Rare earth mother liquor high-efficiency recovery method and system based on surrounding water quality monitoring

Abstract

The application relates to the technical field of water quality monitoring, analysis and treatment, and discloses a rare earth mother liquor efficient recovery method and system based on surrounding water quality monitoring. The method comprises the following steps: performing component analysis on the rare earth mother liquor, and establishing a surrounding water quality monitoring network; based on the components and water quality data, the mother liquor is divided into three types A, B and C; type A is subjected to magnesium bicarbonate precipitation, type B is mixed with filtrate and subjected to gradient extraction through a DOAM-PPA extractant, and type C is separated through ion exchange resin, and finally, high-purity rare earth products are prepared through pH grading elution crystallization. The application solves the problems of low resource utilization efficiency caused by the lack of differentiated processing in the traditional technology and the technical problem that the existing process and environmental monitoring lack a linkage mechanism and cannot realize dynamic optimization of process parameters.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of water quality monitoring analysis and processing, and particularly relates to a rare earth mother liquor efficient recovery method and system based on surrounding water quality monitoring. BACKGROUND

[0002] Rare earth elements have a wide range of applications in modern industrial production and high-tech fields due to their unique optical, electronic and magnetic properties. The mother liquor produced in the rare earth separation process contains a large amount of rare earth elements and other valuable metals, and is an important secondary resource. At present, the treatment methods of rare earth mother liquor mainly include chemical precipitation method, extraction method, ion exchange method, etc. These methods recover and utilize the rare earth elements in the mother liquor through different physical and chemical principles.

[0003] However, the existing rare earth mother liquor recovery technology has obvious deficiencies. Although the traditional chemical precipitation method is simple to operate, it has low recovery rate and is difficult to achieve complete recovery of rare earth elements; the conventional extraction method has the advantage of high separation efficiency, but emulsified phases are easily formed in the extraction process, and the extraction agent is consumed in large quantities; the ion exchange method has good selectivity for rare earth elements, but the treatment cycle is long and the cost of resin regeneration is high. More seriously, the existing technology generally has high energy consumption, large reagent consumption, and serious secondary pollution, and lacks an effective linkage mechanism between the recovery process and environmental impact, resulting in serious pollution of the surrounding water environment during the treatment of rare earth mother liquor.

[0004] With the increasing strictness of environmental protection standards, the technical route that only focuses on the recovery rate of rare earth elements and ignores the environmental impact cannot meet the requirements of sustainable development. How to link the surrounding water quality monitoring data with the real-time recovery process parameters of rare earth mother liquor to achieve a balance between efficient recovery and environmental protection has become a technical problem to be solved. In addition, the existing technology lacks differentiated treatment strategies for rare earth mother liquors of different properties, resulting in low resource utilization efficiency; and lacks an adaptive regulation mechanism based on water environmental capacity, which cannot dynamically optimize the process parameters according to the environmental carrying capacity. Therefore, it is of great significance to develop a rare earth mother liquor efficient recovery method that can intelligently classify, differentially process and real-time regulate based on water quality monitoring data. SUMMARY

[0005] The present application provides a rare earth mother liquor efficient recovery method and system based on surrounding water quality monitoring, which solves the problems of low resource utilization efficiency caused by lack of differentiated processing in traditional technology, and the technical problem that the existing process and environmental monitoring lack a linkage mechanism and cannot dynamically optimize the process parameters.

[0006] In a first aspect, the application provides a rare earth mother liquor efficient recovery method based on surrounding water quality monitoring, which comprises: performing inductively coupled plasma mass spectrometry analysis on a rare earth mother liquor, simultaneously establishing a water quality monitoring network around a factory site to obtain mother liquor composition data and water quality baseline data; based on the mother liquor composition data and the water quality baseline data, dividing the rare earth mother liquor into A-type mother liquor, B-type mother liquor and C-type mother liquor, and adopting different processing routes for different types; adding magnesium bicarbonate carbonate to the A-type mother liquor for precipitation, mixing the B-type mother liquor with filtrate and then performing gradient extraction by using DOAM-PPA extractant, and dynamically adjusting process parameters according to water quality data; mixing extractive enrichment liquid with the C-type mother liquor and then separating by using a D151 and D301 ion exchange resin system, and crystallizing to obtain high-purity rare earth products after pH fractional elution.

[0007] In a second aspect, the application provides a rare earth mother liquor efficient recovery system based on surrounding water quality monitoring, which comprises:

[0008] an analysis module configured to perform inductively coupled plasma mass spectrometry analysis on a rare earth mother liquor, simultaneously establish a water quality monitoring network around a factory site to obtain mother liquor composition data and water quality baseline data;

[0009] a processing module configured to divide the rare earth mother liquor into A-type mother liquor, B-type mother liquor and C-type mother liquor based on the mother liquor composition data and the water quality baseline data, and adopt different processing routes for different types;

[0010] an extraction module configured to add magnesium bicarbonate carbonate to the A-type mother liquor for precipitation, mix the B-type mother liquor with filtrate and then perform gradient extraction by using DOAM-PPA extractant, and dynamically adjust process parameters according to water quality data;

[0011] a separation module configured to mix extractive enrichment liquid with the C-type mother liquor and then separate by using a D151 and D301 ion exchange resin system, and crystallize to obtain high-purity rare earth products after pH fractional elution.

[0012] In a third aspect, a rare earth mother liquor efficient recovery device based on surrounding water quality monitoring is provided, which comprises: a memory and at least one processor, the memory having instructions stored therein; the at least one processor invokes the instructions in the memory to enable the rare earth mother liquor efficient recovery device based on surrounding water quality monitoring to perform the above-mentioned rare earth mother liquor efficient recovery method based on surrounding water quality monitoring.

[0013] In a fourth aspect, a computer readable storage medium is provided, the computer readable storage medium having instructions stored therein, which, when executed on a computer, enable the computer to perform the above-mentioned rare earth mother liquor efficient recovery method based on surrounding water quality monitoring.

[0014] The technical scheme provided in the application realizes accurate monitoring and analysis of mother liquor components and surrounding water quality through inductively coupled plasma mass spectrometry analysis and water quality monitoring network establishment, and provides a data basis for intelligent classification and differential treatment of rare earth mother liquor. By using mother liquor component data and water quality baseline data, the rare earth mother liquor is divided into A, B and C classes, and a differential treatment route is adopted, which significantly improves the resource utilization efficiency and avoids the problem of low treatment efficiency caused by traditional single process. The self-adaptive K-means clustering algorithm and hierarchical decision tree algorithm are used for mother liquor classification, fully considering the multi-dimensional characteristics of rare earth element content, acidity and impurity content, so that the classification result is more scientific and reasonable. The introduction of artificial intelligence algorithms such as bidirectional long short-term memory network, wavelet transform and convolutional neural network realizes deep mining and feature extraction of water quality data. The contribution of these algorithms in specific functional applications is reflected in the effective processing ability of time series data and the accurate capture of spatial distribution characteristics. The neural network time series model can identify the periodic fluctuation characteristics and trends in water quality data, providing decision basis for dynamic adjustment of process parameters. The A-class mother liquor adopts magnesium bicarbonate precipitation process, the B-class mother liquor is mixed with filtrate and then gradient extraction is carried out by DOAM-PPA extractant, and the C-class mother liquor adopts ion exchange process. This targeted treatment strategy fully utilizes the advantages of each process and improves the overall recovery efficiency. In the magnesium bicarbonate precipitation process, accurate temperature control and accurate calculation of the amount of magnesium bicarbonate ensure the completeness of the precipitation reaction, and at the same time realizes the recycling of magnesium salt and CO2, significantly reducing production cost and environmental load; the gradient extraction adopts a three-stage process of DOAM-PPA extractant, which realizes efficient separation of rare earth elements under different pH values and temperature conditions, and has higher separation coefficient and saturation capacity than traditional P507 extractant; the ion exchange system combines the characteristics of D151 and D301 resins to realize effective treatment of low-concentration rare earth mother liquor.

[0015] The application realizes real-time linkage adjustment of water quality monitoring data and process parameters, dynamically optimizes extraction process parameters according to the change of rare earth ion content in water quality data, and avoids waste of reagents and environmental pollution caused by excessive extraction. The pH gradient elution technology realizes fine separation of rare earth elements, and the crystallization and calcination process ensures the high purity of the final product. The introduction of graph neural network takes the monitoring station as a node and the water flow direction as an edge, and simulates the migration and diffusion process of pollutants through the message passing mechanism. This innovative application realizes accurate evaluation and prediction of water environment capacity, and provides a scientific basis for environmentally friendly adjustment of process parameters. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort on the basis of these drawings.

[0017] Figure 1 An embodiment schematic diagram of the rare earth mother liquor efficient recovery method based on surrounding water quality monitoring in the present application;

[0018] Figure 2 An embodiment schematic diagram of the rare earth mother liquor efficient recovery system based on surrounding water quality monitoring in the present application;

[0019] Figure 3 An embodiment schematic diagram of the rare earth mother liquor efficient recovery system based on surrounding water quality monitoring in the present application; DETAILED DESCRIPTION

[0020] The present application provides a rare earth mother liquor efficient recovery method and system based on surrounding water quality monitoring. The terms "first", "second", "third", "fourth" and the like (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily mean a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the term "comprising" or "having" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device comprising a series of steps or units does not necessarily limit to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0021] For the sake of understanding, the specific process of the embodiments of the present application will be described below. Please refer to Figure 1 An embodiment of the rare earth mother liquor efficient recovery method based on surrounding water quality monitoring in the present application includes:

[0022] Step S101, inductively coupled plasma mass spectrometry analysis is performed on the rare earth mother liquor, and a water quality monitoring network is established around the factory area to obtain mother liquor composition data and water quality baseline data;

[0023] Step S102, based on the mother liquor composition data and the water quality baseline data, the rare earth mother liquor is divided into A-type mother liquor, B-type mother liquor and C-type mother liquor, and different types of processing routes are adopted;

[0024] Step S103, add magnesium bicarbonate to the A-type mother liquor for precipitation, mix the B-type mother liquor with the filtrate, and then perform gradient extraction by using a DOAM-PPA extractant, and dynamically adjust process parameters according to water quality data;

[0025] Step S104, mix the extraction enrichment liquid with the C-type mother liquor, and then separate the mixture by using a D151 and D301 ion exchange resin system, and then crystallize to obtain a high-purity rare earth product after pH fractionation elution.

[0026] It can be understood that the execution subject of the present application can be a rare earth mother liquor efficient recovery system based on surrounding water quality monitoring, and can also be a terminal or a server, and the specific implementation is not limited herein. The server is taken as an example for description in the embodiments of the present application.

[0027] Specifically, before the rare earth mother liquor recovery treatment, a multi-element analysis is performed on the mother liquor by using an inductively coupled plasma mass spectrometry, the mother liquor sample is pretreated by using a multi-spectral excitation sequence, and the content of each element is obtained by using a step ionization scanning. Meanwhile, an adaptive monitoring station array is established at key hydrological nodes upstream and downstream of the plant, and a dual-wavelength fluorescence spectrum technology is used to monitor the content change of rare earth elements in the water body in real time. The obtained multi-element spectrum distribution data is subjected to noise reduction processing by using a self-correcting multivariate analysis algorithm, background interference and random noise are eliminated, and the data reliability is improved. The rare earth element content dynamic distribution curve is analyzed by using a neural network time sequence model, the model includes a bidirectional long short-term memory network, a wavelet transform multi-scale decomposition, and a convolutional neural network, and finally a water environment capacity digital twin model is constructed to realize the prediction of water quality change.

[0028] Based on the obtained mother liquor composition data and water quality baseline data, the rare earth mother liquor is classified and treated. Specifically, a multivariate clustering analysis system is used, a three-dimensional feature of rare earth content, acidity and impurity content is spatially divided by using a self-adaptive K-means clustering algorithm, and a feature distribution map is formed. The map is subjected to boundary optimization by using a hierarchical decision tree algorithm, a classification standard is set in combination with an environmental capacity threshold in the water quality baseline data, a mother liquor classification decision model is established. Real-time intelligent classification is performed by using the model, the rare earth mother liquor is divided into A-type (high rare earth and low impurity), B-type (medium rare earth and medium impurity), and C-type (low rare earth and high impurity), and physical diversion is realized by using an intelligent control pipeline system, and different treatment routes are allocated.

[0029] Different processing methods are adopted for different categories of mother liquor. The A-type mother liquor is subjected to constant temperature control in a double-layer structure reaction kettle, the temperature is accurately adjusted to 40±1°C, a calculated amount of magnesium bicarbonate solution which is 1.8 times the molar number of rare earth ions is added for precipitation reaction, and rare earth carbonate precipitate, filtrate and recyclable CO2 gas are obtained, realizing the recycling of magnesium salt. The filtrate is mixed with the B-type mother liquor, the pH value is adjusted to 1.5±0.1, and after being treated with an oxidizing agent, it is introduced into a three-stage countercurrent extraction device, and DOAM-PPA extractant is used as the extractant to form a gradient extraction environment. The extractant has a higher separation coefficient and saturation capacity. According to the real-time changes of the rare earth ion content in the water quality data, the extraction process parameters, including extraction time, phase ratio and pH value, are dynamically adjusted through a multivariate adaptive control system.

[0030] The extraction and enrichment liquid is mixed with the C-type mother liquor in a specific ratio, filtered through a filter after adding a reducing agent for pretreatment, and a clear and uniform ion exchange feed liquid is obtained. The feed liquid is introduced into a series system composed of D151 strong acid cation exchange resin and D301 weak base anion exchange resin, and is subjected to exchange adsorption. The saturated resin bed is subjected to pH gradient elution technology, and different pH value acidic solutions are used for fractional elution in turn, and rare earth enrichment liquids of different compositions are obtained. These enrichment liquids are divided into light rare earth group, medium rare earth group and heavy rare earth group, and are subjected to differential temperature control crystallization treatment, and high-purity rare earth products are prepared through subsequent calcination conversion.

[0031] In a specific embodiment, the process of performing step S101 can specifically include the following steps:

[0032] The rare earth mother liquor is pretreated by a multi-spectral excitation sequence, and the pretreated mother liquor is subjected to stepwise ionization scanning by an inductively coupled plasma mass spectrometer to obtain multi-element spectral distribution data;

[0033] Self-adaptive monitoring station arrays are set at upstream and downstream key hydrological nodes in the factory area, and high-frequency dual-wavelength fluorescence spectrum real-time monitoring is performed on the water body to obtain a dynamic distribution curve of rare earth element content;

[0034] The multi-element spectral distribution data is subjected to noise reduction processing by a self-correcting multivariate analysis algorithm, and is quantitatively converted in combination with a standard substance calibration curve to obtain mother liquor composition data;

[0035] The dynamic distribution curve of rare earth element content is analyzed and processed by a neural network time series model, the periodic fluctuation characteristics and trends are identified, a water environment capacity digital twin model is constructed, and water quality baseline data with prediction function are obtained.

[0036] Specifically, before the rare earth mother liquor recovery treatment, multi-element analysis is performed on the mother liquor by inductively coupled plasma mass spectrometry. This method uses a multi-spectral excitation sequence to pretreat the mother liquor sample and obtains the content of each element through a step ionization scan. At the same time, an adaptive monitoring station array is established at the key hydrological nodes upstream and downstream of the plant, and a dual-wavelength fluorescence spectrum technology is used to monitor the changes in the content of rare earth elements in the water body in real time. The multi-element spectral distribution data obtained are processed by a self-correcting multivariate analysis algorithm to eliminate background interference and random noise and improve data reliability. The dynamic distribution curve of the rare earth element content is analyzed by a neural network time series model, which includes a bidirectional long short-term memory network, a wavelet transform multi-scale decomposition, and a convolutional neural network. Finally, a water environment capacity digital twin model is built to realize the prediction of water quality changes.

[0037] Based on the obtained mother liquor composition data and water quality baseline data, the rare earth mother liquor is classified and processed. Specifically, a multivariate clustering analysis system is used to divide the three-dimensional characteristics of rare earth content, acidity, and impurity content in space by an adaptive K-means clustering algorithm to form a characteristic distribution map. The map is optimized by a hierarchical decision tree algorithm, and a classification standard is set in combination with the environmental capacity threshold in the water quality baseline data to establish a mother liquor classification decision model. Through this model, real-time intelligent classification is performed to divide the rare earth mother liquor into three categories: high-rare-earth low-impurity A, medium-rare-earth medium-impurity B, and low-rare-earth high-impurity C. Physical diversion is realized through an intelligent control pipeline system to distribute different processing routes.

[0038] Different processing methods are used for different categories of mother liquor. The A-type mother liquor is subjected to constant temperature control in a double-layer structure reaction kettle, the temperature is accurately adjusted to 40±1℃, and a calculated amount of magnesium bicarbonate solution 1.8 times the molar number of rare earth ions is added for precipitation reaction to obtain rare earth carbonate precipitate, filtrate, and recoverable CO2 gas, realizing the recycling of magnesium salt. The filtrate is mixed with B-type mother liquor, the pH value is adjusted to 1.5±0.1, and after treatment with an oxidizing agent, it is introduced into a three-stage countercurrent extraction device. DOAM-PPA extractant is used as the extractant to form a gradient extraction environment. This extractant has a wider pH operating window and higher separation selectivity. According to the real-time changes in the content of rare earth ions in the water quality data, the extraction process parameters, including extraction time, phase ratio, and pH value, are dynamically adjusted by a multivariate adaptive control system.

[0039] Finally, the extraction enrichment liquid and the C-type mother liquor are mixed in a specific ratio, a reducing agent is added, and then filtered through a filter to obtain a clear and uniform ion exchange feed liquid. The feed liquid is introduced into a series system composed of D151 strong acid cation exchange resin and D301 weak base anion exchange resin for exchange adsorption. The saturated resin bed is eluted using acidic solutions with different pH values in a stepwise manner to obtain rare earth enrichment liquids with different compositions. These enrichment liquids are divided into light rare earth group, medium rare earth group and heavy rare earth group, which are subjected to differential temperature control crystallization treatment, and high-purity rare earth products are prepared through subsequent calcination conversion. The entire process realizes efficient recycling of magnesium salt and CO2 through the magnesium bicarbonate method, eliminates the environmental pollution problem of traditional processes, and significantly improves the separation efficiency of heavy rare earth elements through the application of DOAM-PPA extractant.

[0040] In a specific embodiment, the process of performing step-by-step analysis and processing of the rare earth element content dynamic distribution curve through a neural network time series model can specifically include the following steps:

[0041] The rare earth element content dynamic distribution curve is input into a bidirectional long short-term memory network for time series feature extraction. The bidirectional long short-term memory network contains a three-layer hidden layer structure, and the number of memory units in each layer is set to twice the number of rare earth element species. The time series encoding feature vector is obtained;

[0042] Wavelet transform is applied to the time series encoding feature vector for multi-scale decomposition to extract wave characteristics in different frequency domains. Each frequency component is assigned a weight through a self-attention mechanism to obtain a multi-frequency domain weighted feature matrix;

[0043] The multi-frequency domain weighted feature matrix is input into a convolutional neural network for spatial feature learning. The convolutional neural network is composed of three convolutional layers and two pooling layers in an alternating manner, with convolution kernel sizes of 3x3, 5x5 and 7x7. The water quality spatial distribution pattern feature is obtained;

[0044] Based on the water quality spatial distribution pattern feature, a graph neural network model is constructed. The monitoring stations are used as nodes, and the water flow direction is used as edges. The message passing mechanism is used to simulate the migration and diffusion process of pollutants to obtain water quality baseline data with prediction function.

[0045] Specifically, the dynamic distribution curve of rare earth element content collected from the upstream and downstream monitoring station array of the plant is input into a bidirectional long short-term memory network (BiLSTM) for time series feature extraction. The bidirectionality of the BiLSTM network reflects the consideration of information at historical and future time points at the same time, and the time series features are abstracted layer by layer through a three-layer hidden layer structure. For the detected 14 rare earth elements (including lanthanum, cerium, neodymium, praseodymium, etc.), 28 memory units (i.e., twice the number of rare earth element types) are set at each layer to ensure that sufficient time series information is captured. Each memory unit of the BiLSTM contains three components: an input gate, a forget gate, and an output gate, which realize long-time dependency learning through selective memory and forgetting, and finally output a time series encoding feature vector. The obtained time series encoding feature vector is applied to wavelet transform for multi-scale decomposition, which decomposes the time series signal into wave features in different frequency domains. Compared with the traditional Fourier transform, the wavelet transform can provide both time domain and frequency domain information, and is more suitable for analyzing non-stationary signals. In specific operations, the Daubechies wavelet basis function is selected for 5-level decomposition to obtain approximate coefficients and detail coefficients in different frequency bands. Then, the self-attention mechanism is used to assign weights to each frequency component. The self-attention mechanism calculates the correlation scores between each frequency component, and the frequency components with high correlation obtain higher weights, thereby forming a multi-frequency domain weighted feature matrix.

[0046] The multi-frequency domain weighted feature matrix is input into a convolutional neural network (CNN) for spatial feature learning. The CNN is composed of three convolutional layers and two pooling layers alternately. The first convolutional layer uses a 3x3 convolutional kernel to extract local fine features, the second convolutional layer uses a 5x5 convolutional kernel to capture medium-scale features, and the third convolutional layer uses a 7x7 convolutional kernel to obtain large-scale spatial patterns. The pooling layer uses the maximum pooling operation to reduce the dimension of the feature map and retain significant features. Through this progressive feature extraction, the CNN outputs water quality spatial distribution pattern features, reflecting the spatial correlation between different monitoring points.

[0047] Based on the water quality spatial distribution pattern features, a graph neural network (GNN) model is constructed, with monitoring sites as nodes and water flow direction as edges to form a water area topology. GNN simulates the migration and diffusion process of pollutants in the water system through a message passing mechanism, and the information transmission between nodes follows the water flow direction, with the upstream node state affecting the downstream node state. After multiple rounds of iteration and update, GNN generates water quality baseline data with prediction function, which can predict the changes of rare earth element content at different monitoring points under different working conditions.

[0048] In a rare earth separation plant application case, rare earth element content data for 30 consecutive days were collected from five monitoring points located 500 meters and 1000 meters upstream and 500 meters, 1000 meters and 2000 meters downstream of the plant, forming time-series curves for 14 rare earth elements. After processing with a BiLSTM network, the original approximately 10080-dimensional data (24 hours × 30 days × 14 elements) was compressed into a 392-dimensional time-series encoded feature vector. Wavelet transform decomposed the feature vector into sub-bands of five scales. A self-attention mechanism calculated the weights of low-frequency components (approximately 0.6), mid-frequency components (approximately 0.3), and high-frequency components (approximately 0.1), forming a multi-frequency domain weighted feature matrix. CNN processing yielded feature maps reflecting the spatial relationships between the five monitoring points. The final constructed GNN model achieved a rare earth element content prediction accuracy of over 90% for downstream monitoring points within a 24-hour prediction window, providing reliable data support for accurate classification of rare earth mother liquor and optimization of process parameters.

[0049] In one specific embodiment, the process of performing step S102 may specifically include the following steps:

[0050] The mother liquor composition data were input into a multivariate clustering analysis system, and the adaptive K-means clustering algorithm was used to spatially divide the three-dimensional features of rare earth element content, acidity and impurity content to obtain the feature distribution map of rare earth mother liquor.

[0051] A hierarchical decision tree algorithm was applied to the feature distribution map for boundary optimization, and a classification standard was set by combining the environmental capacity threshold in the water quality baseline data to obtain the mother liquor classification decision model.

[0052] The rare earth mother liquor is intelligently classified in real time using a mother liquor classification decision model. Based on the parameters of total rare earth content, acidity, and impurity content, it is automatically divided into three categories: high rare earth and low impurity (Class A), medium rare earth and medium impurity (Class B), and low rare earth and high impurity (Class C), resulting in the classified and labeled mother liquor stream.

[0053] The mother liquor streams after classification and labeling are physically diverted through an intelligent control pipeline system. A magnesium bicarbonate precipitation treatment route is assigned to the mother liquor of type A, a DOAM-PPA extraction treatment route is assigned to the mother liquor of type B, and an ion exchange treatment route is assigned to the mother liquor of type C, resulting in three differentiated treatment process routes.

[0054] Specifically, the mother liquor composition data obtained by inductively coupled plasma mass spectrometry is input into a multivariate clustering analysis system, which uses an adaptive K-means clustering algorithm to process the data. The adaptive K-means clustering algorithm is an improvement over the traditional K-means algorithm, which can automatically adjust the number of cluster centers according to the data distribution characteristics without manually specifying the K value. The algorithm takes the total rare earth content, acidity, and impurity content in the rare earth mother liquor as three-dimensional features, constructs a feature space, iteratively calculates the Euclidean distance from each sample point to the cluster center, and continuously adjusts the cluster center position to finally form a feature distribution map of the rare earth mother liquor. The obtained feature distribution map is optimized for boundaries using a hierarchical decision tree algorithm, which is a tree-based classification algorithm that selects the optimal partition attribute based on information gain or Gini impurity index and constructs decision rules layer by layer. In this method, the hierarchical decision tree first examines the total rare earth content, the main attribute, and then examines the acidity and impurity content in turn, forming a hierarchical classification structure. At the same time, the environmental capacity threshold in the water quality baseline data is combined during the construction of the decision tree to ensure that the classification result takes into account both the recovery value and the environmental impact, thereby obtaining an environmentally friendly mother liquor classification decision model.

[0055] The rare earth mother liquor is classified in real time by the above-mentioned mother liquor classification decision model, and according to the combined values of the total rare earth content, acidity, and impurity content, it is automatically divided into three categories: high rare earth low impurity A type with total rare earth content greater than 10 g / L, acidity less than 1 mol / L, and impurity content less than 5%; medium rare earth medium impurity B type with total rare earth content between 5-10 g / L, acidity between 1-3 mol / L, and impurity content between 5-15%; and low rare earth high impurity C type with total rare earth content less than 5 g / L, acidity greater than 3 mol / L, and impurity content greater than 15%. Through this multi-parameter combined classification method, the value and processing difficulty of the mother liquor are comprehensively evaluated, and the classified mother liquor stream is obtained.

[0056] The classified labeled mother liquor stream is physically split by an intelligent control pipeline system, which is composed of solenoid valves, flow meters and multi-channel distributors. According to the classification label signal, the valve opening and closing state is controlled, and the mother liquor of different categories is guided into the corresponding processing route. Specifically, the A-type mother liquor is allocated to the magnesium bicarbonate precipitation treatment route. The magnesium bicarbonate precipitant has the characteristics of strong environmental protection and recyclability, and is suitable for direct precipitation recovery of high rare earth content mother liquor; the B-type mother liquor is allocated to the DOAM-PPA extraction treatment route. DOAM-PPA is a new type of extractant, which has a wider pH operation window and higher separation selectivity, and has excellent extraction performance for medium concentration rare earth elements; the C-type mother liquor is allocated to the ion exchange treatment route. Ion exchange resin can effectively enrich rare earth elements under the condition of low concentration and high impurities. Through this differentiated treatment strategy, the optimal process path is adopted for mother liquor with different characteristics, realizing the recycling of magnesium salt and CO2, and improving the overall recovery efficiency and environmental performance.

[0057] In a specific embodiment, the process of performing step S103 can specifically include the following steps:

[0058] The A-type mother liquor is subjected to constant temperature control by a double-layer structure reaction kettle, the temperature is accurately adjusted to 40±1℃, and a calculated amount of magnesium bicarbonate solution which is 1.8 times the molar number of rare earth ions is added for precipitation reaction, to obtain rare earth carbonate precipitate, filtrate and recyclable CO2 gas;

[0059] The filtrate is mixed with the B-type mother liquor, and the pH value is adjusted to 1.5±0.1. An oxidizing agent is added to the mixed solution for pretreatment, to obtain a liquid with stable valence before extraction;

[0060] The liquid before extraction is introduced into a three-stage countercurrent extraction device, DOAM-PPA is used as the extractant, different pH values and temperature conditions are set in the first, second and third stages respectively, to form a gradient extraction environment, and an organic phase enriched with rare earth elements is obtained;

[0061] According to the real-time change of the rare earth ion content in the water quality data, the extraction process parameters are dynamically adjusted by a multivariate adaptive control system, including adjusting the extraction time, phase ratio and pH value, while recycling magnesium salt and CO2, to obtain an extraction enriched liquid and recycled auxiliary materials.

[0062] Specifically, the treatment is carried out by a double-layer structure reaction kettle, the inner layer of which is made of polytetrafluoroethylene material for corrosion resistance, the outer layer is made of stainless steel material to provide mechanical strength, and the middle interlayer is circulated with constant temperature water to accurately control the reaction temperature to 40±1℃. Temperature control is a key factor, too low will lead to slow reaction rate, too high will promote the formation of soluble complexes of part of light rare earth elements. When adding magnesium bicarbonate solution, the dosage is calculated according to 1.8 times of the molar number of rare earth ions, which takes into account the stoichiometric ratio of magnesium bicarbonate to rare earth elements and the consumption in the reaction process. The specific calculation method is first to obtain the concentration of each rare earth element in the mother liquor by inductively coupled plasma mass spectrometry, convert it into total moles, and then multiply by 1.8 to obtain the required moles of magnesium bicarbonate. After the magnesium bicarbonate precipitation reaction is completed, the rare earth carbonate precipitate, filtrate and recyclable CO2 gas are separated by a vacuum filtration device, realizing the recycling of magnesium salt. The filtrate still contains part of the unprecipitated rare earth elements, which is mixed with the B type mother liquor containing impurities in medium rare earth, and the pH value of the mixed solution is adjusted to 1.5±0.1 by adding acid or base in real time monitoring, which is the balance point of extraction efficiency and stability. Then, the mixed solution is pretreated by adding an oxidizing agent (such as H2O2 or NaClO) to ensure that the divalent iron is oxidized to trivalent iron and the trivalent cerium remains stable, avoiding the formation of precipitate or emulsion in the extraction process, and obtaining the stable extraction liquid before extraction.

[0063] The pretreated liquid before extraction is introduced into a three-stage countercurrent extraction device composed of 20 extraction units connected in series and divided into three functional sections. DOAM-PPA ((di-n-octyl amido methyl) phenyl phosphinic acid) is used as the extractant, which has excellent selectivity and higher extraction capacity for heavy rare earth elements, and wider pH operation window. The key of three-stage gradient extraction is to set different process parameters in different sections: the first section (units 1-8) is the rough extraction section, the pH value is controlled at 2.5±0.1, and the temperature is 40±1℃; the second section (units 9-16) is the fine extraction section, the pH value is 3.0±0.1, and the temperature is 45±1℃; the third section (units 17-20) is the washing section, which uses dilute nitric acid solution, the pH value is 1.5±0.1, and the temperature is 25±2℃. This gradient setting makes the rare earth elements gradually enriched in the organic phase during the extraction process, and the DOAM-PPA extractant can maintain high extraction performance in a wide pH range.

[0064] During the extraction process, the rare earth ion content data collected in real time by the plant water quality monitoring network is used as a feedback signal to input into a multivariate adaptive control system. Based on fuzzy logic control algorithm, the system dynamically adjusts the extraction process parameters according to the water quality data trend. The specific adjustment logic is as follows: when the rare earth content at the downstream monitoring point increases, the system automatically extends the extraction time or reduces the processing rate; when the rare earth ion content in the water phase at the outlet of the extraction unit exceeds the set threshold, the pH value or phase ratio (volume ratio of organic phase to water phase) of the corresponding unit is adjusted. The phase ratio adjustment range is usually 0.8:1 to 1.2:1, and the pH value adjustment amplitude is ±0.2 units. Through this closed-loop control strategy, the linkage between the extraction process and environmental monitoring is achieved, and the technical advantages of magnesium bicarbonate method and DOAM-PPA extractant are fully utilized, ensuring efficient recovery while minimizing environmental impact and achieving circular economy benefits.

[0065] In a specific embodiment, the process of performing step S104 can specifically include the following steps:

[0066] The extraction enrichment liquid is mixed with the C-type mother liquor at a specific ratio, a reducing agent is added to the mixed liquid for pretreatment, and the mixed liquid is filtered through a filter to obtain a clear and uniform ion exchange feed liquid;

[0067] The ion exchange feed liquid is introduced into a series system composed of D151 and D301 ion exchange resins, and exchange adsorption is performed at a suitable flow rate to obtain a saturated resin bed enriched with rare earth elements;

[0068] The saturated resin bed is subjected to pH gradient elution technology, and different pH value acidic solutions are used for fractional elution in sequence to obtain rare earth enrichment liquids of different compositions;

[0069] The rare earth enrichment liquids are divided into light rare earth group, medium rare earth group and heavy rare earth group, and are subjected to differential temperature control crystallization treatment, respectively, and high-purity rare earth products are obtained after subsequent calcination conversion.

[0070] Specifically, the enrichment liquid obtained by the previous extraction is mixed with the C-type mother liquor with low rare earth content and high impurities at a volume ratio of 2:1. This ratio is determined based on the characteristics of the C-type mother liquor with low rare earth content but large volume, which not only ensures the processing efficiency of the ion exchange system, but also fully utilizes the rare earth resources in the C-type mother liquor. A reducing agent Na2SO3 is added to the mixed liquid, and the addition amount is 1% of the volume of the mixed liquid. The main role is to reduce part of the rare earth elements to stable valence, especially to reduce tetravalent cerium to trivalent cerium, to avoid the formation of precipitates in the subsequent ion exchange process. The pretreated mixed liquid is filtered through a 0.22 μm membrane filter to remove insoluble impurities, and a clear and uniform ion exchange feed liquid is obtained.

[0071] The prepared ion exchange feed liquid is introduced into a series ion exchange system composed of D151 strong acid cation exchange resin column and D301 weak base anion exchange resin column. The D151 resin is a sulfonic acid type macroporous cation exchange resin with strong acid group -SO3H, which has good selective adsorption capacity for trivalent rare earth ions; the D301 resin is a tertiary amine type weak base anion exchange resin containing -NR2 functional groups, which is mainly used for removing anion impurities in the solution. The flow rate of the feed liquid is controlled to be 2.0 BV / h (BV represents the volume of the resin bed, i.e. the liquid volume passing through per hour is twice the volume of the resin bed) during the ion exchange process, which ensures sufficient exchange contact time and also takes into account the processing efficiency. As the exchange process proceeds, the D151 resin gradually saturates, and when the outlet rare earth ion concentration reaches 5% of the inlet concentration, it indicates that the resin is close to saturation, at which point the adsorption stops and the elution stage begins.

[0072] The saturated D151 resin bed is eluted by pH gradient elution technology using hydrochloric acid solutions with pH values of 4.0, 3.0, 2.0 and 1.0 in sequence for fractional elution. The principle of pH gradient elution is to take advantage of the difference in the desorption ability of different rare earth elements from the resin at different pH values to achieve the separation of rare earth elements. The specific operation is to first elute with a hydrochloric acid solution with a pH of 4.0, which mainly elutes light rare earth elements such as lanthanum, cerium, praseodymium and neodymium; then elute with a hydrochloric acid solution with a pH of 3.0, which mainly elutes medium rare earth elements such as samarium, europium and gadolinium; then elute with a hydrochloric acid solution with a pH of 2.0, which mainly elutes part of medium rare earth and heavy rare earth elements; finally elute with a hydrochloric acid solution with a pH of 1.0, which mainly elutes heavy rare earth elements such as dysprosium, holmium, erbium, thulium, ytterbium and lutetium. During the elution process, the online monitoring device is used to track the change of rare earth ion concentration in the eluate in real time, and when the rare earth ion concentration in the eluate of a certain stage drops to below 0.1 g / L, the next stage eluate is switched to, realizing precise segmented collection and obtaining rare earth enrichment liquids with different compositions.

[0073] The collected rare earth-rich liquid is divided into light rare earth group, medium rare earth group and heavy rare earth group according to the composition, and the differential temperature control crystallization treatment is adopted for different groups. The pH value of the light rare earth group is adjusted to 1.2±0.1, magnesium bicarbonate is added at 60±1℃ for precipitation, and the molar ratio of magnesium bicarbonate to rare earth ions is 2.8:1; the pH value of the medium rare earth group is adjusted to 1.5±0.1, magnesium bicarbonate is added at 65±1℃, and the molar ratio of magnesium bicarbonate to rare earth ions is 3.0:1; the pH value of the heavy rare earth group is adjusted to 1.8±0.1, magnesium bicarbonate is added at 70±1℃, and the molar ratio of magnesium bicarbonate to rare earth ions is 3.2:1. The crystallization process lasts for 12-16 hours, by controlling the crystallization temperature, time and pH value and other parameters, the growth rate and morphology of the crystal are affected, and the rare earth carbonate precipitate with uniform particle size and high purity is obtained, and the recycling of CO2 gas and magnesium salt solution is realized. After crystallization, filtration, washing and drying, the carbonate is decomposed and converted into rare earth oxide at 800±10℃ for 4 hours, which is the final high-purity rare earth product. The whole process realizes the recycling of magnesium salt and CO2, significantly reduces the production cost and eliminates the environmental pollution problem of traditional process.

[0074] In a specific embodiment, the process of performing step-by-step elution using acid solutions with different pH values can specifically include the following steps:

[0075] The saturated resin bed is pre-washed with deionized water at a constant flow rate to flush the surface of the saturated resin bed and remove free ions that are not fully adsorbed, obtaining a washed and balanced resin bed system;

[0076] The high-pH hydrochloric acid solution is introduced into the washed and balanced resin bed system for first-stage elution, and the eluent collection point is accurately controlled under the real-time tracking of the outlet rare earth ion concentration change by the online monitoring device, obtaining a first eluent component rich in light rare earth elements;

[0077] The medium-pH hydrochloric acid solution is introduced into the resin bed system that has completed the first-stage elution for second-stage elution, and the eluent flow rate is dynamically adjusted according to the change trend of the outlet liquid conductivity, obtaining a second eluent component rich in medium rare earth elements;

[0078] The low-pH hydrochloric acid solution is introduced into the resin bed system that has completed the second-stage elution for final elution, and the elution efficiency is optimized by controlling the temperature and flow rate parameters, obtaining a third eluent component rich in heavy rare earth elements, and realizing accurate separation of rare earth-rich liquids with different compositions.

[0079] Specifically, the pH gradient elution technique is the key to achieve fine separation of rare earth elements. First, the saturated resin bed is prewashed with deionized water with a resistivity greater than 18 MΩ·cm at a constant flow rate of 1.0 BV / h to flush the surface of the saturated resin bed. The purpose of prewashing is to remove the residual feed liquid and free ions that are not fully adsorbed in the gap between the resin beds, so as to avoid interference with the separation effect during elution. The flushing process continues until the water conductivity is less than 50 μS / cm and stable, indicating that the resin bed has reached a washing equilibrium state, at which point the resin bed system is ready for subsequent elution operations.

[0080] A high-pH hydrochloric acid solution with a pH of 4.0 is introduced into the washed and balanced resin bed system for the first stage of elution, with an elution flow rate controlled at 0.8 BV / h. The lower flow rate ensures sufficient contact between the eluent and the resin bed. The online monitoring device tracks the changes in rare earth ion concentration in real time. The device includes a conductivity sensor, an ultraviolet absorption detector, and an automatic sampling analysis system. When the concentration of rare earth ions in the outlet liquid is detected to start rising rapidly, the automatic collection system is started. The collection standard is set to the total concentration of rare earth ions in the eluent being greater than 0.5 g / L, and the collection continues until the concentration drops below 0.5 g / L. At this time, most of the light rare earth elements (mainly lanthanum, cerium, praseodymium, and neodymium) have been eluted, forming a first eluent component rich in light rare earth elements.

[0081] A medium-pH hydrochloric acid solution with a pH of 3.0 is introduced into the resin bed system that has completed the first stage of elution for the second stage of elution. At this time, the eluent flow rate is dynamically adjusted according to the trend of the outlet liquid conductivity. The specific adjustment strategy is as follows: when the conductivity rapidly rises, the flow rate is reduced to 0.6 BV / h to increase the contact time between the eluent and the resin; when the conductivity reaches a peak and starts to decline, the flow rate is adjusted to 1.0 BV / h to speed up the elution. The conductivity change curve collects data through the online monitoring system, and the slope change is calculated in real time based on the preset algorithm to determine the inflection point of the conductivity change, thereby accurately controlling the collection time. The second stage of elution mainly obtains medium rare earth elements (samarium, europium, and gadolinium), forming a second eluent component rich in medium rare earth elements.

[0082] The low pH hydrochloric acid solution with pH value of 1.5 is introduced into the resin bed system after the second stage elution is completed for final elution, and the two parameters of temperature and flow rate are controlled at the same time, the temperature is increased to 35±2℃, and the flow rate is controlled at 1.2BV / h. The increase of temperature can enhance the breaking of the complex bond between heavy rare earth elements and resin, accelerate the desorption of heavy rare earth elements, and the higher flow rate can improve the elution efficiency. The standard for collecting the final elution stage is also based on the concentration and conductivity changes of rare earth ions in the outlet liquid, but an additional spectral analysis confirmation step is added to ensure that the collected is the heavy rare earth element (dysprosium, holmium, erbium, thulium, ytterbium, lutetium) enriched liquid, forming the third elution component, and realizing the accurate separation of rare earth enriched liquid.

[0083] In a certain rare earth separation plant application case, the pH gradient elution is carried out on the completely saturated D151 resin bed (resin volume 5L), 20L deionized water is used for pre-washing, and the conductivity of the outlet water is monitored to decrease from the initial 3200μS / cm to 45μS / cm, and then the first stage elution is started. The elution is carried out using hydrochloric acid solution with pH value of 4.0, and the online monitoring device records that the concentration of rare earth ions first rapidly increases to a peak value of 2.8g / L, and then starts to decrease, and when the concentration decreases to 0.48g / L, the collection is stopped, and a total of 12L of elution liquid rich in light rare earth is obtained. The second stage elution is carried out using hydrochloric acid solution with pH value of 3.0, and according to the slope change of the conductivity curve, the flow rate is first decreased from the standard 0.8BV / h to 0.65BV / h, and about 8L of elution liquid rich in medium rare earth is collected at the concentration peak. The final stage elution is carried out at a temperature of 35℃ using hydrochloric acid solution with pH value of 1.5 at a flow rate of 1.2BV / h, and 6L of elution liquid rich in heavy rare earth is collected. Through ICP-MS analysis, the separation degree of rare earth elements in the three elution components is high, the proportion of light rare earth elements in the first component is 90%, the proportion of medium rare earth elements in the second component is more than 80%, and the proportion of heavy rare earth elements in the third component is more than 75%, effectively solving the technical problem of serious element cross contamination in traditional rare earth separation.

[0084] The above describes the rare earth mother liquor high-efficiency recovery method based on surrounding water quality monitoring in the embodiments of the present application, and the following describes the rare earth mother liquor high-efficiency recovery system based on surrounding water quality monitoring in the embodiments of the present application. Please refer to Figure 2 , one embodiment of the rare earth mother liquor high-efficiency recovery system based on surrounding water quality monitoring in the embodiments of the present application includes:

[0085] The analysis module 201 is used for inductively coupled plasma mass spectrometry analysis of the rare earth mother liquor, and a water quality monitoring network is established around the factory area to obtain mother liquor component data and water quality baseline data;

[0086] The processing module 202 is used for classifying the rare earth mother liquor into A-type mother liquor, B-type mother liquor and C-type mother liquor based on the mother liquor component data and the water quality baseline data, and adopting different processing routes for different types;

[0087] The extraction module 203 is used for adding magnesium bicarbonate to the A-type mother liquor, mixing the B-type mother liquor with filtrate, and then performing gradient extraction through a DOAM-PPA extractant, and dynamically adjusting process parameters according to water quality data.

[0088] The separation module 204 is used for mixing the extraction enriched liquid with the C-type mother liquor, and then separating through a D151 and D301 ion exchange resin system, and crystallizing to obtain high-purity rare earth products after pH grading elution.

[0089] The above Figure 2 The above

[0090] Referring to Figure 3 , the embodiment of the present application also provides a rare earth mother liquor efficient recovery device based on surrounding water quality monitoring. The device can be a server, and its internal structure can be as shown in Figure 3 The device includes a processor, a memory, a display screen, an input device, a network interface and a database connected through a system bus. The processor of the computer is used to provide computing and control capabilities. The memory of the device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the device is used to store the corresponding data in the embodiment. The network interface of the device is used to communicate with external terminals through network connection. The computer program is executed by the processor to realize the above method.

[0091] Those skilled in the art can understand Figure 3 that the structure shown in the embodiment is only a block diagram of part of the structure related to the present application, and does not constitute a limitation on the rare earth mother liquor efficient recovery device based on surrounding water quality monitoring to which the present application is applied.

[0092] The application further provides a computer readable storage medium, which can be a nonvolatile computer readable storage medium or a volatile computer readable storage medium, and the computer readable storage medium stores instructions, and the instructions make a computer execute the steps of the rare earth mother liquor efficient recovery method based on surrounding water quality monitoring when the instructions are run on the computer.

[0093] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, system and unit can refer to the corresponding processes in the foregoing method embodiments, and will not be described here.

[0094] The integrated unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the application or the whole or part of the technical solutions that essentially contribute to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for making a rare earth mother liquor efficient recovery device based on surrounding water quality monitoring (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the application. The foregoing storage medium includes a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program code storage media.

[0095] The above embodiments are only used to illustrate the technical solutions of the application, rather than limit them; although the application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the application.

Claims

1. A method for efficient recovery of rare earth mother liquor based on peripheral water quality monitoring, characterized by, The method includes: Inductively coupled plasma mass spectrometry (ICP-MS) analysis was performed on the rare earth mother liquor, and a water quality monitoring network was established around the plant area to obtain mother liquor composition data and water quality baseline data. This included: preprocessing the rare earth mother liquor using a multispectral excitation sequence; performing stepwise ionization scanning on the preprocessed mother liquor using an ICP-MS to obtain multi-element spectral distribution data; setting up an adaptive monitoring station array at key hydrological nodes upstream and downstream of the plant area to perform real-time high-frequency dual-wavelength fluorescence spectroscopy monitoring of the water body to obtain dynamic distribution curves of rare earth element content; applying a self-calibrating multivariate analysis algorithm to denoise the multi-element spectral distribution data; and performing quantitative conversion using standard substance calibration curves to obtain mother liquor composition data. The dynamic distribution curve of rare earth element content is analyzed and processed using a neural network time-series model to identify periodic fluctuation characteristics and trends, construct a digital twin model of water environmental capacity, and obtain water quality baseline data with predictive capabilities. This includes: inputting the dynamic distribution curve of rare earth element content into a bidirectional long short-term memory (LSTM) network for time-series feature extraction. The LSM network contains a three-layer hidden layer structure, with each layer having twice the number of memory units as the rare earth element types, resulting in a time-series encoded feature vector; applying wavelet transform to the time-series encoded feature vector for multi-scale decomposition to extract fluctuations in different frequency domains. Features are obtained by assigning weights to each frequency component through a self-attention mechanism to obtain a multi-frequency domain weighted feature matrix. This multi-frequency domain weighted feature matrix is ​​then input into a convolutional neural network (CNN) for spatial feature learning. The CNN consists of three alternating convolutional layers and two pooling layers, with kernel sizes of 3×3, 5×5, and 7×7, respectively, to obtain water quality spatial distribution pattern features. Based on these water quality spatial distribution pattern features, a graph neural network model is constructed, using monitoring stations as nodes and water flow direction as edges. A message passing mechanism is used to simulate the pollutant migration and diffusion process, resulting in water quality baseline data with predictive capabilities. Based on the mother liquor component data and the water quality baseline data, the rare earth mother liquor is divided into A type mother liquor, B type mother liquor and C type mother liquor, and different processing routes are adopted for different categories, including: inputting the mother liquor component data into a multivariate clustering analysis system, using an adaptive K-means clustering algorithm to divide the three-dimensional characteristics of rare earth element content, acidity and impurity content in space, and obtaining a characteristic distribution map of the rare earth mother liquor; applying a hierarchical decision tree algorithm to the characteristic distribution map to optimize the boundary, setting a classification standard in combination with the environmental capacity threshold in the water quality baseline data, and obtaining a mother liquor classification decision model; the rare earth mother liquor is classified in real time through the mother liquor classification decision model, and is automatically divided into A type of high rare earth and low impurity, B type of medium rare earth and medium impurity, and C type of low rare earth and high impurity according to the total rare earth content, acidity and impurity content parameters, and a classified mother liquor stream is obtained; wherein the A type of high rare earth and low impurity has a total rare earth content greater than 10 g / L, an acidity less than 1 mol / L and an impurity content less than 5%; the B type of medium rare earth and medium impurity has a total rare earth content between 5-10 g / L, an acidity between 1-3 mol / L and an impurity content between 5-15%; and the C type of low rare earth and high impurity has a total rare earth content less than 5 g / L, an acidity greater than 3 mol / L and an impurity content greater than 15%; the classified mother liquor stream is physically separated through an intelligent control pipeline system, the A type mother liquor is allocated a magnesium bicarbonate carbonate precipitation processing route, the B type mother liquor is allocated a DOAM-PPA extraction processing route, and the C type mother liquor is allocated an ion exchange processing route, and three different processing process paths are obtained; The A type mother liquor is added with magnesium bicarbonate carbonate precipitation, the B type mother liquor is mixed with the obtained filtrate after precipitation, and gradient extraction is performed through a DOAM-PPA extractant, and the process parameters are dynamically adjusted according to the water quality data; The extraction enrichment liquid is mixed with the C type mother liquor, and then separated through a D151 and D301 ion exchange resin system, and a high-purity rare earth product is prepared by crystallization after pH fractionation elution. 2.The method of claim 1, wherein, The A type mother liquor is added with magnesium bicarbonate carbonate precipitation, the B type mother liquor is mixed with the obtained filtrate after precipitation, and gradient extraction is performed through a DOAM-PPA extractant, and the process parameters are dynamically adjusted according to the water quality data; The A type mother liquor is subjected to constant temperature control through a double-layer structure reaction kettle, the temperature is accurately adjusted to 40±1℃, and a calculated amount of magnesium bicarbonate solution which is 1.8 times the molar number of rare earth ions is added for precipitation reaction, to obtain rare earth carbonate precipitate, filtrate and recoverable CO2 gas; The filtrate is mixed with the B type mother liquor, the pH value is adjusted to 1.5±0.1, an oxidizing agent is added to the mixed liquid for pretreatment, and a liquid before extraction with stable valence is obtained; The liquid before extraction is introduced into a three-stage countercurrent extraction device, DOAM-PPA is used as an extractant, different pH values and temperature conditions are set in the first, second and third stages respectively, a gradient extraction environment is formed, and an organic phase enriched in rare earth elements is obtained; According to the real-time change of the rare earth ion content in the water quality data, the extraction process parameters are dynamically adjusted through a multivariate adaptive control system, including adjusting the extraction time, phase ratio and pH value, while recycling magnesium salt and CO2, to obtain an extraction enrichment liquid and recycled auxiliary materials. 3.The method of claim 1, wherein, The extraction enrichment liquid is mixed with the C-type mother liquor, and then separated by a D151 and D301 ion exchange resin system, and a high-purity rare earth product is prepared by crystallization after pH fractionation elution, including: The extraction enrichment liquid is mixed with the C-type mother liquor in a volume ratio of 2:1, a reducing agent is added to the mixed liquid for pretreatment, and the mixed liquid is filtered through a filter to obtain a clear and uniform ion exchange feed liquid; The ion exchange feed liquid is introduced into a series system composed of the D151 and the D301 ion exchange resin, and exchange adsorption is carried out at a suitable flow rate to obtain a saturated resin bed rich in rare earth elements; The saturated resin bed is subjected to pH gradient elution technology, and different pH value acidic solutions are used for fractionation elution in sequence to obtain rare earth enrichment liquids of different compositions; The rare earth enrichment liquids are divided into light rare earth group, medium rare earth group and heavy rare earth group, and are subjected to differential temperature control crystallization treatment, and the high-purity rare earth product is obtained after subsequent calcination conversion.

4. The method according to claim 3, wherein, The saturated resin bed is subjected to pH gradient elution technology, and different pH value acidic solutions are used for fractionation elution in sequence to obtain rare earth enrichment liquids of different compositions, including: The saturated resin bed is subjected to pre-washing treatment, and deionized water is used to flush the surface of the saturated resin bed at a constant flow rate to remove free ions that are not fully adsorbed, and a washing equilibrium resin bed system is obtained; A high-pH value hydrochloric acid solution is introduced into the washing equilibrium resin bed system for first-stage elution, and the eluent collection point is accurately controlled under the real-time tracking of the outlet rare earth ion concentration change by an online monitoring device, to obtain a first eluent component rich in light rare earth elements; A medium-pH value hydrochloric acid solution is introduced into the resin bed system that has completed the first-stage elution for second-stage elution, and the eluent flow rate is dynamically adjusted according to the outlet liquid conductivity change trend, to obtain a second eluent component rich in medium rare earth elements; A low-pH value hydrochloric acid solution is introduced into the resin bed system that has completed the second-stage elution for final elution, and the elution efficiency is optimized by controlling the temperature and flow rate parameters, to obtain a third eluent component rich in heavy rare earth elements, and the accurate separation of the rare earth enrichment liquids of different compositions is realized.

5. A rare earth mother liquor high-efficiency recovery system based on peripheral water quality monitoring, characterized in that, A system for realizing the high-efficiency recovery method of rare earth mother liquor based on surrounding water quality monitoring according to any one of claims 1-4, the system comprising: an analysis module for inductively coupled plasma mass spectrometry analysis of the rare earth mother liquor, and for establishing a water quality monitoring network around the plant to obtain mother liquor composition data and water quality baseline data; a processing module for dividing the rare earth mother liquor into A-type mother liquor, B-type mother liquor and C-type mother liquor based on the mother liquor composition data and the water quality baseline data, and for adopting different processing routes for different types; An extraction module is used to add magnesium bicarbonate to the A type mother liquor for precipitation, the B type mother liquor is mixed with the filtrate and gradient extraction is performed by using a DOAM-PPA extractant, and process parameters are dynamically adjusted according to water quality data; A separation module is used to mix the extraction enriched liquid with the C type mother liquor, separate the mixture by using a D151 and D301 ion exchange resin system, and crystallize to obtain a high-purity rare earth product after pH grading elution.

6. A rare earth mother liquor high-efficiency recovery equipment based on peripheral water quality monitoring, characterized in that, The computer program is stored in the memory and can be run on the processor, and the processor executes the computer program to realize the method for efficiently recycling rare earth mother liquor based on surrounding water quality monitoring according to any one of claims 1 to 4.

7. A computer-readable storage medium having stored thereon a computer program, characterized in that The computer program is stored in the memory and can be run on the processor, and the processor executes the computer program to realize the method for efficiently recycling rare earth mother liquor based on surrounding water quality monitoring according to any one of claims 1 to 4.

Images