Quartz sand chlorination purification method

Abstract

The invention relates to the technical field of quartz sand purification, in particular to a quartz sand chlorination purification method which comprises the following steps: filling quartz sand to be purified into a chlorination reaction furnace; introducing hydrogen chloride gas into the chlorination reaction furnace in a pulse injection mode; temperature abnormal points are identified based on the temperature data in the quartz sand material layer, and first-order temperature abnormal reasons of the temperature abnormal points are judged based on historical data of the temperature abnormal points; based on the change of the judgment index in the preset time window, screening first-order temperature anomaly reasons, and judging second-order temperature anomaly reasons of the temperature anomaly points; according to the second-order temperature abnormity reason, an adjusting strategy is formulated; according to the method, the property and severity of temperature anomaly can be accurately distinguished, and a regulation and control instruction is generated, so that the process efficiency is optimized on the premise of ensuring reaction safety and product purity, and stable and efficient production of high-quality quartz sand is realized.

Description

Technical Field

[0001] This invention relates to the field of quartz sand purification technology, and more particularly to a method for chlorination purification of quartz sand. Background Technology

[0002] High-purity silica sand is a key basic material for high-end industries such as semiconductors, photovoltaics, and fiber optic communications. Its purity directly determines the performance and yield of downstream products. Among them, alkali metal impurities such as sodium are the core factors affecting the insulation and thermal stability of silica sand. High-temperature chlorination purification technology is the mainstream process for removing these impurities. Its principle is to introduce hydrogen chloride gas at high temperature, causing the impurity elements to be converted into volatile chlorides and removed.

[0003] In existing chlorination purification processes, a fixed process parameter-based operation mode is generally adopted, supplemented by basic temperature monitoring. A common control method is to arrange a limited number of thermocouples in the reactor to monitor the overall or local temperature. When the temperature exceeds a certain safety threshold, general adjustment measures such as reducing heating power or reducing gas flow are taken.

[0004] However, this traditional monitoring and control method has significant drawbacks. First, it typically only detects the surface symptom of temperature anomalies and cannot diagnose the underlying causes. Local fluctuations or increases in the temperature field within the reactor can be caused by a variety of factors, such as uneven distribution of impurities in the quartz sand raw material, uneven distribution of hydrogen chloride gas in the packing layer forming channeling, uneven thermal field of the heating system itself, or mismatch between pulsed gas inlet parameters and the current reaction state. Different underlying causes require drastically different control strategies.

[0005] Existing technologies often lack the ability to deeply diagnose the causes of anomalies, resulting in blind and crude control actions. For example, regardless of whether the anomaly is caused by uneven raw material distribution or uneven airflow, the system may simply perform a single operation of cooling or reducing airflow. This adjustment method may not only fail to effectively eliminate the real risk points, but may also excessively suppress the reaction efficiency of normal areas, leading to decreased production efficiency, increased energy consumption, and even the introduction of new instability factors due to misadjustment. Furthermore, alarm mechanisms relying solely on instantaneous temperature exceeding the threshold cannot distinguish between short-term fluctuations and continuously deteriorating anomalies, easily causing false alarms or missed alarms, making it difficult to accurately guarantee the safety of the production process and the consistency of product quality.

[0006] The information disclosed in this background section is intended only to enhance the understanding of the general background of this disclosure and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0007] This invention provides a method for chlorination and purification of quartz sand, which can accurately distinguish the nature and severity of temperature anomalies and generate control instructions, thereby optimizing process efficiency and achieving stable and efficient production of high-quality quartz sand while ensuring reaction safety and product purity.

[0008] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method for chlorinating and purifying quartz sand, the method comprising: S1. The quartz sand to be purified is loaded into the chlorination reactor; S2 introduces hydrogen chloride gas into the chlorination reactor via pulse injection; S3 identifies temperature anomalies based on temperature data inside the quartz sand layer, and determines the first-order cause of temperature anomalies based on historical data of each anomaly. S4 filters the causes of first-order temperature anomalies based on the changes in the indicators within a preset time window, and determines the causes of second-order temperature anomalies at the temperature anomaly points. S5 formulates a regulation strategy based on the causes of second-order temperature anomalies.

[0009] Furthermore, the material storage capacity of the furnace tubes in the chlorination reactor is 20-30% of the furnace tube volume.

[0010] Furthermore, the flow rate of hydrogen chloride gas is 300~400L / h.

[0011] Furthermore, the method for obtaining the cause of a first-order temperature anomaly includes: S31 acquires historical temperature data for each temperature anomaly point and its surrounding reference points; S32 obtains the characteristic parameters of each temperature anomaly point based on historical temperature data. The characteristic parameters include: the intensity of the historical temperature difference sequence fluctuation of the temperature anomaly point and the energy proportion of the spectral characteristics of its historical temperature data in the preset frequency band. S33 inputs the feature parameters into a pre-trained cause classification model to obtain the cause of temperature anomalies at each temperature anomaly point; S34 Calculates the proportion of each cause category among all temperature anomalies and sorts them in descending order of proportion; selects the cause category with the highest proportion as the main suspected cause; S35 determines the first-order cause of temperature anomalies based on the proportion of the main suspected causes and the spatial distribution pattern of the corresponding temperature anomalies.

[0012] Furthermore, the method for obtaining the intensity of fluctuations in historical temperature difference sequences is as follows: S321 acquires temperature data of temperature anomaly points and surrounding reference points over a historical period, resulting in multiple historical temperature difference sequences; S322 calculates the standard deviation of each historical temperature difference series and uses the average of multiple standard deviations as the fluctuation intensity of the historical temperature difference series.

[0013] Furthermore, the method for obtaining the energy proportion of the spectral characteristics of historical temperature data within a preset frequency band is as follows: S32a performs a Fourier transform on the historical temperature data of each temperature anomaly point to obtain the spectrum; The S32b calculates the ratio of the energy of the spectrum within a preset frequency band to the total energy of the spectrum, which is the energy percentage.

[0014] Furthermore, the cause of the first-order temperature anomaly is at least one of the following types: (1) If the quantity ratio exceeds the first threshold and the spatial distribution has no specific pattern, it is determined that the impurity elements in the quartz sand raw material are unevenly distributed; (2) If the proportion of the number exceeds the second threshold and the abnormal points are concentrated in the area downstream of the gas inlet or near the furnace wall, it is determined that the hydrogen chloride gas flow rate is uneven or the gas flow is short-circuited. (3) If the proportion of abnormal points exceeds the third threshold and the abnormal points are distributed in a ring or layer around the heating area, it is determined that the axial or radial temperature gradient in the reactor is inaccurate.

[0015] Furthermore, the judgment indicators include: temperature difference persistence index and temperature pulse correlation index; Among them, the temperature difference persistence index is: the cumulative percentage of time during which the temperature of an abnormal point is continuously higher than a set threshold within a preset time window. The set threshold is a value determined based on the average temperature of its surrounding reference points. The temperature pulse correlation index is the correlation coefficient between the temperature change sequence of temperature anomalies and the timing signal of hydrogen chloride gas pulse injection within a preset time window.

[0016] Furthermore, the methods for calculating the correlation coefficient include: S41 samples the temperature change sequence of the temperature anomaly point at equal time intervals within a preset time window to obtain the first discrete time sequence; simultaneously acquires the hydrogen chloride gas pulse injection status of the corresponding time period, marks the ventilation status as the first value, and marks the gas cut-off status as the second value to form the second discrete time sequence. S42 calculates the cross-correlation coefficient between the first discrete-time series and the second discrete-time series, which is used as the correlation coefficient; the cross-correlation coefficient is calculated as the ratio of the product of the covariance of the two series to their respective standard deviations.

[0017] Furthermore, if it is (1), the adjustment strategy is to increase the reaction temperature and increase the pulse frequency; if it is (2), the adjustment strategy is to reduce the total gas flow rate and adjust the pulse duty cycle; if it is (3), the adjustment strategy is to adjust the heating power distribution of the reactor and increase the pulse gas preheating temperature.

[0018] The technical solution of this invention achieves the following technical effects: This invention monitors temperature data and identifies temperature anomalies using temperature sensors deployed inside the quartz sand layer. Based on historical temperature data from each anomaly point and its surrounding reference points, characteristic parameters are obtained and input into a pre-trained cause classification model to determine the first-order cause of the temperature anomaly. Subsequently, within a preset time window, based on changes in temperature difference persistence and temperature pulse correlation indicators, the first-order cause of the temperature anomaly is screened to identify the second-order cause requiring final control. Finally, based on the second-order cause of the temperature anomaly, a control strategy is matched and formulated. For example, pulse injection parameters are adjusted for airflow distribution problems identified through screening, rather than performing general cooling or flow reduction operations. The control logic is transformed from a response to temperature anomalies to targeted control based on hierarchical diagnosis and cause screening, thereby maintaining the stability of the overall response process while identifying and eliminating specific risk sources. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of the process for chlorination purification of quartz sand. Detailed Implementation

[0021] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0023] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application.

[0024] Example: like Figure 1 As shown, this application provides a method for chlorinating and purifying quartz sand, the method comprising: S1. The quartz sand to be purified is loaded into the chlorination reactor; S2 introduces hydrogen chloride gas into the chlorination reactor via pulse injection; S3 identifies temperature anomalies based on temperature data inside the quartz sand layer, and determines the first-order cause of temperature anomalies based on historical data of each anomaly. S4 filters the causes of first-order temperature anomalies based on the changes in the indicators within a preset time window, and determines the causes of second-order temperature anomalies at the temperature anomaly points. S5 formulates a regulation strategy based on the causes of second-order temperature anomalies.

[0025] This invention monitors temperature data and identifies temperature anomalies using temperature sensors deployed within a quartz sand layer. Based on historical temperature data from each anomaly and surrounding reference points, characteristic parameters are obtained and input into a pre-trained cause classification model to determine the primary causes of the temperature anomalies. Subsequently, within a preset time window, the primary causes are filtered based on changes in temperature difference persistence and temperature pulse correlation indices to identify secondary causes requiring final intervention. Finally, adjustment strategies are matched and formulated based on the secondary causes of the anomalies. For example, pulse injection parameters are adjusted to address identified airflow distribution issues, rather than performing general cooling or flow rate reduction operations. This transforms the control logic from responding to temperature anomalies to targeted regulation based on hierarchical diagnosis and cause screening, thereby maintaining the stability of the overall response process while identifying and eliminating specific risk sources.

[0026] Furthermore, the material storage capacity of the furnace tubes in the chlorination reactor is 20-30% of the furnace tube volume.

[0027] When the material level is below 20%, the quality of quartz sand in the effective reaction zone per unit time is insufficient, resulting in a decrease in single-pass processing capacity and insufficient utilization of equipment capacity. When the material level exceeds 30%, the material in the furnace tube is too densely packed, which will hinder the uniform diffusion and penetration of hydrogen chloride gas in the material layer, resulting in uneven gas distribution and insufficient local reaction.

[0028] Furthermore, the flow rate of hydrogen chloride gas is 300~400L / h.

[0029] When the gas flow rate is below 300 L / h, the reactant supply is insufficient, the reaction rate decreases, and the sodium is not completely removed. When the gas flow rate exceeds 400 L / h, although the reaction rate can be increased, the excess gas will exacerbate the exothermic reaction and easily cause local overheating.

[0030] Furthermore, the method for obtaining the cause of a first-order temperature anomaly includes: S31 acquires historical temperature data for each temperature anomaly point and its surrounding reference points; Specifically, from the distributed temperature sensor network, the temperature data of each identified temperature anomaly point is read at a fixed frequency over a recent period. Simultaneously, centered on the anomaly point, all other sensors not marked as anomalies are selected within a preset spatial range as surrounding reference points. The arithmetic mean of the temperature data of these reference points over the same period is calculated to generate an average temperature sequence representing the local environmental background, which serves as historical temperature data. S32 obtains the characteristic parameters of each temperature anomaly point based on historical temperature data. The characteristic parameters include: the intensity of the historical temperature difference sequence fluctuation of the temperature anomaly point and the energy proportion of the spectral characteristics of its historical temperature data in the preset frequency band. Specifically, the intensity of historical temperature difference series fluctuations is used to quantify the stability of temperature deviations from their local environment at anomalies. Anomalies caused by uneven local impurity concentrations in raw materials tend to have relatively stable exothermic reactions, resulting in lower temperature difference fluctuation intensity values. Anomalies caused by dynamic mass transfer issues such as uneven airflow distribution result in intermittent reactant supply, leading to fluctuations in reaction rates and exothermic reactions, thus resulting in higher temperature difference fluctuation intensity values. The energy proportion of the spectral characteristics of historical temperature data is used to quantify the periodic correlation between temperature fluctuations and pulsed injection operations. If the anomaly is directly related to periodic disturbances in the pulsed airflow, its temperature fluctuation spectrum will show energy concentration near the pulse frequency, resulting in a higher proportion. If the anomaly originates from static factors unrelated to the pulse rhythm, this proportion will be lower. These two parameters provide complementary criteria for distinguishing between essential and mass transfer anomalies from the perspectives of temporal stability and frequency domain periodicity, respectively.

[0031] S33 inputs the feature parameters into a pre-trained cause classification model to obtain the cause of temperature anomalies at each temperature anomaly point; Specifically, the two feature parameter values ​​of each anomaly point calculated in step S32 are input into a pre-trained cause classification model. The cause classification model is trained based on historical production data, and the training samples consist of a large number of temperature anomaly cases whose causes have been confirmed by process experts, along with their corresponding feature parameters. The model outputs the preliminary cause category to which the temperature anomaly point belongs, such as uneven distribution of raw material impurities or uneven airflow distribution. This result serves as the first-order cause of the temperature anomaly point.

[0032] S34 Calculates the proportion of each cause category among all temperature anomalies and sorts them in descending order of proportion; selects the cause category with the highest proportion as the main suspected cause; Specifically, after step S33, the number of all temperature anomalies in the current batch that belong to each cause category is counted, and the proportion of each category is calculated. All proportions are sorted from highest to lowest, and the cause category corresponding to the highest proportion is directly selected as the primary suspected cause that needs to be prioritized and verified.

[0033] S35 determines the first-order temperature anomaly cause of the temperature anomaly point based on the proportion of the main suspected causes and the spatial distribution pattern of the corresponding temperature anomaly points. Specifically, based on quantitative proportions, the dominant potential causes among all anomalies can be identified, thus converging the analytical focus from scattered individual phenomena to the most likely systemic root cause. Furthermore, combining this with spatial distribution analysis allows the statistical conclusions to be correlated with the specific physical structure of the reactor, airflow path, or heating mode. The combination of these two approaches confirms the universality of the problem through proportions and verifies the correlation between the cause and the process equipment through spatial distribution, thus providing both statistical and physical evidence for pinpointing the true cause of the first-order temperature anomaly.

[0034] Furthermore, the method for obtaining the intensity of fluctuations in historical temperature difference sequences is as follows: S321 acquires temperature data of temperature anomaly points and surrounding reference points over a historical period, resulting in multiple historical temperature difference sequences; Specifically, for an identified temperature anomaly, temperature sampling data for that point within the most recent set time period is retrieved from the process database, with a sampling interval of 1 second. Simultaneously, within a 150 mm radius of the anomaly, eight sensors not marked as anomalies and evenly distributed are identified as independent peripheral reference points. Temperature sampling data for each of these eight reference points within the same most recent set time period is retrieved. For each peripheral reference point, its temperature data sequence is aligned with the temperature data sequence of the temperature anomaly point at each time point, and the difference at each sampling moment is calculated, thereby generating eight independent historical temperature difference sequences. Each temperature difference sequence represents the change in temperature deviation between the anomaly point and a specific peripheral reference point over time.

[0035] S322 calculates the standard deviation of each historical temperature difference series and uses the average of multiple standard deviations as the fluctuation intensity of the historical temperature difference series. Specifically, for the aforementioned historical temperature difference series, the standard deviation is calculated for each data point. The formula for calculating the standard deviation is the square root of the sum of the squares of the deviations of each data point from its mean. This yields eight standard deviation values. Subsequently, the arithmetic mean of these eight standard deviations is calculated. This arithmetic mean is ultimately defined as the fluctuation intensity of the historical temperature difference series at that temperature anomaly point.

[0036] The above method generates multiple historical temperature difference sequences by using multiple independent surrounding reference points and calculates the average of their standard deviations to obtain the fluctuation intensity of the historical temperature difference sequence of the temperature anomaly point. This reduces the measurement deviation caused by the accidental fluctuations of a single reference point or local transient interference, and makes the obtained fluctuation intensity value more stably reflect the fluctuation level of the temperature deviation of the temperature anomaly point relative to its overall surrounding environment.

[0037] Furthermore, the method for obtaining the energy proportion of the spectral characteristics of historical temperature data within a preset frequency band is as follows: S32a performs a Fourier transform on the historical temperature data of each temperature anomaly point to obtain the spectrum; Specifically, for an identified temperature anomaly, temperature data sampled at 1-second intervals within the most recent set time period for that point are retrieved from the process database, resulting in 1800 sequentially arranged temperature values. This set of time-ordered temperature values ​​is then input into a Fast Fourier Transform (FFT) algorithm. This algorithm converts the time-dimensional temperature change signal into a frequency-dimensional representation, outputting a result characterizing the distribution of signal energy at different frequencies, thus obtaining the frequency spectrum of the historical temperature data for that anomaly.

[0038] S32b calculates the ratio of the energy of the spectrum within a preset frequency band to the total energy of the spectrum, which is the energy percentage. Specifically, from the frequency spectrum obtained in step S32a, the energy values ​​corresponding to all frequency points within the preset frequency band are extracted and summed to obtain the cumulative energy within the frequency band. Simultaneously, the sum of the energy values ​​of all frequency points in the entire frequency spectrum is calculated to obtain the total spectral energy. Finally, the ratio of the cumulative energy of the preset frequency band to the total spectral energy is calculated; this ratio is defined as the energy proportion of the spectral characteristics of the historical temperature data of the temperature anomaly point within the preset frequency band.

[0039] Furthermore, the cause of the first-order temperature anomaly is at least one of the following types: (1) If the quantity ratio exceeds the first threshold and the spatial distribution has no specific pattern, it is determined that the impurity elements in the quartz sand raw material are unevenly distributed; Specifically, the first threshold is obtained by statistically analyzing batches with irregular spatial distribution of outliers in historical production data. The specific method involves collecting production data from multiple batches of normal raw materials and batches with unevenly distributed known impurities, and calculating the proportion of temperature anomalies in each batch. By comparing the differences in the proportion distribution of the two types of batches, a critical proportion value that can effectively distinguish between them is selected as the first threshold. Alternatively, based on a large amount of historical data, this proportion value can be set at a specific high percentage of the irregularly distributed batch proportion distribution.

[0040] (2) If the proportion of the number exceeds the second threshold and the abnormal points are concentrated in the area downstream of the gas inlet or near the furnace wall, it is determined that the hydrogen chloride gas flow rate is uneven or the gas flow is short-circuited. Specifically, the determination of the second threshold combines computational fluid dynamics simulation and actual gas distribution testing. First, CFD simulations of the gas flow field within the reactor under different operating conditions identify critical conditions that lead to the formation of significant low-velocity zones or short-circuit paths downstream of the inlet or near the furnace wall. The proportion of abnormal virtual monitoring points in the corresponding operating conditions is statistically analyzed. Second, gas tracing or pressure distribution tests are conducted in actual equipment to verify the simulation results and fine-tune the proportions. Finally, the proportion of critical abnormal points, verified through simulation and experimentation and characterizing the beginning of systematic unevenness in gas flow distribution, is set as the second threshold.

[0041] (3) If the proportion of abnormal points exceeds the third threshold and the abnormal points are distributed in a ring or layer around the heating area, it is determined that the axial or radial temperature gradient in the reactor is inaccurate. Specifically, the third threshold is determined based on reactor hot-state calibration data and heating system performance testing. During reactor commissioning or periodic maintenance, the three-dimensional temperature field inside the furnace is measured under different heater power configurations using a high-density temperature sensor array or thermal imager. The patterns of annular or layered temperature anomalies around the heating area are analyzed, and the lower limit of the proportion of anomaly points that cause this anomaly pattern to recur stably is determined. This lower limit, combining the heating system's design uniformity indicators and the process's minimum requirements for temperature field uniformity, is established as the third threshold.

[0042] Furthermore, the judgment indicators include: temperature difference persistence index and temperature pulse correlation index; Among them, the temperature difference persistence index is: the cumulative percentage of time during which the temperature of an abnormal point is continuously higher than a set threshold within a preset time window. The set threshold is a value determined based on the average temperature of its surrounding reference points. The temperature pulse correlation index is the correlation coefficient between the temperature change sequence of temperature anomalies and the timing signal of hydrogen chloride gas pulse injection within a preset time window.

[0043] Specifically, the temperature difference persistence index directly verifies the stability of abnormally high-temperature states. If the first-order cause is a real and persistent problem, its temperature difference persistence index value is expected to remain high. If the high-temperature state is only temporary or intermittent, the temperature difference persistence index will be low, thus helping to filter out suspected anomalies caused by temporary disturbances and improving the accuracy of identifying persistent risks.

[0044] Temperature pulse correlation indicators can independently verify whether an anomaly is strongly correlated with external periodic process operations. If the cause of a first-order temperature anomaly is a mass transfer problem caused by uneven pulsed airflow, its temperature fluctuation should show a high positive or negative correlation with the pulse signal. If the correlation is very weak, it indicates that the temperature change at that point has a low correlation with the pulse operation, thereby weakening the credibility of the mass transfer anomaly judgment or supporting a shift to other cause hypotheses.

[0045] Furthermore, the methods for calculating the correlation coefficient include: S41 samples the temperature change sequence of the temperature anomaly point at equal time intervals within a preset time window to obtain the first discrete time sequence; simultaneously acquires the hydrogen chloride gas pulse injection status of the corresponding time period, marks the ventilation status as the first value, and marks the gas cut-off status as the second value to form the second discrete time sequence. Specifically, for a first-order temperature anomaly to be screened, its original temperature sampling sequence within a preset time window is obtained, with a sampling interval of 1 second. In order to highlight the temperature change trend and reduce the influence of baseline drift, the original temperature sequence is subjected to first-order difference processing, that is, the difference between the temperature of each sampling point and the temperature of the previous sampling point is calculated, thereby obtaining a sequence reflecting the rate of temperature change, which serves as the first discrete time sequence.

[0046] Within the same precise time interval, the pulse valve command signal controlling the on / off state of hydrogen chloride gas is read. This signal is a binary state sequence synchronized with the temperature sampling clock. This state sequence is aligned with the temperature data using the same sampling interval. At each sampling moment, if the command is to open the valve, the state at that moment is marked with a value of 1; if the command is to close the valve, it is marked with a value of 0. This generates a second discrete time sequence that is exactly the same length as the first discrete time sequence and corresponds one-to-one with its time points.

[0047] S42 calculates the cross-correlation coefficient between the first discrete-time series and the second discrete-time series, which is used as the correlation coefficient; the cross-correlation coefficient is calculated as the ratio of the product of the covariance of the two series to their respective standard deviations.

[0048] The aforementioned method effectively eliminates baseline drift interference by processing the temperature series using first-order differencing, allowing the analysis to focus on the instantaneous dynamic changes in temperature and improving the accuracy of subsequent analyses. Encoding pulse states as binary sequences transforms process operations into computational objects that can be precisely time-sequentially aligned with temperature data. Finally, by calculating the standard cross-correlation coefficient between the two sequences, a statistic with a well-defined numerical range and mathematical definition is obtained. This provides an objective, threshold-based quantitative basis for determining whether temperature anomalies are statistically significantly associated with pulse operations, enhancing the rigor of the decision-making process from first-order to second-order cause screening.

[0049] Furthermore, if it is (1), the adjustment strategy is to increase the reaction temperature and increase the pulse frequency; if it is (2), the adjustment strategy is to reduce the total gas flow rate and adjust the pulse duty cycle; if it is (3), the adjustment strategy is to adjust the heating power distribution of the reactor and increase the pulse gas preheating temperature.

[0050] Regarding (1), increasing the overall reaction temperature aims to accelerate the reaction rate in the low-impurity region, reduce the progress difference with the high-impurity region, and promote uniformity. Increasing the pulse frequency, by enhancing gas disturbance and penetration, improves reactant distribution and enhances heat diffusion, helping to alleviate local overheating. The two measures work synergistically to promote the overall reaction while taking into account thermal stability.

[0051] Regarding (2), reducing the total flow rate can weaken the airflow short-circuiting tendency and make the flow more gentle; adjusting the pulse duty cycle can make the airflow shorter and more impactful, breaking the original stable flow channel and improving the uniformity of the lateral distribution of gas in the material layer.

[0052] Regarding (3), adjusting the heating power distribution can directly correct the unevenness of the axial or radial thermal field; increasing the pulse gas preheating temperature can avoid the periodic cooling interference of cold gas injection on the local temperature field, ensuring that the set temperature gradient is maintained stably.

[0053] Although this application has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made thereto without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely exemplary illustrations of the application as defined herein, and are to be considered as covering any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Thus, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method for chlorinating and purifying quartz sand, characterized in that, The method includes: S1. The quartz sand to be purified is loaded into the chlorination reactor; S2 introduces hydrogen chloride gas into the chlorination reactor via pulse injection; S3 identifies temperature anomalies based on temperature data inside the quartz sand layer, and determines the first-order cause of temperature anomalies based on historical data of each anomaly. S4 filters the causes of first-order temperature anomalies based on the changes in the indicators within a preset time window, and determines the causes of second-order temperature anomalies at the temperature anomaly points. S5 formulates a regulation strategy based on the causes of second-order temperature anomalies.

2. The method for chlorinating and purifying quartz sand according to claim 1, characterized in that, The material storage capacity of the furnace tubes in the chlorination reactor is 20-30% of the furnace tube volume.

3. The method for chlorinating and purifying quartz sand according to claim 1, characterized in that, The flow rate of the hydrogen chloride gas is 300~400L / h.

4. The method for chlorinating and purifying quartz sand according to claim 1, characterized in that, The method for obtaining the cause of the first-order temperature anomaly includes: S31 acquires historical temperature data for each temperature anomaly point and its surrounding reference points; S32 obtains the characteristic parameters of each temperature anomaly point based on the historical temperature data. The characteristic parameters include: the intensity of the historical temperature difference sequence fluctuation of the temperature anomaly point and the energy proportion of the spectral characteristics of its historical temperature data in the preset frequency band. S33 inputs the feature parameters into a pre-trained cause classification model to obtain the cause of temperature anomaly at each temperature anomaly point; S34 Calculates the proportion of each cause category among all temperature anomalies and sorts them in descending order of proportion; selects the cause category with the highest proportion as the main suspected cause; S35 determines the first-order cause of temperature anomalies based on the proportion of the main suspected causes and the spatial distribution pattern of the corresponding temperature anomalies.

5. The method for chlorinating and purifying quartz sand according to claim 4, characterized in that, The method for obtaining the intensity of the historical temperature difference sequence fluctuations is as follows: S321 acquires temperature data of temperature anomaly points and surrounding reference points over a historical period, resulting in multiple historical temperature difference sequences; S322 calculates the standard deviation of each historical temperature difference sequence and uses the average of multiple standard deviations as the fluctuation intensity of the historical temperature difference sequence.

6. The method for chlorinating and purifying quartz sand according to claim 4, characterized in that, The method for obtaining the energy proportion of the spectral characteristics of the historical temperature data within the preset frequency band is as follows: S32a performs a Fourier transform on the historical temperature data of each temperature anomaly point to obtain the spectrum; S32b calculates the ratio of the energy of the spectrum within the preset frequency band to the total energy of the spectrum, which is the energy percentage.

7. The method for chlorinating and purifying quartz sand according to claim 4, characterized in that, The cause of a first-order temperature anomaly is at least one of the following types: (1) If the quantity ratio exceeds the first threshold and the spatial distribution has no specific pattern, it is determined that the impurity elements in the quartz sand raw material are unevenly distributed; (2) If the proportion of the number exceeds the second threshold and the abnormal points are concentrated in the area downstream of the gas inlet or near the furnace wall, it is determined that the hydrogen chloride gas flow rate is uneven or the gas flow is short-circuited. (3) If the proportion of abnormal points exceeds the third threshold and the abnormal points are distributed in a ring or layer around the heating area, it is determined that the axial or radial temperature gradient in the reactor is inaccurate.

8. The method for chlorinating and purifying quartz sand according to claim 1, characterized in that, The judgment indicators include: temperature difference persistence index and temperature pulse correlation index; The temperature difference persistence index is: the percentage of time during which the temperature of an abnormal point remains higher than a set threshold within the preset time window, where the set threshold is a value determined based on the average temperature of its surrounding reference points. The temperature pulse correlation index is the correlation coefficient between the temperature change sequence of the temperature anomaly point and the hydrogen chloride gas pulse injection timing signal within the preset time window.

9. The method for chlorinating and purifying quartz sand according to claim 8, characterized in that, The method for calculating the correlation coefficient includes: S41 samples the temperature change sequence of the temperature anomaly point at equal time intervals within the preset time window to obtain the first discrete time sequence; simultaneously acquires the hydrogen chloride gas pulse injection status of the corresponding time period, marks the ventilation status as the first value, and marks the gas cut-off status as the second value to form the second discrete time sequence. S42 calculates the cross-correlation coefficient between the first discrete-time series and the second discrete-time series as the correlation coefficient; wherein, the cross-correlation coefficient is calculated as the ratio of the product of the covariance of the two series to their respective standard deviations.

10. The method for chlorinating and purifying quartz sand according to claim 7, characterized in that, If it is (1), the adjustment strategy is to increase the reaction temperature and increase the pulse frequency; if it is (2), the adjustment strategy is to reduce the total gas flow rate and adjust the pulse duty cycle; if it is (3), the adjustment strategy is to adjust the heating power distribution of the reactor and increase the pulse gas preheating temperature.

Images