicator. For example, if the real-time turbidity is 60 NTU and the predetermined turbidity is 50 NTU, the turbidity deviation is 60 - 50 = 10 NTU. Then, converting the water quality indicator deviation degree into a deviation coefficient to consider the deviation impact of different water quality indicators. For example, dividing the deviation value by the predetermined value to obtain a relative deviation coefficient. The relative deviation coefficient of turbidity is 10 / 50 = 0.2. Finally, comprehensively processing the deviation coefficients of each water quality indicator (such as weighted average, etc.) to obtain the real-time water quality deviation coefficient, reflecting the deviation degree of the current real-time water quality from the predetermined water quality.

[0035] Preferably, further extracting predetermined dosage information from the sedimentation control benchmark is the amount of various chemicals (such as flocculants, magnetic powder, etc.) required to achieve an ideal sedimentation effect under predetermined water quality conditions. For example, the sedimentation control benchmark may stipulate that the predetermined dosage of the flocculant is 5 mg / L and the predetermined dosage of the magnetic powder is 10 mg / L; then comparing the sedimentation dosage information with the predetermined dosage information to calculate the deviation degree of the dosage. For example, if the actual analyzed sedimentation dosage of the flocculant is 6 mg / L and the predetermined dosage is 5 mg / L, the dosage deviation of the flocculant is 6 - 5 = 1 mg / L. Similarly, converting the dosage deviation into a deviation coefficient. For example, the relative deviation coefficient of the flocculant dosage is 1 / 5 = 0.2; then comprehensively processing the dosage deviation coefficients of various chemicals to obtain the dosage deviation coefficient, reflecting the deviation degree of the current actual dosage from the predetermined dosage. For example, as Figure 2 shown, it is the curve of chemical consumption and power consumption savings statistically calculated by a sewage treatment plant on a quarterly basis.

[0036] Preferably, the real-time water quality deviation coefficient and the dosage deviation coefficient are weightedly calculated. Specifically, different weights are assigned to the real-time water quality deviation coefficient and the dosage deviation coefficient according to actual experience and experimental data. For example, water quality has a greater impact on the sedimentation effect. The weight of the real-time water quality deviation coefficient is 0.6, and the weight of the dosage deviation coefficient is 0.4. The two deviation coefficients are multiplied by their respective weights and then added to obtain the dynamic variable factor. Assuming that the real-time water quality deviation coefficient is 0.3 and the dosage deviation coefficient is 0.2, the dynamic variable factor is 0.3×0.6+0.2×0.4=0.26. The dynamic variable factor comprehensively considers the changes in water quality and dosage, and can more comprehensively reflect the dynamic changes of the current magnetic coagulation sedimentation process.

[0037] Preferably, the sedimentation control benchmark is finally adjusted according to the dynamic variable factor, that is, the various parameters in the sedimentation control benchmark are adjusted. For example, if the dynamic variable factor is greater than 1, it means that the current water quality and dosage changes make the actual situation deviate greatly from the predetermined standard, and the dosage of the reagent is increased, the sedimentation time is extended, or the stirring speed is adjusted; if the dynamic variable factor is less than 1, the dosage of the reagent is appropriately reduced or the sedimentation time is shortened. By dynamically adjusting the sedimentation control benchmark, a real-time sedimentation control strategy is obtained, which can timely and accurately adjust the various parameters of the magnetic coagulation sedimentation process according to the real-time changes in water quality and dosage, thereby improving the sedimentation effect and reducing the treatment cost.

[0038] Furthermore, the specific configuration of the real-time sedimentation control strategy acquisition module 30 also includes obtaining any water quality index; traversing and matching the real-time water quality information and the predetermined water quality information to obtain the real-time water quality parameters and the predetermined water quality parameters corresponding to the arbitrary water quality index; introducing a deviation evaluation function to evaluate and analyze the real-time water quality parameters and the predetermined water quality parameters to obtain the real-time water quality deviation coefficient; wherein the expression of the deviation evaluation function is: ; refers to the real-time water quality deviation coefficient, It refers to The value of the real-time water quality parameter, It refers to The value of a predetermined water quality parameter, It refers to The weight coefficient of the parameter, is the sum of water quality parameters, and is an integer greater than or equal to 3.

[0039] Preferably, select one from the water quality indicators, for example, select turbidity. Then, in the real-time water quality information, find the specific value corresponding to any selected water quality indicator through search, lookup, etc., that is, the real-time water quality parameter. For example, the currently real-time monitored water turbidity is 20 NTU. Similarly, in the predetermined water quality information (the water quality parameter preset in the sedimentation control standard), find the predetermined value corresponding to the same water quality indicator, that is, the predetermined water quality parameter. For example, the turbidity specified in the sedimentation control standard should not exceed 15 NTU, and 15 NTU is the predetermined water quality parameter; use the deviation evaluation function to measure the deviation degree between the real-time water quality parameter and the predetermined water quality parameter, and calculate the real-time water quality deviation coefficient, where, and n is an integer greater than or equal to 3, indicating that at least 3 different water quality parameters will be considered during the evaluation.

[0040] Furthermore, the specific configuration of the real-time sedimentation control strategy obtaining module 30 further includes extracting the mixing control information in the sedimentation control standard, where the mixing control information includes a predetermined stirring intensity and a predetermined reaction duration; using the dynamic variable factor as a weight to perform weighted adjustment on the predetermined stirring intensity and the predetermined reaction duration, respectively obtaining the real-time stirring intensity and the real-time reaction duration; forming the real-time sedimentation control strategy based on the real-time stirring intensity and the real-time reaction duration.

[0041] Preferably, extracting the mixing control information from the sedimentation control standard specifically includes a predetermined stirring intensity and a predetermined reaction duration. The predetermined stirring intensity refers to the standard of the operation intensity of the stirring equipment under normal circumstances to fully mix the precipitant with the water body. For example, it is stipulated that the rotation speed of the mixer is 200 revolutions per minute, which is the predetermined stirring intensity; the predetermined reaction duration refers to the time required for the precipitant to react with the water body under normal conditions to achieve the expected sedimentation effect. For example, setting the reaction duration to 30 minutes is the predetermined reaction duration; then, using the dynamic variable factor as a weight to perform weighted adjustment on the predetermined stirring intensity and the predetermined reaction duration. Specifically, multiply the predetermined stirring intensity by the dynamic variable factor. If the dynamic variable factor is greater than 1, it indicates that the current water quality or dosing amount change makes the mixing process require stronger stirring, and at this time the real-time stirring intensity will be higher than the predetermined stirring intensity; if the dynamic variable factor is less than 1, the real-time stirring intensity will decrease. For example, the predetermined stirring intensity is 200 revolutions per minute and the dynamic variable factor is 1.2, then the real-time stirring intensity is 200×1.2 = 240 revolutions per minute.

[0042] Preferably, multiply the predetermined reaction duration by a dynamic variable factor. If the dynamic variable factor is greater than 1, due to reasons such as complex water quality or changes in dosing amounts, the precipitation reaction may require a longer time to proceed fully, and the real-time reaction duration will be extended. If the dynamic variable factor is less than 1, the real-time reaction duration will be shortened. For example, if the predetermined reaction duration is 30 minutes and the dynamic variable factor is 0.8, the real-time reaction duration is 30×0.8 = 24 minutes. Finally, the adjusted real-time stirring intensity and real-time reaction duration, together with other parameters during the precipitation process (such as the determined dosing amount of the precipitant, etc.), constitute the real-time precipitation control strategy, which is a specific operation plan for the current actual water quality and precipitant dosing situation, and is used to guide the operation of the precipitation equipment. For example, the real-time precipitation control strategy may stipulate that in the next precipitation operation, the stirring equipment operates at a real-time stirring intensity of 240 revolutions per minute, the reaction duration is controlled at 24 minutes, and the chemical agent is added according to the calculated dosing amount of the precipitant to achieve an efficient and stable precipitation effect.

[0043] Furthermore, the specific configuration of the real-time precipitation control strategy acquisition module 30 further includes dynamically collecting the real-time microscopic image of the target magnetic flocs in the target water body; performing feature evaluation and analysis on the real-time microscopic image according to the microscopic state evaluation mechanism to obtain the real-time microscopic state; obtaining the any information value coefficient of the real-time microscopic state for any water quality index; combining the any information value coefficient to perform predictive analysis on the real-time water quality parameters to obtain the predicted real-time water quality parameters; and calibrating the real-time water quality parameters with the predicted real-time water quality parameters.

[0044] Preferably, use a USB microscope with a price not exceeding 5000 yuan and a resolution of 5μm, combined with a mobile phone camera for transformation, to replace the industrial-grade microscopic camera, and dynamically capture the target magnetic flocs in the target water body to obtain the real-time microscopic image. Specifically, the transformed imaging unit can achieve the image acquisition function similar to that of the industrial-grade microscopic camera at a lower cost. For example, in the precipitation link of sewage treatment, by photographing the microscopic morphology of the magnetic flocs in the water, relevant information about the precipitation effect can be observed. Then, use the PLC embedded AI module to evaluate and analyze the characteristics (such as size, shape, aggregation degree, etc.) of the target magnetic flocs in the image according to the pre-set microscopic state evaluation mechanism, so as to obtain the current real-time microscopic state of the target water body. Among them, the AI module uses a lightweight neural network model (MobileNetV3). For example, if the magnetic flocs are large and closely aggregated, it indicates that the precipitation effect is good; if the magnetic flocs are small and dispersed, it indicates that the precipitation effect may be poor. Exemplarily, Table 1 shows the relevant data of the reuse transformation of a certain urban sewage treatment plant (treatment capacity: 10,000 tons per day): Table 1 List of Relevant Hardware Transformations for a Certain Treatment Preferably, the information value coefficient of any water quality index for the real-time microscopic state is then obtained. Specifically, different real-time microscopic states have different reference values for each water quality index, that is, the information value coefficient. Analyze the real-time microscopic image according to the microscopic state evaluation mechanism to determine the information value coefficient of the current real-time microscopic state for any water quality index. For any one water quality index (such as turbidity, pH value, etc.), determine the information value coefficient of the current real-time microscopic state for this index. For example, for the turbidity index, when the microscopic state shows that the magnetic flocs are closely aggregated, its information value coefficient for turbidity reduction may be relatively high; conversely, if the magnetic flocs are dispersed, the information value coefficient is relatively low.

[0045] Preferably, the obtained information value coefficient is combined with the real-time water quality parameters for predictive analysis. Specifically, through a large number of experiments or historical data, a relationship model (such as a linear regression model, a neural network model, etc.) between the information value coefficient corresponding to the real-time microscopic state and the water quality parameters is established. For example, taking the information value coefficient corresponding to the microscopic state (such as size, aggregation degree, etc.) of the target magnetic flocs as the independent variable and the water quality index (such as turbidity, dissolved oxygen, etc.) as the dependent variable, use historical data for training to obtain the mathematical relationship between the two; then substitute the information value coefficient into the mathematical relationship to calculate the predicted real-time water quality parameters; then compare the predicted real-time water quality parameters with the real-time water quality parameters obtained by monitoring, calculate the difference between the two, and finally determine the adjustment ratio according to historical data to calibrate the real-time water quality parameters to improve the accuracy of water quality monitoring and treatment.

[0046] Furthermore, the specific configuration of the real-time sedimentation control strategy acquisition module 30 further includes collecting the real-time image feature information of the real-time microscopic image; normalizing and weighting the real-time image color parameters and real-time image structure parameters in the real-time image feature information to obtain a real-time image feature value; combining the microscopic state coefficient stored in the microscopic state evaluation mechanism to perform standardization correction on the real-time image feature value to obtain the real-time microscopic state.

[0047] Preferably, after dynamically collecting real-time microscopic images of target magnetic flocs in the target water body using a modified USB microscope and a mobile phone camera, feature information in the image is extracted, including real-time image color parameters and real-time image structure parameters. Extracting real-time image color parameters specifically includes obtaining hue information by converting the image from an RGB color space to an HSV (hue, saturation, brightness) color space, and also obtaining saturation information in the HSV color space, wherein saturation (S) reflects the vividness of the color, and the value range is 0 to 1. Colors with higher saturation indicate higher purity, while colors with lower saturation are more inclined to gray. In the brightness (V) in the HSV space and in the RGB color space, the brightness information of the image is obtained by calculating the average value of pixel values ​​and other methods to reflect the overall brightness of the image. Factors such as the concentration and size of the magnetic flocs and the turbidity of the surrounding water body will affect the brightness of the image.

[0048] Preferably, extracting the real-time image structural parameters specifically includes: using an edge detection algorithm, such as the Canny algorithm, to extract the edge information of the magnetic floccules, and then analyzing and obtaining its shape characteristics; converting the pixel size in the image into the actual physical size according to the resolution of the image and the known magnification of the microscope to obtain the size characteristics; describing the texture information of the magnetic floccules by calculating the statistics of the gray level co-occurrence matrix GLCM, such as contrast, correlation, energy and entropy; calculating the average distance between the magnetic floccules, aggregation density and other parameters to analyze the distribution characteristics of the magnetic floccules in the image. Since the value ranges and dimensions of the real-time image color parameters and the real-time image structural parameters may be different, different real-time image feature parameters have different importance for judging the microscopic state of the magnetic floccules, and then the real-time image color parameters and real-time image structural parameters in the real-time image feature information are normalized and weighted, that is, different weights are given to the color parameters and the structural parameters, and the real-time image feature values ​​are calculated, which can preliminarily reflect the characteristics of the magnetic floccules presented in the image.

[0049] Preferably, the microstate evaluation mechanism pre-stores microstate coefficients corresponding to different microstates, and these coefficients are standard reference values ​​determined through a large number of experiments and analyses. Finally, the calculated real-time image characteristic values ​​are compared and calculated with these microstate coefficients (such as performing linear transformation), and the real-time image characteristic values ​​are standardized and corrected. The corrected result is the real-time microstate, which more accurately corresponds to the actual microstate of the target magnetic flocs in the target water body, such as whether it is in a good coagulation state, a state of beginning to disperse, etc.

[0050] Furthermore, the specific configuration of the real-time sedimentation control strategy acquisition module 30 also includes extracting a first historical data set from the magnetic coagulation sedimentation database; performing mutual information analysis on a first historical indicator parameter corresponding to a first historical microscopic state of a first magnetic floc in the first historical data set and a first historical arbitrary water quality indicator to obtain a first historical mutual information coefficient; and using the first historical mutual information coefficient as the arbitrary information value coefficient.

[0051] Preferably, a first historical data set is extracted from a magnetic coagulation sedimentation database, wherein the magnetic coagulation sedimentation database stores a large amount of data related to the magnetic coagulation sedimentation process, and the first historical data set includes various information about the first magnetic flocs and related data of the corresponding first historical arbitrary water quality indicators, including the state of the magnetic flocs and the changes in the water quality indicators under different conditions; then, a mutual information analysis is performed on the first historical microstate of the first magnetic flocs and the first historical indicator parameters corresponding to the first historical arbitrary water quality indicators in the first historical data set, that is, the degree of correlation between the two variables is determined by calculating the mutual information between them, wherein the mutual information is used to measure the degree of mutual dependence between two random variables, and the first historical microstate of the first magnetic flocs and the first historical indicator parameters corresponding to the first historical arbitrary water quality indicators are two random variables. For example, if the particle size distribution of the magnetic flocs in the first historical microstate is different from that in the first historical indicator parameters. There is a high mutual information between the turbidity and the real-time microscopic state of the first magnetic flocs, which means that the particle size distribution of the magnetic flocs has a strong influence on the turbidity, that is, the microscopic state of the magnetic flocs can reflect the changes in the water quality indicators to a certain extent, and then the first historical mutual information coefficient is obtained, which quantifies the degree of mutual dependence between the two. The value range is usually 0 to positive infinity. The larger the value, the higher the degree of mutual dependence. Finally, the calculated first historical mutual information coefficient is used as the arbitrary information value coefficient, which means that when predicting and analyzing the real-time water quality parameters, it is used to measure the importance or value of the real-time microscopic state of the first magnetic flocs for predicting any water quality indicator. For example, if the first historical mutual information coefficient is high, it means that the microscopic state of the first magnetic flocs has a high information value for the prediction of the corresponding water quality indicators. When predicting the real-time water quality parameters, it is necessary to pay more attention to the information provided by the real-time microscopic state. On the contrary, if the coefficient is low, it means that the microscopic state has a relatively small predictive effect on the water quality indicator.

[0052] The magnetic coagulation and sedimentation control processing module 40 is used to perform magnetic coagulation and sedimentation control processing on the target water body according to the real-time sedimentation control strategy.

[0053] Preferably, the target water body is subjected to magnetic coagulation and sedimentation control treatment according to the real-time sedimentation control strategy to achieve better sedimentation effect and water quality treatment goals. Specifically, according to the precipitant dosage (including flocculants, magnetic powder, etc.) determined in the real-time sedimentation control strategy, the dosage of the reagents is accurately controlled. For example, the strategy stipulates that under the current real-time water quality conditions, the dosage of the flocculant is 8 mg per liter of water and the dosage of the magnetic powder is 12 mg per liter of water. Then, the corresponding reagents are accurately added to the target water body according to this amount. By accurately controlling the dosage of the reagents, the sedimentation effect can be guaranteed, the waste of the reagents can be avoided, and the treatment cost can be reduced. According to the requirements of the real-time stirring intensity, the operating parameters of the stirring equipment are adjusted. If the real-time stirring intensity determined by the real-time sedimentation control strategy is that the stirrer speed is 250 revolutions per minute, the stirring equipment will be controlled. The speed of the equipment is adjusted to this set value. The appropriate stirring intensity helps to fully mix the agent with the water body, so that the magnetic flocs can be better formed and aggregated, and the sedimentation efficiency is improved; the time of the magnetic coagulation reaction in the target water body is strictly controlled. For example, the real-time sedimentation control strategy stipulates that the reaction time is 25 minutes. Then the timing device is started from the beginning of the agent addition. When the reaction time reaches 25 minutes, the subsequent sedimentation and separation operations are entered. The accurate reaction time can ensure that the agent fully reacts with the impurities in the water body to form larger and easy-to-precipitate magnetic floc particles; after the reaction is completed, according to the guidance of the real-time sedimentation control strategy, the sedimentation and separation operation is carried out, including controlling the operating parameters of the sedimentation equipment, such as the liquid level height of the sedimentation tank, the drainage speed, etc., to ensure that the magnetic flocs can be smoothly precipitated to the bottom of the sedimentation tank and the treated supernatant is discharged. During the entire magnetic coagulation sedimentation control treatment process, water quality parameters (such as turbidity, pH value, dissolved oxygen, etc.) and equipment operating status (such as current and voltage of stirring equipment, liquid level of sedimentation equipment, etc.) are continuously monitored in real time. If the actual situation is found to be inconsistent with the expectations of the real-time sedimentation control strategy, for example, the monitored turbidity does not achieve the expected reduction effect, the system will re-analyze and calculate based on the new monitoring data, and adjust the real-time sedimentation control strategy to ensure the stability and effectiveness of the treatment process, and ultimately achieve the ideal water quality treatment effect for the target water body. For example, Table 2 shows the economic advantages of this application: Table 2 Data related to the economic advantages of this application In the above, refer to Figure 1 The control system of low-cost magnetic coagulation precipitation dynamically controlled by AI according to an embodiment of the present invention is described in detail. Figure 2 The present invention describes a control method for low-cost magnetic coagulation precipitation dynamically regulated by AI according to an embodiment of the present invention.

[0054] AI dynamically controls low-cost magnetic coagulation sedimentation control methods, such as Figure 2As shown, the method includes: dynamically monitoring the water quality of the target water body through a predetermined probe to obtain real-time water quality information; reading the types of substances to be added, and analyzing to obtain the first precipitation dosage of the first type in the types of substances to be added based on the real-time water quality information, forming precipitation dosage information; forming a dynamic variable constraint based on the real-time water quality information and the precipitation dosage information, and dynamically adjusting the precipitation control benchmark under the dynamic variable constraint to obtain a real-time precipitation control strategy; performing magnetic coagulation precipitation control treatment on the target water body according to the real-time precipitation control strategy.

[0055] In a possible implementation manner, the control method for AI dynamic regulation of low-cost magnetic coagulation precipitation further includes: the real-time water quality information at least includes real-time turbidity, real-time pH value, and real-time temperature.

[0056] In a possible implementation manner, the control method for AI dynamic regulation of low-cost magnetic coagulation precipitation further includes: the types of substances to be added include flocculants and magnetic powder.

[0057] In a possible implementation manner, the control method for AI dynamic regulation of low-cost magnetic coagulation precipitation further includes: extracting predetermined water quality information in the precipitation control benchmark; comparing the real-time water quality information with the predetermined water quality information to obtain a real-time water quality deviation coefficient; extracting predetermined dosage information in the precipitation control benchmark; comparing the precipitation dosage information with the predetermined dosage information to obtain a dosage deviation coefficient; performing weighted calculation on the real-time water quality deviation coefficient and the dosage deviation coefficient to obtain a dynamic variable factor; adjusting the precipitation control benchmark based on the dynamic variable factor to obtain the real-time precipitation control strategy.

[0058] In a possible implementation manner, the control method for AI dynamic regulation of low-cost magnetic coagulation precipitation further includes: obtaining any water quality index; respectively traversing and matching in the real-time water quality information and the predetermined water quality information to obtain the real-time water quality parameter and the predetermined water quality parameter corresponding to the any water quality index; introducing a deviation evaluation function to evaluate and analyze the real-time water quality parameter and the predetermined water quality parameter to obtain the real-time water quality deviation coefficient; wherein, the expression of the deviation evaluation function is: ; refers to the real-time water quality deviation coefficient, refers to the value of the th real-time water quality parameter, value of the th predetermined water quality parameter, refers to the weight coefficient of the th parameter, and is an integer greater than or equal to 3.

[0059] In a possible implementation, the control method for AI dynamic regulation of low-cost magnetic coagulation precipitation further includes: extracting the mixing control information in the precipitation control benchmark, where the mixing control information includes a predetermined stirring intensity and a predetermined reaction duration; using the dynamic variable factor as a weight to perform weighted adjustment on the predetermined stirring intensity and the predetermined reaction duration to obtain a real-time stirring intensity and a real-time reaction duration respectively; forming the real-time precipitation control strategy based on the real-time stirring intensity and the real-time reaction duration.

[0060] In a possible implementation, the control method for AI dynamic regulation of low-cost magnetic coagulation precipitation further includes: dynamically collecting a real-time microscopic image of target magnetic flocs in the target water body; performing feature evaluation and analysis on the real-time microscopic image according to a microscopic state evaluation mechanism to obtain a real-time microscopic state; obtaining an arbitrary information value coefficient of the real-time microscopic state for any water quality index; combining the arbitrary information value coefficient to perform prediction analysis on the real-time water quality parameters to obtain predicted real-time water quality parameters; calibrating the real-time water quality parameters with the predicted real-time water quality parameters.

[0061] In a possible implementation, the control method for AI dynamic regulation of low-cost magnetic coagulation precipitation further includes: collecting real-time image feature information of the real-time microscopic image; performing normalized weighting on the real-time image color parameters and the real-time image structure parameters in the real-time image feature information to obtain a real-time image feature value; combining the microscopic state coefficient stored in the microscopic state evaluation mechanism to perform standardized correction on the real-time image feature value to obtain the real-time microscopic state.

[0062] In a possible implementation, the control method for AI dynamic regulation of low-cost magnetic coagulation precipitation further includes: extracting a first historical data set from a magnetic coagulation precipitation database; performing mutual information analysis on the first historical microscopic state of the first magnetic flocs and the first historical index parameters corresponding to the first historical arbitrary water quality index in the first historical data set to obtain a first historical mutual information coefficient; using the first historical mutual information coefficient as the arbitrary information value coefficient.

[0063] The control system for AI dynamic regulation of low-cost magnetic coagulation precipitation provided by the embodiments of the present invention can execute the control method for AI dynamic regulation of low-cost magnetic coagulation precipitation provided by any embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the method.

[0064] Although this application makes various references to certain modules in the system according to the embodiments of this application, however, any number of different modules can be used and run on the user terminal and / or the server. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of mutual distinction and are not used to limit the protection scope of the present invention.

[0065] The above specific implementation manners do not constitute a limitation on the protection scope of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of this application shall be included within the protection scope of this application.

Claims

1. AI dynamically controls the low-cost magnetic coagulation sedimentation control system, characterized in that: The system comprises: Dynamic water quality monitoring module, used to dynamically monitor the water quality of the target water body through a predetermined probe to obtain real-time water quality information; The precipitation dosage information composition module is used to read the category to be added, and with the real-time water quality information as a constraint, analyze and obtain the first precipitation dosage of the first category in the category to be added to form the precipitation dosage information; A real-time sedimentation control strategy acquisition module, used to form a dynamic variable constraint based on the real-time water quality information and the sedimentation dosage information, and dynamically adjust the sedimentation control benchmark under the dynamic variable constraint to obtain a real-time sedimentation control strategy; The magnetic coagulation and sedimentation control processing module is used to perform magnetic coagulation and sedimentation control processing on the target water body according to the real-time sedimentation control strategy.

2. The AI ​​dynamic control system for low-cost magnetic coagulation precipitation according to claim 1 is characterized in that: The dynamic water quality monitoring module includes: The real-time water quality information includes at least real-time turbidity, real-time pH value and real-time temperature.

3. The AI ​​dynamic control system for low-cost magnetic coagulation precipitation according to claim 1 is characterized in that: The precipitation dosage information component module includes: The products to be added include flocculants and magnetic powder.

4. The AI ​​dynamic control system for low-cost magnetic coagulation precipitation according to claim 1 is characterized in that: The real-time precipitation control strategy acquisition module includes: A predetermined water quality information extraction unit, used to extract the predetermined water quality information in the sedimentation control benchmark; A real-time water quality deviation coefficient obtaining unit, used for comparing the real-time water quality information with the predetermined water quality information to obtain a real-time water quality deviation coefficient; A predetermined dosage information extraction unit, used to extract the predetermined dosage information in the sedimentation control benchmark; A dosage deviation coefficient obtaining unit, used for comparing the precipitation dosage information with the predetermined dosage information to obtain a dosage deviation coefficient; A dynamic variable factor obtaining unit, used for performing weighted calculation on the real-time water quality deviation coefficient and the dosage deviation coefficient to obtain a dynamic variable factor; A real-time sedimentation control strategy obtaining unit is used to adjust the sedimentation control benchmark based on the dynamic variable factor to obtain the real-time sedimentation control strategy.

5. The AI ​​dynamic control system for low-cost magnetic coagulation precipitation according to claim 4 is characterized in that: The real-time precipitation control strategy acquisition module includes: An arbitrary water quality index acquisition unit, used to acquire an arbitrary water quality index; A water quality parameter obtaining unit, used to traverse and match the real-time water quality information and the predetermined water quality information to obtain the real-time water quality parameter and the predetermined water quality parameter corresponding to the arbitrary water quality index; A water quality parameter evaluation and analysis unit, used for introducing a deviation evaluation function to evaluate and analyze the real-time water quality parameter and the predetermined water quality parameter to obtain the real-time water quality deviation coefficient; Wherein, the expression of the deviation evaluation function is: ； refers to the real-time water quality deviation coefficient, It refers to The value of the real-time water quality parameter, It refers to The value of a predetermined water quality parameter, It refers to The weight coefficient of the parameter, is the sum of water quality parameters, and is an integer greater than or equal to 3.

6. The AI ​​dynamic control system for low-cost magnetic coagulation precipitation according to claim 4 is characterized in that: The real-time precipitation control strategy acquisition module includes: A mixing control information extraction unit, used to extract mixing control information in the sedimentation control benchmark, wherein the mixing control information includes a predetermined stirring intensity and a predetermined reaction time; A weighted adjustment unit, used to perform weighted adjustment on the predetermined stirring intensity and the predetermined reaction time using the dynamic variable factor as a weight, to obtain a real-time stirring intensity and a real-time reaction time respectively; A sedimentation control strategy forming unit is used to form the real-time sedimentation control strategy based on the real-time stirring intensity and the real-time reaction time.

7. The AI ​​dynamic control system for low-cost magnetic coagulation precipitation according to claim 5 is characterized in that: The real-time precipitation control strategy acquisition module also includes: A real-time microscopic image acquisition unit, used for dynamically acquiring a real-time microscopic image of the target magnetic floccules in the target water body; A feature evaluation and analysis unit, used to perform feature evaluation and analysis on the real-time microscopic image according to a microscopic state evaluation mechanism to obtain a real-time microscopic state; An information value coefficient acquisition unit, used to acquire any information value coefficient of the real-time microscopic state to any water quality index; A prediction and analysis unit, used to perform prediction and analysis on the real-time water quality parameter in combination with the arbitrary information value coefficient to obtain a predicted real-time water quality parameter; A water quality parameter calibration unit is used to calibrate the real-time water quality parameter with the predicted real-time water quality parameter.

8. The AI ​​dynamic control system for low-cost magnetic coagulation precipitation according to claim 7 is characterized in that: The real-time precipitation control strategy acquisition module also includes: A real-time image feature information acquisition unit, used to acquire real-time image feature information of the real-time microscopic image; A real-time image characteristic value obtaining unit, used for normalizing and weighting the real-time image color parameters and the real-time image structure parameters in the real-time image characteristic information to obtain the real-time image characteristic value; The standardization correction unit is used to perform standardization correction on the real-time image characteristic value in combination with the microscopic state coefficient stored in the microscopic state evaluation mechanism to obtain the real-time microscopic state.

9. The AI ​​dynamic control system for low-cost magnetic coagulation precipitation according to claim 7 is characterized in that: The real-time precipitation control strategy acquisition module also includes: A first historical data set extraction unit, used to extract a first historical data set from a magnetic coagulation sedimentation database; a mutual information analysis unit, configured to perform mutual information analysis on a first historical microscopic state of a first magnetic floc in the first historical data set and a first historical indicator parameter corresponding to a first historical arbitrary water quality indicator to obtain a first historical mutual information coefficient; The arbitrary information value coefficient obtaining unit is used to use the first historical mutual information coefficient as the arbitrary information value coefficient.

10. AI dynamically controls low-cost magnetic coagulation precipitation control method, characterized in that: The method is applied to the control system of low-cost magnetic coagulation precipitation dynamically controlled by AI according to any one of claims 1 to 9, and the method comprises: Dynamically monitor the water quality of the target water body through a predetermined probe to obtain real-time water quality information; Read the category to be added, and with the real-time water quality information as a constraint, analyze and obtain the first precipitation dosage of the first category in the category to be added, and form precipitation dosage information; Based on the real-time water quality information and the precipitation dosage information, a dynamic variable constraint is formed, and the precipitation control benchmark is dynamically adjusted under the dynamic variable constraint to obtain a real-time precipitation control strategy; The target water body is subjected to magnetic coagulation sedimentation control treatment according to the real-time sedimentation control strategy.

Images