A method for treating fluorine-containing waste acid generated in a quartz sand pickling process

Abstract

The application provides a treatment and utilization method of fluorine-containing waste acid generated in a quartz sand pickling process, comprising the following steps: if the protective film coverage rate reaches a preset threshold, extracting a waste acid sample from a pickling tank, analyzing the fluorine ion concentration in the waste acid, and obtaining first fluorine content data; according to a third barrier agent formula, extracting a fluorine barrier agent, re-injecting the fluorine barrier agent into the pickling tank, controlling the temperature of the acid liquid, obtaining the size distribution data of the settled particles after the surface treatment of the quartz sand, and judging whether the particle size is lower than a preset threshold; if the particle size is lower than the preset threshold, analyzing the fluorine ion concentration in the waste acid after the circulation treatment, obtaining third fluorine content data, combining the waste acid viscosity measurement data, and determining the circulation treatment efficiency; according to the circulation treatment efficiency, obtaining the separation liquid recovery rate and the settled particle distribution data, and generating a second waste acid circulation scheme. The application realizes efficient circulation and utilization of the fluorine barrier agent, improves the pickling effect, reduces environmental pollution, and has significant economic and environmental benefits.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a method for treating and utilizing fluorine-containing waste acid generated in a quartz sand pickling process. Background Technology

[0002] Acid washing of quartz sand is a key process for preparing high-purity quartz materials, widely used in high-end manufacturing fields such as photovoltaics and semiconductors. The quality of these products directly impacts the performance of downstream industries. However, the fluorine-containing waste acid generated during the acid washing process, due to its high corrosiveness and environmental hazards, has become a pressing problem for the industry. The rational treatment and utilization of fluorine-containing waste acid is not only crucial for environmental protection but also significantly reduces production costs and improves resource utilization efficiency, making it a core element in promoting the green development of the quartz sand industry. Existing methods for treating fluorine-containing waste acid mainly include neutralization precipitation and distillation recovery. While neutralization precipitation is simple to operate, it generates a large amount of solid waste, increasing the risk of secondary pollution. Distillation recovery can partially recover the acid, but its high energy consumption and complex equipment make it difficult to promote. These methods have significant shortcomings in terms of waste acid treatment efficiency and resource recycling, making it difficult to balance economic efficiency and environmental protection. The core challenge in treating fluorine-containing waste acid lies in how to efficiently remove fluorine from the waste acid while reducing the amount of acid used. During the acid washing process, fluorine easily reacts with the surface of the quartz sand to form leachates, resulting in excessively high fluorine content in the waste acid, increasing the difficulty of subsequent treatment. Controlling the fluorine leaching rate can reduce the fluorine concentration in waste acid and simplify the treatment process. However, conventional pickling processes struggle to precisely control fluorine leaching, requiring the introduction of additional energy or chemical methods to accelerate the reaction. This often leads to increased acid consumption, further exacerbating the burden of waste acid treatment. Excessive acid consumption not only raises costs but also significantly increases the difficulty of fluorine containment and recovery due to the increased volume of waste acid. Therefore, how to reduce the fluorine leaching rate and acid consumption in quartz sand pickling by introducing controllable auxiliary energy and chemical containment methods, while simultaneously achieving the recycling of containment agents, has become a key issue in promoting the efficient treatment and resource utilization of fluorine-containing waste acid. Summary of the Invention

[0003] This invention provides a method for treating and utilizing fluorine-containing waste acid generated in a quartz sand pickling process, mainly comprising:

[0004] Microwaves are generated by a microwave generator and emitted onto the surface of quartz sand in the pickling tank to obtain data on the changes in the dissolution rate of surface oxides and metal ions, and to determine the first dissolution rate.

[0005] If the first dissolution rate is lower than the preset threshold v, the microwave frequency is adjusted, the acid temperature is controlled, the adjusted dissolution rate data is obtained, and the second dissolution rate is determined.

[0006] Based on the second dissolution rate, the fluorine barrier agent formulation is matched from the recycling database, the molecular structure fingerprint data is extracted, and the first barrier agent addition scheme is generated.

[0007] Using the first barrier agent addition scheme, a fluorinated barrier agent is injected into the pickling tank, the pH value of the acid solution is controlled, the formation status data of the temporary protective film on the surface of the quartz sand is obtained, and it is determined whether the coverage of the protective film reaches the preset threshold n.

[0008] If the protective film coverage reaches the preset threshold n, then a waste acid sample is extracted from the pickling tank, the fluoride ion concentration in the waste acid is analyzed, and the first fluoride content data is obtained.

[0009] Spectral scanning was performed on the first fluorine content data to obtain molecular vibrational modes and characteristic peak intensity data. Combined with fluorine atom coordination information, the molecular structure fingerprint of the first barrier agent was generated.

[0010] Based on the molecular structure fingerprint of the first barrier agent, the fluorine barrier agent is separated from the waste acid, and the purity of the separated phase and the viscosity of the waste acid are measured to determine whether the purity of the first separation is higher than the preset threshold c.

[0011] If the purity of the first separation is higher than the preset threshold c, a second spectral scan is performed on the separated fluorine barrier to obtain molecular vibration mode and characteristic peak intensity data, and to generate the molecular structure fingerprint of the second barrier.

[0012] Based on the molecular structure fingerprint of the second barrier agent, the fluorine barrier agent formula is extracted, re-injected into the pickling tank, the microwave generator is controlled to operate, and the data of impurity residue after the surface treatment of quartz sand is obtained to determine whether the pickling effect reaches the point where the residual amount is lower than the preset threshold m.

[0013] If the pickling effect reaches the point where the residual amount is lower than the preset threshold m, then the waste acid in the pickling tank is subjected to spectral scanning to obtain the data on the size distribution of settled particles and the recovery rate of the separated liquid, and a first waste acid recycling scheme is generated.

[0014] Based on the first waste acid recycling scheme, adjust the centrifugal separation parameters, obtain the fluoride ion concentration data in the recycled waste acid, and determine the second fluoride content;

[0015] Spectral scanning was performed on the second fluorine content data. The data was processed and combined with characteristic peak intensity analysis and fluorine atom coordination information to generate the molecular structure fingerprint of the third barrier agent and determine the formulation of the third barrier agent.

[0016] Based on the third barrier agent formula, the fluorine barrier agent is extracted, re-injected into the pickling tank, the acid temperature is controlled, and the sedimentation particle size distribution data after the quartz sand surface treatment is obtained to determine whether the particle size is lower than the preset threshold d.

[0017] If the particle size is lower than the preset threshold d, the concentration of fluoride ions in the waste acid after recycling is analyzed to obtain the tertiary fluoride content data. Combined with the waste acid viscosity measurement data, the recycling efficiency is determined.

[0018] Based on the recycling efficiency, data on the recovery rate of the separated liquid and the distribution of settled particles are obtained to generate a second waste acid recycling scheme.

[0019] The above-mentioned method for treating and utilizing fluorine-containing waste acid generated in a quartz sand pickling process includes the following steps: First, a barrier agent is added to the pickling tank, a fluorine barrier agent is injected, the pH value of the acid solution is controlled, data on the formation state of a temporary protective film on the quartz sand surface is obtained, and it is determined whether the protective film coverage reaches a preset threshold n.

[0020] The real-time pH value data of the acid solution in the pickling tank is obtained by using sensors to obtain the pH value fluctuation range;

[0021] If the pH value deviates from the preset range, the injection rate of the fluoride barrier agent is adjusted by a proportional-integral-differential algorithm to ensure that the pH value of the acid solution is stable within the preset range.

[0022] Optical scanning technology was used to acquire images of the temporary protective film formation state on the surface of quartz sand, thus obtaining the first image data;

[0023] The first image data is denoised and edge detected using image processing algorithms to obtain the second image data;

[0024] Based on the second image data, the coverage rate of the temporary protective film is calculated using a convolutional neural network algorithm to obtain the coverage rate value;

[0025] If the coverage value does not reach the preset threshold n, the amount of fluorine barrier agent injected will be increased through the feedback control system to obtain a new coverage value.

[0026] Based on multiple coverage rate values, determine whether the coverage rate of the temporary protective film has stably reached the preset threshold n, and determine the final formation state.

[0027] The above-mentioned method for treating and utilizing fluoride-containing waste acid generated in a quartz sand pickling process includes the following steps: if the protective film coverage reaches a preset threshold n, a waste acid sample is extracted from the pickling tank, the fluoride ion concentration in the waste acid is analyzed, and first fluoride content data is obtained; including:

[0028] If the protective film coverage reaches the preset threshold n, waste acid samples are extracted from the pickling tank, and a fixed volume of waste acid samples is obtained using an automated sampling device to obtain waste acid sample data.

[0029] The concentration of fluoride ions in waste acid samples was analyzed by ion chromatography to obtain the primary fluoride content data and determine the primary fluoride content.

[0030] If the first fluoride content exceeds the preset concentration threshold, a data filtering algorithm is used to denoise the first fluoride content data to obtain the second fluoride content data.

[0031] Based on the second fluoride content data, the fluoride ion concentration of the waste acid samples was classified and the concentration level was determined using the support vector machine algorithm.

[0032] If the concentration level is high, the corresponding treatment parameters are obtained by querying the database to retrieve the pre-established waste acid treatment rules and determine the treatment plan.

[0033] According to the treatment plan, an automated control system is used to adjust the waste acid discharge rate of the pickling tank, obtain the adjusted discharge data, and determine the discharge status.

[0034] Emissions data are analyzed using time series analysis algorithms to obtain emission trend data and determine emission stability.

[0035] The above-mentioned method for treating and utilizing fluorine-containing waste acid generated in a quartz sand pickling process includes: performing a spectral scan on the first fluorine content data to obtain molecular vibrational modes and characteristic peak intensity data, and combining this with fluorine atom coordination information to generate a first barrier agent molecular structure fingerprint; including:

[0036] The fluorine content data was scanned using a molecular spectroscopy analysis device, employing a wavelength range from ultraviolet to visible light to obtain the raw spectral data;

[0037] The original spectral data is processed using a background subtraction algorithm. If environmental spectral interference is detected, the interference signal is separated by Fourier transform to obtain clean spectral data.

[0038] Molecular vibrational modes are extracted from pure spectral data. If the intensity of the vibrational mode signal is lower than a preset threshold z1, the signal is enhanced by Gaussian fitting to determine the molecular vibrational mode.

[0039] The intensity of characteristic peaks is analyzed based on molecular vibrational modes. If the peak intensity distribution conforms to a preset model, key peak data are extracted through principal component analysis to obtain the intensity of characteristic peaks.

[0040] By combining the first fluorine content data and fluorine atom coordination information, if the coordination information matches the vibration mode, the data is fused through a convolutional neural network to generate a preliminary molecular structure fingerprint.

[0041] Data extraction and processing are performed on the preliminary molecular structure fingerprint. If the fingerprint data integrity is higher than the preset threshold w1, the fingerprint features are optimized through cluster analysis to obtain the molecular structure fingerprint of the first barrier agent.

[0042] Key feature data are extracted from the molecular structure fingerprint of the first barrier agent. Database comparison technology is used. If the feature data matches the preset template, the final molecular structure fingerprint is determined.

[0043] The above-mentioned method for treating and utilizing fluorine-containing waste acid generated in a quartz sand pickling process includes separating the fluorine barrier agent from the waste acid based on the molecular structure fingerprint of the first barrier agent, obtaining measurement data of the purity of the separated phase and the viscosity of the waste acid, and determining whether the first separation purity is higher than a preset threshold c; including:

[0044] Using a molecular structure fingerprint database, the characteristics of fluoride barrier agents in waste acid were matched to identify the target separation substance;

[0045] Waste acid is treated by centrifugation at a preset speed and separation time to obtain a separated phase and residual waste acid;

[0046] The purity data of the separated phase was obtained, and the concentration of the fluorine barrier agent was measured by high performance liquid chromatography to determine the separation purity.

[0047] Viscosity data was extracted from the residual waste acid and measured using a digital viscometer to obtain the viscosity value of the waste acid.

[0048] If the separation purity is higher than the preset threshold c, the separation effect is judged to be qualified and a qualified label is generated.

[0049] If the value is below the preset threshold c, adjust the centrifugation speed and repeat the separation process;

[0050] If the viscosity value is within the preset viscosity range, the waste acid is confirmed to be in normal condition.

[0051] If it exceeds the range, it will be marked as an abnormal state;

[0052] By using qualified labels and waste acid status, comprehensive evaluation data of the separation process is generated to determine the final treatment result.

[0053] The above-mentioned method for treating and utilizing fluorine-containing waste acid generated in a quartz sand pickling process includes, if the first separation purity is higher than a preset threshold c, performing a second spectral scan on the separated fluorine barrier agent to obtain molecular vibrational modes and characteristic peak intensity data, and generating a second barrier agent molecular structure fingerprint; including:

[0054] If the purity of the first separation is higher than the preset threshold c, a second scan of the fluorine barrier is performed using a molecular spectroscopy analysis device, employing the ultraviolet to near-infrared band to obtain the original spectral data;

[0055] The original spectral data is processed by a background subtraction algorithm to remove environmental noise and baseline drift, resulting in clean spectral data.

[0056] The Fourier transform algorithm is used to perform frequency domain transformation on the pure spectral data, extract molecular vibrational modes, and determine the characteristic frequency distribution;

[0057] Based on the characteristic frequency distribution, the characteristic peak intensity is calculated, and an intensity dataset is generated;

[0058] Principal component analysis algorithm is used to reduce the dimensionality of the intensity dataset, extract the main molecular structure features, and generate the molecular structure fingerprint of the second barrier agent.

[0059] If the similarity between the molecular structure fingerprint and the preset standard fingerprint is higher than the threshold, the chemical composition information of the second barrier agent is obtained by comparing with the database.

[0060] Based on the chemical composition information, a three-dimensional structural model of the second barrier agent was generated using molecular simulation tools to determine its molecular configuration characteristics.

[0061] The above-mentioned method for treating and utilizing fluoride-containing waste acid generated in a quartz sand pickling process includes the following steps: if the pickling effect achieves a residual amount below a preset threshold m, a spectral scan is performed on the waste acid in the pickling tank to obtain data on the size distribution of settled particles and the recovery rate of the separation liquid, thereby generating a first waste acid recycling scheme; including:

[0062] If the residual amount of acid washing is lower than the preset threshold m, the waste acid is spectrally scanned using a molecular spectral analysis device, covering the ultraviolet to visible spectrum, to obtain the original spectral data;

[0063] The raw spectral data were processed by data normalization, and features were extracted using principal component analysis algorithm to obtain data on sedimentation particle size distribution and separation liquid recovery rate.

[0064] Based on the size distribution of settled particles, a clustering analysis algorithm is used to classify the particles and obtain the particle category distribution;

[0065] If the proportion of large particles in the particle size distribution is higher than the preset proportion, the spectral data is processed again by digital filtering technology to obtain the optimized particle size distribution.

[0066] Based on the optimized particle size distribution and separation liquid recovery rate data, a linear regression algorithm is used to predict the waste acid recycling efficiency and obtain the recycling efficiency parameter.

[0067] By analyzing the circulation efficiency parameters and particle type distribution, a first waste acid circulation scheme is generated, and the circulation scheme parameters are determined.

[0068] Based on the parameters of the recycling scheme, adjust the waste acid treatment process and obtain updated pickling residue data.

[0069] The above-mentioned method for treating and utilizing fluoride-containing waste acid generated in a quartz sand pickling process includes, according to a first waste acid recycling scheme, adjusting centrifugal separation parameters, obtaining fluoride ion concentration data in the recycled waste acid, and determining a second fluoride content; comprising:

[0070] Acquire waste acid viscosity measurement data; obtain the first viscosity data by collecting viscosity values ​​in real time through sensors.

[0071] If the first viscosity data exceeds the preset threshold η, the centrifugation separation parameters are adjusted using a linear regression algorithm to obtain the first separation parameters;

[0072] Based on the first separation parameters, the centrifugal speed is set and the separation time is controlled to perform centrifugal separation operation and obtain the first treated waste acid;

[0073] The concentration of fluoride ions in the first-treatment waste acid was detected by ion chromatography to obtain the first concentration data.

[0074] If the first concentration data does not meet the preset standard, the number of cycles is adjusted according to the concentration deviation to obtain the second waste acid.

[0075] The concentration of fluoride ions in the waste acid from the second treatment was determined again by ion chromatography to confirm the second fluoride content.

[0076] Based on the deviation between the second fluorine content and the preset target, the separation parameters are optimized through a feedback control algorithm to obtain the second separation parameters.

[0077] The above-mentioned method for treating and utilizing fluorine-containing waste acid generated in a quartz sand pickling process includes: extracting a fluorine barrier agent according to a third barrier agent formula, re-injecting it into the pickling tank, controlling the acid temperature, obtaining data on the distribution of settled particle size after quartz sand surface treatment, and determining whether the particle size is below a preset threshold d; including:

[0078] Extract fluorine barrier agent formulation data from the recycling database and generate formulation injection instructions;

[0079] The pickling tank equipment is controlled by formula injection instructions to complete the injection of fluorine barrier agent;

[0080] The temperature of the acid solution in the pickling tank is monitored by a real-time temperature control device and adjusted to a stable state.

[0081] Obtain sedimentation particle samples of quartz sand after surface treatment and generate particle size data;

[0082] Particle size data were processed using a laser particle size analyzer to obtain the particle size distribution;

[0083] If the maximum value in the particle size distribution is lower than the preset threshold d, a qualified judgment result is generated;

[0084] The data acquisition module stores the pass / fail judgment results and updates the recycling database.

[0085] The above-mentioned method for treating and utilizing fluoride-containing waste acid generated in a quartz sand pickling process includes, if the particle size is lower than a preset threshold d, analyzing the fluoride ion concentration in the recycled waste acid to obtain tertiary fluoride content data, and combining this with waste acid viscosity measurement data to determine the recycling efficiency; including:

[0086] If the particle size is lower than the preset threshold d, the concentration of fluoride ions in the waste acid after recycling is analyzed by ion chromatography to obtain the third fluoride content data.

[0087] The viscosity of the waste acid after recycling is measured by a viscometer to obtain the waste acid viscosity data;

[0088] A data fusion algorithm was used to standardize the tertiary fluoride content data and the waste acid viscosity data to obtain a fused feature vector;

[0089] If the Euclidean distance between the fused feature vectors is less than a preset distance threshold, the efficiency of the loop processing is determined by the support vector machine classifier to obtain the efficiency level.

[0090] Based on efficiency levels, the K-means clustering algorithm is used to group the running parameters of the loop processing to obtain a set of optimized parameters;

[0091] By adjusting the operating parameters of the cyclic processing equipment through parameter optimization sets, updated processing efficiency data can be obtained.

[0092] If the updated processing efficiency data does not reach the preset efficiency threshold, the data fusion and efficiency judgment steps are repeated to obtain a new efficiency level.

[0093] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0094] This invention discloses a microwave-assisted method for recycling fluorine barrier agents. A specific frequency microwave is generated by a pre-set microwave device and focused onto the surface of quartz sand in an pickling tank. Power output is controlled, and dissolution rate data is acquired. Based on the dissolution rate, a suitable fluorine barrier agent formulation is matched from a database and injected into the pickling tank to form a temporary protective film. The fluorine barrier agent is separated from the waste acid using centrifugal separation technology. Molecular structural fingerprints are obtained through molecular spectroscopy analysis, and separation parameters are optimized to achieve high-purity recovery. This invention, through real-time monitoring and feedback control, achieves efficient recycling of fluorine barrier agents, improves pickling efficiency, reduces environmental pollution, and has significant economic and environmental benefits. Attached Figure Description

[0095] Figure 1 This is a flowchart of a method for treating and utilizing fluorine-containing waste acid generated in the quartz sand pickling process of the present invention. Detailed Implementation

[0096] The technical solutions of the present invention will be clearly and thoroughly described below with reference to the accompanying drawings. The described embodiments are merely some embodiments of the present invention.

[0097] like Figure 1This embodiment of a method for treating and utilizing fluoride-containing waste acid generated in a quartz sand pickling process may specifically include:

[0098] S1. Microwaves are generated by a microwave generator and emitted to the surface of quartz sand in the pickling tank to obtain data on the change in the dissolution rate of surface oxides and metal ions, and to determine the first dissolution rate.

[0099] In practice, 2.4 GHz microwaves are generated and focused onto the surface of quartz sand in the pickling tank using a directional antenna array. The power output is controlled at 100 W to acquire data on the dissolution rate changes of surface oxides and metal ions, thus determining the first dissolution rate. The specific process includes:

[0100] Microwaves of a specific frequency are generated by a pre-set microwave generator and focused onto the surface of quartz sand in an acid pickling tank using a directional antenna array. Initial dissolution rate data is obtained to determine the first dissolution rate. If the initial dissolution rate data exceeds a preset threshold, the power output of the microwave generator is adjusted to obtain new dissolution rate data, resulting in a second dissolution rate. Based on the second dissolution rate, a pre-established linear regression model is used to analyze the trend of power output and dissolution rate changes, determining the rate change parameters. Using the rate change parameters, a preset threshold comparison method is used to determine the dissolution state of surface oxides and metal ions, resulting in a dissolution state classification. If the dissolution state classification is unstable, the focusing angle of the directional antenna array is adjusted to obtain new dissolution rate data, resulting in a third dissolution rate. Based on the third dissolution rate, a linear regression model is used to analyze the correlation between the focusing angle and the dissolution rate, determining the angle adjustment parameters. Using the angle adjustment parameters, a preset optimization algorithm is used to adjust the operating parameters of the microwave generator and antenna array to obtain a stable dissolution rate.

[0101] S2. If the first dissolution rate is lower than the preset threshold v, adjust the microwave frequency, control the acid temperature to stabilize at T, obtain the adjusted dissolution rate data, and determine the second dissolution rate.

[0102] In practice, if the first dissolution rate is lower than the preset threshold of 0.1 mg / cm² / min, the microwave frequency is adjusted and increased incrementally in 10 MHz steps to 2.5 GHz. Infrared thermometry feedback is used to stabilize the acid solution temperature at 50°C. The adjusted dissolution rate data is sampled every minute to determine the second dissolution rate. The specific process includes:

[0103] If the first dissolution rate is lower than a preset threshold v, the microwave frequency is increased incrementally in fixed steps by the control module to obtain an adjusted microwave frequency value. An infrared thermometer is used to monitor the acid solution temperature in real time, and a feedback control algorithm is used to adjust the heating power to obtain a stable acid solution temperature. Dissolution rate data is sampled every minute, and the adjusted dissolution rate value is extracted from the sampled data to determine the second dissolution rate. If the second dissolution rate is still lower than the preset threshold, a regression analysis algorithm is used to fit the relationship between the microwave frequency and the dissolution rate to obtain the frequency adjustment trend. Based on the frequency adjustment trend, the microwave frequency increment for the next round is calculated to obtain a new microwave frequency setpoint. The control module applies the new microwave frequency setpoint and repeatedly samples the dissolution rate data to obtain an updated second dissolution rate. If the updated second dissolution rate reaches the preset threshold, the current microwave frequency and acid solution temperature are recorded to determine the final process parameters.

[0104] Specifically, when the first dissolution rate is below 0.1 mg / cm² / min, the system automatically activates the microwave frequency adjustment module, increasing the frequency from an initial 500 MHz in 10 MHz increments. After each adjustment, the temperature of the acid solution inside the reactor is monitored in real time by an infrared temperature sensor. If the temperature deviates from 50°C, a PID control algorithm (proportional coefficient Kp = 1.2, integral time Ti = 30 s, derivative time Td = 5 s) is triggered to adjust the microwave power output, stabilizing the temperature within the range of 50 ± 0.5°C. At the 60-second mark after each frequency adjustment, a high-precision electronic balance (resolution 0.01 mg) automatically collects the sample mass loss data and calculates the second dissolution rate value based on the reaction area of ​​2.5 cm². For example, when the frequency rises to 1.2 GHz, the measured mass loss is 15.6 mg, which is calculated using the formula v = Δm / (S·t) = 15.6 / (2.5 × 1) = 6.24 mg / cm² / min. The system compares the collected dissolution rate data with a preset threshold. If the rate is still lower than the threshold, the frequency is increased in increments of 10MHz until the upper limit of 2.5GHz is reached. During this process, the frequency-rate correspondence dataset is automatically recorded for subsequent dissolution kinetic model building. For example, after obtaining a peak rate of 12.8mg / cm² / min at 1.8GHz, the system determines the optimal response frequency as 1.76GHz (R²=0.982) through quadratic polynomial fitting.

[0105] S3. Based on the second dissolution rate, match the fluorine barrier agent formulation from the preset recycling database, extract the molecular structure fingerprint data, and generate the first barrier agent addition scheme.

[0106] In specific implementation, based on the second dissolution rate, fluorine barrier agent formulations with oxide dissolution rates ranging from 0.1 mg / cm² / min to 0.5 mg / cm² / min are matched from a pre-set recycling database. Molecular structure fingerprint data is extracted, and characteristic peak intensity analysis is used to generate the first barrier agent addition scheme. The specific process includes:

[0107] Fluorine barrier agent formulation data with oxide dissolution rates within a preset range are obtained from a recycling database. Data matching analysis is used to obtain a set of formulations that meet the criteria. Molecular structure fingerprint data is extracted from the formulation set, and a molecular structure feature set is obtained through molecular structure data parsing. Characteristic peak extraction technology is used to analyze the characteristic peak intensity of the molecular structure feature set to obtain the characteristic peak intensity distribution. If the characteristic peak intensity distribution meets a preset threshold, the first barrier agent addition amount is generated based on the characteristic peak intensity distribution, resulting in a first barrier agent scheme. The first barrier agent scheme is optimized using the formulation generation scheme to obtain optimized barrier agent formulation data. Formulation composition information is extracted from the optimized barrier agent formulation data, and data matching analysis is used to generate a final barrier agent addition scheme.

[0108] Specifically, firstly, data records with oxide dissolution rates ranging from 0.1 mg / cm² / min to 0.5 mg / cm² / min are selected from the recycling database, and then the SQL query "SELECT ... The query "FROM oxide_tableWHERE dissolution_rate BETWEEN 0.1 AND 0.5" matched 12 fluorine barrier agent formulations. The molecular structure fingerprint data of these formulations was analyzed using the RDKit toolkit, extracting a 128-dimensional Morgan fingerprint feature vector. Through feature peak intensity analysis, principal component analysis was used to reduce the dimensionality to 3, with a variance explanation rate of 95%. The main feature peaks were found to be concentrated in three wavenumber regions: 1200 cm⁻¹, 1450 cm⁻¹, and 1650 cm⁻¹. Based on the k-means clustering algorithm (k=3, Euclidean distance, 100 iterations), these formulations were divided into three categories. The second type of formulation, whose center of gravity is closest to the target dissolution rate of 0.3 mg / cm² / min, was selected as the basis. The characteristic peak intensity ratio of this type of formulation is 1:1.2:0.8. Based on the dissolution kinetic model y=0.25x+0.15 (R²=0.92), the proportions of each component in the formulation were adjusted to achieve a predicted dissolution rate of 0.35 mg / cm² / min. The final first barrier agent addition scheme contains 45% CF2, 30% SiO2, and 25% Al2O3, with an optimized characteristic peak intensity ratio of 1:1.1:0.9.

[0109] S4. Using the first barrier agent addition scheme, inject fluorine barrier agent into the pickling tank, control the pH value of the acid solution through real-time pH value feedback, obtain the formation status data of the temporary protective film on the surface of the quartz sand, and determine whether the coverage rate of the protective film reaches the preset threshold n.

[0110] In practice, the first barrier agent addition scheme is adopted, injecting fluorinated barrier agent into the pickling tank. The pH of the acid solution is maintained between 3.0 and 3.5 through real-time pH feedback. Data on the formation status of the temporary protective film on the quartz sand surface is obtained to determine whether the protective film coverage reaches the preset threshold of 95%. The specific process includes:

[0111] Real-time pH data of the acid solution in the pickling tank is acquired using sensors to determine the pH fluctuation range. If the pH value deviates from the preset range, the injection rate of the fluorine barrier agent is adjusted using a proportional-integral-differential algorithm to stabilize the acid solution pH within the preset range. Optical scanning technology is used to acquire images of the temporary protective film formation state on the quartz sand surface, obtaining the first image data. An image processing algorithm is used to denoise and perform edge detection on the first image data to obtain the second image data. Based on the second image data, a convolutional neural network algorithm is used to calculate the coverage rate of the temporary protective film, obtaining a coverage rate value. If the coverage rate value does not reach the preset threshold n, the fluorine barrier agent injection amount is increased through a feedback control system to obtain a new coverage rate value. Based on multiple coverage rate values, it is determined whether the coverage rate of the temporary protective film has stably reached the preset threshold n, thus determining the final formation state.

[0112] Specifically, when injecting fluorine barrier agent into the pickling tank, the pH value is monitored in real time, and a PID control algorithm is used to adjust the injection amount of fluorine barrier agent to ensure that the pH value of the acid solution is stable between 3.0 and 3.5. In practice, the pH sensor collects data every 10 seconds, and the control system calculates the injection rate of fluorine barrier agent based on the deviation between the current pH value and the target value of 3.25.

[0113] For example, when the pH value deviates from the target value by more than 0.1, the control system increases the injection rate of fluorine barrier agent at a rate of 0.5 liters per minute until the pH value returns to the set range. Simultaneously, image analysis technology is used to acquire data on the formation state of the temporary protective film on the quartz sand surface. A convolutional neural network (CNN) is used to process the surface image and calculate the coverage of the protective film. Assuming an image resolution of 1024×1024 pixels, the trained CNN model can identify the protective film area with 95% accuracy and output coverage data in real time. When the coverage reaches a preset threshold of 95%, the system automatically stops the injection of fluorine barrier agent and records relevant data for subsequent analysis. Throughout the entire process, data acquisition, algorithm calculation, and decision execution are all completed through an automated system, ensuring high efficiency and accuracy.

[0114] S5. If the protective film coverage reaches the preset threshold n, then extract waste acid samples from the pickling tank, analyze the fluoride ion concentration in the waste acid, obtain the first fluoride content data, and determine the first fluoride content.

[0115] In practice, if the protective film coverage reaches a preset threshold of 95%, a waste acid sample is extracted from the pickling tank, and the fluoride ion concentration in the waste acid is analyzed using ion chromatography to obtain the first fluoride content data and determine the first fluoride content. The specific process includes:

[0116] If the protective film coverage reaches a preset threshold n, waste acid samples are extracted from the pickling tank. A fixed volume of waste acid sample is obtained using an automated sampling device to obtain waste acid sample data. The fluoride ion concentration in the waste acid sample is analyzed by ion chromatography to obtain the first fluoride content data and determine the first fluoride content. If the first fluoride content exceeds a preset concentration threshold, a data filtering algorithm is used to denoise the first fluoride content data to obtain the second fluoride content data. Based on the second fluoride content data, a support vector machine algorithm is used to classify the fluoride ion concentration of the waste acid sample and determine the concentration level. If the concentration level is high, the corresponding treatment parameters are obtained by querying the database according to pre-established waste acid treatment rules to determine the treatment plan. According to the treatment plan, an automated control system is used to adjust the waste acid discharge rate of the pickling tank, and the adjusted discharge data is obtained to determine the discharge status. A time series analysis algorithm is used to perform trend analysis on the discharge data to obtain discharge trend data and determine discharge stability.

[0117] Specifically, when the protective film coverage reaches a preset threshold of 95%, the system automatically triggers the process of extracting waste acid samples from the pickling tank. The extracted waste acid samples are analyzed using ion chromatography. First, the sample is injected into a chromatographic column, and separation is performed using a mobile phase (e.g., 0.1 mol / L sodium hydroxide solution). The retention time of fluoride ions in the column is 2.3 minutes. The peak area of ​​fluoride ions is measured using a detector (e.g., a conductivity detector). The concentration of fluoride ions in the waste acid is calculated based on a standard curve (e.g., for fluoride ion concentrations ranging from 0.1 mg / L to 10 mg / L, the peak area and concentration have a linear relationship, with a regression equation of y = 0.98x + 0.02, R² = 0.999). Assuming the measured peak area of ​​fluoride ions is 1500, substituting it into the regression equation yields a fluoride ion concentration of 1.53 mg / L, which is the first fluoride content data. The system records this data and uses it for subsequent process adjustments or environmental monitoring to ensure that the waste acid treatment process complies with environmental standards.

[0118] S6. Perform spectral scanning on the first fluorine content data to obtain molecular vibration mode and characteristic peak intensity data, and combine fluorine atom coordination information to generate the molecular structure fingerprint of the first barrier agent.

[0119] In practice, molecular spectroscopy analysis equipment is used to perform spectral scanning on the first fluorine content data, with a scanning range of 200 nm to 800 nm. A background subtraction algorithm is employed to remove environmental spectral interference, and molecular vibrational modes and characteristic peak intensity data are obtained. Combined with fluorine atom coordination information, a molecular structural fingerprint of the first barrier agent is generated. The specific process includes:

[0120] The first fluorine content data was spectrally scanned using a molecular spectroscopy analysis device, employing the ultraviolet to visible light wavelength range to acquire raw spectral data. A background subtraction algorithm was used to process the raw spectral data. If environmental spectral interference was detected, Fourier transform was used to separate the interference signal, yielding pure spectral data. Molecular vibrational modes were extracted from the pure spectral data. If the vibrational mode signal intensity was below a preset threshold z1, Gaussian fitting was used to enhance the signal and determine the molecular vibrational mode. Characteristic peak intensities were analyzed based on the molecular vibrational modes. If the peak intensity distribution conformed to a preset model, principal component analysis was used to extract key peak data, obtaining characteristic peak intensities. Combining the first fluorine content data and fluorine atom coordination information, if the coordination information matched the vibrational modes, a convolutional neural network was used to fuse the data, generating a preliminary molecular structure fingerprint. Data extraction and processing were performed on the preliminary molecular structure fingerprint. If the fingerprint data integrity was above a preset threshold w1, cluster analysis was used to optimize the fingerprint features, obtaining the molecular structure fingerprint of the first barrier agent. Key feature data were extracted from the molecular structure fingerprint of the first barrier agent. Database comparison technology was used; if the feature data matched a preset template, the final molecular structure fingerprint was determined.

[0121] Specifically, using molecular spectroscopy analysis equipment, the first fluorine content data is scanned in the wavelength range of 200nm to 800nm. A high-resolution spectrometer is used to acquire data in 1nm increments to ensure the accuracy of the spectral data. During the scanning process, a background subtraction algorithm is used to remove environmental spectral interference. This algorithm acquires the background spectrum when there is no sample and subtracts it from the sample spectrum, thereby eliminating interference factors such as ambient light and equipment noise.

[0122] For example, in the ultraviolet region of 200 nm to 300 nm, the background spectral intensity is 0.05, the sample spectral intensity is 0.12, and after background subtraction, the actual sample spectral intensity is 0.07. Next, Fourier transform infrared spectroscopy (FTIR) analysis is used to obtain molecular vibrational modes and characteristic peak intensity data, for example, at 1200 cm⁻¹. - A stretching vibration peak of the CF bond was observed at position ¹, with an intensity of 0.8, indicating a strong coordination between the fluorine and carbon atoms. Based on the fluorine coordination information, the molecular structure was optimized and vibrational frequencies were calculated using quantum chemical calculation software such as Gaussian, generating the molecular structural fingerprint of the first barrier agent.

[0123] For example, calculations revealed that the CF bond length is 1.35 Å and the bond angle is 109.5°. These data closely match the characteristic peaks of the experimental spectra, further validating the accuracy of the molecular structure. Finally, the spectral data and calculation results are integrated to generate a unique molecular structural fingerprint, providing a reliable basis for subsequent molecular identification and functional studies.

[0124] S7. Based on the molecular structure fingerprint of the first barrier agent, separate the fluorine barrier agent from the waste acid, obtain the separation phase purity and waste acid viscosity measurement data, and determine whether the first separation purity is higher than the preset threshold c.

[0125] In practice, based on the molecular structure fingerprint of the first barrier agent, centrifugal separation technology is used. A centrifugal speed of 8000 rpm and a separation time of 30 minutes are set to separate the fluoride barrier agent from the waste acid. Measurement data of the purity of the separated phase and the viscosity of the waste acid are obtained to determine whether the first separation purity is higher than a preset threshold of 98%. The specific process includes:

[0126] Using a molecular structure fingerprint database, the characteristics of fluoride barrier agents in waste acid were matched to identify the target separation substance. Waste acid was treated by centrifugation at preset speeds and separation times to obtain a separated phase and residual waste acid. Purity data of the separated phase was obtained, and the concentration of the fluoride barrier agent was measured using high-performance liquid chromatography (HPLC) to determine the separation purity. Viscosity data was extracted from the residual waste acid and measured using a digital viscometer to obtain the waste acid viscosity value. If the separation purity was higher than a preset threshold c, the separation effect was considered satisfactory, and a satisfactory label was generated; if it was lower than the preset threshold c, the centrifugation speed was adjusted, and the separation was repeated. The viscosity value was compared with a preset viscosity range; if the viscosity value was within the range, the waste acid state was confirmed as normal; if it exceeded the range, it was marked as abnormal. Based on the satisfactory label and the waste acid state, comprehensive evaluation data of the separation process was generated to determine the final processing result.

[0127] Specifically, the chemical composition and functional group distribution of the first barrier agent were first determined through molecular structure fingerprinting analysis. High-performance liquid chromatography-mass spectrometry (HPLC-MS / MS) was used to obtain its molecular weight of 356.2 g / mol, with characteristic peaks at m / z 357.1 and 359.3. The sedimentation coefficient was calculated to be 4.2 × 10⁻⁶ according to Stokes' law. -At 13 s, combined with the waste acid density of 1.28 g / cm³ and viscosity of 12.5 cP, the theoretical separation efficiency can reach 99.7% under the centrifugation condition of 8000 rpm. In the actual separation process, a Thermo Scientific Sorvall LYNX 6000 centrifuge was used, with the rotation speed set at 8000 ± 50 rpm, the temperature maintained at 25 ± 0.5 °C, and the supernatant was obtained after centrifugation for 30 minutes. The absorbance of the separated phase was measured at 0.342 at a wavelength of 278 nm using an ultraviolet spectrophotometer. The concentration of the fluorine barrier agent was calculated to be 98.4% through the pre-established calibration curve y = 0.1567x + 0.0213 (R² = 0.998). At the same time, a Brookfield DV2T viscometer was used to measure the residual viscosity of the waste acid to be 3.8 cP, a 69.6% reduction compared to the initial value. The purity data was input into the decision tree model. When the purity ≥ 98% and the viscosity reduction ≥ 50% were met, it was判定 as qualified. The current data nodes all passed the threshold test, and the purity index exceeded the preset threshold by 0.4 percentage points. To verify the reliability of the results, Monte Carlo simulation was used for 1000 repeated calculations. The purity data conforms to a normal distribution (μ = 98.2%, σ = 0.3%), and the 98% confidence interval is [97.9%, 98.5%], confirming that the separation effect is stable and meets the standards.

[0128] S8. If the purity of the first separation is higher than the preset threshold c, perform a secondary spectral scan with optimized spectral resolution on the separated fluorine barrier agent to obtain data on molecular vibration modes and characteristic peak intensities, and generate a second fingerprint of the barrier agent molecular structure;

[0129] In specific implementation, if the purity of the first separation is higher than the preset threshold of 98%, perform a secondary scan with optimized spectral resolution on the separated fluorine barrier agent using a molecular spectroscopy analysis device. The scanning range is from 200 nm to 800 nm, and the data is processed using a background subtraction algorithm to obtain data on molecular vibration modes and characteristic peak intensities, and generate a second fingerprint of the barrier agent molecular structure. The specific process includes:

[0130] If the initial purity exceeds a preset threshold c, a secondary scan of the fluorine barrier agent is performed using a molecular spectroscopy analysis device in the ultraviolet to near-infrared band to acquire raw spectral data. The raw spectral data is processed using a background subtraction algorithm to remove environmental noise and baseline drift, resulting in pure spectral data. A Fourier transform algorithm is then used to perform frequency domain transformation on the pure spectral data to extract molecular vibrational modes and determine the characteristic frequency distribution. Based on the characteristic frequency distribution, the intensity of characteristic peaks is calculated, generating an intensity dataset. Principal component analysis is used to reduce the dimensionality of the intensity dataset, extracting key molecular structural features and generating a molecular structural fingerprint for the second barrier agent. If the similarity between the molecular structural fingerprint and a preset standard fingerprint exceeds a threshold, the chemical composition information of the second barrier agent is obtained through database comparison. Based on the chemical composition information, a three-dimensional structural model of the second barrier agent is generated using molecular simulation tools to determine its molecular configuration characteristics.

[0131] Specifically, when the initial separation purity reaches 98% or higher, the system automatically triggers the secondary scanning process of the molecular spectroscopy analysis equipment. The scanning range is set to the ultraviolet-visible band from 200 nm to 800 nm, and full-band data acquisition is performed using a 1 nm step resolution. In the data processing stage, high-frequency noise is first eliminated using the Savitzky-Golay smoothing algorithm (window width 15, polynomial order 3). Then, a baseline correction algorithm is applied to subtract solvent background interference, specifically using the asymmetric least squares method (λ=1e5, p=0.01) to fit the baseline. For characteristic peak identification, the system performs second-order derivative transformation combined with Gaussian fitting (half-width threshold ≥20 nm, R²>0.995) to accurately locate vibrational modes. For example, a characteristic peak of benzene ring skeleton vibration (intensity ≥1200 a.u., half-width 25.3 nm) is identified at 650 nm. By establishing a peak area integral matrix (integration interval ±15nm) and comparing it with a standard database, feature peaks with a matching degree exceeding 90% are included in the fingerprint database. For example, quantified data of CF bond stretching vibration (relative intensity ratio 1:0.78) and out-of-plane bending vibration (peak asymmetry factor 0.92) are captured at 420nm and 580nm, respectively. The final generated molecular structure fingerprint contains 12 feature vectors. Each vector consists of normalized intensity values ​​(range 0-1), peak shape parameters (skewness ≤0.3, kurtosis 2.8-3.2), and vibrational assignment codes. After dimensionality reduction through principal component analysis (cumulative variance contribution rate >85%), these vectors are stored in the barrier agent feature database.

[0132] S9. Based on the molecular structure fingerprint of the second barrier agent, extract the fluorine barrier agent formula, re-inject it into the pickling tank, control the operation of the microwave generator, use the rotor material corrosion resistance equipment with centrifugal force distribution optimization, obtain the impurity residue data after the surface treatment of quartz sand, and determine whether the pickling effect reaches the point where the residue is lower than the preset threshold m.

[0133] In practice, based on the molecular structure fingerprint of the second barrier agent, the fluorine barrier agent formula is extracted from the recycling database and re-injected into the pickling tank. The microwave generator is controlled to operate at a frequency of 2.5 GHz and a power of 200 W. A rotor material corrosion resistance device with optimized centrifugal force distribution is used to obtain data on residual impurities after quartz sand surface treatment, and to determine whether the pickling effect has reached the preset threshold of residual amount below 0.01 mg / cm². The specific process includes:

[0134] Fluorine barrier agent formulations are extracted from a recycling database, and formulation data is generated based on the molecular structure fingerprint of the second barrier agent. A molecular structure fingerprint matching algorithm is used to verify the consistency between the formulation data and the database records, obtaining verification results. Based on the verification results, the pickling tank injection system is controlled to inject the fluorine barrier agent corresponding to the formulation data, confirming the injection completion status. A microwave generator is run to treat the quartz sand surface in the completed pickling tank, acquiring surface treatment data. If the residual impurities in the surface treatment data are lower than a preset threshold, a centrifugal force optimization algorithm is used to adjust the rotor material corrosion resistance equipment parameters to obtain optimized parameters. Based on the optimized parameters, the surface treatment equipment is run to generate residual impurities data on the quartz sand surface, judging the pickling effect. By comparing the residual impurities data with a preset threshold, if it is lower than the threshold, a pickling effect compliance indicator is generated, confirming the treatment completion.

[0135] Specifically, firstly, a molecular structure fingerprint matching algorithm is used to extract the fluorine barrier agent formulation from the recycling database. A deep learning-based graph neural network model is then used to calculate the structural similarity between the second barrier agent and each formulation in the database. When the similarity exceeds 0.85, the corresponding formulation parameters are automatically retrieved, including a main component system consisting of 12% fluorocarbon polymer and 8% silane coupling agent by mass. When the formulation is injected into the pickling tank, the injection rate is monitored in real-time by a flow sensor to maintain a rate of 2.0 L / min, while a pH probe detects and maintains the tank solution's pH within the range of 3.5 ± 0.2. The microwave generator employs PID closed-loop control, operating at a frequency of 2.5 GHz and a power of 200 W. Finite element analysis algorithms are used to optimize the electromagnetic field distribution, ensuring that the standing wave ratio within the tank does not exceed 1.3 and the temperature gradient is controlled within ± 2℃. The centrifuge equipment uses a 316L stainless steel rotor. ANSYS simulation determined the optimal rotational speed to be 1500 rpm. The centrifugal acceleration distribution curve conforms to a Weibull distribution (shape parameter k=2.1, scale parameter λ=1200). Corrosion resistance testing showed an annual corrosion rate of less than 0.005 mm / a. Impurity detection after quartz sand surface treatment was performed using laser-induced breakdown spectroscopy. Elemental concentration data were acquired at 5×5 grid sampling points. A residual distribution model was obtained by fitting the data using the least squares method. When the residual amount in 90% of the entire area after cubic spline interpolation was below 0.008 mg / cm², and the maximum extreme value did not exceed 0.012 mg / cm², the pickling effect was deemed satisfactory. Throughout the process, data was transmitted to the MES system in real time via the OPC-UA protocol. An automatic compensation mechanism was triggered when any parameter deviated from the set threshold by more than 5%.

[0136] S10. If the pickling effect reaches the point where the residual amount is lower than the preset threshold m, then the waste acid in the pickling tank is subjected to spectral scanning to obtain the data on the size distribution of settled particles and the recovery rate of the separation liquid, and a first waste acid recycling scheme is generated.

[0137] In practice, if the pickling effect reaches a preset threshold where the residual amount is below 0.01 mg / cm², the waste acid in the pickling tank is spectrally scanned using a molecular spectroscopy analysis device. The scanning range is 200 nm to 800 nm. Data normalization is then performed to obtain data on the size distribution of settled particles and the recovery rate of the separated liquid, generating the first waste acid recycling scheme. The specific process includes:

[0138] If the residual acid pickling amount is below a preset threshold m, the waste acid is spectrally scanned using a molecular spectral analysis device, covering the ultraviolet to visible spectrum, to obtain raw spectral data. The raw spectral data is normalized, and principal component analysis is used to extract features, yielding data on the sedimentation particle size distribution and the recovery rate of the separated liquid. Based on the sedimentation particle size distribution, a clustering analysis algorithm is used to classify the particles, resulting in a particle category distribution. If the proportion of large particles in the particle category distribution is higher than a preset ratio, digital filtering technology is used to perform secondary processing on the spectral data to obtain an optimized particle size distribution. Based on the optimized particle size distribution and the recovery rate of the separated liquid, a linear regression algorithm is used to predict the waste acid recycling efficiency, obtaining recycling efficiency parameters. Using the recycling efficiency parameters and the particle category distribution, a first waste acid recycling scheme is generated, and the recycling scheme parameters are determined. Based on the recycling scheme parameters, the waste acid treatment process is adjusted, and updated residual acid pickling data is obtained.

[0139] Specifically, after the pickling effect reaches a preset threshold where the residual amount is below 0.01 mg / cm², the system automatically activates a molecular spectroscopy analysis device to perform a spectral scan of the waste acid in the pickling tank. The scanning range is set to 200 nm to 800 nm to ensure coverage of the spectral characteristics of various chemical components that may be present in the waste acid. During the scan, the device uses a high-resolution spectrometer, acquiring data every 0.1 nm to ensure data accuracy. The acquired spectral data undergoes data normalization processing to unify the absorbance values ​​of different wavelengths to the range of 0 to 1 for subsequent analysis. The normalization processing uses a minimum-maximum normalization algorithm, with the formula: Normalized value = (Original value - Minimum value) / (Maximum value - Minimum value). The processed data is then analyzed for particle size distribution using machine learning algorithms. A Gaussian mixture model (GMM) is used to classify particle sizes, and the model parameters are optimized using the expectation-maximum (EM) algorithm to finally obtain the particle size distribution curve. Meanwhile, the system calculates the recovery rate of the separated liquid and uses a linear regression model to predict the relationship between the recovery rate and particle size, with a goodness of fit R² exceeding 0.95. Based on the above analysis results, the system generates a first waste acid recycling scheme, which details the waste acid treatment strategies for different particle sizes. For example, for particles smaller than 10 μm, centrifugal separation technology can achieve a recovery rate of over 90%; for particles larger than 10 μm, filtration technology can achieve a recovery rate of over 85%. The scheme also includes specific parameters for waste acid recycling, such as the number of cycles and cycle time, ensuring the high efficiency and environmental friendliness of waste acid treatment.

[0140] S11. Based on the first waste acid recycling scheme, adjust the centrifugal separation parameters using waste acid viscosity measurement data, obtain fluoride ion concentration data in the recycled waste acid, and determine the second fluoride content.

[0141] In specific implementation, based on the first waste acid recycling scheme, the centrifugal separation parameters are adjusted using waste acid viscosity measurement data. The centrifugal speed is set to 8500 rpm, and the separation time is controlled at 25 minutes. The fluoride ion concentration data in the recycled waste acid is obtained to determine the second fluoride content. The specific process includes:

[0142] Viscosity measurement data of waste acid is acquired by real-time acquisition of viscosity values ​​using sensors to obtain the first viscosity data. If the first viscosity data exceeds a preset threshold η, the centrifugation separation parameters are adjusted using a linear regression algorithm to obtain the first separation parameters. Based on the first separation parameters, the centrifugation speed is set and the separation time is controlled to perform centrifugation separation, obtaining the first treated waste acid. The fluoride ion concentration in the first treated waste acid is detected by ion chromatography to obtain the first concentration data. If the first concentration data does not meet the preset standard, the number of cycles is adjusted according to the concentration deviation to obtain the second treated waste acid. The fluoride ion concentration in the second treated waste acid is detected again using ion chromatography to determine the second fluoride content. Based on the deviation between the second fluoride content and the preset target, the separation parameters are optimized using a feedback control algorithm to obtain the second separation parameters.

[0143] Specifically, in the waste acid recycling process, the viscosity data of the waste acid is first collected in real time by an online viscosity sensor. For example, if the viscosity is measured to be 12.5 mPa·s, the baseline speed is calculated as 7680 rpm by combining the preset viscosity-speed correspondence model y=0.08x²+150 (where x is the viscosity value and y is the base speed). Then, a correction value of 820 rpm is added according to the process requirements, and the centrifuge speed is finally set to 8500 rpm. The separation time control module matches the current material characteristics according to the waste acid composition database. When the fluoride content is detected to be the initial value of 1500 mg / L, the time optimization algorithm t=0.02C+5 (where C is the fluoride ion concentration in mg / L) is automatically called to calculate the theoretical separation time of 22 minutes. After adding a safety redundancy of 3 minutes, a 25-minute separation program is executed. After centrifugation, the treated solution was analyzed using ion chromatography. The fluoride ion concentration was analyzed using a standard curve method. For example, when the absorbance was measured to be 0.345, the fluoride ion concentration was calculated to be 158.6 mg / L by substituting it into the calibration equation y = 0.0021x + 0.012 (R² = 0.998). This value was then input into the process control system as the second fluoride content value. Throughout the process, the data acquisition system recorded the rotation speed fluctuations at a frequency of 1 Hz, maintaining them within ±15 rpm. A temperature sensor ensured that the waste acid was kept at a suitable separation condition of 45 ± 2℃. When the fluoride ion removal rate reached 89.4%, the next processing unit was automatically triggered.

[0144] S12. Perform spectral scanning on the second fluorine content data, process the data and combine characteristic peak intensity analysis with fluorine atom coordination information to generate the molecular structure fingerprint of the third barrier agent and determine the formulation of the third barrier agent.

[0145] In practice, molecular spectroscopy analysis equipment is used to perform spectral scanning on the second fluorine content data, with a scanning range of 200nm to 800nm. A background subtraction algorithm is used to process the data, and combined with characteristic peak intensity analysis and fluorine atom coordination information, a molecular structure fingerprint of the third barrier agent is generated to determine the formulation of the third barrier agent. The specific process includes:

[0146] The second fluorine content data was scanned using a molecular spectroscopy analysis device, covering the ultraviolet to visible light region, to obtain raw spectral data. A background subtraction algorithm was used to process the raw spectral data, removing background noise to obtain pure spectral data. Characteristic peak intensities were extracted from the pure spectral data, and combined with principal component analysis, the coordination characteristics of fluorine atoms were determined. If the characteristic peak intensity was higher than a preset threshold, a molecular structure fingerprint of the third barrier agent was generated based on the fluorine atom coordination characteristics; if it was lower than the threshold, the spectral scanning parameters were adjusted, and the raw spectral data was re-acquired. Based on the molecular structure fingerprint, a clustering analysis algorithm was used to classify the molecular structure of the third barrier agent, obtaining molecular structure categories. The molecular structure categories were matched with a pre-established formulation database to determine the formulation of the third barrier agent. The formulation composition information was obtained from the formulation database to generate the final formulation data of the third barrier agent.

[0147] Specifically, in the molecular spectroscopy analysis equipment, the spectral scanning range was first set to 200 nm to 800 nm, and a high-resolution spectrometer was used for data acquisition to ensure that the spectral intensity data for each wavelength was accurate to 0.1 nm. A background subtraction algorithm was used, and a Savitzky-Golay filter was applied to smooth the raw spectral data, removing noise interference and retaining the effective signal. In the characteristic peak intensity analysis, for the second fluorine content data, characteristic peaks at 250 nm, 350 nm, and 450 nm were identified, corresponding to different coordination states of fluorine atoms. Combining the fluorine atom coordination information, principal component analysis (PCA) was used to reduce the dimensionality of the characteristic peak intensities and extract key feature vectors. Based on these feature vectors, the support vector machine (SVM) algorithm in machine learning was used to generate the molecular structure fingerprint of the third barrier agent, determining its chemical structure. Finally, through optimization algorithms, combined with the molecular structure fingerprint and known barrier agent performance data, the optimal formulation of the third barrier agent was determined, with the fluorine content controlled at 12% ± 0.5% to ensure optimal barrier performance.

[0148] S13. According to the third barrier agent formula, extract the fluorine barrier agent, re-inject it into the pickling tank, control the acid temperature to stabilize at T, obtain the sedimentation particle size distribution data after the quartz sand surface treatment, and determine whether the particle size is lower than the preset threshold d.

[0149] In practice, based on the third barrier agent formula, a fluorinated barrier agent is extracted from the recycling database and re-injected into the pickling tank. A real-time acid temperature control device is used to maintain the temperature at 50°C. Data on the settling particle size distribution after quartz sand surface treatment is obtained to determine if the particle size is below a preset threshold of 5μm. The specific process includes:

[0150] Extract fluorine barrier agent formulation data from the recycling database and generate formulation injection instructions;

[0151] In practice, the formula data is retrieved from the recycling database, which includes the chemical composition and proportion of the fluorine barrier agent; the query results are converted into formula injection instructions through the data parsing module, which include the injection amount and injection rate.

[0152] The pickling tank equipment is controlled by formula injection instructions to complete the injection of fluorine barrier agent;

[0153] In practice, the formula injection command is transmitted to the control unit of the pickling tank equipment; the control unit adjusts the injection pump according to the command, and the fluorine barrier agent is injected from the storage tank into the pickling tank, generating an injection completion signal;

[0154] The temperature of the acid solution in the pickling tank is monitored by a real-time temperature control device and adjusted to a stable state.

[0155] In practice, the injection completion signal triggers the real-time temperature control device; the device acquires acid temperature data through a temperature sensor, and uses a proportional-integral-derivative algorithm to adjust the heating element to generate a stable temperature signal;

[0156] Obtain sedimentation particle samples of quartz sand after surface treatment and generate particle size data;

[0157] In practice, a stable temperature signal initiates the surface treatment process, and the quartz sand undergoes surface treatment in the pickling tank; the treated sedimented particles are extracted by a sampling device to generate particle size data.

[0158] Particle size data were processed using a laser particle size analyzer to obtain the particle size distribution;

[0159] In practice, the particle size data is transmitted to a laser particle size analyzer. The instrument analyzes the particle size using the principle of light scattering and generates particle size distribution data, including the statistical distribution of particle diameter.

[0160] If the maximum value in the particle size distribution is lower than the preset threshold d, a qualified judgment result is generated;

[0161] In practice, particle size distribution data is input into the threshold judgment module; if the diameter of the largest particle in the distribution is lower than the preset threshold d, the module generates a qualified judgment result; otherwise, an unqualified signal is generated.

[0162] The data acquisition module stores the pass / fail judgment results and updates the recycling database.

[0163] In practice, the qualification determination result is transmitted to the recycling database through the data acquisition module; the database updates the formula data and processing records, and generates an update completion signal.

[0164] Specifically, when extracting the fluorinated barrier agent from the recycling database based on the third barrier agent formulation, the system will call the preset SQL query statement "SELECT The algorithm uses the formula "FROM barrier_agent_table WHERE agent_type='fluorine' AND concentration>12.5%" to screen for fluorine barrier agent solutions with a concentration higher than 12.5%. Only solutions with a purity of 99.2% or higher, as determined by liquid chromatography, are allowed to be injected into the pickling tank. During injection, a PID control algorithm is employed, with a flow rate set at 2.3 L / min. The flow rate is adjusted via a solenoid valve to ensure an injection accuracy error of no more than ±0.05 L. A real-time acid temperature control device uses three PT100 temperature sensors for multi-point monitoring at a sampling frequency of 10 Hz. When a temperature deviation of 50 ± 0.5 °C is detected, the heating power is adjusted using a fuzzy control algorithm. The control parameters Kp=12.5, Ki=0.8, and Kd=3.2 ensure that the standard deviation of temperature fluctuation is less than 0.3℃. The treated quartz sand particles are tested using a laser particle size analyzer. The particle size distribution is calculated using Mie scattering theory. Statistical analysis is performed on more than 10,000 particles. When the d50 value is lower than 3.8μm and the d90 value is lower than 5.2μm, an early warning mechanism is triggered. The system automatically calls a BP neural network model to predict the treatment effect. When the prediction result shows that the upper limit of the 95% confidence interval is lower than the 5μm threshold, the data is stored in the quality database and a treatment qualification report is generated.

[0165] S14. If the particle size is lower than the preset threshold d, the concentration of fluoride ions in the waste acid after recycling is analyzed by ion chromatography to obtain the tertiary fluoride content data. Combined with the waste acid viscosity measurement data, the recycling efficiency is determined.

[0166] In practice, if the particle size is below the preset threshold of 5μm, the concentration of fluoride ions in the recycled waste acid is analyzed by ion chromatography to obtain the tertiary fluoride content data. Combined with the waste acid viscosity measurement data, the recycling efficiency is determined. The specific process includes:

[0167] If the particle size is below a preset threshold d, the fluoride ion concentration in the waste acid after recycling is analyzed by ion chromatography to obtain the tertiary fluoride content data. The viscosity of the waste acid after recycling is measured using a viscometer to obtain the waste acid viscosity data. A data fusion algorithm is used to standardize the tertiary fluoride content data and the waste acid viscosity data to obtain a fused feature vector. If the Euclidean distance of the fused feature vector is less than a preset distance threshold, a support vector machine classifier is used to determine the recycling efficiency and obtain an efficiency level. Based on the efficiency level, the operating parameters of the recycling process are grouped using a K-means clustering algorithm to obtain a parameter optimization set. The operating parameters of the recycling equipment are adjusted using the parameter optimization set to obtain updated processing efficiency data. If the updated processing efficiency data does not reach the preset efficiency threshold, the data fusion and efficiency judgment steps are repeated to obtain a new efficiency level.

[0168] Specifically, in the waste acid recycling process, the particle size is first detected by a particle size analyzer. If the detection result shows that the particle size is lower than the preset threshold of 5μm, then ion chromatography is started to analyze the fluoride ion concentration.

[0169] For example, using ion chromatography to analyze the waste acid sample, the fluoride ion concentration was measured to be 12.3 mg / L, which was taken as the third fluoride content data. Simultaneously, the viscosity of the waste acid was measured using a viscometer, and the viscosity was found to be 1.45 mPa·s. Combining the fluoride ion concentration and viscosity data, the recycling efficiency was calculated using an algorithm, assuming the formula η = (C0 - C1) / C0 × 100%, where C0 is the initial fluoride ion concentration and C1 is the fluoride ion concentration after treatment. If the initial fluoride ion concentration is 25.6 mg / L, then the recycling efficiency is (25.6 - 12.3) / 25.6 × 100% = 51.95%. Through this series of data collection, analysis, and calculations, the efficiency of waste acid recycling can be accurately evaluated, providing data support for subsequent process optimization.

[0170] S15. Based on the recycling efficiency, centrifugal separation technology is used to obtain data on the recovery rate of the separated liquid and the distribution of settled particles, and a second waste acid recycling scheme is generated.

[0171] In practice, based on the recycling efficiency, centrifugal separation technology is employed, with a centrifugal speed set at 9000 rpm and a separation time controlled at 20 minutes. Data on the recovery rate of the separated liquid and the distribution of settled particles are obtained to generate a second waste acid recycling scheme. The specific process includes:

[0172] Centrifugal separation technology was employed, with predefined rotation speed and separation time set to obtain initial separation liquid recovery rate and sedimentation particle distribution data, resulting in a first separation dataset. Particle distribution analysis was used to extract particle size distribution characteristics from the first separation dataset, determining particle distribution parameters. If the particle distribution parameters exceeded a preset threshold, the centrifugal rotation speed control parameters were adjusted, resulting in a second separation dataset to assess the degree of particle distribution optimization. Based on the second separation dataset, a recovery rate evaluation method was used to calculate the separation liquid recovery rate, obtaining the recovery rate evaluation result. A separation time optimization algorithm was used to extract the time influence factor from the recovery rate evaluation result, determining the optimized separation time parameters. Using the optimized separation time parameters and centrifugal rotation speed control parameters, a second waste acid recycling scheme was generated, resulting in the final recycling treatment scheme. Based on the final recycling treatment scheme, recycling efficiency data was obtained to assess the treatment effect of the waste acid recycling scheme.

[0173] Specifically, in centrifugal separation technology, the centrifugal speed is set to 9000 rpm. By calculating the centrifugal force F = 1.118 × 10^(-5) × r × (rpm)^2, where r is the centrifugal radius, assuming r is 10 cm, the centrifugal force F = 1.118 × 10^(-5) × 10 × (9000)^2 = 904.86 g, ensuring that the particles effectively settle in the centrifugal field. The separation time is controlled at 20 minutes. According to Stokes' Law, v = (2r^2(ρ_p-ρ_f)g) / (9η), where v is the settling velocity, r is the particle radius, ρ_p is the particle density, ρ_f is the fluid density, and η is the fluid viscosity. Assuming the particle radius is 1μm, the particle density is 2.5g / cm³, the fluid density is 1g / cm³, and the fluid viscosity is 0.01Pa·s, the settling velocity v = (2×(1×10^(-6))^2×(2.5-1)×9.8) / (9×0.01) = 2.18×10^(-6) m / s. The particle settling distance within 20 minutes is 2.18×10^(-6)×1200 = 2.616mm, ensuring sufficient particle settling. After centrifugation, the recovery rate of the separated liquid and the distribution of settled particles were obtained. Image analysis technology was used to statistically analyze the particle size distribution. Assuming the particle size distribution range is 0.5-5 μm, a particle size distribution curve was obtained through Gaussian fitting. The calculated average particle size was 2.5 μm, and the standard deviation was 0.8 μm, analyzing the uniformity of particle distribution. Based on the recovery rate and particle distribution data, a second waste acid recycling scheme was generated, setting the waste acid recycling rate at 85%. Through material balance calculations, the waste acid recycling volume was determined to be 85% of the initial waste acid volume, optimizing the waste acid treatment process, reducing waste acid emissions, and improving resource utilization.

[0174] It should also be noted that the various specific technical features described in the above embodiments can be combined in any suitable manner without contradiction. To avoid unnecessary repetition, the present invention will not describe the various possible combinations separately. Furthermore, various different embodiments of the present invention can also be arbitrarily combined, as long as they do not violate the spirit of the present invention, and should also be regarded as the content disclosed by the present invention.

Claims

1. A method for treating and utilizing fluoride-containing waste acid generated in a quartz sand pickling process, characterized in that, The method includes: Microwaves are generated by a microwave generator and emitted onto the surface of quartz sand in the pickling tank to obtain data on the changes in the dissolution rate of surface oxides and metal ions, and to determine the first dissolution rate. If the first dissolution rate is lower than the preset threshold v, the microwave frequency is adjusted, the acid temperature is controlled, the adjusted dissolution rate data is obtained, and the second dissolution rate is determined. Based on the second dissolution rate, the fluorine barrier agent formulation is matched from the recycling database, the molecular structure fingerprint data is extracted, and the first barrier agent addition scheme is generated. Using the first barrier agent addition scheme, a fluorinated barrier agent is injected into the pickling tank, the pH value of the acid solution is controlled, the formation status data of the temporary protective film on the surface of the quartz sand is obtained, and it is determined whether the coverage of the protective film reaches the preset threshold n. If the protective film coverage reaches the preset threshold n, then a waste acid sample is extracted from the pickling tank, the fluoride ion concentration in the waste acid is analyzed, and the first fluoride content data is obtained. Spectral scanning was performed on the first fluorine content data to obtain molecular vibrational modes and characteristic peak intensity data. Combined with fluorine atom coordination information, the molecular structure fingerprint of the first barrier agent was generated. Based on the molecular structure fingerprint of the first barrier agent, the fluorine barrier agent is separated from the waste acid, and the purity of the separated phase and the viscosity of the waste acid are measured to determine whether the purity of the first separation is higher than the preset threshold c. If the purity of the first separation is higher than the preset threshold c, a second spectral scan is performed on the separated fluorine barrier to obtain molecular vibration mode and characteristic peak intensity data, and to generate the molecular structure fingerprint of the second barrier. Based on the molecular structure fingerprint of the second barrier agent, the fluorine barrier agent formula is extracted, re-injected into the pickling tank, the microwave generator is controlled to operate, and the data of impurity residue after the surface treatment of quartz sand is obtained to determine whether the pickling effect reaches the point where the residual amount is lower than the preset threshold m. If the pickling effect reaches the point where the residual amount is lower than the preset threshold m, then the waste acid in the pickling tank is subjected to spectral scanning to obtain the data on the size distribution of settled particles and the recovery rate of the separated liquid, and a first waste acid recycling scheme is generated. Based on the first waste acid recycling scheme, adjust the centrifugal separation parameters, obtain the fluoride ion concentration data in the recycled waste acid, and determine the second fluoride content; Spectral scanning was performed on the second fluorine content data. The data was processed and combined with characteristic peak intensity analysis and fluorine atom coordination information to generate the molecular structure fingerprint of the third barrier agent and determine the formulation of the third barrier agent. Based on the third barrier agent formula, the fluorine barrier agent is extracted, re-injected into the pickling tank, the acid temperature is controlled, and the sedimentation particle size distribution data after the quartz sand surface treatment is obtained to determine whether the particle size is lower than the preset threshold d. If the particle size is lower than the preset threshold d, the concentration of fluoride ions in the waste acid after recycling is analyzed to obtain the tertiary fluoride content data. Combined with the waste acid viscosity measurement data, the recycling efficiency is determined. Based on the recycling efficiency, data on the recovery rate of the separated liquid and the distribution of settled particles are obtained to generate a second waste acid recycling scheme.

2. The method for treating and utilizing fluoride-containing waste acid generated in a quartz sand pickling process according to claim 1, characterized in that, The first barrier agent addition scheme involves injecting a fluorinated barrier agent into the pickling tank, controlling the pH value of the acid solution, obtaining data on the formation state of the temporary protective film on the quartz sand surface, and determining whether the protective film coverage reaches a preset threshold n; including: The real-time pH value data of the acid solution in the pickling tank is obtained by using sensors to obtain the pH value fluctuation range; If the pH value deviates from the preset range, the injection rate of the fluoride barrier agent is adjusted by a proportional-integral-differential algorithm to ensure that the pH value of the acid solution is stable within the preset range. Optical scanning technology was used to acquire images of the temporary protective film formation state on the surface of quartz sand, thus obtaining the first image data; The first image data is denoised and edge detected using image processing algorithms to obtain the second image data; Based on the second image data, the coverage rate of the temporary protective film is calculated using a convolutional neural network algorithm to obtain the coverage rate value; If the coverage value does not reach the preset threshold n, the amount of fluorine barrier agent injected will be increased through the feedback control system to obtain a new coverage value. Based on multiple coverage rate values, determine whether the coverage rate of the temporary protective film has stably reached the preset threshold n, and determine the final formation state.

3. The method for treating and utilizing fluoride-containing waste acid generated in a quartz sand pickling process according to claim 1, characterized in that, If the protective film coverage reaches a preset threshold n, a waste acid sample is extracted from the pickling tank, the fluoride ion concentration in the waste acid is analyzed, and the first fluoride content data is obtained; including: If the protective film coverage reaches the preset threshold n, waste acid samples are extracted from the pickling tank, and a fixed volume of waste acid samples is obtained using an automated sampling device to obtain waste acid sample data. The concentration of fluoride ions in waste acid samples was analyzed by ion chromatography to obtain the primary fluoride content data and determine the primary fluoride content. If the first fluoride content exceeds the preset concentration threshold, a data filtering algorithm is used to denoise the first fluoride content data to obtain the second fluoride content data. Based on the second fluoride content data, the fluoride ion concentration of the waste acid samples was classified and the concentration level was determined using the support vector machine algorithm. If the concentration level is high, the corresponding treatment parameters are obtained by querying the database to retrieve the pre-established waste acid treatment rules and determine the treatment plan. According to the treatment plan, an automated control system is used to adjust the waste acid discharge rate of the pickling tank, obtain the adjusted discharge data, and determine the discharge status. Emissions data are analyzed using time series analysis algorithms to obtain emission trend data and determine emission stability.

4. The method for treating and utilizing fluoride-containing waste acid generated in a quartz sand pickling process according to claim 1, characterized in that, The step of performing a spectral scan on the first fluorine content data to obtain molecular vibrational modes and characteristic peak intensity data, and combining this with fluorine atom coordination information to generate a molecular structure fingerprint of the first barrier agent, includes: The fluorine content data was scanned using a molecular spectroscopy analysis device, employing a wavelength range from ultraviolet to visible light to obtain the raw spectral data; The original spectral data is processed using a background subtraction algorithm. If environmental spectral interference is detected, the interference signal is separated by Fourier transform to obtain clean spectral data. Molecular vibrational modes are extracted from pure spectral data. If the intensity of the vibrational mode signal is lower than a preset threshold z1, the signal is enhanced by Gaussian fitting to determine the molecular vibrational mode. The intensity of characteristic peaks is analyzed based on molecular vibrational modes. If the peak intensity distribution conforms to a preset model, key peak data are extracted through principal component analysis to obtain the intensity of characteristic peaks. By combining the first fluorine content data and fluorine atom coordination information, if the coordination information matches the vibration mode, the data is fused through a convolutional neural network to generate a preliminary molecular structure fingerprint. Data extraction and processing are performed on the preliminary molecular structure fingerprint. If the fingerprint data integrity is higher than the preset threshold w1, the fingerprint features are optimized through cluster analysis to obtain the molecular structure fingerprint of the first barrier agent. Key feature data are extracted from the molecular structure fingerprint of the first barrier agent. Database comparison technology is used. If the feature data matches the preset template, the final molecular structure fingerprint is determined.

5. The method for treating and utilizing fluoride-containing waste acid generated in a quartz sand pickling process according to claim 1, characterized in that, The step of separating the fluorine barrier agent from the waste acid based on the molecular structure fingerprint of the first barrier agent, obtaining measurement data on the purity of the separated phase and the viscosity of the waste acid, and determining whether the first separation purity is higher than a preset threshold c; includes: Using a molecular structure fingerprint database, the characteristics of fluoride barrier agents in waste acid were matched to identify the target separation substance; Waste acid is treated by centrifugation at a preset speed and separation time to obtain a separated phase and residual waste acid; The purity data of the separated phase was obtained, and the concentration of the fluorine barrier agent was measured by high performance liquid chromatography to determine the separation purity. Viscosity data was extracted from the residual waste acid and measured using a digital viscometer to obtain the viscosity value of the waste acid. If the separation purity is higher than the preset threshold c, the separation effect is judged to be qualified and a qualified label is generated. If the value is below the preset threshold c, adjust the centrifugation speed and repeat the separation process; If the viscosity value is within the preset viscosity range, the waste acid is confirmed to be in normal condition. If it exceeds the range, it will be marked as an abnormal state; By using qualified labels and waste acid status, comprehensive evaluation data of the separation process is generated to determine the final treatment result.

6. The method for treating and utilizing fluoride-containing waste acid generated in a quartz sand pickling process according to claim 1, characterized in that, If the first separation purity is higher than a preset threshold c, then a second spectral scan is performed on the separated fluorine barrier to obtain molecular vibrational modes and characteristic peak intensity data, generating a second barrier molecular structure fingerprint; including: If the purity of the first separation is higher than the preset threshold c, a second scan of the fluorine barrier is performed using a molecular spectroscopy analysis device, employing the ultraviolet to near-infrared band to obtain the original spectral data; The original spectral data is processed by a background subtraction algorithm to remove environmental noise and baseline drift, resulting in clean spectral data. The Fourier transform algorithm is used to perform frequency domain transformation on the pure spectral data, extract molecular vibrational modes, and determine the characteristic frequency distribution; Based on the characteristic frequency distribution, the characteristic peak intensity is calculated, and an intensity dataset is generated; Principal component analysis algorithm is used to reduce the dimensionality of the intensity dataset, extract the main molecular structure features, and generate the molecular structure fingerprint of the second barrier agent. If the similarity between the molecular structure fingerprint and the preset standard fingerprint is higher than the threshold, the chemical composition information of the second barrier agent is obtained by comparing with the database. Based on the chemical composition information, a three-dimensional structural model of the second barrier agent was generated using molecular simulation tools to determine its molecular configuration characteristics.

7. The method for treating and utilizing fluoride-containing waste acid generated in a quartz sand pickling process according to claim 1, characterized in that, If the pickling effect reaches the point where the residual amount is lower than the preset threshold m, then the waste acid in the pickling tank is subjected to spectral scanning to obtain the data on the size distribution of settled particles and the recovery rate of the separation liquid, and a first waste acid recycling scheme is generated. include: If the residual amount of acid washing is lower than the preset threshold m, the waste acid is spectrally scanned using a molecular spectral analysis device, covering the ultraviolet to visible spectrum, to obtain the original spectral data; The raw spectral data were processed by data normalization, and features were extracted using principal component analysis algorithm to obtain data on sedimentation particle size distribution and separation liquid recovery rate. Based on the size distribution of settled particles, a clustering analysis algorithm is used to classify the particles and obtain the particle category distribution; If the proportion of large particles in the particle size distribution is higher than the preset proportion, the spectral data is processed again by digital filtering technology to obtain the optimized particle size distribution. Based on the optimized particle size distribution and separation liquid recovery rate data, a linear regression algorithm is used to predict the waste acid recycling efficiency and obtain the recycling efficiency parameter. By analyzing the circulation efficiency parameters and particle type distribution, a first waste acid circulation scheme is generated, and the circulation scheme parameters are determined. Based on the parameters of the recycling scheme, adjust the waste acid treatment process and obtain updated pickling residue data.

8. The method for treating and utilizing fluoride-containing waste acid generated in a quartz sand pickling process according to claim 1, characterized in that, The step of adjusting the centrifugal separation parameters according to the first waste acid recycling scheme, obtaining fluoride ion concentration data in the recycled waste acid, and determining the second fluoride content includes: Acquire waste acid viscosity measurement data; obtain the first viscosity data by collecting viscosity values ​​in real time through sensors. If the first viscosity data exceeds the preset threshold η, the centrifugation separation parameters are adjusted using a linear regression algorithm to obtain the first separation parameters; Based on the first separation parameters, the centrifugal speed is set and the separation time is controlled to perform centrifugal separation operation and obtain the first treated waste acid; The concentration of fluoride ions in the first-treatment waste acid was detected by ion chromatography to obtain the first concentration data. If the first concentration data does not meet the preset standard, the number of cycles is adjusted according to the concentration deviation to obtain the second waste acid. The concentration of fluoride ions in the waste acid from the second treatment was determined again by ion chromatography to confirm the second fluoride content. Based on the deviation between the second fluorine content and the preset target, the separation parameters are optimized through a feedback control algorithm to obtain the second separation parameters.

9. The method for treating and utilizing fluoride-containing waste acid generated in a quartz sand pickling process according to claim 1, characterized in that, The step involves extracting a fluorine barrier agent according to the third barrier agent formulation, re-injecting it into the pickling tank, controlling the acid temperature, obtaining the sedimentation particle size distribution data after quartz sand surface treatment, and determining whether the particle size is lower than a preset threshold d; including: Extract fluorine barrier agent formulation data from the recycling database and generate formulation injection instructions; The pickling tank equipment is controlled by formula injection instructions to complete the injection of fluorine barrier agent; The temperature of the acid solution in the pickling tank is monitored by a real-time temperature control device and adjusted to a stable state. Obtain sedimentation particle samples of quartz sand after surface treatment and generate particle size data; Particle size data were processed using a laser particle size analyzer to obtain the particle size distribution; If the maximum value in the particle size distribution is lower than the preset threshold d, a qualified judgment result is generated; The data acquisition module stores the pass / fail judgment results and updates the recycling database.

10. A method for treating and utilizing fluoride-containing waste acid generated in a quartz sand pickling process according to claim 1, characterized in that, If the particle size is lower than a preset threshold d, the concentration of fluoride ions in the waste acid after recycling is analyzed to obtain the tertiary fluoride content data. Combined with the waste acid viscosity measurement data, the recycling efficiency is determined; including: If the particle size is lower than the preset threshold d, the concentration of fluoride ions in the waste acid after recycling is analyzed by ion chromatography to obtain the third fluoride content data. The viscosity of the waste acid after recycling was measured by a viscometer to obtain the waste acid viscosity data; A data fusion algorithm was used to standardize the tertiary fluoride content data and the waste acid viscosity data to obtain a fused feature vector; If the Euclidean distance between the fused feature vectors is less than a preset distance threshold, the efficiency of the loop processing is determined by the support vector machine classifier to obtain the efficiency level. Based on efficiency levels, the K-means clustering algorithm is used to group the running parameters of the loop processing to obtain a parameter optimization set; By adjusting the operating parameters of the cyclic processing equipment through parameter optimization sets, updated processing efficiency data can be obtained. If the updated processing efficiency data does not reach the preset efficiency threshold, the data fusion and efficiency judgment steps are repeated to obtain a new efficiency level.

Images