Intelligent integrated targeted plant powder adsorbent water purification system and method

Abstract

The application discloses an intelligent integrated targeted plant powder adsorbent water purification system and method, and constructs a multi-unit cooperative intelligent treatment system to provide a water pollution treatment system and method which are exclusive, customized, green, repaired, efficient and economical.The plant powder adsorbent water purification system of the application provides a complete and efficient scheme design process for water pollution treatment.For different water areas with different pollution conditions, the system can adjust the most suitable plant powder adsorbent formula, and optimizes the design to improve the adsorption efficiency and realize the maximum economic benefit of the matching device.In addition, the water pollution treatment scheme provided by the system is dynamic, and the most suitable treatment scheme for the next stage can be recommended according to the change of the treatment progress.The application is designed and optimized in the aspects of scheme design, plant powder adsorbent ratio optimization, treatment equipment optimization, whole purification process supervision and dynamic adjustment, and is helpful to reduce the cost and difficulty of water pollution treatment and improve the treatment efficiency.

Description

Technical Field

[0001] This invention relates to an intelligent integrated targeted plant powder adsorbent water purification system and method, belonging to the field of water pollution control technology. Background Technology

[0002] my country's numerous river basins exhibit diverse sources of pollution. In the modern industrial system, chemical, dyeing, and papermaking industries all contribute to water pollution. The chemical and dyeing industries, in particular, have complex production processes involving numerous chemical reactions, often producing wastewater containing heavy metals such as mercury, cadmium, and lead, as well as large amounts of recalcitrant organic matter. Domestic sewage contains significant amounts of organic matter, as well as nutrients like nitrogen and phosphorus, making it a major cause of eutrophication. Agricultural production is also a significant source of water pollution, characterized by its wide reach, dispersed nature, and difficulty in control. The accumulation of nutrients in water bodies leads to eutrophication, triggering algal blooms and disrupting the ecological balance of aquatic environments.

[0003] Currently, traditional water pollution control methods mainly include physical, chemical, and biological treatment. Physical methods, primarily using filtration, sedimentation, and centrifugation, are simple and easy to implement, but most can only filter larger impurities and cannot completely remove pollutants. Chemical treatment requires chemical reactions and mass transfer to remove pollutants, making it extremely expensive, complex, and prone to secondary pollution due to the various chemical reactions involved, hindering field operations. Biological treatment mainly utilizes microorganisms or living plants. Different microorganisms have varying abilities to treat water pollutants. When dealing with different water bodies, it is necessary to identify the key areas for treatment and then use different microorganisms accordingly. However, the symbiotic relationships between different microorganisms remain a complex issue, and the survival requirements of microorganisms are extremely demanding. Therefore, using microorganisms to treat water pollution is an extremely large and complex project. Living biological methods typically require the combination of multiple plants to achieve a certain water purification effect. However, this not only necessitates considering the symbiotic relationships between different plants but also requires strict control of their growth environment. Furthermore, the planting, maintenance, and management of living plants involve a significant amount of engineering work, especially in waters with complex pollution conditions, making implementation difficult and costly. In general, traditional water pollution treatment methods have certain limitations due to their high technical requirements, high costs, poor targeting, and tendency to cause secondary pollution.

[0004] The construction of traditional integrated wastewater treatment facilities requires huge capital investment. From the construction of wastewater treatment plants to the laying of wastewater pipelines, every stage, from planning and design to actual construction, and then to subsequent equipment maintenance and upgrades, involves high costs. Furthermore, traditional integrated equipment cannot target complex and diverse pollutants in water bodies. Traditional processes rely on manual operation, and their treatment effectiveness is easily affected by external factors, making them unsuitable for small-scale, decentralized scenarios. Constructed wetlands generally have the drawback of requiring large land areas. Improving the purification efficiency of constructed wetlands under the constraint of limited land resources is a major challenge. Especially under stressful environments such as low temperatures in the north and salinization in inland and coastal areas, the operating efficiency of constructed wetlands declines, posing a significant challenge to their widespread application.

[0005] Traditional treatment methods can no longer meet current needs; therefore, the adoption of more advanced intelligent technologies for water pollution treatment has attracted widespread attention. Intelligent technologies have already played a crucial role in water pollution control, demonstrating outstanding performance in improving water treatment efficiency and water quality. This invention proposes an intelligent control system and method for in-situ water purification using plant powder adsorbents, comprising seven units: a sampling unit, a water sample analysis unit, an intelligent control center, an adsorbent preparation unit, an environmental analysis unit, a device manufacturing unit, and an operational unit. Compared to traditional methods, this system achieves intelligent, automated, highly efficient, and green pollution removal. By inputting sampling data and water sample analysis data, the intelligent control center can derive a suitable adsorbent formula based on an algorithm, simultaneously determining the watershed type of the sampling point and selecting appropriate supporting products to create a customized pollution removal solution. Furthermore, adsorbent recovery can also be achieved. It can achieve full coverage of water samples without blind spots in the sampling unit, realize intelligent management of the system in the intelligent control center, ensure real-time monitoring of water quality and environment in the water sample analysis unit and environmental analysis unit, realize specific, efficient, green and economical water pollution treatment in the adsorbent preparation unit and device manufacturing unit, and complete dynamic adjustment in the input unit to realize the closed loop of the system.

[0006] Chinese Patent Publication No. CN 119551765 A discloses a constant temperature circulating water treatment system, which includes the following steps: constructing a sensor module to collect, preliminarily process, and transmit data from wastewater treatment equipment; designing a feature fusion module to receive data from the sensor module, select, extract, and fuse data features; developing an adaptive multimodal feature fusion fault prediction model to automatically adjust the feature fusion weights based on actual and historical operating data to calculate the probability of equipment failure; and building a real-time monitoring and early warning module to analyze real-time and fault prediction results to monitor whether the equipment has abnormal operating status and issue an early warning when abnormality occurs.

[0007] However, this AI-based wastewater treatment equipment operation monitoring system is expensive to build and maintain, resulting in high operating costs. It is more suitable for optimizing large-scale wastewater treatment plants or complex processes. It cannot adapt to various watersheds in real-world applications, nor can it address specific pollution situations within those watersheds. Its widespread adoption and application are significantly limited in resource-constrained environments such as small wastewater treatment plants, parks, lakes, and urban rivers.

[0008] Existing water pollution treatment technologies have the following main defects and shortcomings in practical applications:

[0009] (1) The core problem of physical treatment technology lies in its low efficiency in treating dissolved pollutants and its high operating cost. Precipitation is limited by Stokes' law and can only effectively remove suspended particles with a density greater than water. It is not effective in separating dissolved organic pollutants and low-concentration heavy metal ions. Although filtration and membrane separation technologies (such as microfiltration and reverse osmosis) can retain micron-sized particles, membrane modules are easily clogged by colloidal substances or organic fouling, leading to flux decline and frequent chemical cleaning, which significantly increases energy consumption and maintenance costs. In addition, physical methods generally generate the risk of secondary pollution. For example, if saturated activated carbon is not regenerated in time, it will release adsorbates to form new pollution sources, and the treatment of reverse osmosis concentrate still requires additional process support.

[0010] (2) Chemical treatment technologies are prone to causing secondary environmental risks during the degradation of pollutants. Strong oxidation processes (such as ozone and Fenton's reagent) may generate more toxic intermediate products through free radical chain reactions, such as carcinogenic halogenated hydrocarbons produced by chlorine disinfection; metal salt flocculants (aluminum / iron salts) used in chemical precipitation methods can cause fluctuations in water pH and residual metal ions, which may disrupt the balance of aquatic ecosystems in the long term. More importantly, chemical methods require precise control of reaction kinetic parameters (pH, redox potential, and reagent dosage ratio). For industrial wastewater containing multiple pollutants, the complexity of reagent compatibility increases exponentially, resulting in a non-linear relationship between treatment costs and pollutant concentration.

[0011] (3) The limitations of biological treatment technology stem from the environmental sensitivity of microbial metabolic activities. Nitrifying / denitrifying bacteria have low tolerance thresholds to temperature fluctuations (±5℃), dissolved oxygen concentrations (<0.5 mg / L), and toxic substances (such as heavy metal ion concentrations >1 ppm). They are prone to inactivation in the presence of high salt or recalcitrant organic matter (polycyclic aromatic hydrocarbons, antibiotics, etc.). In addition, the start-up phase of a biological treatment system requires a acclimatization period of several weeks, and even a small deviation in the carbon-nitrogen-phosphorus ratio (usually required to be 100:5:1) can lead to a decrease in nitrogen and phosphorus removal efficiency. The residual sludge generated during the treatment process contains extracellular polymers and pathogenic microorganisms. The cost of its dewatering, drying, and harmless disposal accounts for 30%-40% of the total operating cost of the system. Improper landfilling or incineration may cause secondary pollution.

[0012] (4) There are ecological safety risks in the engineering application of living bio-adsorption systems. If the hyperaccumulating plants (such as water hyacinth and reed) selected for ecological floating islands are not genetically modified, they may form biological invasions through pollen dispersal, leading to a 30%-60% decrease in the diversity index of local aquatic plant communities. In waters with heavy metal pollution (such as Cd>0.1 mg / L + Pb>0.5 mg / L), the tolerance period of aquatic plants is usually shortened to 2-3 months. After exceeding the critical value, the organic acids secreted by the roots will promote the desorption of heavy metals. The mineralization and decomposition of dead plants will also release nutrients such as N and P, causing the chlorophyll a concentration in the water to increase by 50%-100%, inducing eutrophication.

[0013] (6) Constructed wetland technology is limited by space efficiency and long-term operation and maintenance costs. The hydraulic loading rate of traditional surface flow constructed wetlands is only 0.1-0.5 m³ / (m²·d), and it requires 10-15 hectares of land to treat tens of thousands of tons of water. The application cost in urban built-up areas increases by 3-5 times. The quantitative relationship between matrix layer design parameters (particle size distribution, porosity, permeability coefficient) and pollutant removal rate has not yet been established. In actual projects, the TN / TP removal rate fluctuates by ±25%. More seriously, if the biomass (dry weight of about 5-8 kg / m²) produced by wetland plants each year is not harvested in time, the decay process will increase the leaching rate of bound heavy metals in the sediment by 2-3 orders of magnitude, resulting in pollutant rebound after 3-5 years of system operation. Summary of the Invention

[0014] To overcome the shortcomings of existing technologies, this invention proposes an intelligent integrated targeted plant powder adsorbent water purification system and method. This system utilizes plant powder adsorbents to achieve efficient adsorption of water pollution while simultaneously realizing ecological and environmental protection principles. The plant powder adsorbent water treatment system of this invention provides a complete and efficient solution for water pollution control. For water bodies with different pollution levels, the system can formulate the most suitable plant powder adsorbent formula and, depending on the specific conditions of the water body, can be equipped with supporting devices that improve adsorption efficiency and maximize economic benefits. Furthermore, the water pollution control solution provided by this system is dynamic, adjusting the optimal treatment plan for the next stage based on changes in the treatment progress. This novel plant powder adsorbent water purification system solves the problems of high difficulty and high cost associated with traditional water pollution control, maximizing economic benefits while achieving efficient treatment.

[0015] This invention is achieved through the following technical solution: an intelligent integrated targeted plant powder adsorbent water purification system and method, wherein the water purification system includes a sampling unit, a water sample analysis unit, an intelligent control center, an adsorbent preparation unit, an environmental analysis unit, a device manufacturing unit, and an adsorbent recovery unit.

[0016] The sampling unit inputs the natural geographic information, ecological environment information, and human activity information obtained from the field survey into the intelligent control center. The intelligent control center connects the water sample analysis unit, the adsorbent preparation unit, and the environmental analysis unit. The water sample analysis unit is equipped with a full-parameter water quality analyzer, which transmits the measured heavy metal concentration, nutrient concentration, pH value, and soluble pollutant data to the intelligent control center. The intelligent control center analyzes the soluble pollutant data, calculates the required adsorbent mass, and transmits the data to the adsorbent preparation unit in real time. Simultaneously, the intelligent control center performs environmental analysis using GIS and satellite imagery to determine the watershed type and transmits the watershed type data to the environmental analysis unit. The watershed type data is also transmitted to the device manufacturing unit. The device manufacturing unit determines whether to use its self-circulating ecological floating vessel or a detachable powder-carrying device for water purification based on the watershed type. After water purification, the adsorbent enters the adsorbent recovery unit.

[0017] The specific steps for a water purification system are as follows:

[0018] Step 1: The sampling unit is a full-coverage sampling unit, including the natural geographic information of the sampling points: watershed area index N1, narrowest river index N2, average slope index N3, river network density index N4; ecological environment information: vegetation coverage index E1, biodiversity index E2, water pollution index E3, wind speed E4; and human activity information: population density index H1, land development intensity index H2, water resource utilization intensity index H3, and vessel activity intensity index H4. The data is then entered into the intelligent control center. Drones or remote sensing technology can be added for large-scale, high-precision auxiliary sampling. Simultaneously, more types of sensors, such as pH sensors and dissolved oxygen sensors, can be introduced to obtain more comprehensive water quality data.

[0019] Step 2: The water sample analysis unit's full-parameter water quality analyzer integrates digestion and measurement, intelligently monitoring COD, ammonia nitrogen, total phosphorus, total nitrogen content, and pH value. The heavy metal detector in the water quality analyzer has 360° rotating colorimetric tube detection and cuvette detection, accurately monitoring the content of heavy metals such as copper, cadmium, zinc, and lead in the water. The water sample analysis unit inputs the measured data into the intelligent control center in real time.

[0020] Step 3: The intelligent control center uses the dynamic multidimensional collaborative watershed classification algorithm DMS-WCS to derive three watershed types from the data input in Step 1 and Step 2: N-type natural geographical indicators, E-type ecological environment indicators, and H-type human activity indicators.

[0021] Step 4: The intelligent control center determines the matching device for the plant powder adsorbent to be a self-circulating ecological floating boat or a detachable powder loading tank based on the above three watershed types.

[0022] Step 5: The device manufacturing unit determines the device size based on the watershed type and application scenario, and manufactures the device.

[0023] Step 6: The intelligent control center uses the Dynamic Robust Synergistic Adsorption Optimization Algorithm (DRSAO) to calculate the effective adsorption capacity of the data input from Step 1 and Step 2, and then performs dynamic calculations and batch iterative processing of the plant powder adsorbent to analyze the synergistic adsorption optimization of multiple metal ions and organic particles, and finally obtains the plant powder adsorbent formulation.

[0024] Step 7: The adsorbent preparation unit prepares plant powder adsorbents according to the obtained optimal formula;

[0025] Step 8: The adsorbent preparation unit puts the prepared adsorbent into the device prepared in Step 5 for water treatment;

[0026] Step 9: The intelligent control center uses the Dynamic Economic Collaborative Recycling Optimization Algorithm (DESRO) to calculate the time required for the plant powder adsorbent to be recycled after reaching its maximum adsorption efficiency, determines the maximum adsorption efficiency and recycling time of the plant powder adsorbent, and transmits the data to the adsorbent recycling unit.

[0027] Step 10: The adsorbent recovery unit replaces and recovers the adsorbent in the device at regular intervals.

[0028] The intelligent control center has been enhanced with an intelligent early warning function. When water quality data shows abnormalities, the early warning mechanism can be automatically triggered to promptly notify relevant personnel for handling.

[0029] Furthermore, the Step 3 intelligent control center performs the following calculations:

[0030] Based on the sampling data from the sampling units and water sample analysis units, the intelligent control center classifies water bodies into three index types, and first calculates the comprehensive water pollution index P:

[0031]

[0032] Where P represents the comprehensive water pollution index. The weighting coefficient ranges from [0, 1]. Ck represents the measured concentration of substances after water sample analysis, and Sk represents the standard concentration limit of heavy metals in Class V water (see Table 1).

[0033] Table 1 (Unit: mg / L)

[0034]

[0035] Based on the P-value, the indicators are further classified: if P = 0-4.5, the water body is classified as a natural geographical indicator (N); if P = 4.5-7.0, the water body is classified as an ecological environmental indicator (E); if P > 7.0, the water body is classified as a human activity indicator (H). After determining the indicator classification, each indicator type is calculated and corrected.

[0036] ① Type N Natural Geographic Indicators: Composed of natural geographic indicators including drainage area N1, narrowest river N2, average slope N3, and river network density N4. The calculation formula is as follows:

[0037] Units are standardized for each indicator, i.e., each indicator is converted to the range [0,1].

[0038]

[0039] maxNi and minNi are the maximum and minimum values ​​of Ni, respectively, and are calculated using the following formulas:

[0040]

[0041] Where ai is the weighting coefficient. Furthermore, considering the influence of the natural environment, a natural correction coefficient αN = 0.8-1.2 is introduced, and the corrected natural geographical indicators are as follows: ;

[0042] like [0, 3.0), then the water body is of high ecological integrity type, suitable for flexible and convenient detachable powder loading tank (this type of device has the advantages of simple operation and high flexibility); if [3.0, 4.8) is a moderately ecologically stable type, compatible with a flexible and convenient detachable powder loading tank and a flexible and convenient self-circulating ecological floating boat (capable of freely navigating narrow waters, easy to use, and meeting the needs of rapid travel and operation); if [4.8, +∞) is a lightweight, flexible, and convenient self-circulating ecological floating vessel;

[0043] ② Type E Ecological Environment Indicators: Composed of vegetation coverage index E1, biodiversity index E2, water pollution index E3, and wind speed index E4. These indicators are standardized in units, meaning each indicator is converted to a value within the range [0,1].

[0044]

[0045] maxEi and minEi are the maximum and minimum values ​​of Ei, respectively, and are calculated using the following formulas:

[0046]

[0047] Where ai is the weighting coefficient. Introducing the ecological correction coefficient αE (0.7-1.3)

[0048] The revised ecological and environmental indicators are as follows:

[0049] [0, 2.8), then the water body is ecologically coordinated and is suitable for a versatile and practical detachable powder carrying tank (good stability, better able to resist wind and waves, and also has a certain capacity to carry people). [2.8, 5.2), then the water body is of a balanced type for human activities, and is suitable for a versatile and practical detachable powder loading tank and a versatile and practical self-circulating ecological floating boat (which takes into account a certain transportation capacity and can meet the needs of more time for navigation). [5.2, +∞), then the water body is a human activity-adaptable, all-purpose, practical self-circulating ecological floating vessel;

[0050] ③ Human activity index H type:

[0051] The human activity index is composed of population density index H1, land development intensity index H2, water resource utilization intensity index H3, and ship activity intensity index H4. The units of these indicators are standardized, meaning each indicator is converted to a value within the range [0,1].

[0052]

[0053] maxHi and minHi are the maximum and minimum values ​​of Hi, respectively, and are calculated using the following formulas:

[0054]

[0055] Where ai is the weighting coefficient. Introducing a human activity correction factor αH (0.9-1.1),

[0056] The revised human activity index is

[0057] [0, 3.2), then the water body is of human development type, suitable for heavy-duty long-distance detachable powder carrying tank (with high safety performance to cope with slightly larger watersheds, and sufficient carrying capacity). [3.2, 4.4), then the water body is of moderate disturbance type, suitable for heavy-duty long-range detachable powder loading tank and heavy-duty long-range self-circulating ecological floating vessel (higher safety performance, stronger endurance, able to carry out large-scale watershed operations for a longer period of time, and stronger loading capacity, and can achieve economies of scale, etc.). If [4.4, +∞), then the water body is of low-level restoration type and is suitable for an all-around practical self-circulating ecological floating vessel.

[0058] Furthermore, in Step 6, the intelligent control center (3) uses the Dynamic Robust Collaborative Adsorption Optimization Algorithm (DRSAO) to calculate the effective adsorption capacity based on the input parameters of the data from Step 1 and Step 2, as follows:

[0059] ① Input parameters and calculation of effective adsorption capacity

[0060] The water sample analysis unit measures the heavy metal content and organic particle concentration C0 in the water. Based on the national water quality discharge standards, the estimated heavy metal content and organic particle concentration Ct in the treated water are determined. The volume V of the treated water is measured, and the maximum adsorption capacity Qmax of different plant powder adsorbents is input. Combined with environmental correction factors α, β, and γ, the actual effective adsorption capacity is calculated.

[0061] Wherein, the environmental correction factor is:

[0062] α is the pH correction factor. The optimal adsorption of plant powder adsorbents usually occurs within a specific pH range (e.g., pH=5~6). When the pH deviates from the range, it needs to be multiplied by the correction factor α (0~1). If the pH deviates from the specific range by more than or equal to 1.5, take α=0~0.5. If the pH deviates from the specific range by less than 1.5, take α=0.5~1.

[0063] β is the temperature correction factor. The adsorption capacity of plant powder adsorbent is closely related to the temperature. The adsorption efficiency is best at 25℃. High or low temperatures inhibit the adsorption. When adsorption is carried out under different temperature conditions, the correction factor β (0.8~1) needs to be multiplied. If the temperature deviates from 25℃ by more than or equal to 20℃, β=0.8~0.9 is taken. If the temperature deviates from 25℃ by less than 20℃, β=0.9~1 is taken.

[0064] γ is the competitive remediation inhibition coefficient. The presence of other heavy metal ions or organic particles in the water body affects the adsorption capacity of plant powder adsorbents for specific heavy metal ions or organic particles, so a competitive inhibition coefficient γ (0~1) needs to be introduced. If there are more than 17 other heavy metal ions or organic particles in the water body, γ is taken as 0~0.5. If there are less than 17 other heavy metal ions or organic particles in the water body, γ is taken as 0.5~1.

[0065] The environmental correction coefficients α, β, and γ are extended to a robust optimization model:

[0066]

[0067]

[0068]

[0069] Dynamic calculation of plant powder adsorbent:

[0070] Introducing a safety factor k, calculate the mass of plant powder:

[0071]

[0072] The safety factor k is used to ensure that the adsorbent can cope with various uncertainties and potential efficiency losses in actual adsorption. Based on engineering experience and experimental data, it aims to ensure the safety and reliability of the system. k is 1.2 to 2.0. According to the degree, it is divided into mild k=1.2 to 1.4, moderate k=1.5 to 1.7, and severe k=1.8 to 2.0.

[0073] Introduce constraints:

[0074]

[0075] The emission standard constraint refers to the requirement that the expected heavy metal content and organic particulate concentration (Ct) of the treated water be less than or equal to the emission standard value. ;

[0076] The maximum batch limit refers to the maximum number of experimental batches set to prevent excessive experimental energy consumption and save costs; the maximum number of experimental batches must be less than or equal to [a certain value]. ;

[0077] The budget constraint refers to the quality of the novel plant powder adsorbent produced. Must be less than or equal to the constraint value This helps avoid waste and loss of raw materials and controls costs.

[0078] ② Batch iterative processing

[0079] Various influencing factors affect the adsorption capacity of plant powder adsorbents, resulting in the water failing to reach the target concentration Ct after a single treatment. Therefore, the remaining concentration is calculated.

[0080]

[0081] Then, using Cremaining as the new C0, iterate until the target is met;

[0082] Supplementary recurrence relation:

[0083]

[0084] The transfer equation is an iterative equation for calculating the remaining concentration with added noisy terms. The noisy terms are intended to make the model more realistic and to take into account random disturbances when analyzing the robustness of the system, predicting the error range, or designing control strategies, so as to avoid the model becoming completely deterministic and unable to reflect the random behavior of the real system.

[0085] ③ Optimization of synergistic adsorption of multiple metal ions and organic particles

[0086] For water bodies containing multiple heavy metal ions and organic particles, calculate the mass Mi of plant powder adsorbent required to adsorb each type of metal ion and organic particle, and take the maximum value M=max(M1,M2,M3,...,Mn) to ensure that the concentrations of all heavy metal ions and organic particles meet the standards.

[0087] Introducing a coupling adsorption matrix A∈Rm×n, we describe the adsorption efficiency of m plant powder adsorbents for n heavy metal ions and organic particles:

[0088]

[0089] constraint:

[0090]

[0091] The first item (adsorption effect fitting item): The concentration of pollutant j at the initial time. The concentration of the j-th pollutant after treatment time t. This refers to the adsorption efficiency of plant powder adsorbent for the j-th pollutant. For each metal target concentration vector, this term represents minimizing the difference between the actual removal concentration and the theoretical adsorption efficiency. The sum of squares form aims to make the adsorption effect as close as possible to the expectation.

[0092] The second term (regularization term): ρ refers to the sum of squares of all elements in matrix A, and ρ refers to the regularization coefficient, which is used to balance fitting accuracy and model complexity. This term prevents the model from overfitting the experimental data, ensures that the element values ​​of the adsorption matrix A are not too large, and enhances the generalization ability.

[0093] Step 9, the recovery of plant powder adsorbent, specifically involves:

[0094] Plant powder adsorbents can be recycled after reaching maximum adsorption efficiency. By constructing a "mixed integer nonlinear programming (MINLP)" model, the dosage and recycling time of plant powder adsorbents can be optimized simultaneously.

[0095]

[0096]

[0097]

[0098] The first item is the cost of adsorbent usage: This indicates the amount of the i-th type of adsorbent used. This refers to the weighting coefficient for adsorbent dosage, ranging from 0.5 to 0.7. It reflects the importance of adsorbent cost in the overall objective. This item represents minimizing the total cost of adsorbent usage. If the cost is relatively high, A larger value, 0.6 to 0.7, should be chosen to reduce the amount of adsorbent used. If the cost is relatively low, Then take the smaller value, 0.5~0.6;

[0099] The second item is the time cost of recycling: This refers to the adsorption efficiency coefficient of the i-th adsorbent. This refers to the initial adsorption efficiency of the i-th adsorbent. The recovery time of the i-th adsorbent is the decision variable. The weighting coefficient for recovery time reflects the importance of recovery time cost in the overall objective. This factor represents minimizing the recovery time cost of all adsorbents. If the time cost is relatively high, A larger value, 0.2 to 0.3, should be chosen to shorten the recovery time. If the time cost is relatively low, Then take the smaller value, 0.1 to 0.2;

[0100] Constraints: The upper limit of the total available amount of adsorbent, the total amount constraint equation (②) is that the total amount of all adsorbents must not exceed the available resources. The upper limit of single recovery time is defined by the recovery time constraint equation (③), which states that the recovery time of each adsorbent cannot exceed the maximum allowable value.

[0101] For water bodies containing various heavy metal ions and organic particles, the time required ti for the plant powder adsorbent to adsorb various metal ions and organic particles was calculated, and the average time required to adsorb water bodies containing various heavy metal ions and organic particles was obtained:

[0102]

[0103] The Step 10 adsorbent recovery unit will humify the plant powder adsorbents that are saturated in the self-circulating ecological floating boats or detachable powder-carrying devices used in the watershed, thereby achieving the recovery of heavy metals and the acquisition of organic fertilizer.

[0104] The formula will be imported in real time into the plant powder adsorbent preparation system, which is the system disclosed in patent application number 2025204041678. This system allows for the customization of highly efficient, green, and economical plant powder adsorbents, which can then be used in the powder loading tank of a self-circulating ecological floating vessel or a detachable powder loading device manufactured by the device manufacturing unit. The powder loading tank in the detachable powder loading device is the loading tank in the loading device claimed in patent application number 2025102823124. The self-circulating ecological floating vessel is the ecological floating vessel device disclosed in application number 2024111928915.

[0105] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0106] (1) The present invention significantly improves pollutant removal efficiency and reduces costs. Based on intelligent proportioning technology for dynamic water quality monitoring, the adsorption capacity of plant powder adsorbent for dissolved heavy metals (such as Cu²⁺, Zn²⁺, Cd²⁺) is greatly increased, and the over-humification recovery system allows for the secondary utilization of heavy metals and organic matter in the adsorbent. At the same time, the adsorbent selectivity coefficient (Ks) is increased to 8.6-12.4, effectively inhibiting competitive adsorption of Ca²⁺ / Mg²⁺ ions, and the removal rate of target pollutants remains stable at over 95%.

[0107] (2) The modular device design of this invention breaks through the spatial and operational limitations of traditional treatment methods. The self-circulating ecological floating vessel has a water throughput of 11t / d, and the device volume is compact, making it suitable for various watershed conditions, thus reducing operational energy consumption (solar power supply accounts for ≥90%). The detachable powder-carrying device can be attached to existing devices or mounted on the riverbank. Utilizing the water's own hydrodynamics and combined with GIS environmental adaptation technology, the system reduces the fluctuation range of pollutant removal rate in complex waters (N / E / H type) from ±25% to ±8%.

[0108] (3) The whole-process resource utilization technology of this invention eliminates the risk of secondary pollution. The adsorbent humification treatment improves the recovery rate of heavy metals, and the total nutrients (N+P2O2+K2O) of the organic fertilizer generated simultaneously are ≥12%, which meets the agricultural standard (NY / T 525-2021). The overall operating cost of the system is reduced by 45%-50% compared with the traditional method, and no chemical reagents are added, avoiding the generation of toxic byproducts such as halogenated hydrocarbons (detection limit <0.001μg / L), realizing a dual closed loop of pollutant reduction and resource regeneration.

[0109] (4) The present invention uses the Dynamic Multidimensional Collaborative Watershed Classification Algorithm (DMS-WCS) to determine the type of watershed to be applied, uses the Dynamic Robust Collaborative Adsorption Optimization Algorithm (DRSAO) to calculate the formulation of the new plant powder adsorbent, and uses the Dynamic Economic Collaborative Recovery Optimization Algorithm (DESRO) to determine the maximum adsorption efficiency and recovery time of the plant powder adsorbent, thus realizing integrated, intelligent and efficient water pollution treatment. Attached Figure Description

[0110] Figure 1 This is a schematic diagram of the connection structure of each unit of the present invention;

[0111] Figure 2 This is a flowchart of the water area classification calculation process of this invention;

[0112] Figure 3 This is a flowchart for calculating the formulation of plant powder adsorbents. Detailed Implementation

[0113] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments, but the scope of protection of the present invention is not limited to the content described.

[0114] Example 1: As Figure 1 As shown, this invention relates to an intelligent integrated targeted plant powder adsorbent water purification system and method, characterized in that the water purification system includes a sampling unit 1, a water sample analysis unit 2, an intelligent control center 3, an adsorbent preparation unit 4, an environmental analysis unit 5, a device manufacturing unit 6, and an adsorbent recovery unit 7.

[0115] The sampling unit 1 inputs the natural geographical information, ecological environment information, and human activity information obtained from the field survey into the intelligent control center 3. The intelligent control center 3 is connected to the water sample analysis unit 2, the adsorbent preparation unit 4, and the environmental analysis unit 5. The water sample analysis unit 2 is equipped with a full-parameter water quality analyzer, which transmits the measured heavy metal concentration, nutrient concentration, pH value, and soluble pollutant data to the intelligent control center 3. The intelligent control center 3 analyzes the soluble pollutant data and calculates the required adsorbent mass, transmitting the data to the adsorbent preparation unit 4 in real time. At the same time, the intelligent control center 3 performs environmental analysis using GIS and satellite images to determine the watershed type and transmits the watershed type data to the environmental analysis unit 5. The watershed type data is also transmitted to the device manufacturing unit 6. The device manufacturing unit 6 determines whether to use its self-circulating ecological floating boat or detachable powder-carrying device for water purification based on the watershed type. After water purification, the adsorbent enters the adsorbent recovery unit 7.

[0116] The specific steps of the water purification system are as follows:

[0117] Step 1: Point sampling unit 1 is a full-range coverage sampling unit, including the natural geographic information of the points: watershed area index N1, river narrowest point index N2, average slope index N3, river network density index N4; ecological environment information: vegetation coverage index E1, biodiversity index E2, water pollution index E3, wind speed E4; and human activity information: population density index H1, land development intensity index H2, water resource utilization intensity index H3, and vessel activity intensity index H4. The data is then entered into the intelligent control center 3.

[0118] Step 2: The full-parameter water quality analyzer in water sample analysis unit 2 integrates digestion and measurement, and intelligently monitors COD, ammonia nitrogen, total phosphorus, total nitrogen content, and pH value. The heavy metal detector in the water quality analyzer has 360° rotating colorimetric tube detection and cuvette detection, accurately monitoring the content of heavy metals such as copper, cadmium, zinc, and lead in the water. Water sample analysis unit 2 inputs the measured data into the intelligent control center 3 in real time.

[0119] Step 3: The intelligent control center (3) uses the data input from Step 1 and Step 2 to derive three watershed types based on the Dynamic Multidimensional Collaborative Watershed Classification Algorithm (DMS-WCS): Natural Geographic Indicators (N-type), Ecological Environment Indicators (E-type), and Human Activity Indicators (H-type), such as... Figure 2 As shown, the calculation is performed as follows:

[0120] Based on the sampling data from the sampling units and water sample analysis units, the intelligent control center classifies water bodies into three index types, and first calculates the comprehensive water pollution index P:

[0121]

[0122] Where P represents the comprehensive water pollution index. The weighting coefficient ranges from [0, 1]. Ck represents the measured concentration of substances after water sample analysis, and Sk represents the standard concentration limit of heavy metals in Class V water (see Table 1).

[0123] Table 1 (Unit: mg / L)

[0124]

[0125] Based on the P-value, the indicators are further classified: if P = 0-4.5, the water body is classified as a natural geographical indicator (N); if P = 4.5-7.0, the water body is classified as an ecological environmental indicator (E); if P > 7.0, the water body is classified as a human activity indicator (H). After determining the indicator classification, each indicator type is calculated and corrected.

[0126] ① Type N Natural Geographic Indicators: Composed of natural geographic indicators including drainage area N1, narrowest river N2, average slope N3, and river network density N4. The calculation formula is as follows:

[0127] Units are standardized for each indicator, i.e., each indicator is converted to the range [0,1].

[0128]

[0129] maxNi and minNi are the maximum and minimum values ​​of Ni, respectively, and are calculated using the following formulas:

[0130]

[0131] Where ai is the weighting coefficient. Furthermore, considering the influence of the natural environment, a natural correction coefficient αN = 0.8-1.2 is introduced, and the corrected natural geographical indicators are as follows: ;

[0132] like If [0, 3.0), then the water body is of a high ecological integrity type; if [3.0, 4.8) represents a moderately ecologically stable type; if [4.8, +∞) indicates a mild ecological stress type;

[0133] ② Type E Ecological Environment Indicators: Composed of vegetation coverage index E1, biodiversity index E2, water pollution index E3, and wind speed index E4. These indicators are standardized in units, meaning each indicator is converted to a value within the range [0,1].

[0134]

[0135] maxEi and minEi are the maximum and minimum values ​​of Ei, respectively, and are calculated using the following formulas:

[0136]

[0137] Where ai is the weighting coefficient. Introducing the ecological correction coefficient αE (0.7-1.3)

[0138] The revised ecological and environmental indicators are as follows:

[0139] If [0, 2.8), then the water body is of the ecologically harmonious type; [2.8, 5.2), then the water body is of a balanced type in terms of human activities; If [5.2, +∞), then the water body is suitable for human activities;

[0140] ③ Human activity index H type:

[0141] The human activity index is composed of population density index H1, land development intensity index H2, water resource utilization intensity index H3, and ship activity intensity index H4. The units of these indicators are standardized, meaning each indicator is converted to a value within the range [0,1].

[0142]

[0143] maxHi and minHi are the maximum and minimum values ​​of Hi, respectively, and are calculated using the following formulas:

[0144]

[0145] Where ai is the weighting coefficient. Introducing a human activity correction factor αH (0.9-1.1),

[0146] The revised human activity index is

[0147] If [0, 3.2), then the water body is of human development type; [3.2, 4.4), then the water body is of moderate disturbance type; If [4.4, +∞), then the water body is a low-level remediation type;

[0148] Water samples were taken from a tailings mine and analyzed to obtain the concentrations of various metal ions (Cu2+ concentration 1.0 mg / L, Zn2+ concentration 2.0 mg / L, Cr6+ ion concentration 0.1 mg / L, Ni2+ ion concentration 0.2 mg / L, etc.). The calculated value was P=4.72, and the water body was classified as an ecological environment indicator type (Type E). Data analysis of the river yielded Eeffective=5.13. The system is compatible with a versatile and practical detachable powder loading tank and a versatile and practical self-circulating ecological floating vessel, enabling more accurate watershed management and achieving purification.

[0149] Step 4: Intelligent control center (3) Determine the matching device for the plant powder adsorbent as a self-circulating ecological floating boat or a detachable powder loading tank according to the above three watershed types.

[0150] Step 5: Device manufacturing unit (9) determines the device size according to the watershed type and application scenario, and manufactures the device;

[0151] Step 6: The intelligent control center (3) uses the Dynamic Robust Synergistic Adsorption Optimization Algorithm (DRSAO) to calculate the effective adsorption capacity of the data input from Step 1 and Step 2, and then performs dynamic calculations and batch-by-batch iterative processing of the plant powder adsorbent. It analyzes the synergistic adsorption optimization of multiple metal ions and organic particles, and finally obtains the plant powder adsorbent formulation, such as... Figure 3 As shown, the details are as follows:

[0152] ① Input parameters and calculation of effective adsorption capacity

[0153] The water sample analysis unit measures the heavy metal content and organic particle concentration C0 in the water. Based on the national water quality discharge standards, the estimated heavy metal content and organic particle concentration Ct in the treated water are determined. The volume V of the treated water is measured, and the maximum adsorption capacity Qmax of different plant powder adsorbents is input. Combined with environmental correction factors α, β, and γ, the actual effective adsorption capacity is calculated.

[0154] Wherein, the environmental correction factor is:

[0155] α is the pH correction factor. The optimal adsorption of plant powder adsorbents usually occurs within a specific pH range (e.g., pH=5~6). When the pH deviates from the range, it needs to be multiplied by the correction factor α (0~1). If the pH deviates from the specific range by more than or equal to 1.5, take α=0~0.5. If the pH deviates from the specific range by less than 1.5, take α=0.5~1.

[0156] β is the temperature correction factor. The adsorption capacity of plant powder adsorbent is closely related to the temperature. The adsorption efficiency is best at 25℃. High or low temperatures inhibit the adsorption. When adsorption is carried out under different temperature conditions, the correction factor β (0.8~1) needs to be multiplied. If the temperature deviates from 25℃ by more than or equal to 20℃, β=0.8~0.9 is taken. If the temperature deviates from 25℃ by less than 20℃, β=0.9~1 is taken.

[0157] γ is the competitive remediation inhibition coefficient. The presence of other heavy metal ions or organic particles in the water body affects the adsorption capacity of plant powder adsorbents for specific heavy metal ions or organic particles, so a competitive inhibition coefficient γ (0~1) needs to be introduced. If there are more than 17 other heavy metal ions or organic particles in the water body, γ is taken as 0~0.5. If there are less than 17 other heavy metal ions or organic particles in the water body, γ is taken as 0.5~1.

[0158] The environmental correction coefficients α, β, and γ are extended to a robust optimization model:

[0159]

[0160]

[0161]

[0162] Dynamic calculation of plant powder adsorbent:

[0163] Introducing a safety factor k, calculate the mass of plant powder:

[0164]

[0165] The safety factor k is used to ensure that the adsorbent can cope with various uncertainties and potential efficiency losses in actual adsorption. Based on engineering experience and experimental data, it aims to ensure the safety and reliability of the system. k is 1.2 to 2.0. According to the degree, it is divided into mild k=1.2 to 1.4, moderate k=1.5 to 1.7, and severe k=1.8 to 2.0.

[0166] Introduce constraints:

[0167]

[0168] The emission standard constraint refers to the requirement that the expected heavy metal content and organic particulate concentration (Ct) of the treated water be less than or equal to the emission standard value. ;

[0169] The maximum batch limit refers to the maximum number of experimental batches set to prevent excessive experimental energy consumption and save costs; the maximum number of experimental batches must be less than or equal to [a certain value]. ;

[0170] The budget constraint refers to the quality of the novel plant powder adsorbent produced. Must be less than or equal to the constraint value This is to avoid the problem of raw material waste and loss;

[0171] ② Batch iterative processing

[0172] Various influencing factors affect the adsorption capacity of plant powder adsorbents, resulting in the water failing to reach the target concentration Ct after a single treatment. Therefore, the remaining concentration is calculated.

[0173]

[0174] Then, using Cremaining as the new C0, iterate until the target is met;

[0175] Supplementary recurrence relation:

[0176]

[0177] The transfer equation is an iterative equation for calculating the remaining concentration with added noisy terms. The noisy terms are intended to make the model more realistic and to take into account random disturbances when analyzing the robustness of the system, predicting the error range, or designing control strategies, so as to avoid the model becoming completely deterministic and unable to reflect the random behavior of the real system.

[0178] ③ Optimization of synergistic adsorption of multiple metal ions and organic particles

[0179] For water bodies containing multiple heavy metal ions and organic particles, calculate the mass Mi of plant powder adsorbent required to adsorb each type of metal ion and organic particle, and take the maximum value M=max(M1,M2,M3,...,Mn) to ensure that the concentrations of all heavy metal ions and organic particles meet the standards.

[0180] Introducing a coupling adsorption matrix A∈Rm×n, we describe the adsorption efficiency of m plant powder adsorbents for n heavy metal ions and organic particles:

[0181]

[0182] constraint:

[0183]

[0184] The first item (adsorption effect fitting item): The concentration of pollutant j at the initial time. The concentration of the j-th pollutant after treatment time t. This refers to the adsorption efficiency of plant powder adsorbent for the j-th pollutant. For each metal target concentration vector, this term represents minimizing the difference between the actual removal concentration and the theoretical adsorption efficiency. The sum of squares form aims to make the adsorption effect as close as possible to the expectation.

[0185] The second term (regularization term): ρ refers to the sum of squares of all elements in matrix A, and ρ refers to the regularization coefficient, which is used to balance fitting accuracy and model complexity. This term prevents the model from overfitting the experimental data, ensures that the element values ​​of the adsorption matrix A are not too large, and enhances the generalization ability.

[0186] Step 7: Adsorbent preparation unit 4 prepares plant powder adsorbents according to the obtained optimal formula;

[0187] Step 8: The adsorbent preparation unit 4 puts the prepared adsorbent into the device prepared in Step 5 for water treatment;

[0188] Step 9: The intelligent control center 3 calculates the time required for the plant powder adsorbent to be recycled after reaching its maximum adsorption efficiency using the Dynamic Economic Collaborative Recycling Optimization Algorithm (DESRO), determines the maximum adsorption efficiency and recycling time of the plant powder adsorbent, and transmits the data to the adsorbent recycling unit 12, as detailed below:

[0189] Plant powder adsorbents can be recycled after reaching maximum adsorption efficiency. By constructing a "mixed integer nonlinear programming (MINLP)" model, the dosage and recycling time of plant powder adsorbents can be optimized simultaneously.

[0190]

[0191]

[0192]

[0193] The first item is the cost of adsorbent usage: This indicates the amount of the i-th type of adsorbent used. This refers to the weighting coefficient for adsorbent dosage, ranging from 0.5 to 0.7. It reflects the importance of adsorbent cost in the overall objective. This item represents minimizing the total cost of adsorbent usage. If the cost is relatively high, A larger value, 0.6 to 0.7, should be chosen to reduce the amount of adsorbent used. If the cost is relatively low, Then take the smaller value, 0.5~0.6;

[0194] The second item is the time cost of recycling: This refers to the adsorption efficiency coefficient of the i-th adsorbent. This refers to the initial adsorption efficiency of the i-th adsorbent. The recovery time of the i-th adsorbent is the decision variable. The weighting coefficient for recovery time reflects the importance of recovery time cost in the overall objective. This factor represents minimizing the recovery time cost of all adsorbents. If the time cost is relatively high, A larger value, 0.2 to 0.3, should be chosen to shorten the recovery time. If the time cost is relatively low, Then take the smaller value, 0.1 to 0.2;

[0195] Constraints: The upper limit of the total available amount of adsorbent, the total amount constraint equation (②) is that the total amount of all adsorbents must not exceed the available resources. The upper limit of single recovery time is defined by the recovery time constraint equation (③), which states that the recovery time of each adsorbent cannot exceed the maximum allowable value.

[0196] For water bodies containing various heavy metal ions and organic particles, the time required ti for the plant powder adsorbent to adsorb various metal ions and organic particles was calculated, and the average time required to adsorb water bodies containing various heavy metal ions and organic particles was obtained:

[0197] ;

[0198] Step 10: The adsorbent recovery unit 12 replaces and recovers the adsorbent in the device at regular intervals.

[0199] In this embodiment, tailings water sampling and water sample data analysis were performed. The following data was input into the intelligent control system:

[0200] Sampling water volume: V=1000ml, temperature: 20℃, pH value: 6.5;

[0201] Determination of CO of four pollutants:

[0202] Cu2+: 1.0mg / L, Zn2+: 2.0mg / L, Cr6+: 0.1mg / L, PAHs: 0.1mg / L;

[0203] Determine the expected post-treatment concentration Ct:

[0204] Cu2+: 0.01mg / L, Zn2+: 0.05mg / L, Cr6+: 0.01mg / L, PAHs: 0.02mg / L;

[0205] Maximum adsorption capacity of the adsorbent, Qmax:

[0206] Cu2+: 0.9mg / g, Zn2+: 1.1mg / g, Cr6+: 0.6mg / g, PAHs: 0.5mg / g;

[0207] The intelligent control center concluded that:

[0208] pH correction factor α: 0.8 (pH=6.5, deviation from optimum range 1.5), temperature correction factor β: 0.9 (temperature=20℃, deviation from 25℃), competitive retrieval inhibition factor γ: 0.9 (4 competing ions exist), safety factor k: 1.5 (moderate uncertainty), budget constraint Mbudget: 5000 g, maximum batch limit Nmax: 5 times;

[0209] Calculate the effective adsorption capacity Qeffective:

[0210]

[0211] Cu2+: 0.9×0.8×0.9×0.9=0.5832mg / g

[0212] Zn2+: 1.1×0.8×0.9×0.9=0.7128mg / g

[0213] Cr6+: 0.6×0.8×0.9×0.9=0.3888mg / g

[0214] PAHs: 0.5×0.8×0.9×0.9=0.3240mg / g

[0215] Calculate the adsorbent mass M:

[0216]

[0217]

[0218]

[0219]

[0220]

[0221] Take the maximum value M = 4090g (satisfying the budget constraint M ≤ 5000g).

[0222] Batch iterative processing:

[0223] The first treatment used an adsorbent mass of M = 4090 g, and the remaining concentration was:

[0224]

[0225]

[0226]

[0227]

[0228] All pollutant concentrations met the standards, and no further iterations were required.

[0229] Adsorbent recovery time:

[0230] The adsorption efficiency of the adsorbent in water is v0 = 0.05 mg / g / d, and the recovery time is:

[0231]

[0232]

[0233]

[0234]

[0235] Average recovery time:

[0236]

[0237] Example 2: Same as Example 1, except that the sampling unit 1 is equipped with a drone or remote sensing technology for large-scale, high-precision auxiliary sampling, and a pH sensor and dissolved oxygen sensor are introduced to obtain more comprehensive water quality data.

[0238] Example 3: Same as Example 1, except that the intelligent control center 3 adds an intelligent early warning function. When water quality data is abnormal, it can automatically trigger the early warning mechanism and notify relevant personnel to handle it in a timely manner.

[0239] The specific embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

Claims

1. A smart integrated targeted plant powder adsorbent water purification method, characterized in that, The water purification system used in the method includes a sampling unit, a water sample analysis unit, an intelligent control center, an adsorbent preparation unit, an environmental analysis unit, a device manufacturing unit, and an adsorbent recovery unit. The sampling unit inputs the natural geographic information, ecological environment information, and human activity information obtained from the field survey into the intelligent control center. The intelligent control center connects the water sample analysis unit, the adsorbent preparation unit, and the environmental analysis unit. The water sample analysis unit is equipped with a full-parameter water quality analyzer, which transmits the measured heavy metal concentration, nutrient concentration, pH value, and soluble pollutant data to the intelligent control center. The intelligent control center analyzes the soluble pollutant data, calculates the required adsorbent mass, and transmits the data to the adsorbent preparation unit in real time. Simultaneously, the intelligent control center performs environmental analysis using GIS and satellite imagery to determine the watershed type and transmits the watershed type data to the environmental analysis unit. The watershed type data is also transmitted to the device manufacturing unit. The device manufacturing unit determines whether to use its self-circulating ecological floating vessel or a detachable powder-carrying device for water purification based on the watershed type. After water purification, the adsorbent then enters the adsorbent recovery unit. The specific steps of the water purification system are as follows: Step 1: The sampling unit is a full-coverage sampling unit, including the natural geographic information of the sampling points: watershed area index N1, narrowest river index N2, average slope index N3, river network density index N4; ecological environment information: vegetation coverage index E1, biodiversity index E2, water pollution index E3, wind speed E4; and human activity information: population density index H1, land development intensity index H2, water resource utilization intensity index H3, and vessel activity intensity index H4. The data is then entered into the intelligent control center. Step 2: The water sample analysis unit's full-parameter water quality analyzer integrates digestion and measurement, intelligently monitoring COD, ammonia nitrogen, total phosphorus, total nitrogen content, and pH value. The heavy metal detector in the water quality analyzer has 360° rotating colorimetric tube detection and cuvette detection, accurately monitoring the content of heavy metals such as copper, cadmium, zinc, and lead in the water. The water sample analysis unit inputs the measured data into the intelligent control center in real time. Step 3: The intelligent control center uses the dynamic multidimensional collaborative watershed classification algorithm DMS-WCS to derive three watershed types from the data input in Step 1 and Step 2: N-type natural geographical indicators, E-type ecological environment indicators, and H-type human activity indicators. Step 4: The intelligent control center determines the matching device for the plant powder adsorbent to be a self-circulating ecological floating boat or a detachable powder loading tank based on the above three watershed types. Step 5: The device manufacturing unit determines the device size based on the watershed type and application scenario, and manufactures the device. Step 6: The intelligent control center uses the Dynamic Robust Synergistic Adsorption Optimization Algorithm (DRSAO) to calculate the effective adsorption capacity of the data input from Step 1 and Step 2, and then performs dynamic calculations and batch iterative processing of the plant powder adsorbent to analyze the synergistic adsorption optimization of multiple metal ions and organic particles, and finally obtains the plant powder adsorbent formulation. Step 7: The adsorbent preparation unit prepares plant powder adsorbents according to the obtained optimal formula; Step 8: The adsorbent preparation unit puts the prepared adsorbent into the device prepared in Step 5 for water treatment; Step 9: The intelligent control center uses the Dynamic Economic Collaborative Recycling Optimization Algorithm (DESRO) to calculate the time required for the plant powder adsorbent to be recycled after reaching its maximum adsorption efficiency, determines the maximum adsorption efficiency and recycling time of the plant powder adsorbent, and transmits the data to the adsorbent recycling unit. Step 10: The adsorbent recovery unit replaces and recovers the adsorbent in the device at regular intervals.

2. The intelligent integrated targeted plant powder adsorbent water purification method according to claim 1, characterized in that: Step 3: The intelligent control center performs the calculations as follows: Based on the sampling data from the sampling units and water sample analysis units, the intelligent control center classifies water bodies into three index types, and first calculates the comprehensive water pollution index P: Where P represents the comprehensive water pollution index, C is the weighting coefficient, with a value range of [0, 1]. k S represents the measured concentration of a substance after water sample analysis. k This indicates the standard concentration limits for heavy metals in Class V water. Based on the P-value, the indicators are further classified: if P = 0-4.5, the water body is classified as a natural geographical indicator (N); if P = 4.5-7.0, the water body is classified as an ecological environmental indicator (E); if P > 7.0, the water body is classified as a human activity indicator (H). After determining the indicator classification, each indicator type is calculated and corrected. ① Type N Natural Geographic Indicators: Composed of natural geographic indicators such as watershed area N1, narrowest river N2, average slope N3, and river network density N4. Units for each indicator are standardized, and each indicator is converted to a value within the range [0,1]. The formula for calculating the natural geographical indicator N is as follows: in, These are the weighting coefficients. Furthermore, taking into account the influence of the natural environment, a natural correction coefficient α is introduced. N= The revised natural geographical indicators are 0.8-1.2: ; like [0, 3.0), then the water body is of high ecological integrity type, suitable for flexible and convenient detachable powder loading tank; if [3.0, 4.8) are moderately ecologically stable, compatible with flexible and convenient detachable powder loading tanks and flexible and convenient self-circulating ecological floating boats; if [4.8, +∞) is a mild ecological stress type, suitable for flexible and convenient self-circulating ecological floating boats; ② Type E Ecological Environment Indicators: Composed of vegetation coverage index E1, biodiversity index E2, water pollution index E3, and wind speed index E4. These indicators are standardized in units, with each index converted to a value within the range [0,1]. The ecological environment indicator E is calculated using the following formula: in, These are the weighting coefficients. Introducing an ecological correction coefficient (0.7-1.3), The revised ecological and environmental indicators are as follows: ; [0, 2.8), then the water body is of the ecologically harmonious type and is suitable for the versatile and practical detachable powder loading tank; [2.8, 5.2), then the water body is of a balanced type for human activities, and is suitable for a versatile and practical detachable powder loading tank and a versatile and practical self-circulating ecological floating boat; [5.2, +∞), then the water body is a human-activity type, adaptable, versatile and practical self-circulating ecological floating vessel; ③ Human activity index H type: The human activity index is composed of population density index H1, land development intensity index H2, water resource utilization intensity index H3, and ship activity intensity index H4. The units of these indices are standardized, and each index is converted to a value within the range [0,1]. The human activity index H is calculated using the following formula: in, These are the weighting coefficients. Introducing a human activity correction factor (0.9-1.1), The revised human activity index is ; [0, 3.2), then the water body is of human development type and is suitable for heavy-duty long-distance detachable powder carrying tank; [3.2, 4.4), then the water body is of moderate disturbance type, suitable for heavy-duty long-distance detachable powder loading tank and heavy-duty long-distance self-circulating ecological floating vessel; If [4.4, +∞), then the water body is of low-level restoration type and is suitable for an all-around practical self-circulating ecological floating vessel.

3. The intelligent integrated targeted plant powder adsorbent water purification method according to claim 1, characterized in that: In Step 6, the intelligent control center uses the Dynamic Robust Collaborative Adsorption Optimization Algorithm (DRSAO) to calculate the effective adsorption capacity based on the data input from Step 1 and Step 2, as detailed below: ① Input parameters and calculation of effective adsorption capacity The water sample analysis unit measured the heavy metal content and organic particulate concentration C0 in the water. Based on the national water quality discharge standards, the expected heavy metal content and organic particulate concentration C of the treated water were determined. t Measure the volume V of the treated water and input the maximum adsorption capacity Q of different plant powder adsorbents. max The actual effective adsorption capacity is calculated by combining environmental correction factors α, β, and γ: Wherein, the environmental correction factor is: α is the pH correction factor. The optimal adsorption of plant powder adsorbents usually occurs within a specific pH range. If the pH deviates from the specific range, the correction factor α needs to be multiplied. If the pH deviates from the specific range by more than or equal to 1.5, α is taken as 0~0.

5. If the pH deviates from the specific range by less than 1.5, α is taken as 0.5~1. β is the temperature correction factor. The adsorption capacity of plant powder adsorbent is closely related to the temperature. The adsorption efficiency is best at 25℃. High or low temperatures inhibit the adsorption. When adsorption is carried out under different temperature conditions, the correction factor β needs to be multiplied. If the temperature deviates from 25℃ by more than or equal to 20℃, β is taken as 0.8~0.

9. If the temperature deviates from 25℃ by less than 20℃, β is taken as 0.9~1. γ is the competitive remediation inhibition coefficient. The presence of other heavy metal ions or organic particles in the water body affects the adsorption capacity of plant powder adsorbents for specific heavy metal ions or organic particles, so a competitive inhibition coefficient γ needs to be introduced. If there are more than 17 other heavy metal ions or organic particles in the water body, γ is taken as 0~0.

5. If there are less than 17 other heavy metal ions or organic particles in the water body, γ is taken as 0.5~1. The environmental correction coefficients α, β, and γ are extended to a robust optimization model: Dynamic calculation of plant powder adsorbent: Introducing a safety factor k, calculate the mass of the plant powder adsorbent: The safety factor k is used to ensure that the adsorbent can cope with various uncertainties and potential efficiency losses in actual adsorption. Based on engineering experience and experimental data, it aims to ensure the safety and reliability of the system. k is 1.2 to 2.

0. According to the degree, it is divided into mild k=1.2 to 1.4, moderate k=1.5 to 1.7, and severe k=1.8 to 2.

0. Introduce constraints: Emission standards constraints refer to the set targets for the expected heavy metal content and particulate matter concentration (C) in the treated water. t The value must be less than or equal to the emission standard value. ; Maximum batch limit refers to the maximum number of experimental batches set to prevent excessive energy consumption and save costs. ; Budget constraints refer to the quality of the plant powder adsorbent produced. Must be less than or equal to the constraint value This is to avoid the problem of raw material waste and loss; ② Batch iterative processing Various influencing factors affect the adsorption capacity of plant powder adsorbents, resulting in the water failing to reach the target concentration C after a single treatment. t Then, the remaining concentration is calculated: And with C remaining Iterate as a new C0 until the target is met; Supplementary recurrence relation: The transfer equation is that a noisy term is added to the iterative equation for calculating the remaining concentration. The aim is to make the model more realistic, to consider random disturbances when analyzing the robustness of the system, the range of prediction error, or designing control strategies, and to avoid the model becoming completely deterministic and unable to reflect the random behavior of the real system. ③ Optimization of synergistic adsorption of multiple metal ions and organic particles For water bodies containing multiple heavy metal ions and organic particles, calculate the mass M of plant powder adsorbent required to adsorb each type of metal ion and organic particle. i Find the maximum value M = max(M1, M2, M3, ..., M n Ensure that the concentrations of all heavy metal ions and organic particles meet the standards. Introducing the coupling adsorption matrix A∈R m×n Describe the adsorption efficiency of m kinds of plant powder adsorbents for n kinds of heavy metal ions and organic particles: constraint: Among them, the first adsorption effect fitting term is: The concentration of pollutant j at the initial time. The concentration of the j-th pollutant after treatment time t. This refers to the adsorption efficiency of plant powder adsorbent for the j-th pollutant. For each metal target concentration vector, this term represents minimizing the difference between the actual removal concentration and the theoretical adsorption efficiency. The sum of squares form aims to make the adsorption effect as close as possible to the expectation. The second regularization term: ρ refers to the sum of squares of all elements in matrix A, and ρ refers to the regularization coefficient, which is used to balance fitting accuracy and model complexity. This term prevents the model from overfitting the experimental data, ensures that the element values ​​of the adsorption matrix A are not too large, and enhances the generalization ability.

4. The intelligent integrated targeted plant powder adsorbent water purification method according to claim 1, characterized in that: In Step 10, the adsorbent recovery unit will humify the plant powder adsorbents that have been saturated with adsorption in the self-circulating ecological floating boats or detachable powder-carrying devices used in the watershed, thereby achieving the recovery of heavy metals and the acquisition of organic fertilizer.

5. The intelligent integrated targeted plant powder adsorbent water purification method according to claim 1, characterized in that: The sampling unit incorporates drones or remote sensing technology for large-scale, high-precision auxiliary sampling, and introduces pH and dissolved oxygen sensors to obtain more comprehensive water quality data.

6. The intelligent integrated targeted plant powder adsorbent water purification method according to claim 1, characterized in that: The intelligent control center has been enhanced with an intelligent early warning function. When water quality data shows abnormalities, the early warning mechanism can be automatically triggered to promptly notify relevant personnel for handling.

Images