A seawater full-component high-efficiency separation system based on intelligent regulation

Abstract

This application discloses a high-efficiency separation system for all components of seawater based on intelligent control, relating to the field of seawater treatment technology. This application achieves closed-loop control of the entire process from raw seawater sample collection, multi-parameter water quality monitoring, ion concentration detection, component identification and path prediction, confined nanofiltration and air flotation impurity removal, temperature-controlled degassing and ion threshold separation, high-value ion enrichment and multi-path extraction, to real-time feedback adjustment. Through adaptive generation of separation paths and precise adjustment of process parameters, it solves the problems of poor component adaptability, high energy consumption, and poor target analyte selectivity in traditional seawater resource extraction methods. The overall separation system not only significantly improves the extraction efficiency and selectivity of trace high-value ions but also reduces operating energy consumption and the degree of human intervention, possessing significant advantages such as modular structure, intelligent operation, and wide applicability.

Description

Technical Field

[0001] This application relates to the field of seawater treatment technology, and in particular to a highly efficient seawater separation system based on intelligent control. Background Technology

[0002] Seawater resources contain a large amount of recyclable metal ions, organic matter and inorganic salts, which have broad prospects for extraction and reuse. Traditional seawater resource extraction methods mostly use a single process stage to separate the target components, such as nanofiltration, electrodialysis, extraction and precipitation. However, due to the complex composition of seawater, the variety of ions and the large concentration differences, traditional separation processes often face problems such as low processing efficiency, high energy consumption and difficulty in accurately distinguishing multiple components, making it difficult to achieve efficient and comprehensive recovery of various resources in seawater.

[0003] Existing seawater separation systems generally suffer from the following two key problems: First, the process path lacks an intelligent control mechanism, making it impossible to dynamically adjust the separation process according to the characteristics of the raw water components. This results in poor adaptability to different types of seawater samples and low component recovery efficiency. Second, most systems lack multi-dimensional data feedback closed-loop control methods and process control mechanisms based on multi-parameter optimization algorithms and reinforcement learning. This makes it difficult to achieve real-time adaptive optimization of operating parameters, which can easily lead to increased energy consumption, drift of operating parameters, and other phenomena, seriously affecting the overall operating performance. Summary of the Invention

[0004] The main objective of this application is to propose a highly efficient seawater separation system based on intelligent control to improve the separation efficiency of seawater.

[0005] To achieve the above objectives, this application proposes a highly efficient seawater full-component separation system based on intelligent control, the separation system comprising:

[0006] The raw water acquisition module is used to collect seawater samples; collect the water quality parameters of the seawater samples and detect the concentration of representative ions in the seawater samples; identify the component characteristics of the seawater samples, and then generate a separation path topology map;

[0007] The path identification module is used to configure the confined nanofiltration component and the multi-stage microbubble dissolved air flotation unit, configure the temperature-controlled degassing chamber and the ion threshold separation unit, and configure the online adjustment function module and feedback adjustment method.

[0008] The process execution module is used to configure electric field-induced chelation extraction and gradient solvent extraction, as well as to configure metal ion co-precipitation and parameter self-tuning methods.

[0009] The feedback optimization module is used to configure the full-process data integration platform and feedback optimization unit, dynamically generate operation instructions, and execute closed-loop control and strategy buffer methods.

[0010] The product collection module is used to configure the component storage and zoned collection system, configure terminal desalination and pure water recovery, and configure residual liquid value assessment and waste liquid safe treatment.

[0011] In some embodiments, the raw water collection module is used to collect seawater samples through a seawater sampling system set in the target sea area.

[0012] In some embodiments, the seawater sampling system includes a sampling arm, a sampling depth adjustment component, a sampling pump unit, and a seawater sample delivery pipeline;

[0013] The sampling depth adjustment component is set to a sampling depth of 5 to 10 meters to obtain representative seawater samples.

[0014] The sampling pump unit transports the seawater sample to the raw water sampling pool through the seawater sample delivery pipeline under constant negative pressure.

[0015] The raw water sampling pool is made of polyvinylidene fluoride and has an integrated constant temperature control module.

[0016] The constant temperature control module adopts a combination structure of temperature control board and heat exchange circulating water circuit, which is used to stably maintain the temperature of the seawater sample within the set temperature range.

[0017] The raw water sampling tank is equipped with a multi-parameter water quality monitoring array, which includes various sensors, including a temperature sensor, a conductivity sensor, a pH sensor, a turbidity sensor, and a redox potential sensor.

[0018] A temperature sensor is used to detect the temperature of the seawater sample; a conductivity sensor is used to check the total concentration of dissolved ions in the seawater sample; a pH sensor is used to check the acidity or alkalinity of the seawater sample solution; a turbidity sensor is used to check the concentration of particulate matter in the seawater sample; and a redox potential sensor is used to reflect the oxidizing and reducing environmental states of the seawater sample.

[0019] The temperature sensor, conductivity sensor, pH sensor, turbidity sensor, and redox potential sensor collect real-time data at a set frequency to form a set of raw water environmental parameters.

[0020] The set of raw water environmental parameters includes the temperature, conductivity, pH, electrochemical turbidity, and redox potential of the seawater sample.

[0021] In some embodiments, the outlet of the raw water sampling pool is provided with a seawater ion concentration detection module, which includes a micro ion-selective electrode array system and a plasma emission spectroscopy analysis subsystem.

[0022] The micro ion-selective electrode array system is used to dynamically detect the mass concentration of sodium ions, potassium ions, calcium ions, magnesium ions, lithium ions, and bromide ions in the seawater sample; each ion is provided with an independent electrode channel, and the electrode head of the electrode channel adopts a solid film structure.

[0023] The plasma emission spectroscopy analysis subsystem is used to analyze the mass concentrations of borate ions, strontium ions, and barium ions in the seawater sample; the plasma emission spectroscopy analysis subsystem employs a dual-channel sample introduction system and a CCD high-resolution detection module;

[0024] The micro ion-selective electrode array system and the plasma emission spectroscopy analysis subsystem are used to collaboratively detect and obtain a set of raw water ion component parameters, which includes multiple ion mass concentration indicators.

[0025] In some embodiments, when the raw water acquisition module performs the step of identifying the component characteristics of the seawater sample and generating a separation path topology map, it performs the following steps:

[0026] The raw water environmental parameter set and the raw water ionic component parameter set are transmitted to the edge computing module for processing via high-speed Ethernet; the edge computing module is equipped with a seawater component identification model and a separation process prediction model.

[0027] The seawater component identification model is used to construct input variables including sodium ion concentration, potassium ion concentration, calcium ion concentration, magnesium ion concentration, lithium ion concentration, bromide ion concentration, borate ion concentration, strontium ion concentration, barium ion concentration, conductivity value and ion concentration ratio through a decision tree structure. The output results are component labels, and the label types include lithium-rich type, magnesium-rich type, high sodium type, high hardness type and multiple impurities type.

[0028] The separation process prediction model is used to construct a matching structure based on path rules. It takes lithium-rich, magnesium-rich, high-sodium, high-hardness, and multi-impurity types as inputs and outputs complete process path structure data.

[0029] After the seawater component identification model is run, a unique corresponding component feature label is generated. The separation process prediction model uses this label as the decision entry point, calls the built-in path module library, and performs the optimal path recommendation based on the current parameters.

[0030] The result generated by the separation process prediction model is a structured separation path topology map, which includes: process path hierarchy, functional module configuration order, and set of control parameter initialization values.

[0031] The process path hierarchy is the division of functional segments involved in the separation process. The functional segment division includes: pretreatment segment, ion fractionation segment, high-value enrichment segment, functional module configuration order, and the order of operation and logical relationship of each functional module.

[0032] The functional modules include: an electro-driven ion migration module, a gradient solvent extraction module, and a chelation precipitation module. The set of initial control parameter values ​​includes: the target electric field strength value, membrane flux setting value, reaction residence time, extractant concentration setting value, flow rate setting value, and control cycle for each functional module.

[0033] After generating the separation path topology map, it is output in standard XML control format and automatically transmitted to the central control system. The central control system initializes and configures each module in the separation system and prepares the start command based on the structure and parameters in the separation path topology map.

[0034] In some embodiments, when the path identification module performs the step of configuring the confined nanofiltration component and the multi-stage microbubble dissolved air flotation unit, it performs the following steps:

[0035] The seawater sample is first introduced into the confined nanofiltration assembly, which contains a nanofiltration membrane unit made of polyamide material, the pore size of which ranges from 0.001 μm to 0.01 μm.

[0036] The nanofiltration membrane unit is used to retain organic particles and metal ions with divalent or trivalent charges in the seawater sample, with a particle size greater than 0.001 μm.

[0037] The configuration of the nanofiltration membrane unit includes: a pre-membrane pressure sensor, a post-membrane pressure sensor, and a flow sensor. The central control system adjusts the speed of the water pump in front of the membrane module and the membrane operating pressure based on the pressure difference before and after the membrane, the flux of the nanofiltration membrane module, and the concentration polarization.

[0038] The confined nanofiltration component performs bypass operation or skip configuration according to the limiting conditions in the separation path topology diagram;

[0039] The seawater sample after the confined nanofiltration treatment is introduced into the multi-stage microbubble dissolved air flotation unit. The multi-stage microbubble dissolved air flotation unit consists of three sets of flotation modules arranged in series. Each set of flotation modules includes: an ozone-air mixed microbubble injection device, a reaction chamber, and a separation tank. The ozone-air mixed microbubble injection device is used to generate ozone-air bubbles with a particle size of less than 20 μm. The injection pressure is controlled between 0.1 MPa and 0.3 MPa, and the ozone volume fraction is controlled within the range of 2% to 5%.

[0040] Microbubbles adhere to the surfaces of colloidal particles and nonpolar organic molecules in the seawater sample through interfacial adsorption mechanisms, thereby promoting their rise and entry into the top gas-liquid separation zone. The impurity-carrying capacity of microbubbles is calculated as follows:

[0041] ;

[0042] in, It is the mass of impurities carried by microbubbles rising per unit time, with units of 1. , It is the first Microbubble number concentration, in units of , It is the first The surface area of ​​microbubbles, in units of , It is the first The average contact time between the microbubble-like object and the target material, in seconds;

[0043] When the path identification module executes the step of configuring the temperature-controlled degassing chamber and the ion threshold separation unit, it performs the following steps:

[0044] The seawater sample treated with microbubble flotation is introduced into a temperature-controlled degassing chamber. The temperature-controlled degassing chamber adopts a two-section structure design. The first section is set to a temperature of 25°C, and the second section is set to a temperature of 45°C, which release bromide and hydrogen sulfide respectively. The degassing process is controlled by a multi-point temperature-controlled resistor module and a thermistor array.

[0045] The escaping gas is sent to the condensation and recovery device through the top gas duct, and the moisture and target components in the escaping gas are captured by the reflux condensation structure.

[0046] The central control system adjusts the heating power and water flow rate based on the gas mass concentration feedback from the condensation recovery device, so that the volatile component removal rate reaches more than 90%.

[0047] The seawater sample after temperature-controlled degassing is introduced into the ion threshold separation unit. The ion threshold separation unit adopts a double-layer anti-charge membrane structure, including a cation-selective permeable membrane and an anion-selective permeable membrane. Positive and negative electrode plates are respectively set on both sides of the membrane module, and a DC voltage is applied and controlled within the range of 3V to 5V to form a uniform electric field.

[0048] Under the influence of an electric field, multivalent ions are preferentially driven to the concentration-side collection tank due to their large charge number and fast migration rate, while monovalent ions partially enter the downstream module through the dialysis-side release channel; wherein the multivalent ions include magnesium ions and calcium ions, and the monovalent ions include sodium ions and potassium ions.

[0049] When the path identification module executes the steps of the online configuration adjustment function module and the feedback adjustment method, it performs the following steps:

[0050] Each functional module is equipped with an online adjustment module at the liquid outlet. The online adjustment module integrates an ion-selective electrode detection submodule, a spectral detection submodule, and a flow detection submodule.

[0051] The ion-selective electrode detection submodule is used to monitor changes in sodium ion concentration, calcium ion concentration, magnesium ion concentration and lithium ion concentration; the spectral detection submodule is used to analyze the degree of organic residue and color intensity in the solution; and the flow detection submodule is used to monitor the current outflow rate and internal pressure difference.

[0052] The central control system adjusts the system based on the real-time feedback from the online adjustment module using a proportional-integral-derivative closed-loop control strategy. The PID controller control logic is as follows:

[0053] ;

[0054] in, It is a feedback control system in time Adjust the output value, It is time The systematic error at any given time represents the difference between the set value and the measured value. It is the proportional adjustment coefficient, reflecting the response strength of the current error. It is an integral adjustment factor used to eliminate long-term accumulated errors. It is the differential adjustment coefficient, used to suppress overshoot caused by an excessively fast system response. It is an integral variable used to integrate errors over a past time period.

[0055] The central control system dynamically selects to skip or activate target functional modules before operation based on the module activation field in the separation path topology diagram. If the component characteristic label of the seawater sample is high sodium type, the central control system skips the multi-stage microbubble dissolved air flotation unit and the temperature-controlled degassing chamber. If the component characteristic label is multi-impurity type, the central control system prioritizes the activation of the confined nanofiltration component and the multi-stage microbubble dissolved air flotation unit.

[0056] In some embodiments, when the process execution module performs the steps of configured electric field-induced chelation extraction and gradient solvent extraction, it performs the following steps:

[0057] The concentrated solution of the seawater sample is first introduced into the electric field-induced chelation extraction module, which is filled with nanoscale chelating particles modified with functional groups of polyethyleneimine and pyrrolidone. The particle size of the chelating particles is in the range of 50nm to 100nm and they are uniformly distributed in three parallel electric field gradient cavities.

[0058] Each of the electric field gradient cavities is provided with parallel electrode plates for applying DC voltage to form an electric field gradient of 0.5V / cm to 3.0V / cm. The electric field induces target metal ions to migrate towards the direction of high electric field strength and to undergo a specific binding reaction with functional sites in the chelating particle material. The target metal ions include lithium ions, rubidium ions, and strontium ions.

[0059] The outlet of the electric field-induced chelation extraction module is equipped with a gradient solvent extraction module. The gradient solvent extraction module includes an organic phase addition unit, a static mixing reaction chamber, and a phase separation channel. The solvent system consists of n-hexane and ethanol mixed at a volume ratio of 3:1. The target chelating agent is sulfosalicylic acid, and the concentration is set to 0.05 mol / L.

[0060] The hexane-ethanol mixed solvent is injected into the reaction chamber at a flow rate of 0.1 mL / min via a high-precision peristaltic pump, and fully contacts the concentrated liquid entering the module at the phase interface. Bromine ions, borate ions and neutral or weakly polar trace substances preferentially enter the organic phase to achieve selective extraction.

[0061] When the process execution module performs the steps of configuring the metal ion co-precipitation and parameter self-tuning method, it performs the following steps:

[0062] The remaining concentrate that was not recovered by the electric field-induced chelation extraction module or the gradient solvent extraction module is fed into the metal ion coprecipitation module. The metal ion coprecipitation module is equipped with a pH adjustment unit and a dynamic dosing unit. The pH adjustment unit adjusts the pH value of the solution to the range of 9.0 to 9.5 through an alkaline precision control pump.

[0063] Using a dynamic dosing unit, phosphate or oxalate ions at a concentration of 0.05 mol / L are added based on detection feedback to co-precipitate with the target metal ions.

[0064] The precipitation reaction chamber is equipped with a swirling structure to control the particle size of the precipitate to be above 5μm. The precipitate is discharged through the slag discharge channel to achieve highly selective removal of calcium and magnesium ions.

[0065] An online concentration ratio calculation module is configured at the liquid outlet of the electric field-induced chelation extraction module, the gradient solvent extraction module, and the metal ion co-precipitation module, respectively. The online concentration ratio calculation module integrates a concentration sensor, a flow meter, and a temperature compensation module to collect liquid outlet concentration data.

[0066] When the concentration factor is lower than the minimum threshold set in the separation path topology diagram, the central control system automatically performs adjustment operations, including extending the chelation residence time, reducing the extraction flow rate, increasing the dosing concentration, or adjusting the pH value.

[0067] The central control system determines the enrichment path configuration scheme for the current operating cycle based on the component feature labels output by the seawater component identification model and the module activation fields in the separation path topology diagram. The enrichment path configuration scheme includes: when the component feature label is lithium-rich, the central control system only activates the electric field-induced chelation extraction module; when the component feature label is multi-impurity, the central control system activates the gradient solvent extraction module and the metal ion co-precipitation module; when the target label involves both high magnesium and high boron types, the three modules are activated synchronously and run in parallel.

[0068] In some embodiments, when the feedback optimization module executes the step of configuring the full-process data integration platform and the feedback optimization unit, it performs the following steps:

[0069] A full-process data integration platform is configured in the central control system to collect and process data in a unified manner, and to integrate key parameter data during the operation of the seawater full-component separation system. The full-process data integration platform includes: a sensor fusion center, a feedback optimization core module, and an operation command generator.

[0070] The sensor fusion center is equipped with a high-speed data acquisition channel, which connects to all deployed sensor nodes in the system. The parameters acquired include: raw water temperature, conductivity, pH value, oxidation-reduction potential, pressure difference before and after the membrane module, instantaneous energy consumption of the module, change in target ion concentration, and effluent flow rate. The fusion center integrates a data cleaning subroutine based on mean filtering and wavelet denoising algorithm, and performs Z-score standardization operation to generate a set of structured process monitoring parameters.

[0071] The feedback optimization core module is loaded into the end-to-end data integration platform. The feedback optimization core module is composed of a multi-objective genetic algorithm sub-module and an intelligent control agent based on deep reinforcement learning strategy.

[0072] The multi-objective genetic algorithm submodule uses the objectives of maximizing the recovery rate of the target ion, minimizing the energy consumption per unit of processing, and minimizing the residual concentration of the end reaction as the objective functions. Input variables include: membrane module operating pressure, electric field-induced voltage, chelating agent concentration, solvent injection rate, and module running time. The output is a parameter control vector set, specifically:

[0073] ;

[0074] in, It is the first The comprehensive optimization objective function value of each control scheme is used to evaluate the overall merits of the scheme. The recovery rate of the target component The weighting coefficients reflect the degree of preference for the recovery rate indicator. It is the first The target component recovery rate under each control scheme Energy consumption per unit of output The weighting coefficients reflect the degree of importance attached to energy-saving indicators. It is the first The energy consumed per unit mass of the target component under each control scheme, in units of , Residual concentration The weighting coefficients reflect the degree of requirement for the purity of the effluent. It is the first The residual concentration of the target component in the output liquid under each control scheme, in units of ;

[0075] The deep reinforcement learning agent employs a proximal policy optimization algorithm, taking historical running data as the input of the environmental state and the combination of regulatory actions as the output. The regulatory policy is updated based on the real-time reward value fed back by the system. The agent's optimization objectives include: maximizing long-term reward and minimizing policy perturbation.

[0076] When the feedback optimization module executes the dynamically generated operation instructions and performs the closed-loop control and policy buffer method steps, it performs the following steps:

[0077] The optimal parameter solution output by the feedback optimization core module is transmitted to the operation instruction generator. The operation instruction generator parses the optimization result into a system-executable instruction format and automatically completes the following tasks: adjusting the in-membrane pressure and operating pressure difference of the confined nanofiltration module, controlling the voltage value and electric field gradient of the electric field induced chelation extraction module, optimizing the solvent injection flow rate and chelating agent concentration of the gradient solvent extraction module, correcting the dosing rate and pH setpoint of the metal ion coprecipitation module, and dynamically setting the operating time window and control cycle of each module.

[0078] In some embodiments, when the product collection module executes the steps of configuring the component storage and zoned collection system and configuring terminal desalination and pure water recovery, it performs the following steps:

[0079] After completing the selective extraction of high-value trace components, the product streams of various target components are collected in precise zones according to ion type, and a target component storage and zone collection system is constructed. The storage and zone collection system includes lithium ion storage tanks, potassium ion storage tanks, magnesium ion storage tanks and borate ion storage tanks. Each storage tank is composed of a liquid level monitoring module, an ion concentration monitoring module, a temperature stabilization module and a liquid inlet flow rate adjustment module.

[0080] The liquid level monitoring module uses an ultrasonic level gauge for real-time volume detection with an accuracy of ±1%. The ion concentration monitoring module integrates a miniature ion selective electrode sensor to continuously detect the mass concentration of the corresponding target ion in the storage tank. The temperature stabilization module uses a constant temperature water jacket and a PTC heating element for linkage control to keep the storage liquid temperature stable at 25℃±1℃. The liquid inlet flow rate adjustment module adjusts the inlet speed of the target product liquid through a proportional solenoid valve group.

[0081] The remaining product stream after the extraction of all target components is guided into the terminal desalination and pure water recovery module, which consists of a three-stage reverse osmosis membrane treatment unit, a hollow fiber nanofiltration membrane filtration unit, and a pH adjustment and water quality stabilization unit.

[0082] The system utilizes a three-stage reverse osmosis membrane treatment unit to remove residual dissolved ions, small molecule organic matter, and trace impurities; a hollow fiber nanofiltration membrane filtration unit to further remove large molecule residual organic matter; and a pH adjustment and water quality stabilization unit to adjust the pH value of the output water to 6.5~7.0.

[0083] When the product collection module performs the steps of residual liquid value assessment and waste liquid safe treatment, it performs the following steps:

[0084] The remaining liquid that is not collected into the target component storage tank is uniformly entered into the residual liquid value assessment and intelligent classification judgment module. The residual liquid value assessment and intelligent classification judgment module consists of a high-value ion concentration detection subunit, a total organic carbon analysis subunit, and a pH stability judgment subunit.

[0085] The high-value ion concentration detection subunit is used to detect the residual concentration of key high-value ions such as lithium ions, magnesium ions and borate ions in the residual liquid; the total organic carbon analysis subunit monitors the concentration of organic pollutants in the residual liquid through a TOC analyzer; and the pH stability judgment subunit confirms whether the liquid has adjustable properties and is suitable for subsequent reflux treatment.

[0086] The residual liquid is classified according to the judgment logic by the central control system. The judgment logic includes: when the concentration of any high value-added ion exceeds the corresponding reflux threshold, the residual liquid is switched to the reflux path through the automatic control valve and reintroduced into the confined nanofiltration component or the electric field induced chelation extraction module. When the TOC concentration is higher than the preset threshold and the target ion concentration is lower than the minimum economic recovery concentration, the residual liquid is guided to the waste liquid treatment path.

[0087] All residual liquids that are confirmed to be non-recyclable are uniformly guided to the waste liquid safety treatment and neutralization discharge module. The waste liquid safety treatment and neutralization discharge module includes: a solid-liquid separation unit, a two-step neutralization reaction unit, a pre-discharge detection unit, and an intelligent discharge control module.

[0088] The solid-liquid separation unit uses a tubular high-speed centrifuge to physically separate suspended particles and dissolved sediments in the residual liquid. The two-step neutralization reaction unit first adjusts the pH value to 6.5~7.0 through acid-base titration, and then introduces ferric chloride to form ferric hydroxide precipitate to capture residual heavy metal ions. The pre-discharge detection unit detects pH, conductivity, heavy metal ion concentration and TOC value. The intelligent discharge control module automatically opens the liquid outlet channel after receiving a system confirmation signal before discharge.

[0089] The system's resource utilization rate assessment module is used to comprehensively analyze the target component production efficiency and resource utilization throughout the separation process. The system's resource utilization rate assessment module collects the following data: the liquid volume and mass concentration of the target component produced per unit of time per 24 hours, the total power consumption and total operating energy consumption of the system, the target component recovery rate, and the proportion of waste liquid discharged per ton of raw seawater.

[0090] In some embodiments, the components separated from the seawater sample using the separation system are used in post-processing scenarios.

[0091] The embodiments of this application include at least the following beneficial effects:

[0092] This application provides a highly efficient seawater separation system based on intelligent control. The system achieves closed-loop control throughout the entire process, from raw seawater sample collection, multi-parameter water quality monitoring, ion concentration detection, component identification and path prediction, confined nanofiltration and air flotation impurity removal, temperature-controlled degassing and ion threshold separation, high-value ion enrichment and multi-path extraction, to real-time feedback adjustment. Through adaptive generation of separation paths and precise adjustment of process parameters, it solves the problems of poor component adaptability, high energy consumption, and poor target analyte selectivity in traditional seawater resource extraction methods. The overall separation system not only significantly improves the extraction efficiency and selectivity of trace high-value ions but also reduces operating energy consumption and the degree of human intervention, possessing significant advantages such as modular structure, intelligent operation, and wide applicability. Attached Figure Description

[0093] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0094] Figure 1 A schematic diagram of a high-efficiency seawater separation system based on intelligent control, provided for an embodiment of this application;

[0095] Figure 2 This is an application flowchart of a high-efficiency seawater separation system based on intelligent control, provided in an embodiment of this application. Detailed Implementation

[0096] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0097] It is understood that the terms “first,” “second,” etc., used in this application may be used herein to describe various concepts, but unless otherwise stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the words “if,” “when,” or “in response to a determination” as used herein may be interpreted as “when…” or “when…” or “in response to a determination.”

[0098] As used in this application, the terms "at least one", "multiple", "each", "any", etc., "at least one" includes one, two or more, "multiple" includes two or more, "each" refers to each of the corresponding multiples, and "any" refers to any one of the multiples.

[0099] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0100] The purpose of this application is to address the shortcomings of existing technologies by providing a highly efficient seawater separation system based on intelligent regulation. This system integrates multiple functional units, including a component identification and process prediction module, a confined nanofiltration and air flotation pretreatment module, a temperature-controlled degassing and electric field separation module, an electro-induced chelation and gradient extraction module, a co-precipitation concentration module, and a closed-loop feedback regulation module. This allows for intelligent path customization and dynamic parameter adjustment control for different types of seawater, improving the recovery rate and purity of various target ions in seawater, significantly reducing system energy consumption per unit, enhancing the system's adaptability to environmental fluctuations and operational stability, and demonstrating promising prospects for industrial applications.

[0101] Reference Figure 1 and Figure 2 This application provides a highly efficient seawater separation system based on intelligent control, the separation system comprising:

[0102] The raw water acquisition module is used to collect raw seawater samples, collect raw water quality parameters, detect the concentration of representative ions in seawater, identify the component characteristics of seawater samples, and generate a separation path topology map.

[0103] The path identification module is used to configure the confined nanofiltration component and the multi-stage microbubble dissolved air flotation unit, the temperature-controlled degassing chamber and the ion threshold separation unit, and the online adjustment function module and feedback regulation mechanism.

[0104] The process execution module is used to configure electric field-induced chelation extraction and gradient solvent extraction, as well as to configure metal ion co-precipitation and parameter self-adjustment mechanisms.

[0105] The feedback optimization module is used to configure the full-process data integration platform and feedback optimization core, dynamically generate operation instructions, and execute closed-loop control and policy buffer mechanisms.

[0106] The product collection module is used to configure the component storage and zoned collection system, configure terminal desalination and pure water recovery, and configure residual liquid value assessment and waste liquid safe treatment.

[0107] In some specific implementations, the raw water acquisition module specifically includes:

[0108] Raw seawater samples were collected, raw water quality parameters were obtained, and the concentrations of representative ions in the seawater were measured.

[0109] Raw seawater samples are collected using a seawater sampling system set up in the target sea area.

[0110] The seawater sampling system includes a sampling arm, a sampling depth adjustment component, a sampling pump unit, and a seawater sample delivery pipeline. The sampling depth adjustment component is set to a sampling depth of 5 to 10 meters to obtain representative seawater samples.

[0111] The sampling pump unit transports seawater samples to the raw water sampling pool through the seawater sample delivery pipeline under constant negative pressure. The raw water sampling pool is made of polyvinylidene fluoride and integrates a constant temperature control module.

[0112] The constant temperature control module adopts a combination structure of temperature control board and heat exchange circulating water circuit to keep the temperature of the raw water sample stable at 22℃±1℃.

[0113] A multi-parameter water quality monitoring array is installed inside the raw water sampling tank. The multi-parameter water quality monitoring array includes sensor modules: temperature sensor, conductivity sensor, pH sensor, turbidity sensor and redox potential sensor.

[0114] Temperature sensor detects raw water temperature with an accuracy of ±0.1℃; conductivity sensor checks the total concentration of dissolved ions; pH sensor checks the acidity or alkalinity of the solution; turbidity sensor checks particulate matter concentration; and oxidation-reduction potential sensor reflects the oxidizing / reducing environmental conditions.

[0115] Temperature sensors, conductivity sensors, pH sensors, turbidity sensors, and redox potential sensors collect real-time data once per second, forming a complete set of raw water environmental parameters.

[0116] The raw water environmental parameter set consists of five fields: seawater temperature, conductivity, pH, electrochemical turbidity, and redox potential.

[0117] A seawater ion concentration detection module is installed at the outlet of the raw water sampling pool. The seawater ion concentration detection module consists of two core components: a micro ion-selective electrode array system and a plasma emission spectroscopy analysis subsystem.

[0118] The miniature ion-selective electrode array system can detect the mass concentration of sodium, potassium, calcium, magnesium, lithium, and bromide ions in real time. Each ion has an independent electrode channel, and the electrode head adopts a solid film structure with a response time of less than 10 seconds.

[0119] The plasma emission spectrometry analysis subsystem analyzes the mass concentrations of borate ions, strontium ions, and barium ions. The spectrometry analysis subsystem uses a dual-channel sample introduction system and a CCD high-resolution detection module, and the analysis time does not exceed 30 seconds per sample.

[0120] The raw water ion component parameter set is constructed by the coordinated detection of the micro ion-selective electrode array system and the plasma emission spectroscopy analysis subsystem. The raw water ion component parameter set contains nine ion mass concentration indicators, all output in mg / L.

[0121] Identify the component characteristics of seawater samples and generate a separation path topology map.

[0122] The raw water environmental parameter set and the raw water ion component parameter set are transmitted to the edge computing module via high-speed Ethernet. The edge computing module is equipped with a seawater component identification model and a separation process prediction model.

[0123] The seawater component identification model uses a decision tree structure to construct input variables including sodium ion concentration, potassium ion concentration, calcium ion concentration, magnesium ion concentration, lithium ion concentration, bromide ion concentration, borate ion concentration, strontium ion concentration, barium ion concentration, conductivity value to ion concentration ratio, and output results as component labels. The label types include lithium-rich, magnesium-rich, high-sodium, high-hardness, and multi-impurity types.

[0124] The separation process prediction model is constructed by matching the structure according to the path rules. The input is lithium-rich, magnesium-rich, high-sodium, high-hardness, and multi-impurity types, and the output is complete process path structure data.

[0125] After the seawater component identification model runs, it generates a unique corresponding component feature label. The separation process prediction model uses this label as the decision entry point, calls the built-in path module library, and performs optimal path recommendation based on the current parameters.

[0126] The result generated by the separation process prediction model is a structured separation path topology diagram, which includes: process path hierarchy, functional module configuration order, and set of control parameter initialization values.

[0127] The process path hierarchy refers to the functional segment division involved in the separation process. The functional segments are divided into: pretreatment segment, ion fractionation segment, and high-value enrichment segment. The configuration order of functional modules, the order of operation and logical relationship of each functional module, and the module types include: electro-driven ion migration module, gradient solvent extraction module and chelation precipitation module. The set of initial control parameter values ​​includes: target electric field strength value, membrane flux setting value, reaction residence time, extractant concentration setting value, flow rate setting value and control cycle for each functional module.

[0128] After generating the split path topology map, it is output in standard XML control format and automatically transmitted to the central control system. The central control system initializes and configures each functional module in the system and prepares the start command based on the structure and parameters in the received topology map.

[0129] After the central control system completes the parameter writing, it returns an initialization status confirmation signal. The initialization status confirmation signal is used to notify the control layer that the process path configuration is complete and the system enters the separate process operation preparation state.

[0130] In some specific implementations, the path recognition module specifically includes:

[0131] It is equipped with a confined nanofiltration component and a multi-stage microbubble dissolved air flotation unit.

[0132] The raw seawater sample first enters the confined nanofiltration module, which is equipped with a nanofiltration membrane unit made of polyamide, with a pore size ranging from 0.001 μm to 0.01 μm.

[0133] Nanofiltration membrane units retain organic particles and metal ions with divalent or trivalent charges, with a particle size greater than 0.001 μm.

[0134] The nanofiltration membrane module configuration includes: a pre-membrane pressure sensor, a post-membrane pressure sensor, and a flow sensor. The central control system adjusts the speed of the upstream water pump and the membrane operating pressure based on the pressure difference across the membrane, the nanofiltration membrane module flux, and the concentration polarization. The water flux calculation is as follows:

[0135] ;

[0136] In the formula: It is the water flux per unit membrane area per unit time, with units of 1000 liters. , It is the pressure difference across the nanofiltration membrane, measured in bar. It is the osmotic pressure difference across the nanofiltration membrane, measured in bar. It is the dynamic viscosity of a liquid, measured in Pa·s. It is the membrane resistance coefficient of nanofiltration membrane, which is dimensionless or calibrated according to the system.

[0137] The confined nanofiltration component performs bypass operation or skips configuration based on the constraints in the path topology graph.

[0138] Seawater samples treated by confined nanofiltration are fed into a multi-stage microbubble dissolved air flotation unit. The multi-stage microbubble dissolved air flotation unit consists of three sets of flotation modules arranged in series. Each set of flotation modules includes an ozone-air mixed microbubble injection device, a reaction chamber, and a separation tank. The mixed microbubble injection device generates ozone-air bubbles with a particle size of less than 20 μm. The injection pressure is controlled between 0.1 MPa and 0.3 MPa, and the ozone volume fraction is controlled between 2% and 5%.

[0139] Microbubbles adhere to the surfaces of colloidal particles and nonpolar organic molecules in water through interfacial adsorption, promoting their flotation and entry into the top gas-liquid separation zone. The impurity carrying capacity of microbubbles is calculated as follows:

[0140] ;

[0141] In the formula: It is the mass of impurities carried by microbubbles rising per unit time, with units of 1. , It is the first Microbubble number concentration, in units of , It is the first The surface area of ​​microbubbles, in units of , It is the first The average contact time between the microbubble-like object and the target material, in seconds.

[0142] It is equipped with a temperature-controlled degassing chamber and an ion threshold separation unit.

[0143] After microbubble flotation treatment, the seawater sample enters the temperature-controlled degassing chamber. The temperature-controlled degassing chamber adopts a two-section structure design. The first section is set to a temperature of 25°C, and the second section is set to a temperature of 45°C, which release bromide and hydrogen sulfide respectively. The degassing process is controlled by a multi-point temperature-controlled resistor module and a thermistor array.

[0144] The escaped gas is sent to the condensation and recovery unit through the top gas duct. A reflux condensation structure is used to capture the moisture and target components in the escaped gas. The degassing efficiency is calculated as follows:

[0145] ;

[0146] In the formula: This is the degassing efficiency, which indicates the percentage of volatile components removed. This is the mass concentration of dissolved gas before it enters the temperature-controlled degassing chamber, in units of... , This refers to the residual dissolved gas concentration in the effluent from the degassing chamber, expressed in units of... .

[0147] The central control system adjusts the heating power and water flow rate based on the gas mass concentration feedback from the condensation recovery unit, so that the removal rate of volatile components reaches more than 90%.

[0148] After temperature-controlled degassing, the seawater sample enters the ion threshold separation unit. The ion threshold separation unit adopts a double-layer anti-charge membrane structure, including a cation-selective permeable membrane and an anion-selective permeable membrane. Positive and negative electrode plates are set on both sides of the membrane module, and a DC voltage is applied and controlled within the range of 3V to 5V to form a uniform electric field.

[0149] Under the influence of an electric field, multivalent ions such as magnesium and calcium ions are preferentially driven to the concentration-side collection tank due to their large charge number and fast migration rate, while monovalent ions such as sodium and potassium ions partially enter the downstream module through the dialysis-side release channel. The specific calculation of ion migration rate is as follows:

[0150] ;

[0151] In the formula: It is the first The migration rate of ionoids, in units of , It is the first The charge number of ions, It is the fundamental charge constant, with a value of , It is electric field strength, and the unit is 1000 kJ / m². , It is the dynamic viscosity of water, measured in Pa⁻¹s. It is the first Hydration radius of ions, in nm.

[0152] By adjusting the electric field strength and the working area of ​​the membrane module, selective migration control of ions can be achieved, with a separation efficiency of no less than 85%.

[0153] Configure online adjustment function modules and feedback adjustment mechanisms.

[0154] Each functional module is equipped with an online adjustment module at its outlet, which integrates an ion-selective electrode detection submodule, a spectral detection submodule, and a flow detection submodule.

[0155] The ion-selective electrode detection submodule monitors changes in sodium, calcium, magnesium, and lithium ion concentrations; the spectral detection submodule analyzes the degree of organic residue and color intensity in the solution; and the flow detection submodule monitors the current outflow rate and internal pressure differential.

[0156] The central control system adjusts based on real-time feedback from the online adjustment module, employing a proportional-integral-derivative closed-loop control strategy. The PID controller control logic is as follows:

[0157] ;

[0158] In the formula: It is a feedback control system in time Adjust the output value, It is time The systematic error at any given time is the difference between the set value and the measured value. It is the proportional adjustment coefficient, reflecting the response strength of the current error. It is an integral adjustment factor used to eliminate long-term accumulated errors. It is the differential adjustment coefficient, used to suppress overshoot caused by an excessively fast system response. It is an integral variable used to integrate errors over a past time period. The adjustment strategy can automatically correct operational deviations and ensure that each functional module is in a highly efficient and stable operating range.

[0159] Based on the module activation field in the separation path topology diagram, the central control system can dynamically select to skip or activate the target functional module before operation. If the component characteristic label of the seawater sample is high sodium type, the central control system skips the multi-stage microbubble dissolved air flotation unit and the temperature-controlled degassing chamber. If the component characteristic label is multi-impurity type, the central control system prioritizes to activate the confined nanofiltration component and the multi-stage microbubble dissolved air flotation unit.

[0160] In some specific implementations, the process execution module specifically includes:

[0161] The system is configured with electric field-induced chelation extraction and gradient solvent extraction. The concentrate is first introduced into the electric field-induced chelation extraction module, which is filled with nanoscale chelating particles modified with functional groups of polyethyleneimine and pyrrolidone. The particle size of the chelating particles is controlled in the range of 50nm to 100nm and is uniformly distributed in three parallel electric field gradient cavities.

[0162] Parallel electrode plates are placed within each electric field gradient cavity. A DC voltage is applied to create an electric field gradient ranging from 0.5 V / cm to 3.0 V / cm. The electric field induces trace metal ions to migrate towards the direction of higher electric field strength and undergo specific binding reactions with functional sites in the chelating particle material. The target ions include lithium ions, rubidium ions, and strontium ions. The migration selectivity of the target ions is calculated using a migration selectivity factor.

[0163] ;

[0164] In the formula: It is the first The migration selectivity factor of a ion-like ion measures its preferential migration ability relative to other ions under the influence of an electric field. It is the first The charge number of an ion represents the numerical value of the positive or negative charge carried by the ion. It is the first Electromobility of ions, in units of This represents the migration rate of the ion per unit strength in an electric field. It is electric field strength, and the unit is 1000 kJ / m². This represents the voltage gradient of the electric field over a unit length. It is the first The diffusion coefficient of ions, in units of This indicates the ability of the ion to diffuse in a liquid.

[0165] The central control system adjusts the electric field strength and residence time of chelated particles in real time based on feedback from the ion-selective electrode detection submodule to ensure that the selective enrichment efficiency of target ions is not less than 92%.

[0166] A gradient solvent extraction module is configured at the outlet of the electric field-induced chelation extraction module. The module includes an organic phase addition unit, a static mixing reaction chamber, and a phase separation channel. The solvent system consists of n-hexane and ethanol mixed at a volume ratio of 3:1. The target chelating agent is sulfosalicylic acid, and the concentration is set to 0.05 mol / L.

[0167] The hexane-ethanol mixed solvent is injected into the reaction chamber at a flow rate of 0.1 mL / min using a high-precision peristaltic pump, ensuring full contact with the concentrated solution entering the module at the phase interface. Bromide ions, borate ions, and trace amounts of neutral or weakly polar substances preferentially enter the organic phase, achieving selective extraction. The partition coefficient of the target ion is calculated as follows:

[0168] ;

[0169] In the formula: It is the partition coefficient of the target ion, used to measure the affinity of the target ion between the organic and aqueous phases. It is the mass concentration of the target ion in the organic phase, in units of , It is the mass concentration of the target ion in the aqueous phase, in units of .

[0170] The central control system, in conjunction with the spectrophotometric detection submodule, adjusts the organic solvent injection rate and chelating agent concentration in real time based on the distribution coefficient to ensure a separation efficiency of no less than 90%.

[0171] Configure a metal ion coprecipitation and parameter self-adjustment mechanism.

[0172] The remaining concentrate that is not recovered by the electric field-induced chelation extraction module or the gradient solvent extraction module enters the metal ion coprecipitation module. The metal ion coprecipitation module is equipped with a pH adjustment unit and a dynamic dosing unit. The pH adjustment unit adjusts the pH value of the solution to the range of 9.0 to 9.5 through an alkaline precision control pump.

[0173] The dynamic dosing unit adds phosphate or oxalate ions at a concentration of 0.05 mol / L based on the detection feedback, which then co-precipitate with the target ions.

[0174] The precipitation reaction chamber is equipped with a swirling structure to control the particle size of the precipitate to be above 5μm. The precipitate is discharged through the slag discharge channel, achieving highly selective removal of calcium and magnesium ions.

[0175] Online concentration ratio calculation modules are configured at the outlet ends of the electric field-induced chelation extraction module, the gradient solvent extraction module, and the metal ion co-precipitation module, respectively. The online concentration ratio calculation module integrates a concentration sensor, a flow meter, and a temperature compensation module to collect the outlet concentration data.

[0176] The concentration factor is calculated as follows:

[0177] ;

[0178] In the formula: It is the concentration factor of the target component, used to indicate the degree of enrichment of the target ions during the treatment process. This is the mass concentration of the target ions in the module's output liquid, in units of... , This is the mass concentration of the target ions in the module's feed solution, in units of... .

[0179] When the concentration factor is lower than the minimum threshold set in the separation path topology diagram, the central control system automatically performs adjustment operations, including: extending the chelation residence time, reducing the extraction flow rate, increasing the dosing concentration, or adjusting the pH value.

[0180] The central control system determines the enrichment path configuration scheme for the current operating cycle based on the component feature labels output by the seawater component identification model and the module activation field in the separation path topology diagram. The scheme includes: when the component feature label is lithium-rich, the central control system only activates the electric field-induced chelation extraction module; when the component feature label is multi-impurity, the central control system activates the gradient solvent extraction module and the metal ion co-precipitation module; when the target label involves both high magnesium and high boron types, the three modules are activated simultaneously and run in parallel.

[0181] In some specific implementations, the feedback optimization module specifically includes:

[0182] Configure a full-process data integration platform and a feedback optimization core.

[0183] A full-process data integration platform is configured in the central control system to collect and process data in a unified manner, and to integrate key parameter data during the operation of the seawater full-component separation system. The full-process data integration platform includes: a sensor fusion center, a feedback optimization core module, and an operation command generator.

[0184] The sensor fusion center is equipped with a high-speed data acquisition channel, which connects to all deployed sensor nodes in the system. The parameters collected include: raw water temperature, conductivity, pH value, redox potential, pressure difference across the membrane module, instantaneous energy consumption of the module, change in target ion concentration, and effluent flow rate. The fusion center integrates a data cleaning subroutine based on mean filtering and wavelet denoising algorithm, and performs Z-score standardization operation to generate a structured set of process monitoring parameters.

[0185] The feedback optimization core module is loaded into the end-to-end data integration platform. The feedback optimization core module consists of a multi-objective genetic algorithm sub-module and an intelligent control agent based on deep reinforcement learning strategy.

[0186] The multi-objective genetic algorithm submodule aims to maximize the recovery rate of the target ion, minimize the energy consumption per unit of processing, and minimize the residual concentration of the final reaction. Input variables include: membrane module operating pressure, electric field-induced voltage, chelating agent concentration, solvent injection rate, and module running time. The output is a parameter control vector set, specifically:

[0187] ;

[0188] In the formula: It is the first The comprehensive optimization objective function value of each control scheme is used to evaluate the overall merits of the scheme. The recovery rate of the target component The weighting coefficients reflect the degree of preference for the recovery rate indicator. It is the first The target component recovery rate under each control scheme Energy consumption per unit of output The weighting coefficients reflect the degree of importance attached to energy-saving indicators. It is the first The energy consumed per unit mass of the target component under each control scheme, in units of , Residual concentration The weighting coefficients reflect the degree of requirement for the purity of the effluent. It is the first The residual concentration of the target component in the output liquid under each control scheme, in units of .

[0189] Deep reinforcement learning agents employ a proximal policy optimization algorithm, taking historical operational data as input to the environment state and combinations of regulatory actions as output. The regulatory policy is updated based on real-time reward values ​​fed back by the system. The agent's optimization objectives include: maximizing long-term returns and minimizing policy perturbations.

[0190] Dynamically generate operation instructions and execute closed-loop control and policy buffer mechanisms.

[0191] The optimal parameter solution output by the feedback optimization core module is transmitted to the operation instruction generator. The operation instruction generator parses the optimization result into a system-executable instruction format and automatically completes the following tasks: adjusting the membrane inlet pressure and operating pressure difference of the confined nanofiltration module, controlling the voltage value and electric field gradient of the electric field induced chelation extraction module, optimizing the solvent injection flow rate and chelating agent concentration of the gradient solvent extraction module, correcting the dosing rate and pH setpoint of the metal ion coprecipitation module, and dynamically setting the operating time window and control cycle of each module.

[0192] The generated instructions are issued to the central control system in the form of a standardized control instruction set. The central control system then writes and adjusts the parameters of each functional module in real time based on the instructions.

[0193] To balance the system's response speed and operational stability, a parameter update trigger strategy is set, including: in normal mode, a closed-loop parameter optimization refresh is performed every hour; when any abnormal deviation event occurs, an optimization refresh is performed immediately. Abnormal deviation events are: the recovery rate of any target component decreases by more than 5% compared to the previous cycle, the system's unit energy consumption increases by more than 10% compared to the previous cycle, and the operating parameters of any module deviate from the set value threshold by more than ±15%.

[0194] Before the optimization parameters take effect, all instructions to be adopted first enter the policy buffer. The policy buffer is set with a 5-minute evaluation window to make a prediction of system stability and ensure that the implementation of the instructions will not cause instantaneous disturbances.

[0195] If the system is stable, the strategy will automatically enter the execution phase. If the system is in a highly sensitive state, the strategy will be temporarily suspended and an early warning log will be generated.

[0196] In some specific implementations, the product collection module specifically includes:

[0197] Configure a component storage and zoned collection system, and configure terminal desalination and pure water recycling.

[0198] After completing the selective extraction of high-value trace components, the product streams of various target components are collected in precise zones according to ion type, and a target component storage and zone collection system is constructed, including: lithium ion storage tank, potassium ion storage tank, magnesium ion storage tank and borate ion storage tank. Each storage tank consists of a liquid level monitoring module, an ion concentration monitoring module, a temperature stabilization module and an inlet flow rate regulation module.

[0199] The liquid level monitoring module uses an ultrasonic level gauge for real-time volume detection with an accuracy of ±1%. The ion concentration monitoring module integrates a miniature ion selective electrode sensor to continuously detect the mass concentration of the corresponding target ion in the storage tank. The temperature stabilization module uses a constant temperature water jacket and a PTC heating element for linkage control to keep the storage liquid temperature stable at 25℃±1℃. The liquid inlet flow rate adjustment module adjusts the inlet speed of the target product liquid through a proportional solenoid valve group.

[0200] The remaining product stream after the extraction of all target components will be directed into the terminal desalination and pure water recovery module, which consists of a three-stage reverse osmosis membrane treatment unit, a hollow fiber nanofiltration membrane filtration unit, and a pH adjustment and water quality stabilization unit.

[0201] The three-stage reverse osmosis membrane treatment unit removes residual dissolved ions, small molecule organic matter, and trace impurities. The hollow fiber nanofiltration membrane unit further removes large molecule residual organic matter. The pH adjustment and water quality stabilization unit adjusts the pH value of the output water to 6.5~7.0.

[0202] Configure residual liquid value assessment and safe waste liquid treatment.

[0203] For any remaining liquid that is not collected into the target component storage tank, it is uniformly entered into the residual liquid value assessment and intelligent classification judgment module. The residual liquid value assessment and intelligent classification judgment module consists of a high-value ion concentration detection subunit, a total organic carbon analysis subunit, and a pH stability judgment subunit.

[0204] The high-value ion concentration detection subunit detects the residual concentrations of key high-value ions such as lithium, magnesium, and borate ions in the residual liquid. The total organic carbon analysis subunit monitors the concentration of organic pollutants in the residual liquid using a TOC analyzer. The pH stability assessment subunit confirms whether the liquid has adjustable properties and is suitable for subsequent reflux treatment.

[0205] The central control system performs residual liquid path classification according to the judgment logic. The judgment logic includes: when the concentration of any high value-added ion exceeds the corresponding reflux threshold, the residual liquid is switched to the reflux path through the automatic control valve and re-introduced into the confined nanofiltration component or the electric field induced chelation extraction module. When the TOC concentration is higher than the preset threshold and the target ion concentration is lower than the minimum economic recovery concentration, the residual liquid is guided to the waste liquid treatment path.

[0206] For residual liquids that are confirmed to be non-recyclable, the system uniformly guides them to the waste liquid safety treatment and neutralization discharge module, which includes: a solid-liquid separation unit, a two-step neutralization reaction unit, a pre-discharge detection unit, and an intelligent discharge control module.

[0207] The solid-liquid separation unit uses a tubular high-speed centrifuge to physically separate suspended particles and dissolved sediments in the residual liquid. The two-step neutralization reaction unit first adjusts the pH value to 6.5~7.0 through acid-base titration, and then introduces ferric chloride to form ferric hydroxide precipitate to capture residual heavy metal ions. The pre-discharge detection unit detects pH, conductivity, heavy metal ion concentration and TOC value. The intelligent discharge control module automatically opens the liquid outlet channel after receiving a system confirmation signal before discharge.

[0208] Configure a system operation resource utilization rate assessment module to comprehensively analyze the target component production efficiency and resource utilization throughout the entire separation process. The system operation resource utilization rate assessment module collects the following data: liquid volume and mass concentration of the target component produced per unit of output per 24 hours, total system power consumption and total operating energy consumption, target component recovery rate, and the proportion of waste liquid discharged per ton of raw seawater. The target component recovery rate is calculated as follows:

[0209] ;

[0210] In the formula: This is the target component recovery rate, representing the percentage of the total target component retained in the product solution after extraction relative to the total original input. It is the first The mass concentration of the target component in the product liquid, in units of , It is the first The volume of the product liquid of the target component, in liters (L). It is the first The mass concentration of the target component in the influent, in units of , It is the first The volume of the target component introduced into the liquid, in liters (L). It is the total number of target component types being counted.

[0211] All data is uploaded to the central control system database every 8 hours and serves as the input for the parameter optimization module, assisting the system in making intelligent adjustments in the next processing cycle.

[0212] In practical application, the above-mentioned process involves first collecting raw seawater samples at a depth of 5 to 10 meters using a seawater sampling system set up in the target sea area. The sampling system includes a sampling arm, a depth adjustment component, a sampling pump unit, and delivery pipelines. During the sampling process, a negative pressure pump is used to transport seawater to a constant-temperature sampling pool made of polyvinylidene fluoride. Then, a multi-parameter water quality monitoring array integrated in the sampling pool continuously collects temperature, conductivity, pH value, turbidity, and redox potential. In conjunction with an ion-selective electrode and a spectral analysis module, the concentrations of sodium ions, potassium ions, calcium ions, magnesium ions, lithium ions, bromide ions, borate ions, strontium ions, and barium ions are detected.

[0213] Secondly, the raw water environment parameter set and ion component parameter set are transmitted to the edge computing module via high-speed Ethernet. The edge computing module has a built-in seawater component identification model and separation process prediction model. Based on the input data, it generates component feature labels and predicts and recommends the optimal separation path. The prediction results are output in the form of a topology map, including the process path structure, functional module sequence and control parameter settings. The unified format is XML data, which is called by the central control system and configuration instructions are issued.

[0214] Next, the confined nanofiltration module is activated to pretreat the seawater. The module uses polyamide nanofiltration membrane units with a pore size of 0.001μm~0.01μm to retain polymeric impurities and di / trivalent ions. The central control system monitors the pressure difference, flux and polarization state across the membrane in real time and dynamically adjusts the operating parameters according to the membrane module status. After nanofiltration, the seawater sample enters the multi-stage microbubble dissolved air flotation unit. Three sets of series-connected flotation modules use ozone-air mixed microbubbles to remove non-polar impurities and colloidal particles, and improve the separation efficiency through surface adsorption mechanism.

[0215] Then, the treated seawater sample enters a two-stage temperature-controlled degassing chamber, where bromide is released at 25°C and hydrogen sulfide at 45°C. The escaping gas is recovered by a condenser. Subsequently, the sample enters an ion threshold separation unit, which includes a cation- and anion-selective permeable membrane. Multivalent ions are driven to migrate to the concentration side and collected by a 3V~5V DC electric field, while monovalent ions are dialyzed to the downstream channel, achieving a preliminary ion fractionation effect.

[0216] Next, the concentrated solution is guided into the electric field-induced chelation extraction module, which is filled with chelating particles with a particle size of 50nm~100nm. The surface of the material contains functional groups of polyethyleneimine and pyrrolidone. The module is equipped with three sets of electric field gradient cavities, and an electric field of 0.5V / cm to 3.0V / cm is applied to drive target ions such as lithium ions, rubidium ions and strontium ions to bind with the chelating particles. Subsequently, the gradient solvent extraction module selectively extracts neutral and weakly polar substances such as bromide ions and borate ions with a hexane-ethanol mixture to complete the further separation.

[0217] Finally, the concentrate that was not recovered by the upstream module enters the metal ion co-precipitation module. The pH is adjusted to 9.0-9.5 with alkali solution, and phosphate or oxalate is dynamically added to precipitate metal ions such as calcium and magnesium ions. The module outlet is equipped with an online concentration ratio calculation module, which calculates the concentration factor in real time in conjunction with a concentration sensor, flow meter and temperature compensation module. When the detected concentration factor is lower than the threshold, the central control system automatically executes the adjustment action, and updates the control parameters through a closed-loop feedback mechanism combined with a multi-objective genetic algorithm and reinforcement learning strategy to ensure that the system operates stably and efficiently under unattended conditions.

[0218] Beneficial effects:

[0219] This application constructs a highly efficient seawater component separation system based on intelligent regulation. The system realizes closed-loop control of the entire process, from raw seawater sample collection, multi-parameter water quality monitoring, ion concentration detection, component identification and path prediction, confined nanofiltration and air flotation impurity removal, temperature-controlled degassing and ion threshold separation, high-value ion enrichment and multi-path extraction, to real-time feedback adjustment. The system introduces a seawater component identification model and a separation process prediction model, combined with a multi-functional module dynamic activation mechanism and PID control algorithm, to achieve adaptive generation of separation paths and precise adjustment of process parameters. It solves the problems of poor component adaptability, high energy consumption and poor target analyte selectivity of traditional seawater resource extraction methods. The overall system not only significantly improves the extraction efficiency and selectivity of trace high-value ions, but also reduces operating energy consumption and human intervention. It has significant advantages such as modular structure, intelligent operation and wide application scenarios.

[0220] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0221] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0222] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0223] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0224] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A highly efficient seawater full-component separation system based on intelligent control, characterized in that, The separation system includes: The raw water acquisition module is used to collect seawater samples; collect the water quality parameters of the seawater samples and detect the concentration of representative ions in the seawater samples; identify the component characteristics of the seawater samples, and then generate a separation path topology map; The path identification module is used to configure the confined nanofiltration component and the multi-stage microbubble dissolved air flotation unit, configure the temperature-controlled degassing chamber and the ion threshold separation unit, and configure the online adjustment function module and feedback adjustment method. The process execution module is used to configure electric field-induced chelation extraction and gradient solvent extraction, as well as to configure metal ion co-precipitation and parameter self-tuning methods. The feedback optimization module is used to configure the full-process data integration platform and feedback optimization unit, dynamically generate operation instructions, and execute closed-loop control and strategy buffer methods. The product collection module is used to configure the component storage and zoned collection system, configure terminal desalination and pure water recovery, and configure residual liquid value assessment and waste liquid safe treatment. When the raw water acquisition module performs the step of identifying the component characteristics of the seawater sample and generating a separation path topology map, it performs the following steps: The raw water environmental parameter set and the raw water ionic component parameter set are transmitted to the edge computing module for processing via high-speed Ethernet; the edge computing module is equipped with a seawater component identification model and a separation process prediction model. The seawater component identification model is used to construct input variables including sodium ion concentration, potassium ion concentration, calcium ion concentration, magnesium ion concentration, lithium ion concentration, bromide ion concentration, borate ion concentration, strontium ion concentration, barium ion concentration, conductivity value and ion concentration ratio through a decision tree structure. The output results are component labels, and the label types include lithium-rich type, magnesium-rich type, high sodium type, high hardness type and multiple impurities type. The separation process prediction model is used to construct a matching structure based on path rules. It takes lithium-rich, magnesium-rich, high-sodium, high-hardness, and multi-impurity types as inputs and outputs complete process path structure data. After the seawater component identification model is run, a unique corresponding component feature label is generated. The separation process prediction model uses this label as the decision entry point, calls the built-in path module library, and performs the optimal path recommendation based on the current parameters. The result generated by the separation process prediction model is a structured separation path topology map, which includes: process path hierarchy, functional module configuration order, and set of control parameter initialization values. The process path hierarchy is the division of functional segments involved in the separation process. The functional segment division includes: pretreatment segment, ion fractionation segment, high-value enrichment segment, functional module configuration order, and the order of operation and logical relationship of each functional module. The functional modules include: an electro-driven ion migration module, a gradient solvent extraction module, and a chelation precipitation module. The set of initial control parameter values ​​includes: the target electric field strength value, membrane flux setting value, reaction residence time, extractant concentration setting value, flow rate setting value, and control cycle for each functional module. After generating the separation path topology map, it is output in standard XML control format and automatically transmitted to the central control system. The central control system initializes and configures each module in the separation system and prepares the start command based on the structure and parameters in the separation path topology map.

2. The seawater full-component high-efficiency separation system based on intelligent control according to claim 1, characterized in that, The raw water collection module is used to collect seawater samples through a seawater sampling system set in the target sea area.

3. The seawater full-component high-efficiency separation system based on intelligent control according to claim 2, characterized in that, The seawater sampling system includes a sampling arm, a sampling depth adjustment component, a sampling pump unit, and a seawater sample delivery pipeline; The sampling depth adjustment component is set to a sampling depth of 5 to 10 meters to obtain representative seawater samples. The sampling pump unit transports the seawater sample to the raw water sampling pool through the seawater sample delivery pipeline under constant negative pressure. The raw water sampling pool is made of polyvinylidene fluoride and has an integrated constant temperature control module. The constant temperature control module adopts a combination structure of temperature control board and heat exchange circulating water circuit, which is used to stably maintain the temperature of the seawater sample within the set temperature range. The raw water sampling tank is equipped with a multi-parameter water quality monitoring array, which includes various sensors, including a temperature sensor, a conductivity sensor, a pH sensor, a turbidity sensor, and a redox potential sensor. A temperature sensor is used to detect the temperature of the seawater sample; a conductivity sensor is used to detect the total concentration of dissolved ions in the seawater sample; a pH sensor is used to detect the acidity or alkalinity of the seawater sample solution; a turbidity sensor is used to detect the concentration of particulate matter in the seawater sample; and a redox potential sensor is used to reflect the oxidizing and reducing environmental states of the seawater sample. The temperature sensor, conductivity sensor, pH sensor, turbidity sensor, and redox potential sensor collect real-time data at a set frequency to form a set of raw water environmental parameters. The set of raw water environmental parameters includes the temperature, conductivity, pH, electrochemical turbidity, and redox potential of the seawater sample.

4. The seawater full-component high-efficiency separation system based on intelligent control according to claim 3, characterized in that, The raw water sampling pool is equipped with a seawater ion concentration detection module at its outlet. The seawater ion concentration detection module includes a micro ion-selective electrode array system and a plasma emission spectroscopy analysis subsystem. The micro ion-selective electrode array system is used to dynamically detect the mass concentration of sodium ions, potassium ions, calcium ions, magnesium ions, lithium ions, and bromide ions in the seawater sample; each ion is provided with an independent electrode channel, and the electrode head of the electrode channel adopts a solid film structure. The plasma emission spectroscopy analysis subsystem is used to analyze the mass concentrations of borate ions, strontium ions, and barium ions in the seawater sample; the plasma emission spectroscopy analysis subsystem employs a dual-channel sample introduction system and a CCD high-resolution detection module; The micro ion-selective electrode array system and the plasma emission spectroscopy analysis subsystem are used to collaboratively detect and obtain a set of raw water ion component parameters, which includes multiple ion mass concentration indicators.

5. The seawater full-component high-efficiency separation system based on intelligent control according to claim 4, characterized in that, When the path identification module executes the steps of configuring the confined nanofiltration component and the multi-stage microbubble dissolved air flotation unit, it performs the following steps: The seawater sample is first introduced into the confined nanofiltration assembly, which contains a nanofiltration membrane unit made of polyamide material, the pore size of which ranges from 0.001 μm to 0.01 μm. The nanofiltration membrane unit is used to retain organic particles and metal ions with divalent or trivalent charges in the seawater sample, with a particle size greater than 0.001 μm. The configuration of the nanofiltration membrane unit includes: a pre-membrane pressure sensor, a post-membrane pressure sensor, and a flow sensor. The central control system adjusts the speed of the water pump in front of the membrane module and the membrane operating pressure based on the pressure difference before and after the membrane, the flux of the nanofiltration membrane module, and the concentration polarization. The confined nanofiltration component performs bypass operation or skip configuration according to the limiting conditions in the separation path topology diagram; The seawater sample processed by the confined nanofiltration component is then introduced into the multi-stage microbubble dissolved air flotation unit. The multi-stage microbubble dissolved air flotation unit consists of three sets of flotation modules arranged in series. Each set of flotation modules includes: an ozone-air mixed microbubble injection device, a reaction chamber, and a separation tank. The ozone-air mixed microbubble injection device is used to generate ozone-air bubbles with a particle size of less than 20 μm. The injection pressure is controlled between 0.1 MPa and 0.3 MPa, and the ozone volume fraction is controlled within the range of 2% to 5%. Microbubbles adhere to the surfaces of colloidal particles and nonpolar organic molecules in the seawater sample through interfacial adsorption mechanisms, thereby promoting their rise and entry into the top gas-liquid separation zone. The impurity-carrying capacity of microbubbles is calculated as follows: ； in, It is the mass of impurities carried by microbubbles rising per unit time, with units of 1. , It is the first Microbubble number concentration, in units of , It is the first The surface area of ​​microbubbles, in units of , It is the first The average contact time between the microbubble-like object and the target material, in seconds; When the path identification module executes the step of configuring the temperature-controlled degassing chamber and the ion threshold separation unit, it performs the following steps: The seawater sample treated with microbubble flotation is introduced into a temperature-controlled degassing chamber. The temperature-controlled degassing chamber adopts a two-section structure design. The first section is set to a temperature of 25°C, and the second section is set to a temperature of 45°C, which release bromide and hydrogen sulfide respectively. The degassing process is controlled by a multi-point temperature-controlled resistor module and a thermistor array. The escaping gas is sent to the condensation and recovery device through the top gas duct, and the moisture and target components in the escaping gas are captured by the reflux condensation structure. The central control system adjusts the heating power and water flow rate based on the gas mass concentration feedback from the condensation recovery device, so that the volatile component removal rate reaches more than 90%. The seawater sample after temperature-controlled degassing is introduced into the ion threshold separation unit. The ion threshold separation unit adopts a double-layer anti-charge membrane structure, including a cation-selective permeable membrane and an anion-selective permeable membrane. Positive and negative electrode plates are respectively set on both sides of the membrane module, and a DC voltage is applied and controlled within the range of 3V to 5V to form a uniform electric field. Under the influence of an electric field, multivalent ions are preferentially driven to the concentration-side collection tank due to their large charge number and fast migration rate, while monovalent ions partially enter the downstream module through the dialysis-side release channel; wherein the multivalent ions include magnesium ions and calcium ions, and the monovalent ions include sodium ions and potassium ions. When the path identification module executes the steps of the configuration online adjustment function module and the feedback adjustment method, it performs the following steps: Each functional module is equipped with an online adjustment module at the liquid outlet. The online adjustment module integrates an ion-selective electrode detection submodule, a spectral detection submodule, and a flow detection submodule. The ion-selective electrode detection submodule is used to monitor changes in sodium ion concentration, calcium ion concentration, magnesium ion concentration and lithium ion concentration; the spectral detection submodule is used to analyze the degree of organic residue and color intensity in the solution; and the flow detection submodule is used to monitor the current outflow rate and internal pressure difference. The central control system adjusts the system based on the real-time feedback from the online adjustment module using a proportional-integral-derivative closed-loop control strategy. The PID controller control logic is as follows: ； in, It is a feedback control system in time Adjust the output value, It is time The systematic error at any given time represents the difference between the set value and the measured value. It is the proportional adjustment coefficient, reflecting the response strength of the current error. It is an integral adjustment factor used to eliminate long-term accumulated errors. It is the differential adjustment coefficient, used to suppress overshoot caused by an excessively fast system response. It is an integral variable used to integrate errors over a past time period. The central control system dynamically selects to skip or activate target functional modules before operation based on the module activation field in the separation path topology diagram. If the component characteristic label of the seawater sample is high sodium type, the central control system skips the multi-stage microbubble dissolved air flotation unit and the temperature-controlled degassing chamber. If the component characteristic label is multi-impurity type, the central control system prioritizes the activation of the confined nanofiltration component and the multi-stage microbubble dissolved air flotation unit.

6. The seawater full-component high-efficiency separation system based on intelligent control according to claim 5, characterized in that, When the process execution module performs the steps of configured electric field-induced chelation extraction and gradient solvent extraction, it performs the following steps: The concentrated solution of the seawater sample is first introduced into the electric field-induced chelation extraction module, which is filled with nanoscale chelating particles modified with functional groups of polyethyleneimine and pyrrolidone. The particle size of the chelating particles is in the range of 50nm to 100nm and they are uniformly distributed in three parallel electric field gradient cavities. Each of the electric field gradient cavities is provided with parallel electrode plates for applying DC voltage to form an electric field gradient of 0.5V / cm to 3.0V / cm. The electric field induces target metal ions to migrate towards the direction of high electric field strength and to undergo a specific binding reaction with functional sites in the chelating particle material. The target metal ions include lithium ions, rubidium ions, and strontium ions. The outlet of the electric field-induced chelation extraction module is equipped with a gradient solvent extraction module. The gradient solvent extraction module includes an organic phase addition unit, a static mixing reaction chamber, and a phase separation channel. The solvent system consists of n-hexane and ethanol mixed at a volume ratio of 3:

1. The target chelating agent is sulfosalicylic acid, and the concentration is set to 0.05 mol / L. The hexane-ethanol mixed solvent is injected into the reaction chamber at a flow rate of 0.1 mL / min via a high-precision peristaltic pump, and fully contacts the concentrated liquid entering the module at the phase interface. Bromine ions, borate ions and neutral or weakly polar trace substances preferentially enter the organic phase to achieve selective extraction. When the process execution module performs the steps of configuring the metal ion co-precipitation and parameter self-tuning method, it performs the following steps: The remaining concentrate that was not recovered by the electric field-induced chelation extraction module or the gradient solvent extraction module is fed into the metal ion coprecipitation module. The metal ion coprecipitation module is equipped with a pH adjustment unit and a dynamic dosing unit. The pH adjustment unit adjusts the pH value of the solution to the range of 9.0 to 9.5 through an alkaline precision control pump. Using a dynamic dosing unit, phosphate or oxalate ions at a concentration of 0.05 mol / L are added based on detection feedback to co-precipitate with the target metal ions. The precipitation reaction chamber is equipped with a swirling structure to control the particle size of the precipitate to be above 5μm, and the precipitate is discharged through the slag discharge channel. An online concentration ratio calculation module is configured at the liquid outlet of the electric field-induced chelation extraction module, the gradient solvent extraction module, and the metal ion co-precipitation module, respectively. The online concentration ratio calculation module integrates a concentration sensor, a flow meter, and a temperature compensation module to collect liquid outlet concentration data. When the concentration factor is lower than the minimum threshold set in the separation path topology diagram, the central control system automatically performs adjustment operations, including extending the chelation residence time, reducing the extraction flow rate, increasing the dosing concentration, or adjusting the pH value. The central control system determines the enrichment path configuration scheme for the current operating cycle based on the component feature labels output by the seawater component identification model and the module activation fields in the separation path topology diagram. The enrichment path configuration scheme includes: when the component feature label is lithium-rich, the central control system only activates the electric field-induced chelation extraction module; when the component feature label is multi-impurity, the central control system activates the gradient solvent extraction module and the metal ion co-precipitation module; when the target label involves both high magnesium and high boron types, the three modules are activated synchronously and run in parallel.

7. The seawater full-component high-efficiency separation system based on intelligent control according to claim 6, characterized in that, When the feedback optimization module executes the steps of configuring the full-process data integration platform and the feedback optimization unit, it performs the following steps: A full-process data integration platform is configured in the central control system to collect and process data in a unified manner, and to integrate key parameter data during the operation of the seawater full-component separation system. The full-process data integration platform includes: a sensor fusion center, a feedback optimization core module, and an operation command generator. The sensor fusion center is equipped with a high-speed data acquisition channel, which connects to all deployed sensor nodes in the system. The parameters acquired include: raw water temperature, conductivity, pH value, oxidation-reduction potential, pressure difference before and after the membrane module, instantaneous energy consumption of the module, change in target ion concentration, and effluent flow rate. The fusion center integrates a data cleaning subroutine based on mean filtering and wavelet denoising algorithm, and performs Z-score standardization operation to generate a set of structured process monitoring parameters. The feedback optimization core module is loaded into the end-to-end data integration platform. The feedback optimization core module is composed of a multi-objective genetic algorithm sub-module and an intelligent control agent based on deep reinforcement learning strategy. The multi-objective genetic algorithm submodule uses the objectives of maximizing the recovery rate of the target ion, minimizing the energy consumption per unit of processing, and minimizing the residual concentration of the end reaction as the objective functions. Input variables include: membrane module operating pressure, electric field-induced voltage, chelating agent concentration, solvent injection rate, and module running time. The output is a parameter control vector set, specifically: ； in, It is the first The comprehensive optimization objective function value of each control scheme is used to evaluate the overall merits of the scheme. The recovery rate of the target component The weighting coefficients reflect the degree of preference for the recovery rate indicator. It is the first The target component recovery rate under each control scheme Energy consumption per unit of output The weighting coefficients reflect the degree of importance attached to energy-saving indicators. It is the first The energy consumed per unit mass of the target component under each control scheme, in units of , Residual concentration The weighting coefficients reflect the degree of requirement for the purity of the effluent. It is the first The residual concentration of the target component in the output liquid under each control scheme, in units of ; The deep reinforcement learning agent employs a proximal policy optimization algorithm, taking historical running data as the input of the environmental state and the combination of regulatory actions as the output. The regulatory policy is updated based on the real-time reward value fed back by the system. The agent's optimization objectives include: maximizing long-term reward and minimizing policy perturbation. When the feedback optimization module executes the dynamically generated operation instructions and performs the closed-loop control and policy buffer method steps, it performs the following steps: The optimal parameter solution output by the feedback optimization core module is transmitted to the operation instruction generator. The operation instruction generator parses the optimization result into a system-executable instruction format and automatically completes the following tasks: adjusting the in-membrane pressure and operating pressure difference of the confined nanofiltration module, controlling the voltage value and electric field gradient of the electric field induced chelation extraction module, optimizing the solvent injection flow rate and chelating agent concentration of the gradient solvent extraction module, correcting the dosing rate and pH setpoint of the metal ion coprecipitation module, and dynamically setting the operating time window and control cycle of each module.

8. The seawater full-component high-efficiency separation system based on intelligent control according to claim 7, characterized in that, When the product collection module executes the configuration component storage and zoned collection system, and configures the terminal desalination and pure water recovery steps, it performs the following steps: After completing the selective extraction of high-value trace components, the product streams of various target components are collected in precise zones according to ion type, and a target component storage and zone collection system is constructed. The storage and zone collection system includes lithium ion storage tanks, potassium ion storage tanks, magnesium ion storage tanks and borate ion storage tanks. Each storage tank is composed of a liquid level monitoring module, an ion concentration monitoring module, a temperature stabilization module and a liquid inlet flow rate adjustment module. The liquid level monitoring module uses an ultrasonic level gauge for real-time volume detection with an accuracy of ±1%. The ion concentration monitoring module integrates a miniature ion selective electrode sensor to continuously detect the mass concentration of the corresponding target ion in the storage tank. The temperature stabilization module uses a constant temperature water jacket and a PTC heating element for linkage control to keep the storage liquid temperature stable at 25℃±1℃. The liquid inlet flow rate adjustment module adjusts the inlet speed of the target product liquid through a proportional solenoid valve group. The remaining product stream after the extraction of all target components is guided into the terminal desalination and pure water recovery module, which consists of a three-stage reverse osmosis membrane treatment unit, a hollow fiber nanofiltration membrane filtration unit, and a pH adjustment and water quality stabilization unit. The system utilizes a three-stage reverse osmosis membrane treatment unit to remove residual dissolved ions, small molecule organic matter, and trace impurities; a hollow fiber nanofiltration membrane filtration unit to further remove large molecule residual organic matter; and a pH adjustment and water quality stabilization unit to adjust the pH value of the output water to 6.5~7.

0. When the product collection module performs the steps of residual liquid value assessment and waste liquid safe treatment, it performs the following steps: The remaining liquid that is not collected into the target component storage tank is uniformly entered into the residual liquid value assessment and intelligent classification judgment module. The residual liquid value assessment and intelligent classification judgment module consists of a high-value ion concentration detection subunit, a total organic carbon analysis subunit, and a pH stability judgment subunit. The high-value ion concentration detection subunit is used to detect the residual concentration of key high-value ions such as lithium ions, magnesium ions and borate ions in the residual liquid; the total organic carbon analysis subunit monitors the concentration of organic pollutants in the residual liquid through a TOC analyzer; and the pH stability judgment subunit confirms whether the liquid has adjustable properties and is suitable for subsequent reflux treatment. The residual liquid is classified according to the judgment logic by the central control system. The judgment logic includes: when the concentration of any high value-added ion exceeds the corresponding reflux threshold, the residual liquid is switched to the reflux path through the automatic control valve and reintroduced into the confined nanofiltration component or the electric field induced chelation extraction module. When the TOC concentration is higher than the preset threshold and the target ion concentration is lower than the minimum economic recovery concentration, the residual liquid is guided to the waste liquid treatment path. All residual liquids that are confirmed to be non-recyclable are uniformly guided to the waste liquid safety treatment and neutralization discharge module. The waste liquid safety treatment and neutralization discharge module includes: a solid-liquid separation unit, a two-step neutralization reaction unit, a pre-discharge detection unit, and an intelligent discharge control module. The solid-liquid separation unit uses a tubular high-speed centrifuge to physically separate suspended particles and dissolved sediments in the residual liquid. The two-step neutralization reaction unit first adjusts the pH value to 6.5~7.0 through acid-base titration, and then introduces ferric chloride to form ferric hydroxide precipitate to capture residual heavy metal ions. The pre-discharge detection unit detects pH, conductivity, heavy metal ion concentration and TOC value. The intelligent discharge control module automatically opens the liquid outlet channel after receiving a system confirmation signal before discharge. The system's resource utilization rate assessment module is used to comprehensively analyze the target component production efficiency and resource utilization throughout the separation process. The system's resource utilization rate assessment module collects the following data: the liquid volume and mass concentration of the target component produced per unit of time per 24 hours, the total power consumption and total operating energy consumption of the system, the target component recovery rate, and the proportion of waste liquid discharged per ton of raw seawater.

9. A highly efficient seawater full-component separation system based on intelligent control according to any one of claims 1 to 8, characterized in that, The components separated from the seawater sample using the separation system are used in post-processing scenarios.

Images