Blockchain-based fluorine salt recovery state monitoring method and system for fluorine-containing wastewater

Abstract

The application provides a blockchain-based fluorine salt recovery state monitoring method and system for fluorine-containing wastewater, relates to the fluorine-containing wastewater recovery data processing technical field, and through accurate collection of fluorine salt concentration correlation data of initial and end states of each stage, uploading of a blockchain node to complete tamper-proof storage of original data, and ensuring data traceability and integrity. Subsequently, for physical interference such as micro-bubbles and flocs in sampling of high-concentration fluorine-containing wastewater, corresponding abnormal data is removed through deviation correction, and then combined with offline calibration data to compensate for system errors caused by electrode drift and environmental fluctuations, so that accurate optimization data conforming to actual working conditions are finally obtained. Based on the optimization data, the fluorine salt recovery state of each stage can be determined in real time, and fast feedback is realized to the process control system to achieve accurate regulation and control. At the same time, the recovery state results are stored again, which not only solves the core problems of data distortion and difficult removal of interference in traditional monitoring, but also adapts to the stable detection needs of the alkaline stripping agent system.

Description

Technical Field

[0001] This invention relates to the field of fluoride wastewater recovery data processing technology, and in particular to a blockchain-based method and system for monitoring the recovery status of fluoride salts in fluoride wastewater. Background Technology

[0002] In industries such as photovoltaics, electroplating, and semiconductor manufacturing, the treatment of fluoride-containing wastewater and the recovery of fluoride salts are important directions for environmental protection and resource utilization. Currently, the industry generally adopts a complete process chain of pretreatment-core reaction-deep treatment, and its monitoring process has formed a mature system. For example, the treatment and fluoride salt recovery process for high-concentration fluoride-containing wastewater generated by the high-borosilicate glass panel manufacturing industry: Pretreatment is first completed through a screen, sand filter, and homogenization tank. Wastewater containing hydrogen fluoride needs to be neutralized to pH 7-9 with alkali before entering the core reaction stage, where lime slurry, calcium chloride, and other agents are added to generate calcium fluoride precipitate. Subsequent treatment involves coagulation enhancement... Advanced treatment methods include activated alumina adsorption or membrane separation. Monitoring involves using techniques such as ion-selective electrode method and ion chromatography to collect parameters at each stage, including total fluoride concentration, fluoride ion concentration, pH value, flow rate, and reagent dosage. Samples are stored in polyethylene bottles with added fixatives, and laboratory spiked recovery rate tests ensure data quality. After completing instrument calibration and curve verification according to specifications, the collected data are calculated to determine recovery efficiency and treatment compliance. The results are stored in a dedicated database and simultaneously fed back to the control system to adjust process parameters, ensuring the process complies with relevant emission standards.

[0003] Around the aforementioned process chain, a process monitoring system is generally deployed in industrial sites. This system collaborates with the plant's DCS / SCADA system via a standard bus or interface. Within this system, fluoride recovery status monitoring operates as a functional sub-module alongside modules such as discharge compliance monitoring, equipment health monitoring, and energy and chemical consumption assessment. The monitoring system does not directly take over process control; it only outputs reports, alarms, and parameter optimization suggestions.

[0004] The existing monitoring system generally consists of: on-site sampling and flow cells, online sensors and analyzers (fluoride ion selective electrodes, total fluoride / free fluoride ion chromatography, pH, conductivity, turbidity / suspended solids, temperature, flow rate, etc.) and key laboratory channels (standard curves, spiked recovery, quality control samples); acquisition and synchronization layer: PLC / remote terminals, data acquisition devices and time synchronization; data processing and quality control layer: noise reduction, detection / quantitation limit processing, calibration and curve verification, standardization and time alignment; calculation and model layer: material balance, recovery / removal rate calculation, stage target achievement assessment and anomaly identification; and storage and display layer: historical database, reports and visualization, threshold and event management, and interface with DCS / SCADA.

[0005] Within this framework, the monitoring process for fluoride recovery involves setting up monitoring points at each of the homogenization / neutralization, precipitation, coagulation separation, advanced treatment (adsorption / membrane), evaporation crystallization, and permeate / condensate stages to continuously collect key parameters such as free fluoride, total fluoride, pH, flow rate, reagent dosage, transmembrane pressure / flux, concentration ratio, and crystalline phase. Data from each stage are summarized by time slice to complete segmented material balance and stage recovery rate calculations. The results are then fed back to the monitoring interface and control system for verification of process settings such as dosage ratio, membrane operating point, evaporation load, and seeding strategy.

[0006] The above-mentioned technology has the following technical problems:

[0007] However, in scenarios involving fluoride-containing wastewater with high concentrations of target components and coexisting salts, and rapid fluctuations in influent and effluent water quality and operating conditions within a short period, fluoride monitoring faces more specific systemic challenges: the flow cell and pipelines in the sampling link are prone to trapping microbubbles, flocs, or gas-liquid blockages. Two-phase impurities disrupt the stable measurement environment of electrochemical / optical sensors, causing discontinuities in the conductive path, abnormal fluctuations in boundary layer thickness, drift in reference node potential, and superposition of additional resistance voltage drop and flow potential. This results in sudden increases / decreases in the measured fluoride ion readings that are unrelated to the actual state and deviate from the actual concentration. At the same time, two-phase disturbances can also cause a momentary decrease in local flow velocity and increase sampling hysteresis, amplifying data distortion and misleading process parameter adjustments for units such as extraction, precipitation, membrane separation, and evaporation. In back-extraction and deep treatment systems characterized by alkaline or other highly reactive agents, stable, low-noise, and uniformly caliber fluoride monitoring input is particularly required; otherwise, it is easy to cause dosing oscillations and threshold misjudgments. Summary of the Invention

[0008] To address the technical problem in existing technologies where microbubbles, flocs, or gas-liquid blockages easily get trapped in the flow tank and pipeline during sampling, causing sudden increases / decreases in monitoring readings that are unrelated to the actual pollutant state and deviate from the actual concentration, this invention provides a blockchain-based method and system for monitoring the fluoride recovery status in fluoride-containing wastewater. The technical solution is as follows:

[0009] On the one hand, a blockchain-based method for monitoring the recovery status of fluoride salts in fluoride-containing wastewater is provided. This method includes: sequentially monitoring and collecting initial and final fluoride salt concentration correlation data for each stage (extraction, residual liquid alkali addition, and tailwater treatment), and uploading this data to a blockchain node to complete the original data storage. Specifically, the initial stage of the extraction stage refers to the untreated state of the fluoride-containing wastewater before it enters the extraction equipment; the final stage refers to the state after the fluoride-containing wastewater and extractant have completed mixing and reaction, and the residual liquid has been discharged after settling; the initial stage of the residual liquid alkali addition stage refers to the initial state of the residual liquid entering the alkali addition reaction equipment after extraction; and the final stage refers to the state after the effluent is treated in stages. The initial stage of the effluent treatment process refers to the initial state of the clear liquid entering the advanced treatment system after the residual liquid is neutralized by alkali addition and solid-liquid separation is completed. The final stage refers to the state of the effluent meeting the standards after calcium chloride precipitation and electrodialysis treatment. The collected fluoride concentration correlation data is corrected to remove deviations caused by physical interference, and then the accuracy is calibrated by combining it with offline calibration data to obtain optimized fluoride concentration correlation data. Based on the optimized fluoride concentration correlation data, the fluoride recovery status of each treatment stage of fluoride-containing wastewater is monitored and determined, and the fluoride recovery status results are fed back and uploaded to the blockchain node to complete data storage.

[0010] On the other hand, a blockchain-based monitoring system for fluoride recovery in fluoride-containing wastewater is provided. This system includes a data acquisition and storage module, used to sequentially monitor and collect initial and final fluoride concentration data for each stage—extraction, residual liquid alkali addition, and effluent treatment—and upload this data to a blockchain node for raw data storage. Specifically, the initial stage of the extraction stage refers to the untreated state of the fluoride-containing wastewater before it enters the extraction equipment; the final stage refers to the state after the fluoride-containing wastewater and extractant have completed mixing and reaction, and the residual liquid has been discharged. The initial stage of the residual liquid alkali addition stage refers to the initial state of the residual liquid entering the alkali addition reaction equipment after extraction; and the final stage refers to the staged alkali neutralization. The system comprises several modules: the initial stage of the effluent treatment phase (where the reaction is complete, the clear liquid is discharged after solid-liquid separation), the final stage of the effluent treatment phase (where the clear liquid enters the advanced treatment system after alkali addition), and the final stage of the effluent treatment phase (where the effluent meets standards after calcium chloride precipitation and electrodialysis). A data accuracy calibration module is used to correct deviations in the collected fluoride concentration correlation data by eliminating physical interference, and then combines this with offline calibration data to complete accuracy calibration, resulting in optimized fluoride concentration correlation data. A status result feedback module is used to monitor and determine the fluoride recovery status at each stage of fluoride-containing wastewater treatment based on the optimized fluoride concentration correlation data, thereby providing fluoride recovery status result feedback and uploading the fluoride recovery status result to the blockchain node for data storage.

[0011] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0012] (1) In the recovery of fluoride salts from fluoride-containing wastewater in industries such as photovoltaics and electroplating, the existing extraction-residue alkali addition-tailwater treatment process often uses fluoride ion selective electrodes to collect concentration data to guide regulation. However, when treating wastewater containing more than 2% free fluoride, impurities such as microbubbles and flocs in the sampling process can easily damage the electrode detection environment, causing sudden rises and falls in readings and flow rate fluctuations, resulting in data distortion and process adjustment deviations. Moreover, the alkaline back-extraction agent system has higher requirements for detection stability, which seriously restricts the recovery efficiency. This invention collects the initial and final fluoride salt concentration correlation data of each stage according to the process sequence and uploads them to the blockchain for evidence storage. First, physical interference is eliminated through deviation correction, and then the error is compensated by offline calibration to obtain accurate and optimized data. Based on this, the recovery status is determined and feedback is provided for regulation. At the same time, the results are stored on the blockchain. This process not only solves the problem of data distortion in traditional monitoring and adapts to the detection requirements of alkaline systems, but also ensures data credibility through blockchain, achieving accurate and controllable fluoride salt recovery and stable and efficient process.

[0013] (2) This invention accurately matches the monitoring scenario characteristics of high-concentration fluoride-containing wastewater, and specifically selects two parameters for judgment: differential pressure energy and acoustic signal amplitude. Impurities such as microbubbles, flocs, and gas-containing liquid plugs in the flow tank and pipeline will cause changes in pipeline flow resistance (directly reflected as differential pressure energy fluctuations), and at the same time generate characteristic acoustic signals (related to impurity size and flow state). Through these two scenario adaptability parameters, the detection interference caused by impurities can be accurately captured. The real-time differential pressure signal amplitude exceeding the standard corresponds to the flow obstruction caused by impurities, and the acoustic signal amplitude abnormal corresponds to impurity collision or disturbance of the electrode environment. The two dimensions work together to achieve accurate differentiation between data disturbance and abnormality. It avoids misjudgment by a single parameter and can specifically identify the distortion of fluoride ion concentration data caused by different types of interference, providing a clear direction for subsequent data correction. It effectively solves the problems of difficult identification of impurity interference and misuse of distorted data for process control in traditional monitoring, adapts to the stable detection requirements of alkaline back-extraction agent system, and ensures the accuracy and reliability of fluoride salt recovery process control.

[0014] (3) This invention uses time-synchronized mask array weighted calculation to accurately adapt to high-concentration fluoride wastewater monitoring scenarios. Its advantages are in line with actual needs. For data distortion of different degrees caused by impurities such as microbubbles and flocs, it achieves precise control through differentiated weight assignment. Normal data is assigned full weight to ensure authenticity, disturbed data is weighted down by gradient to weaken the impact of slight interference, and abnormal data is assigned the lowest weight to eliminate serious distortion, thus avoiding interference from distorted data caused by impurities in the overall calculation. The average concentrations of fluorosilicic acid and fluoroboric acid are corrected by combining the trend of fluoride ion concentration change. This not only uses the stoichiometric correlation of the three to ensure the logical consistency of the data, but also reduces the impact of single distorted data through weighted algorithm, effectively solving the problem of concentration calculation deviation caused by impurity interference in traditional calculation. The final optimized fluoride salt concentration correlation data is more in line with the actual working conditions, providing accurate basis for subsequent process control, adapting to the stable detection requirements of alkaline back-extraction agent system, and ensuring fluoride salt recovery efficiency and process stability. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention, 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 the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart of a blockchain-based method for monitoring the recovery status of fluoride salts in fluoride-containing wastewater, provided in an embodiment of the present invention.

[0017] Figure 2 This is a schematic diagram of the structure of a blockchain-based fluoride recovery status monitoring system for fluoride-containing wastewater provided in an embodiment of the present invention;

[0018] Figure 3 This is a process flow diagram of fluoride-containing wastewater treatment provided in an embodiment of the present invention;

[0019] Figure 4 This is the XRD characterization spectrum of the potassium fluoride product provided in the embodiments of the present invention;

[0020] Figure 5 This is the XRD characterization spectrum of the potassium fluorosilicate product provided in the embodiments of the present invention;

[0021] Figure 6 This is the XRD characterization spectrum of the potassium fluoroborate product provided in the embodiments of the present invention. Detailed Implementation

[0022] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0023] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0024] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0025] This invention provides a blockchain-based method for monitoring the recovery status of fluoride salts in fluoride-containing wastewater, such as... Figure 1 The flowchart shown is a blockchain-based method for monitoring the recovery status of fluoride in fluoride-containing wastewater. The process can include the following steps: Following the sequence of extraction, residual liquid alkali addition, and effluent treatment, the initial and final fluoride concentration correlation data for each stage are monitored and collected sequentially, and then uploaded to the blockchain node for raw data storage. Specifically, the initial stage of the extraction stage refers to the untreated state of the fluoride-containing wastewater before it enters the extraction equipment; the final stage refers to the state after the fluoride-containing wastewater and extractant have completed mixing and reaction, and the residual liquid has been discharged. The initial stage of the residual liquid alkali addition stage refers to the initial state of the residual liquid after extraction entering the alkali addition reaction equipment; the final stage refers to the state after separation... The treatment process is divided into three stages: the initial stage (after alkali neutralization and solid-liquid separation, the clear liquid is discharged), the final stage (after calcium chloride precipitation and electrodialysis, the effluent reaches the pre-set standards of the technicians), and the final stage (after alkali addition and the clear liquid enters the deep treatment system). The collected fluoride concentration correlation data is corrected to remove physical interference and abnormal data, and then combined with offline calibration data to achieve accuracy calibration, resulting in optimized fluoride concentration correlation data. Based on the optimized fluoride concentration correlation data, the fluoride recovery status of each treatment stage of fluoride-containing wastewater is monitored and determined, and the fluoride recovery status results are fed back and uploaded to the blockchain node for data storage.

[0026] Figure 3This is a process flow diagram of fluoride-containing wastewater treatment provided in this embodiment of the invention. First, the etching waste liquid enters the extraction process, where it is extracted with an organic extractant to generate a fluoride-containing organic phase. Then, it enters the back-extraction section, where it is back-extracted with a 30% potassium hydroxide solution to generate a potassium fluoride solution. The back-extraction turbidity is filtered to remove impurities and then sent to an evaporation unit for concentration and crystallization to obtain potassium fluoroborate crystals. The steam condensate is recycled back into the system. Simultaneously, the extraction residue is stirred and filtered to form a filter residue and a filtrate. The filter residue is washed with water to remove impurities and dried, and can be returned upstream as a byproduct. The filtrate enters the subsequent electrodialysis unit. Electrodialysis separates the filtrate into two streams: desalinated water and concentrated water. The desalinated water is recycled as production water, while the concentrated water enters the evaporation unit for concentration and crystallization. The condensate is recovered, and the impurities are sent to a solid waste treatment center. For the fluoride-containing byproduct portion, some calcium chloride sludge is supplied to a hydrogen fluoride production company, where it reacts to generate hydrogen fluoride. Simultaneously, another portion is dried and returned to the system for reuse, achieving a closed-loop circulation operation. The overall process forms a multi-stage series system of extraction, back-extraction, filtration, evaporation, electrodialysis, evaporation crystallization, and by-product disposal. Fluorine resources are converted into usable products such as potassium fluoride, potassium fluorosilicate, and potassium fluoroborate. Condensate and fresh water are reused, and miscellaneous salts are uniformly sent to the solid waste center for disposal, achieving the goal of full-scale recovery and zero discharge of high-concentration fluoride-containing wastewater.

[0027] Specifically, the initial stage fluoride concentration correlation data and the final stage fluoride concentration correlation data are monitored and collected sequentially for each stage, which refers to the extraction stage, the residual liquid alkali addition stage, and the tailwater treatment stage.

[0028] The extractant used in the extraction stage is a mixture of P204 (diisooctyl phosphate), Cyanex 923 (a mixture of trioctylphosphine oxide and TBP), and TBP (tributyl phosphate) in a 2:1:1 ratio, stirred at 25°C to 60°C to ensure complete miscibility of the components. It is diluted using kerosene or ethyl acetate. Sulfonated kerosene or ethyl acetate is used as the diluent, and the dilution ratio needs to be optimized according to the specific composition of the target wastewater to adjust the viscosity of the extractant, improve phase separation performance, and control costs.

[0029] The filtered etching wastewater, used as the aqueous phase, is added to a mixer-clarifier or stirred reactor at an oil-to-water ratio of 1:1 to 4:1. The mixture is stirred for 20 minutes at a temperature of 25°C to 50°C and a stirring rate of 100–500 rpm. During this process, free fluoride and a small amount of fluoroborate in the wastewater are selectively transferred to the organic phase. After settling and separation, the aqueous phase (raffinate) is sent to subsequent stages for further treatment to ensure it meets standards for reuse within the plant.

[0030] The enriched organic phase loaded with the target component enters the back-extraction section. It is mixed with a 30% potassium hydroxide (KOH) solution at an O / A ratio of 2:1 and reacted at 25°C and 1000 rpm for 10 min. During this process, KOH reacts with free fluorine in the organic phase to form potassium fluoride (KF), which is transferred to the aqueous phase. Simultaneously, a small amount of potassium fluoroborate is formed and precipitated. After the reaction, the phase is allowed to stand and separate. The regenerated organic phase (oil phase) can be recycled back to the extraction stage after simple adjustments.

[0031] The back-extraction liquid was filtered using a plate and frame filter press, and the resulting cake was potassium fluoroborate, which was dried and used as a byproduct. The clarified liquid from the plate and frame filter press was evaporated using a multi-effect evaporator to obtain potassium fluoride.

[0032] In the residual alkali addition stage, the total amount of alkali required to adjust the waste liquid to neutral is calculated based on the acidity of the residual solution. First, 3% to 10% of the alkali solution is added to the waste liquid. The key to this stage is precise control of the amount of alkali added, adjusting the system to the specific acidity range required for the precipitation of potassium fluoroborate (KBF4). The alkali solution is added slowly while stirring at 100-300 rpm to ensure uniform reaction. After alkali addition, the mixture is separated into solid and liquid phases, and the resulting filter cake is the crude potassium fluoroborate byproduct. This step not only recovers valuable components but also avoids the possibility of potassium fluoroborate dissolving or transforming due to excessive alkali during subsequent neutralization, ensuring product purity and yield. The crude potassium fluoroborate is mixed with water at a solid-liquid ratio of 1:1 to 3:1, and washed 1 to 2 times with stirring to remove surface-adsorbed soluble impurities (such as fluorides, chlorides, sulfates, etc.). After drying, the potassium fluoroborate byproduct is obtained.

[0033] Continue to slowly add the remaining alkali solution to the clarified liquid after alkali filtration until the pH of the solution reaches neutrality (6-9). This process requires real-time monitoring of pH changes to ensure precise endpoint control. During neutralization, residual fluorosilicate ions (SiF6) in the clarified liquid will be neutralized. 2- It reacts with potassium ions to form potassium fluorosilicate (K2SiF6) precipitate. The reaction principle can be represented as: H2SiF6 + 2KOH → K2SiF6↓ + 2H2O.

[0034] After the reaction is complete, solid-liquid separation is performed again, and the resulting filter residue is crude potassium fluorosilicate. Similarly, the crude potassium fluorosilicate is mixed with water at a solid-liquid ratio of 1:1 to 3:1, and washed 1 to 2 times with stirring to remove soluble impurities (such as fluorides, chlorides, sulfates, etc.) adsorbed on the surface. After drying, potassium fluorosilicate by-product is obtained. The resulting washing wastewater is concentrated in the electrodialysis section and then evaporated in the evaporator.

[0035] In the effluent treatment stage, the free fluoride content of the filtrate from the alkali addition stage is tested to calculate the calcium chloride dosage, which is then added at 1.2 times the recommended amount. After the reaction, solid-liquid separation is performed using a plate and frame filter press. The resulting calcium fluoride sludge (containing fluoride salts) can be transported to a hydrofluoric acid (HF) production plant. The plant can then use a calcination process to react calcium fluoride with sulfuric acid and other substances to generate HF, thus recovering and utilizing fluoride resources. The reaction equation is: CaF2 + H2SO4 → CaSO4 + 2HF↑.

[0036] After filtration, the effluent is treated by electrodialysis. The purified water meets the Class III surface water quality standard and is then reused. The concentrated water is evaporated and the condensate is reused as greywater. The miscellaneous salts are transported to the solid waste disposal center for treatment.

[0037] In one example embodiment, 7.37 cubic meters of glass panel etching waste liquid with a total fluoride content of 6%, a free fluoride content of 3.8%, and an acidity of 7 mol / L can be processed using this process to produce 1 ton of potassium fluoride, 0.55 tons of potassium fluoroborate, and 2.26 tons of potassium fluorosilicate, consuming 2.03 tons of potassium hydroxide and 0.04 tons of calcium chloride. The characterization results of the potassium fluoride, potassium fluoroborate, and potassium fluorosilicate products are as follows: Figures 4-6 As shown.

[0038] Figure 4 The image shows the XRD pattern of the potassium fluoride product. Figure 5 The image shows the XRD pattern of potassium fluorosilicate. Figure 6 The XRD characterization spectra of potassium fluoroborate products are shown. The horizontal axis represents the diffraction angle 2θ (degrees), and the vertical axis represents the diffraction intensity (counts). Figure 4 The black master curve in the middle represents the measured diffraction intensity of the sample at different diffraction angles of 2θ as a function of angle. The thin red line represents the background / baseline or its fitting result. The red bar chart below the figure corresponds to the characteristic peak position and relative intensity of standard card 00-036-1458 (anhydrous potassium fluoride, Carobbiite), and the blue bar chart corresponds to the characteristic peak position and relative intensity of standard card 00-032-0783 (potassium fluoride dihydrate, KF·2H2O). The phase composition can be determined by matching the sample peak position with the red and blue standard bar charts (identifying anhydrous KF and whether it is accompanied by impurities such as KF·2H2O). Combining the relative relationship between peak intensity, noise and background can assess crystallinity and purity, while the peak half-width at half-maximum can be used to further estimate grain size and micro-strain. As can be seen from the figure, the main peak of the sample is highly consistent with the anhydrous KF standard peak, and the overlap of some positions with the blue bar chart suggests the possible presence of trace amounts of hydrated phases.

[0039] Figure 5The black master curve represents the measured diffraction intensity of the sample at different diffraction angles of 2θ, and the thin red line represents the background / baseline fitting. The red bar chart below corresponds to the characteristic peak positions and relative intensities of standard card 04-006-8962 (K2SiF6, potassium fluorosilicate, Hieratite), the blue bar chart corresponds to standard card 00-005-0613 (K2SO4, potassium sulfate, Arcanite), and the pink bar chart corresponds to standard card 00-004-0587 (KCl, potassium chloride, Sylvite). The matching relationship shows a high degree of agreement between the sample's main diffraction peaks and the red bar chart, indicating that the potassium fluorosilicate phase is dominant. Some peak positions overlap with the blue or pink bar charts, suggesting the possible presence of trace impurities such as potassium sulfate or potassium chloride. Combining peak intensity, background, and peak shape allows for further evaluation of crystallinity and purity, and subsequent calculations of grain size and microstrain can be performed based on the peak half-width at half-maximum (WHM).

[0040] Figure 6 The black master curve in the middle represents the measured diffraction intensity of the sample at different diffraction angles of 2θ, and the thin red line represents the background / baseline fit. The red bar graph below corresponds to the characteristic peak positions and relative intensities of standard card 04-007-5338 (Avogadrite, potassium fluoroborate KBF4). It can be seen that the sample's main diffraction peaks highly match the peak positions of this standard, indicating that the phase is predominantly potassium fluoroborate. The peak intensity and low background indicate good crystallinity of the sample, and no characteristic peaks significantly overlap with other known impurity phase standard peak systems. If further confirmation of impurity phases, grain size, and stress information is needed, quantitative analysis can be performed based on the peak half-width at half-maximum (FWHM) and Rietveld fitting.

[0041] The above three types of monitoring provide verifiable product-side evidence and closed-loop optimization signals: On the one hand, the characteristic peak positions, relative intensities, quantitative proportions of phases, crystallinity, and impurity fingerprints (such as hydrates, potassium chloride, potassium sulfate, etc.) of each batch of XRD are used to generate summaries and hashes, serving as quality fingerprints for on-chain evidence storage and traceability, supporting compliance proofs for processes such as recovery rate and removal rate; on the other hand, the types and contents of impurities precipitated by XRD are back-calculated to process parameters (such as extraction ratio, back-extraction alkalinity and temperature, precipitation section pH and calcium-fluorine ratio, membrane section rejection rate and flux, evaporation crystallization concentration and seeding conditions), driving upstream automatic adjustment and re-optimization, thereby continuously improving product purity and stability in a positive cycle from data to evidence storage to correction to re-production, while reducing energy consumption and pharmaceutical consumption, reducing secondary impurity salt generation, and forming an integrated quality closed loop with controllable process and verifiable results.

[0042] The collected fluoride concentration correlation data were corrected to remove deviations caused by physical interference, and then the accuracy was calibrated by combining it with offline calibration data to obtain optimized fluoride concentration correlation data.

[0043] Specifically, the process involves removing deviations caused by physical interference from the collected fluoride concentration correlation data and correcting for abnormal data. It also includes classifying the fluoride ion concentration sequence in the fluoride concentration correlation data. The specific classification process is as follows: fluoride concentration correlation data includes fluoride ion concentration sequence, mean fluorosilicate ion concentration, and mean fluoroboronic acid ion concentration. Fluoride concentration correlation data refers to the fluoride concentration correlation data within the monitoring window.

[0044] A monitoring window is a fixed time period defined by relevant technical personnel to ensure that data from different stages are comparable and calculable under the same caliber.

[0045] The fluoride ion concentration sequence was obtained by converting the electrode potential into fluoride ion activity and then further converting it into a concentration sequence using a fluoride ion selective electrode based on the Nernst relation. The mean concentration of fluorosilicate ions was preferably obtained by averaging multiple measurements taken within the same window using ion chromatography or potentiometric titration. The mean concentration of fluoroboronic acid ions was obtained in the same way as that of fluorosilicate ions. The reason for using the fluoride ion concentration sequence to correct for deviations in the mean concentrations of fluorosilicate ions and fluoroboronic acid ions is that all three belong to the total fluoride system and are affected by the same unified processes such as extraction, precipitation, membrane separation, and evaporation. Furthermore, there is a reversible and consistent material constraint between fluoride ions and the two types of complexed fluorides. Therefore, the true trend of fluoride ion changes after masking and precision calibration, which has a higher time resolution, can be used to perform feedforward consistency and material balance closure checks on the results of low-frequency sampling of the two types of complexed fluorides. In the event of occasional deviations such as injection baseline drift or chromatographic peak overlap, a small correction can be made according to the stoichiometric relationship. At the same time, process control targets (such as the reduction targets in the precipitation or extraction stages) first directly affect free fluoride ions. Based on the consistency correction of the fluoride ion sequence, it can eliminate false differences introduced by asynchronous sampling and detection noise, and make the classification judgments of normal, disturbance, and abnormal based on the same chemical caliber and material constraints, thereby improving the accuracy and auditability of recovery status identification.

[0046] In the process of fluoride recovery from fluoride-containing wastewater, simultaneous monitoring of the concentrations of fluoride ions, fluorosilicate, and fluoroborate is both necessary and complementary: First, fluoride ions directly reflect the removal and migration of free fluoride, serving as the primary indicator for real-time control of precipitation, extraction, membrane separation, and evaporation. Second, fluorosilicate and fluoroborate represent complexed fluoride sources, significantly affected by pH, ionic strength, and reaction ratios. They may be released as free fluoride under conversion conditions or selectively recovered as byproduct crystals, thus determining the upper limit of recoverable total fluoride and the byproduct yield. Third, the combination of these three components forms a material closure constraint for total fluoride, used to verify the consistency of metering in each stage, identify detection deviations and sampling asynchrony, and establish a unified standard for calculating recovery rate, removal rate, energy consumption, and reagent consumption. Fourth, data on different fluoride forms can also pinpoint process bottlenecks, such as insufficient extraction, inadequate precipitation, or deviations in evaporation crystallization purity, providing actionable decision-making basis for reagent setting, ratio distribution, separation stage selection, and cleaning and maintenance. In summary, the coordinated monitoring of the three concentrations not only supports the accurate determination and traceability of the recovery status, but also improves the reliability of process optimization and compliance management.

[0047] Physical interference refers to the physical interference caused by microbubbles and flocculent impurities present in the system when sampling fluoride-containing wastewater.

[0048] The system collects and monitors the real-time differential pressure signal amplitude and the real-time acoustic signal amplitude within the monitoring window.

[0049] The differential pressure signal amplitude refers to the instantaneous pressure difference between two pressure taps at any given moment within the monitoring window. It is used to characterize the resistance pulsations and flow fluctuations caused by microbubbles, flocs, gas-liquid blockages, etc., within the flow cell or pipeline. The measurement method involves setting up upstream and downstream pressure taps in the process section, connecting them to a differential pressure transmitter for continuous sampling, and directly outputting the instantaneous value after anti-aliasing filtering and zero-point calibration. The pressure taps must be coaxial, with equal-length pressure guide tubes and no air accumulation. The sensor range and accuracy must match the actual operating conditions, and temperature compensation and calibration must be performed periodically. This signal can be acquired synchronously at various stages, including key locations such as influent, post-extraction, post-precipitation, post-membrane treatment, and condensate outlet. Pressure taps and differential pressure sensors are arranged in the corresponding pipe sections or flow cells to achieve real-time differential pressure monitoring throughout the entire process.

[0050] Real-time acoustic signal amplitude refers to the instantaneous vibration intensity caused by acoustic disturbances within the medium at any given moment within the monitoring window. It is used to indicate the instantaneous degree of phenomena such as microbubble rupture, two-phase agglomeration, and turbulent pulsation. Measurement is performed by attaching or clamping piezoelectric acoustic sensors to the pipe wall or the outer wall of the flow cell, or by using an ultrasonic transducer with integrated transmitter and receiver operating continuously at a fixed frequency. After bandpass filtering and envelope extraction (or rectification and smoothing) of the original vibration signal, the instantaneous amplitude is directly output. Installation must ensure good coupling, use a coupling medium, avoid high-noise areas such as bends and valves, and perform regular sensitivity checks and background baseline updates. This signal can also be acquired synchronously at all stages of the entire process, including influent, post-extraction, post-precipitation, post-membrane treatment, and condensate outlet. By deploying acoustic sensors at corresponding locations, real-time monitoring of acoustic disturbances at each stage can be achieved.

[0051] If normal data conditions exist at a certain time point within the monitoring window, the fluoride ion concentration corresponding to that time point will be labeled as normal. The core logic is to accurately match the stable detection requirements of high-concentration fluoride wastewater sampling scenarios. When both signals are less than or equal to the corresponding threshold values, it indicates that there are no obvious physical impurities such as microbubbles and flocs in the flow tank and pipeline, the fluid flow is stable, and the detection environment of the fluoride ion selective electrode is undisturbed. At this time, the fluoride ion concentration data can truly reflect the actual working conditions of the wastewater, so it is labeled as normal.

[0052] If data disturbances exist at a certain time point within the monitoring window, the corresponding fluoride ion concentration at that time point will be labeled with a data disturbance tag. In high-concentration fluoride wastewater sampling, microbubbles and flocs may only cause single-dimensional changes in the detection environment. When only the real-time differential pressure signal amplitude exceeds the corresponding threshold, it is mostly due to impurities causing fluctuations in pipeline flow resistance but not directly interfering with the electrode response. When only the real-time acoustic signal amplitude exceeds the corresponding threshold, it is mostly due to impurities colliding with the pipeline or electrode surface but not affecting fluid stability. Such situations only cause slight data distortion and do not reach the level of serious anomalies. Therefore, data disturbance tags are added to provide a clear basis for subsequent gradient weighting processing and avoid minor interference being misjudged as serious anomalies.

[0053] If abnormal data conditions exist at a certain time point within the monitoring window, the corresponding fluoride ion concentration at that time point will be labeled with an abnormal data tag. When the amplitudes of two signals simultaneously exceed the corresponding threshold values, it indicates the presence of numerous microbubbles, flocs, or gas-containing liquid plugs in the flow cell and pipeline. This causes drastic fluctuations in fluid velocity and directly damages the electrode detection environment, leading to problems such as discontinuous conductive paths and potential drift. In this case, the fluoride ion concentration data is severely distorted and cannot reflect the true operating conditions. Therefore, labeling the data with an abnormal data tag facilitates the subsequent direct assignment of the lowest weight to remove such data, fundamentally preventing distorted data from affecting the accuracy of fluoride salt concentration correlation data calculation.

[0054] Normal data conditions refer to the real-time differential pressure signal amplitude being less than or equal to the defined differential pressure signal amplitude, and the real-time acoustic signal amplitude being less than or equal to the defined acoustic signal amplitude.

[0055] Define the differential pressure signal amplitude, which refers to the maximum allowed value of the differential pressure signal amplitude preset in the database; define the acoustic signal amplitude, which refers to the maximum allowed value of the acoustic signal amplitude preset in the database.

[0056] Data disturbance conditions refer to either data disturbance condition one or data disturbance condition two. Data disturbance condition one means that the amplitude of the real-time differential pressure signal is greater than the amplitude of the defined differential pressure signal, and the amplitude of the real-time acoustic signal is less than or equal to the amplitude of the defined acoustic signal. Data disturbance condition two means that the amplitude of the real-time differential pressure signal is less than or equal to the amplitude of the defined differential pressure signal, and the amplitude of the real-time acoustic signal is greater than the amplitude of the defined acoustic signal.

[0057] Abnormal data conditions refer to situations where the amplitude of the real-time differential pressure signal is greater than the amplitude of the defined differential pressure signal, and the amplitude of the real-time acoustic signal is greater than the amplitude of the defined acoustic signal.

[0058] After the fluoride ion concentration sequence classification is completed, the classification data are integrated.

[0059] Furthermore, after the fluoride ion concentration sequence is classified, the classified data is integrated. The specific integration process is as follows: based on the time axis within the monitoring window, adjacent fluoride ion concentration sequences with the same label are merged into continuous time segments, that is, discrete data points that are continuous in time and have the same label type are merged into a continuous data segment.

[0060] The final result is several non-overlapping continuous time periods, namely, normal data period, data disturbance period, and abnormal data period. The start time of each period is the time when the corresponding label first appears, and the end time is the time when the corresponding label last appears consecutively. The fluoride ion concentration sequence within each period maintains the same label attribute.

[0061] After the classification data is integrated, deviations caused by physical interference are corrected to remove abnormal data.

[0062] In one example embodiment, the time range of a certain monitoring window is set to 0~100s. According to the aforementioned classification rules, the labels for each time point are as follows: 0~20s are all normal labels, 21~45s are all data disturbance labels, 46~60s are all normal labels, 61~85s are all data abnormal labels, and 86~100s are all data disturbance labels.

[0063] After merging adjacent discrete data points with the same label in chronological order, five non-overlapping consecutive time periods are obtained: normal data period (start 0s, end 20s), data disturbance period (start 21s, end 45s), normal data period (start 46s, end 60s), abnormal data period (start 61s, end 85s), and data disturbance period (start 86s, end 100s). The fluoride ion concentration sequence in each time period maintains the corresponding unified label attribute. Based on this integration result, the deviation correction of physical interference abnormal data can be accurately carried out.

[0064] Specifically, after the classification data is integrated, deviation correction is carried out to remove abnormal data caused by physical interference. The specific correction process is as follows: for any two adjacent time periods of the same label type, if the interval between adjacent time periods is less than or equal to the first defined time period, the time period merging strategy is executed.

[0065] The time period merging strategy refers to using the earliest start time point of two adjacent time periods as the start time point of the merged target time period of the same type, and the latest end time point as the end time point of the merged target time period of the same type, thereby merging them into a continuous target time period of the same type.

[0066] After the time period merging strategy is executed, several normal data periods, data disturbance periods, and data abnormal periods are updated; if the duration of a certain period is less than or equal to the second defined duration, all data corresponding to that period is discarded; wherein, the first defined duration is less than the second defined duration.

[0067] In one example embodiment, a first defined duration (i.e., the threshold for merging similar time periods) is set to 2 seconds, and a second defined duration (i.e., the threshold for discarding time periods) is set to 5 seconds. After initial merging, a certain monitoring window forms three consecutive time periods without intervals: a normal data period (0~18s), a data disturbance period (18~21s) (lasting 3 seconds), and a normal data period (21~27s). According to the rules, the time period merging strategy is executed first: the previous normal data period ends at 18s, and the next normal data period begins at 21s. The interval duration = 21s - 18s = 3s > the first defined duration of 2s, which does not meet the merging condition. After executing the time period merging strategy, there are still three consecutive time periods without intervals.

[0068] After implementing the time period merging strategy, the duration of the data disturbance period is 3 seconds, which is less than the second defined duration, so data from 18 to 21 seconds is discarded.

[0069] The first timeframe is shorter than the second timeframe because the former is used to determine whether short intervals between similar time periods need to be merged (to avoid splitting valid data of the same type into short intervals), while the latter is used to filter invalid time periods with excessively short durations (to remove fragmented data with no practical reference value). The difference in thresholds between the two can achieve a logical closed loop of first integrating valid data and then removing invalid data. The time period merging strategy is executed only once to avoid multiple mergings leading to blurred time period boundaries and distorted data correlation, ensuring that the merging results match the original time distribution characteristics of the data within the monitoring window. The core purpose of merging is to integrate similar time periods separated by short intervals into continuous valid data segments, reducing the impact of short-term physical interference on data continuity in the monitoring of high-concentration fluoride wastewater and ensuring the integrity of concentration calculation. The core purpose of discarding is to remove fragmented time periods with excessively short durations. These time periods are mostly caused by interference from instantaneous microbubbles, flocs, etc., and the data is not representative. Removing them can avoid them interfering with the overall concentration calculation accuracy, ultimately providing continuous and reliable basic data for deviation correction.

[0070] By adopting a strategy of unconditionally merging short-term segments of the same type (intervals shorter than the first defined duration, without distinguishing interval labels), scattered fragments can be restored into continuous events that reflect the overall trend, significantly reducing timeline fragmentation and label jitter caused by transient noise. The merged continuous segments can more realistically reflect the macro-operating conditions of that period, so that statistics such as material balance, weighted average, and rate of change are no longer skewed by brief breakpoints, and key indicators such as recovery rate / removal rate and health are thus more stable and comparable.

[0071] By merging two adjacent time periods, each shorter than the second defined duration, into a single segment under the rule of including the interval, the total duration after merging exceeds the threshold. This provides a more objective portrayal of the overall trend and operational occupancy of the time period, avoiding the misjudgment of homogeneous processes fragmented by short intervals as unrepresentative events. The advantages are: first, it restores effective coverage of disturbances or steady states on the time axis, reducing the underestimation of duration and intensity caused by fragmentation, thus lowering missed detections and false negatives; second, it provides a continuous observation window for statistics such as material balance, weighted average, and rate of change, making recovery rate and health assessments more robust and closer to the actual process load; and third, it aligns with the hysteresis / de-jittering strategy on the control side, suppressing frequent starts and stops and parameter fluctuations, and improving the certainty of the linkage strategy.

[0072] The effective sampling rate of fluoride ion concentration is obtained and compared with the preset threshold sampling rate. If the effective sampling rate of fluoride ion concentration is less than or equal to the threshold sampling rate, data resampling is performed; otherwise, accuracy calibration is completed by combining offline calibration data.

[0073] The sampling rate is defined as the minimum allowable effective sampling rate of fluoride ion concentration preset in the database.

[0074] The effective sampling rate of fluoride ion concentration refers to the ratio of the total number of all fluoride ion concentration sampling points within the valid time period that were not discarded to the total number of original sampling points for fluoride ion concentration within the monitoring window.

[0075] Data resampling refers to re-triggering the synchronous acquisition of fluoride ion concentration, real-time differential pressure signal, and real-time acoustic signal based on the original monitoring window time span. During the acquisition process, the monitoring equipment parameters and sampling frequency are kept consistent with the original monitoring process to ensure the comparability of the resampling data with the original data. After the resampling is completed, the newly acquired data is processed according to the aforementioned data classification, time period merging, and short-term invalid data discarding rules, and the effective sampling rate is recalculated. If the effective sampling rate still does not meet the requirements after resampling, technicians can check the interference factors in conjunction with the on-site working conditions and repeat the above resampling process until the effective sampling rate meets the standard to ensure the reliability of subsequent fluoride salt concentration correlation data calculation.

[0076] Furthermore, accuracy calibration is completed by combining offline calibration data. The specific calibration process is as follows: obtain the sequence of fluoride ion concentration changes within the monitoring window.

[0077] The range of fluoride ion concentration variation was obtained from offline calibration data to determine the reasonable fluctuation range of fluoride ion concentration variation.

[0078] Using the range of fluoride ion concentration variation as the accuracy constraint benchmark, the generated fluoride ion concentration variation sequence is verified and corrected point by point.

[0079] For abnormal fluctuation values ​​that do not fall within the range of fluoride ion concentration fluctuations, calibration is performed based on the range of fluoride ion concentration fluctuations. For fluctuation values ​​of fluoride ion concentration that fall within the range of fluoride ion concentration fluctuations, the original data is retained.

[0080] Finally, the accuracy calibration of the fluoride ion concentration sequence was completed.

[0081] The change in fluoride ion concentration refers to the difference between two consecutive sampling concentrations (which can be positive or negative).

[0082] In one example embodiment, the initial concentration array within the monitoring window is [1.00, 1.01, 1.19, 1.54, 1.52, 1.22, 1.32] mg / L, and the corresponding variation range array is [+0.01, +0.18, +0.35, -0.02, -0.30, +0.10] mg / L·step. The reasonable variation range given by offline calibration is [-0.20, +0.20] mg / L·step. Point-by-point verification and correction along the original direction: Point-by-point verification and truncation along the interval boundaries along the original direction: If +0.01 belongs to the fluoride ion concentration variation range, it is retained; +0.35 is greater than the upper limit of the fluoride ion concentration variation range, so the upper limit is truncated to +0.20; -0.30 is less than the lower limit of the fluoride ion concentration variation range, so the lower limit is truncated to -0.20. This process is repeated to obtain the calibrated variation range array [+0.01, +0.18, +0.20, -0.02, -0.20, +0.10]. Starting with an initial concentration of 1.00, the calibrated concentration array [1.00, 1.01, 1.19, 1.39, 1.37, 1.17, 1.27] is obtained; thus, the accuracy calibration of the fluoride ion concentration sequence within this monitoring window is completed.

[0083] Limiting the concentration variation range of adjacent sampling points and correcting it along the interval boundary essentially reduces the peak and limits the rate of instantaneous spurious fluctuations caused by microbubbles, flocs, and gas-liquid plugs at the data level, allowing the sequence changes to follow the true kinetic upper limit of sampling-mixing-transmission, thereby suppressing the interference of sudden increases / decreases unrelated to the true fluoride ion activity on subsequent calculations and control. This approach achieves several advantages: First, it significantly reduces the bias of anomalies caused by discontinuous conductive paths, boundary layer anomalies, reference node drift, and the superposition of additional resistance voltage drop and flow potential on the mean, trend, and anomaly determination, ensuring the usability and comparability of statistics within the monitoring window. Second, it avoids amplifying transient two-phase disturbances into erroneous sudden changes in operating conditions, reducing over-adjustment of extraction, precipitation, and membrane separation stages, and preventing misleading addition of alkaline back-extraction agents leading to over-dosing or oscillations. Third, it stabilizes material balance and the calculation of key indicators such as recovery and removal rates, improving the reliability of compliance determination. Fourth, compared to simply discarding entire data segments, amplitude-based edge-fitting retains more effective information, increases the effective sampling rate, and provides a definitive correction caliber for subsequent traceability and auditing, meeting the long-term stable monitoring needs of high-concentration fluorine-containing and alkaline systems.

[0084] Specifically, the optimized fluoride concentration correlation data includes the optimized fluoride ion concentration sequence, the mean fluorosilicate ion concentration, and the mean fluoroborate ion concentration.

[0085] Based on the fluoride ion concentration sequence after precision calibration, a fluoride ion concentration mask array is generated synchronously according to the time dimension within the monitoring window. The index of this array corresponds one-to-one with the time node of the fluoride ion concentration sequence.

[0086] The weight values ​​of the mask array are assigned differently: for the fluoride ion concentration sequence during normal data periods, the weight value of the corresponding position in the mask array is assigned to 1; for the fluoride ion concentration sequence during data disturbance periods, the weight value of the corresponding position in the mask array is gradient-reduced based on the real-time differential pressure signal amplitude and the real-time acoustic signal amplitude; for the fluoride ion concentration sequence during abnormal data periods, the weight value of the corresponding position in the mask array is assigned to the lowest weight value preset in the database.

[0087] Based on the fluoride ion concentration sequence and the corresponding mask weight array, the total weighted value is obtained by weighted summation, and then divided by the sum of the weighted values ​​to finally obtain the weighted mean of the fluoride ion concentration.

[0088] The gradient reduction process involves applying gradient reduction to the weights at corresponding positions in the mask array based on the real-time differential pressure signal amplitude and the real-time acoustic signal amplitude. This includes integrating the real-time differential pressure signal amplitude during the data disturbance period (marked as the differential pressure disturbance value) and integrating the real-time acoustic signal amplitude during the same period (marked as the acoustic disturbance value). The maximum allowable values ​​for both differential pressure and acoustic disturbances are retrieved from the database, and a first reduction coefficient G is set to... JYR is the maximum allowable value of differential pressure disturbance, YR is the differential pressure disturbance value, and the second reduction factor K is... JSX is the maximum allowable value of acoustic disturbance, SX is the acoustic disturbance value. If the first reduction coefficient G is less than 1, then the result of multiplying 1 by G is used as the reduced weight value. If the second reduction coefficient K is less than 1, then the result of multiplying 1 by K is used as the reduced weight value. If both the first reduction coefficient G and the second reduction coefficient K are less than 1, then the result of multiplying 1 by G and then by K is used as the reduced weight value.

[0089] The core advantage of this gradient weighting method is its ability to accurately match the actual impact of physical interference in the monitoring of high-concentration fluoride wastewater, and to specifically address the data disturbance caused by impurities such as microbubbles and flocs. By integrating the real-time differential pressure signal amplitude and acoustic signal amplitude during the data disturbance period, the differential pressure disturbance value and acoustic disturbance value are quantified. Then, a reduction coefficient is calculated based on the corresponding maximum allowable value preset in the database, so that the weight value is directly linked to the disturbance intensity. The more severe the disturbance (the closer the integral result is to the maximum allowable value, the smaller the coefficient), the lower the weight, which can effectively weaken the interference of strong disturbances on concentration calculation. At the same time, through the calculation of multiple coefficient superposition, it accurately covers different disturbance scenarios such as single signal exceeding the standard and dual signal exceeding the standard, avoiding the weight allocation deviation caused by single-dimensional judgment. This quantitative weighting logic not only abandons the traditional coarse-grained processing of fixed weighting, but also accurately distinguishes the differences in the degree of disturbance, ensuring that the contribution of the disturbance data matches the actual credibility, thereby reducing the error in fluoride ion concentration calculation caused by physical interference, providing a more realistic basis for subsequent optimization of fluoride salt concentration correlation data, and ensuring the accuracy of process control.

[0090] In one example embodiment, a monitoring window time range is set to 1~10s, and the sampling frequency is 1 time / s. After precision calibration, the fluoride ion concentration sequence is: [10.2mg / L, 10.3mg / L, 10.5mg / L, 12.1mg / L, 11.8mg / L, 10.4mg / L, 9.9mg / L, 10.1mg / L, 10.3mg / L, 10.2mg / L]. A 10-bit mask array is generated simultaneously (index 0~9 corresponds to time 1~10s). After classification and integration, the time periods are divided as follows: normal data period (1~3s, 7~10s), data disturbance period (4~5s, real-time differential pressure signal amplitude exceeds the standard), and abnormal data period (6s, both signals exceed the standard). The weighting rules are as follows: during normal periods, the weight is 1; during periods of disturbance, the weight is reduced to 0.6 or 0.5 based on the magnitude of the exceedance; during periods of abnormality, the minimum weight is preset to 0.1. The final mask array is: [1, 1, 1, 0.6, 0.5, 0.1, 1, 1, 1, 1]. Weighted calculation process: Total weighted value = (10.2×1) + (10.3×1) + (10.5×1) + (12.1×0.6) + (11.8×0.5) + (10.4×0.1) + (9.9×1) + (10.1×1) + (10.3×1) + (10.2×1) = 85.7; Total weighted value = 1 + 1 + 1 + 0.6 + 0.5 + 0.1 + 1 + 1 + 1 + 1 = 8.2; Final weighted average fluoride ion concentration = 85.7 ÷ 8.2 ≈ 10.45 mg / L.

[0091] The initial mean value of fluoride ion concentration was obtained and compared with the weighted mean value of fluoride ion concentration to determine the direction and magnitude of fluoride ion concentration change. The obtained mean values ​​of fluorosilicate ion concentration and fluoroboric acid ion concentration were then corrected. Finally, the optimized correlation data of fluoride salt concentration was obtained.

[0092] The initial mean of fluoride ion concentration refers to the result of directly averaging the fluoride ion concentration sequence before bias correction.

[0093] In another example embodiment, assuming the initial average fluoride ion concentration is 10.68 mg / L and the weighted average fluoride ion concentration is 10.42 mg / L, the change characteristics are compared and found to be: the direction of change is decreasing, and the change magnitude = |10.42-10.68|÷10.68×100%≈2.43%; then the average concentrations of fluorosilicate ions and fluoroborate ions both decrease by 2.43%.

[0094] Based on the direction (decreasing) and magnitude (2.44%) of the fluoride ion concentration change, the mean concentrations of fluorosilicate and fluoroboronic acid ions were simultaneously corrected by the same magnitude. The core principle is to align with the process scenario of fluoride-containing wastewater treatment and the characteristics of the total fluoride system: all three belong to the total fluoride system, and under unified processes such as extraction and precipitation, they follow consistent material constraints. Furthermore, there is a reversible relationship between fluoride ions and the two types of complexed fluorides, resulting in a strong correlation in their concentration change trends. Fluoride ion concentration was obtained through a fluoride ion selective electrode, offering higher temporal resolution. After masking and precision calibration, the weighted mean more closely reflects actual operating conditions. In contrast, the concentrations of fluorosilicate and fluoroboronic acid ions rely on low-frequency sampling methods such as ion chromatography or potentiometric titration, which are susceptible to occasional deviations due to factors such as injection baseline drift and peak overlap. By comparing the initial average value of fluoride ions before deviation correction with the weighted average value after correction, the actual change range can be obtained, which can accurately reflect the actual concentration fluctuation of the total fluoride system under the action of the process. In this way, the average concentrations of the two types of complexed fluoride can be corrected in the same trend and with the same magnitude. This can not only achieve material balance closure of the total fluoride system, but also correct the local deviation caused by low-frequency sampling, ensuring that the data related to the three types of fluoride are consistent with the actual effect of the process. This provides logically consistent and accurate correlation data that fits the operating conditions for subsequent fluoride salt recovery process control.

[0095] Based on the optimized fluoride concentration correlation data, the fluoride recovery status of each treatment stage of fluoride-containing wastewater is monitored and determined, thereby providing feedback on the fluoride recovery status results. At the same time, the fluoride recovery status results are uploaded to the blockchain node to complete data storage.

[0096] Specifically, the monitoring and determination of fluoride recovery status at each stage of fluoride-containing wastewater treatment is carried out. The specific monitoring process is as follows: after the optimization of fluoride concentration correlation data, the initial stage fluoride concentration correlation data and the final stage fluoride concentration correlation data are used for each stage.

[0097] Obtain the fluoride ion recovery rate, fluorosilicate ion recovery rate, and fluoroboronic acid ion recovery rate for each stage. The fluoride ion recovery rate is calculated by subtracting the weighted average fluoride ion concentration of the final stage from the weighted average fluoride ion concentration of the initial stage, and then dividing by the weighted average fluoride ion concentration of the initial stage. The fluorosilicate ion recovery rate is calculated by subtracting the optimized average fluorosilicate ion concentration of the final stage from the optimized average fluorosilicate ion concentration of the initial stage, and then dividing by the optimized average fluorosilicate ion concentration of the initial stage. The fluoroboronic acid ion recovery rate is calculated by subtracting the optimized average fluoroboronic acid ion concentration of the final stage from the optimized average fluoroboronic acid ion concentration of the initial stage, and then dividing by the optimized average fluoroboronic acid ion concentration of the initial stage.

[0098] Fluoride recovery status refers to the recovery rates of fluoride ions, fluorosilicate ions, and fluoroboronic acid ions.

[0099] The recovery status of fluoride salts is clearly defined as fluoride ion recovery rate, fluorosilicate ion recovery rate, and fluoroboronic acid ion recovery rate. This approach closely aligns with the actual scenario of total fluoride resource recovery in fluoride-containing wastewater treatment, specifically addressing the technical problem that traditional single-indicator monitoring cannot comprehensively reflect the recovery effect. Fluoride in fluoride-containing wastewater mainly exists in three forms: fluoride ions, fluorosilicate ions, and fluoroboronic acid ions. These three forms are constrained by the same material in recovery processes such as extraction and precipitation, and exhibit a reversible relationship. The recovery rate of a single form cannot reflect the overall recovery efficiency of total fluoride resources. Furthermore, the separation difficulty and recovery characteristics of the three types of ions differ. Monitoring their recovery rates separately allows for precise identification of the root cause of substandard recovery of a particular form of fluoride resource (such as distorted recovery calculations due to chromatographic detection errors in fluorosilicate ions), avoiding misjudgments of the overall process effect based on a single indicator. This multi-dimensional monitoring method not only ensures the integrity and accuracy of the total fluoride recovery rate calculation, but also provides precise and targeted basis for process control (such as optimizing the extractant ratio for low fluoroboric acid ion recovery rate). It effectively solves the problems of one-sided monitoring and blind control in traditional recovery status monitoring, and ensures the stability, efficiency and maximum utilization of resources in the fluoride recovery process.

[0100] Furthermore, feedback on the fluoride recovery status is conducted. The specific feedback process involves extracting the reference recovery rate ranges for fluoride ion concentration, fluorosilicate ion concentration, and fluoroboronic acid ion concentration for each stage from the database and comparing the data.

[0101] The formulation of reference recovery rate ranges for fluoride ion concentration, fluorosilicate ion concentration, and fluoroboronic acid ion concentration at each stage is closely aligned with the process segmentation characteristics and total fluoride system recovery patterns of fluoride-containing wastewater treatment. First, based on the initial fluoride form composition of the fluoride-containing wastewater (such as the initial proportions of fluoride ions and the two types of complexed fluorides) and process design parameters (such as extractant ratio, reaction temperature, residence time, etc.), actual recovery data of the three types of fluoride forms at each stage under different operating conditions are collected through multiple small-scale and pilot-scale experiments to screen out the optimal recovery range for stable process operation. The effective recovery records from historical operating data are then incorporated, and combined with the principle of material balance (total fluoride input = recovery of each form of fluoride + residual amount), after removing abnormal and disturbing data, statistical analysis is performed to determine the benchmark recovery values ​​of the three forms of fluoride at each stage. Finally, taking into account actual influencing factors such as on-site water quality fluctuations and equipment operating errors, reasonable upper and lower fluctuation ranges are set to form the reference recovery rate ranges corresponding to each stage and store them in the database. This provides a scientific and condition-appropriate basis for subsequent real-time recovery status comparison, ensuring the accuracy of recovery effect evaluation and the targeted nature of process control.

[0102] If the comparison result meets the preset stage stop condition, the current stage is determined to be completed, triggering the next stage start command, thus realizing the stage-by-stage feedback of the recycling process.

[0103] The stage-based stop conditions are quantitative judgment criteria set based on the process objectives and recovery efficiency requirements of each stage. The core is to ensure that the recovery effect of the current stage meets the standard before proceeding to the next stage, avoiding resource waste or incomplete recovery caused by premature process progression. Specifically, the actual measured recovery rates of fluoride ions, fluorosilicate ions, and fluoroboronic acid ions in the current stage must all fall within the corresponding reference recovery rate range preset in the database (e.g., if the reference recovery rate range is 90%~95%, the actual recovery rate must be within this range); and the duration of this compliant state must be greater than or equal to the set stabilization time (e.g., 30 seconds).

[0104] The command to trigger the next stage of startup includes three core aspects: First, the process parameter switching command, which automatically adjusts key operating parameters such as extractant ratio, reactor temperature, stirring rate, membrane separation pressure, and evaporation intensity according to the preset process plan for the next stage; second, the equipment linkage control command, which triggers the startup of the corresponding processing unit in the next stage (such as starting the deep filtration equipment or switching the ion exchange column path), while shutting down or switching some redundant equipment in the current stage according to the process logic; and third, the data acquisition synchronization command, which updates the parameter acquisition thresholds and data processing rules in the monitoring window (such as adjusting the fluoride ion concentration threshold signal amplitude and updating the reference range for the recovery rate of the three types of ions) after the command is issued, ensuring that the monitoring data in the next stage is accurately matched with the process status.

[0105] If the comparison results meet the preset stage anomaly conditions, the anomaly warning mechanism will be triggered immediately, and the anomaly occurrence stage, anomaly type, and corresponding recovery rate deviation data will be recorded simultaneously.

[0106] Stage-specific abnormal conditions refer to risk assessment standards set for sudden situations such as process operation failures and water quality fluctuations. The core is to identify abnormal recovery efficiency in a timely manner and avoid the failure from escalating to the point that the total fluoride recovery does not meet the standard. Specifically, it means that the difference between the actual recovery rate of any type of fluoride in the current stage and the lower limit of the reference recovery rate range is greater than or equal to the low abnormality threshold (e.g., 5% or more below the lower limit), or the difference between the actual recovery rate of any type of fluoride in the current stage and the upper limit of the reference recovery rate range is greater than or equal to the high abnormality threshold (e.g., 5% or more above the upper limit).

[0107] If the comparison results do not show any stage termination conditions or stage abnormal conditions, then the reaction process of that stage will be continuously monitored.

[0108] Figure 2 This is a schematic diagram of the structure of a blockchain-based fluoride recovery status monitoring system for fluoride-containing wastewater provided in an embodiment of the present invention. The second aspect of the present invention, a blockchain-based fluoride recovery status monitoring system for fluoride-containing wastewater, includes a data acquisition and storage module, a data accuracy calibration module, a status result feedback module, and a database.

[0109] The data acquisition and storage module is connected to the data accuracy calibration module, which in turn is connected to the status result feedback module. All three modules—data acquisition and storage, data accuracy calibration, and status result feedback—are connected to the database.

[0110] The database stores parameters involved in the blockchain-based monitoring system for fluoride recovery in fluoride-containing wastewater. Its development process closely aligns with the entire process of fluoride recovery in fluoride-containing wastewater and the evidence-keeping characteristics of blockchain. Specifically: First, the core parameter categories required for system operation are identified, covering process design parameters (extractant ratio ranges at each stage, reaction condition thresholds, etc.) and monitoring and calibration parameters (defining differential pressure / acoustic signal amplitude, mask weight assignment rules, etc.). Then, based on the process verification results from pilot and intermediate-scale tests, initial parameter ranges suitable for different water quality conditions and ensuring monitoring accuracy and recovery efficiency are selected. Next, adhering to the immutability requirements of blockchain data, a parameter storage structure is designed, dividing the data into independent blocks and associating them with timestamps and operation logs to ensure traceability of parameter modifications. Finally, through continuous collection of actual operating data during system trial operation, and combined with feedback on recovery effects, the parameter ranges and storage rules are iteratively optimized to form the final database that meets the requirements for automated monitoring and blockchain evidence-keeping in fluoride recovery from fluoride-containing wastewater.

[0111] The data acquisition and storage module is used to monitor and collect the initial and final fluoride concentration data for each stage, following the sequence of extraction, residual liquid alkali addition, and tailwater treatment. This data is then uploaded to the blockchain node to complete the raw data storage. Specifically, the initial stage of the extraction stage refers to the untreated state of the fluoride-containing wastewater before it enters the extraction equipment; the final stage refers to the state after the fluoride-containing wastewater and extractant have completed mixing and reaction, and the residual liquid has been discharged after settling. Similarly, the initial stage of the residual liquid alkali addition stage refers to the initial state of the residual liquid entering the alkali addition reaction equipment after extraction; and the final stage refers to the state after the staged alkali neutralization reaction is completed, solid-liquid separation is achieved, and the clear liquid is discharged. The initial stage of the effluent treatment process refers to the initial state of the clarified liquid entering the advanced treatment system after the residual liquid is alkali-treated. The final stage refers to the state of the effluent meeting the standards after calcium chloride precipitation and electrodialysis treatment. The data accuracy calibration module is used to correct the deviation of abnormal data caused by physical interference in the collected fluoride concentration correlation data. It then combines the data with offline calibration data to complete the accuracy calibration and obtain the optimized fluoride concentration correlation data. The status result feedback module is used to monitor and determine the fluoride recovery status of each treatment stage of fluoride-containing wastewater based on the optimized fluoride concentration correlation data, thereby providing feedback on the fluoride recovery status results and uploading the fluoride recovery status results to the blockchain node to complete data storage.

[0112] The parameters and data involved in this embodiment, such as the first definition duration, the second definition duration, the definition differential pressure signal amplitude, and the definition acoustic signal amplitude, are for illustrative purposes only. Their core function is to clearly explain the implementation logic and operation process of this technical solution. In actual industrial applications, the above parameters need to be comprehensively evaluated by technical personnel based on factors such as the characteristics of the fluoride-containing wastewater treatment process, the performance indicators of the monitoring equipment, and the actual situation of on-site interference factors. Accurate parameter values ​​suitable for specific operating conditions should be scientifically formulated to fully meet the actual control requirements of the fluoride recovery process.

[0113] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for monitoring the recovery state of fluorine salt in fluorine-containing wastewater based on a blockchain, characterized in that, The method includes: Following the sequence of extraction, residual liquid alkali addition, and tailwater treatment, the initial and final fluoride concentration data for each stage are monitored and collected sequentially, and then uploaded to the blockchain node to complete the original data storage. Specifically, the initial stage of the extraction stage refers to the untreated state of the fluoride-containing wastewater before it enters the extraction equipment, and the final stage refers to the state of the residual liquid after the fluoride-containing wastewater and extractant have completed the mixing reaction and settled into layers. The initial stage of the residual liquid alkali addition stage refers to the initial state of the residual liquid after extraction entering the alkali addition reaction equipment, and the final stage refers to the state of the clear liquid after the staged alkali addition neutralization reaction is completed and the solid-liquid separation is completed. The initial stage of the tailwater treatment stage refers to the initial state of the clear liquid after residual liquid alkali addition entering the deep treatment system, and the final stage refers to the state of the effluent meeting the standards after calcium chloride precipitation and electrodialysis treatment. The collected fluoride concentration correlation data were corrected to remove deviations caused by physical interference, and then the accuracy was calibrated by combining it with offline calibration data to obtain optimized fluoride concentration correlation data. The process of removing physical interference and correcting for deviations in the collected fluoride concentration correlation data also includes classifying the fluoride ion concentration sequences in the fluoride concentration correlation data. The specific classification process is as follows: The fluoride salt concentration correlation data includes fluoride ion concentration sequence, mean fluorosilicate ion concentration, and mean fluoroborate ion concentration. The physical interference refers to the physical interference caused by microbubbles and flocculent impurities present in the system when sampling fluoride-containing wastewater. The system collects and monitors the real-time differential pressure signal amplitude and the real-time acoustic signal amplitude within the monitoring window. If the data is within normal range at a certain time point within the monitoring window, then the fluoride ion concentration corresponding to that time point will be labeled as normal. If there is data disturbance at a certain time point within the monitoring window, then the fluoride ion concentration corresponding to that time point will be labeled with a data disturbance label; If there are abnormal data conditions at a certain time point within the monitoring window, the fluoride ion concentration corresponding to that time point will be labeled with an abnormal data label. The normal data conditions refer to the real-time differential pressure signal amplitude being less than or equal to the defined differential pressure signal amplitude, and the real-time acoustic signal amplitude being less than or equal to the defined acoustic signal amplitude. The data disturbance condition refers to satisfying either data disturbance condition one or data disturbance condition two. Data disturbance condition one means that the amplitude of the real-time differential pressure signal is greater than the amplitude of the defined differential pressure signal, and the amplitude of the real-time acoustic signal is less than or equal to the amplitude of the defined acoustic signal. Data disturbance condition two means that the amplitude of the real-time differential pressure signal is less than or equal to the amplitude of the defined differential pressure signal, and the amplitude of the real-time acoustic signal is greater than the amplitude of the defined acoustic signal. The aforementioned abnormal data conditions refer to the real-time differential pressure signal amplitude being greater than the defined differential pressure signal amplitude, and the real-time acoustic signal amplitude being greater than the defined acoustic signal amplitude. After the fluoride ion concentration sequence classification is completed, the classification data are integrated. Based on the optimized fluoride concentration correlation data, the fluoride recovery status of each treatment stage of fluoride-containing wastewater is monitored and determined, and the fluoride recovery status results are fed back and uploaded to the blockchain node to complete data storage. The optimized fluoride salt concentration correlation data specifically includes the optimized fluoride ion concentration sequence, the average fluorosilicate ion concentration, and the average fluoroboronic acid ion concentration. Based on the fluoride ion concentration sequence after precision calibration, a fluoride ion concentration mask array is generated synchronously according to the time dimension within the monitoring window. The index of this array corresponds one-to-one with the time node of the fluoride ion concentration sequence. The weight values ​​of the mask array are assigned differentially, specifically as follows: For the fluoride ion concentration sequence during periods of normal data, assign a weight of 1 to the corresponding position in the mask array; For the fluoride ion concentration sequence during the data perturbation period, gradient reduction processing is performed on the weight values ​​at the corresponding positions in the mask array based on the real-time differential pressure signal amplitude and the real-time acoustic signal amplitude. For the fluoride ion concentration sequence during periods of data anomalies, the weight value of the corresponding position in its mask array is assigned the preset minimum weight value; Based on the fluoride ion concentration sequence and the corresponding mask weight array, the total weighted value is obtained by weighted summation, and then divided by the sum of the weight values ​​to finally obtain the weighted mean of the fluoride ion concentration. The initial average value of fluoride ion concentration is obtained and compared with the weighted average value of fluoride ion concentration to obtain the direction and magnitude of the change in fluoride ion concentration. The obtained average values ​​of fluorosilicate ion concentration and fluoroboric acid ion concentration are then corrected. The optimized correlation data for fluoride salt concentration were finally obtained.

2. The blockchain-based monitoring method of claim 1, wherein, The initial stage fluoride concentration correlation data and the final stage fluoride concentration correlation data are monitored and collected sequentially for each stage, where each stage refers to the extraction stage, the residual liquid alkali addition stage, and the tailwater treatment stage. The extraction stage refers to the process where fluoride-containing wastewater is first fed into a mixer or stirred reactor along with an extractant prepared from functional extractant components and sulfonated kerosene or ethyl acetate diluent. The mixture is then reacted at a set temperature and stirring rate. After the reaction, the mixture is allowed to stand and separate into layers. The aqueous phase of the raffinate is sent to the subsequent alkali addition stage, while the enriched organic phase loaded with the target components enters the back-extraction section. It is then mixed and reacted with an alkaline solution of a set concentration that can ionize hydroxide ions. This process converts free fluoride in the organic phase into potassium fluoride, which is transferred to the aqueous phase, and converts fluoroborate ions into potassium fluoroborate precipitate. After back-extraction, the mixture is allowed to stand and separate into layers. The regenerated organic phase is then returned to the extraction stage for recycling after parameter adjustments, and the back-extraction liquid is sent to the separation and purification unit. The residual liquid alkali addition stage refers to the process of adding alkali to the raffinate after extraction. The total amount of alkali added is calculated based on its acidity, and the alkali solution is added in two stages. After the alkali reaction in the first stage, solid and liquid are separated, and the filter cake is washed and dried to obtain potassium fluoroborate by-product. In the second stage, alkali is added until the solution is neutral, so that the residual fluorosilicate ions react with potassium ions to form potassium fluorosilicate precipitate. After solid-liquid separation, washing and drying, potassium fluorosilicate by-product is obtained. The washing wastewater and the filtered clear liquid are sent to the corresponding treatment units respectively. The tailwater treatment stage refers to the process of adding calcium chloride to the filtered clear liquid after adding alkali to the raffinate to generate calcium fluoride sludge, separating it by plate and frame filter press, and then treating the filtered tailwater by electrodialysis. The clear water is reused after meeting the standards, the concentrated water from electrodialysis is evaporated, the condensate is reused as wastewater, and the miscellaneous salts are sent to the solid waste disposal center to achieve deep purification and resource recovery of wastewater.

3. The blockchain-based monitoring method of claim 1, wherein, After the fluoride ion concentration sequence classification is completed, the classification data is integrated. The specific integration process is as follows: Using the time axis within the monitoring window as the order, the fluoride ion concentration sequences that are directly adjacent and labeled with the same label are merged into a continuous time segment, that is, discrete data points that are continuous in time and have the same label type are merged into a continuous data segment. This results in several non-overlapping consecutive time periods, namely, normal data periods, data disturbance periods, and abnormal data periods; The start time of each period is the time when the corresponding label first appears, and the end time is the time when the corresponding label appears consecutively. The fluoride ion concentration sequence within each period maintains the same label attribute. After the classification data is integrated, deviations caused by physical interference are corrected to remove them.

4. The blockchain-based monitoring method of claim 3, wherein, After the classification data is integrated, a bias correction process is carried out to remove abnormal data caused by physical interference. The specific correction process is as follows: For any two adjacent time periods of the same tag type, if the interval between the adjacent time periods is less than or equal to the first defined interval, the time period merging strategy is executed. The time period merging strategy refers to taking the first occurrence of two adjacent time periods as the starting time point of the merged target time period of the same type, and taking the last occurrence of the ending time point as the ending time point of the merged target time period of the same type, thereby merging them into a continuous target time period of the same type. After the time period merging strategy is executed, several data normal periods, data disturbance periods, and data abnormal periods are updated. If the duration of a certain period is less than or equal to the second defined duration, then all data corresponding to that period should be discarded. The duration of the first definition is shorter than the duration of the second definition. The effective sampling rate of fluoride ion concentration is obtained and compared with the preset threshold sampling rate. If the effective sampling rate of fluoride ion concentration is less than or equal to the threshold sampling rate, data resampling is performed; otherwise, accuracy calibration is completed by combining offline calibration data.

5. The blockchain-based monitoring method of claim 4, wherein, The accuracy calibration is performed by combining offline calibration data. The specific calibration process is as follows: Obtain the sequence of fluoride ion concentration changes within the monitoring window; The range of fluoride ion concentration variation was obtained from offline calibration data to determine the reasonable fluctuation range of fluoride ion concentration variation. Using the range of fluoride ion concentration variation as the accuracy constraint benchmark, the generated fluoride ion concentration variation sequence is verified and corrected point by point. For abnormal fluctuation values ​​that do not fall within the range of fluoride ion concentration fluctuations, calibration is performed based on the range of fluoride ion concentration fluctuations. For fluctuation values ​​of fluoride ion concentration that fall within the range of fluoride ion concentration fluctuations, the original data is retained. Finally, the accuracy calibration of the fluoride ion concentration sequence was completed.

6. The blockchain-based monitoring method of claim 1, wherein, The monitoring and determination of fluoride recovery status at each stage of fluoride-containing wastewater treatment is specifically carried out as follows: After the optimization of the fluoride concentration correlation data is completed, the initial stage fluoride concentration correlation data and the final stage fluoride concentration correlation data are used for each stage. Obtain the fluoride ion recovery rate, fluorosilicate ion recovery rate and fluoroborate ion recovery rate for each stage. Fluoride recovery status refers to the recovery rates of fluoride ions, fluorosilicate ions, and fluoroboronic acid ions.

7. The blockchain-based monitoring method of claim 1, wherein, The feedback process for the fluoride recovery status is as follows: Obtain the reference recovery rate ranges for fluoride ion concentration, fluorosilicate ion concentration, and fluoroboronic acid ion concentration at each stage, and compare the data. If the comparison result meets the preset stage stop condition, the current stage is determined to be completed, and the next stage start command is triggered to realize the stage progressive feedback of the recycling process. If the comparison results meet the preset stage anomaly conditions, the anomaly warning mechanism will be triggered immediately, and the anomaly occurrence stage, anomaly type, and corresponding recovery rate deviation data will be recorded simultaneously.

8. A blockchain-based monitoring system for the recovery status of fluoride salts in fluoride-containing wastewater, employing the blockchain-based monitoring method for the recovery status of fluoride salts in fluoride-containing wastewater as described in any one of claims 1-7, characterized in that: include: The data acquisition and storage module is used to monitor and collect the initial and final fluoride concentration data of each stage in the order of extraction, residual liquid alkali addition, and tailwater treatment. The data is then uploaded to the blockchain node to complete the original data storage. Specifically, the initial stage of the extraction stage refers to the untreated state of the fluoride-containing wastewater before it enters the extraction equipment, and the final stage refers to the state of the residual liquid after the fluoride-containing wastewater and extractant have completed the mixing reaction and settled into layers. The initial stage of the residual liquid alkali addition stage refers to the initial state of the residual liquid after extraction entering the alkali addition reaction equipment, and the final stage refers to the state of the clear liquid after the staged alkali addition neutralization reaction is completed and the solid-liquid separation is completed. The initial stage of the tailwater treatment stage refers to the initial state of the clear liquid after residual liquid alkali addition entering the deep treatment system, and the final stage refers to the state of the effluent meeting the standards after calcium chloride precipitation and electrodialysis treatment. The data accuracy calibration module is used to correct deviations in the collected fluoride concentration correlation data by eliminating physical interference and abnormal data. It then combines the data with offline calibration data to complete the accuracy calibration and obtain optimized fluoride concentration correlation data. The status result feedback module is used to monitor and determine the fluoride recovery status of each treatment stage of fluoride-containing wastewater based on the optimized fluoride concentration correlation data, and then provide feedback on the fluoride recovery status results, uploading the fluoride recovery status results to the blockchain node to complete data storage.

Images