Trichlorosilane hydrogen reduction polycrystalline silicon growth monitoring system fusing multi-source data

Abstract

The invention relates to the technical field of chemical engineering, and discloses a trichlorosilane hydrogen reduction polycrystalline silicon growth monitoring system fusing multi-source data. Comprising a multi-sensor monitoring module, a multi-sensor data acquisition module, a multi-source data preprocessing module, a growth process data analysis module, a growth state online evaluation module, a process parameter online regulation and control module, a regulation and control result feedback module and a system collaborative management module, the multi-sensor monitoring module monitors the growth process of trichlorosilane hydrogen reduction polycrystalline silicon through multiple sensors, the multi-sensor data acquisition module acquires original data, the multi-source data preprocessing module preprocesses the data, the growth process data analysis module carries out data calculation, and the growth state online evaluation module carries out online comprehensive evaluation. The process parameter online regulation and control module regulates and controls process parameters online, the regulation and control result feedback module generates an adjustment strategy, and the system collaborative management module performs system parameter recording and storage, system optimization and exception handling.

Description

Technical Field

[0001] This invention relates to the field of chemical technology, specifically to a monitoring system for the growth of trichlorosilane hydrogen-reduced polycrystalline silicon that integrates multi-source data. Background Technology

[0002] In the production process of polycrystalline silicon using the trichlorosilane (TCS) hydrogen reduction method, the precision of process control directly affects product quality and production efficiency. Traditional technologies mainly rely on operator experience and single-point sensor monitoring, which makes it difficult to capture the complex dynamic changes within the reaction system in real time. Issues such as uneven temperature gradient distribution, pressure fluctuations, or raw material imbalances can easily lead to unstable polycrystalline silicon deposition rates, reaction system instability, and low raw material utilization. For example, relying solely on intermittent manual sampling and analysis of exhaust gas composition cannot promptly reflect deviations in key parameters during the reaction process; data from a single temperature or pressure sensor is also insufficient to comprehensively characterize the overall state of the reaction system. These technical bottlenecks not only restrict the precise control of the production rhythm but can also cause polycrystalline silicon grain defects and decreased uniformity due to drastic reaction fluctuations, severely impacting the yield of electronic or photovoltaic products. Furthermore, insufficient raw material conversion efficiency can lead to unreacted TCS escaping with the exhaust gas, increasing material losses and environmental treatment costs. To address the aforementioned issues, there is an urgent need for a monitoring system that can integrate data from multiple sensor sources, analyze the dynamic characteristics of the reaction system in real time, and possess intelligent decision-making capabilities, in order to achieve precise control over the polycrystalline silicon growth process. Summary of the Invention

[0003] (a) Technical problems to be solved

[0004] To address the shortcomings of existing technologies, this invention provides a trichlorosilane hydrogen reduction polycrystalline silicon growth monitoring system that integrates multi-source data. This system has the advantages of real-time monitoring, precise control, intelligent decision-making, and high efficiency and stability, solving the problems of slow response, crude control, resource waste, and unstable product quality associated with traditional technologies.

[0005] (II) Technical Solution

[0006] To achieve the above objectives, the present invention provides the following technical solution: a trichlorosilane hydrogen reduction polycrystalline silicon growth monitoring system integrating multi-source data, comprising a multi-sensor monitoring module, a multi-sensor data acquisition module, a multi-source data preprocessing module, a growth process data analysis module, a growth status online assessment module, a process parameter online control module, a control result feedback module, and a system collaborative management module;

[0007] The multi-sensor monitoring module monitors the growth process of trichlorosilane hydrogen reduction polycrystalline silicon through various sensors installed in the reduction furnace and its auxiliary systems.

[0008] The multi-sensor data acquisition module is responsible for acquiring raw data from various sensors in real time;

[0009] The multi-source data preprocessing module is responsible for preprocessing the raw data from the multi-sensor data acquisition module.

[0010] The growth process data analysis module receives the preprocessed data and performs polysilicon deposition rate analysis. Stability index of the reaction system Conversion rate with trichlorosilane Calculation;

[0011] The online growth status evaluation module performs an online comprehensive evaluation of the current polysilicon growth process based on the calculation results and a preset parameter range, and outputs the evaluation results.

[0012] The online process parameter control module automatically generates targeted adjustment strategies based on the output results of the growth status assessment module, and controls the process parameters online.

[0013] The control result feedback module implements specific control commands to the reduction furnace and its auxiliary systems based on the adjustment strategy generated by the process parameter adjustment module, and then feeds back the adjusted results.

[0014] The system collaborative management module records and stores system parameters, optimizes the system, and handles anomalies based on feedback information.

[0015] Preferably, the multi-sensor monitoring module is equipped with a temperature sensor, a pressure sensor, a flow sensor, a component sensor, and an optical sensor, specifically monitoring the following objects:

[0016] The temperature sensor monitors the real-time temperature of different deposition zones within the reactor online.

[0017] The pressure sensor monitors the real-time pressure of the hydrogen-trichlorosilane mixed gas phase inside the reactor online.

[0018] The flow sensor monitors the feed flow rate of trichlorosilane and the feed flow rate of hydrogen online.

[0019] The component sensor monitors the concentrations of unreacted trichlorosilane and hydrogen chloride byproducts in the reaction tail gas online.

[0020] The optical sensor monitors the thickness and surface flatness of the polycrystalline silicon deposited layer online.

[0021] Preferably, the multi-sensor data acquisition module includes a temperature-pressure data acquisition unit, a gas-liquid flow rate data acquisition unit, an exhaust gas composition data acquisition unit, and a deposition state data acquisition unit.

[0022] Preferably, the temperature-pressure data acquisition unit synchronously acquires real-time data from each temperature sensor and pressure sensor;

[0023] The gas-liquid flow data acquisition unit acquires real-time data from the trichlorosilane feed flow sensor and the hydrogen feed flow sensor, and simultaneously records the opening degree of the feed valve.

[0024] Preferably, the exhaust gas composition data acquisition unit is connected to a composition sensor to collect data on the exhaust gas composition. , , Real-time concentration data.

[0025] The deposition state data acquisition unit acquires data on the polycrystalline silicon deposition thickness and surface flatness monitored by the optical sensor.

[0026] Preferably, the growth process data analysis module receives the preprocessed data and performs polycrystalline silicon deposition rate analysis. The calculation formula is as follows:

[0027]

[0028] In the formula, Indicates the polysilicon deposition rate. This indicates the thickness of polysilicon deposited per unit time. Represents the reaction rate constant. This indicates the molar flow rate of trichlorosilane. This represents the absolute pressure of the system, expressed in Pascals. Indicates the temperature of the dominant reaction zone. Indicates the activation energy of the reaction. denoted by , where a and b represent the ideal gas constant, and respectively, the empirical indices of the molar flow rate of trichlorosilane and the absolute pressure of the system.

[0029] Preferably, the growth process data analysis module is responsible for performing a reaction system stability index analysis on the preprocessed data. The calculation formula is as follows:

[0030]

[0031] In the formula, Indicates the stability index of the reaction system. This represents the standard deviation of the average temperature inside the reactor after pretreatment. This indicates the reaction temperature setpoint. This represents the standard deviation of the pressure inside the reactor after pretreatment. This indicates the reaction pressure setpoint. , These represent the weights of temperature fluctuations and pressure fluctuations, respectively. + =1.

[0032] Preferably, the growth process data analysis module is responsible for calculating the trichlorosilane conversion rate of the preprocessed data. The calculation formula is as follows:

[0033]

[0034] In the formula, Indicates the trichlorosilane conversion rate. This indicates the feed flow rate of trichlorosilane after pretreatment. This indicates the initial concentration of the trichlorosilane raw material. This represents the equivalent flow rate of unreacted trichlorosilane in the pretreated exhaust gas. This indicates the concentration of unreacted trichlorosilane in the exhaust gas after pretreatment.

[0035] Preferably, the online growth status evaluation module performs an online comprehensive evaluation of the current polycrystalline silicon growth process based on the calculation results and a preset parameter range, and outputs the evaluation results. The specific process is as follows:

[0036] S1.1 Setting preset range: Set the benchmark threshold and fluctuation range of core indicators according to the polysilicon product grade;

[0037] S1.2 Determining the optimal state: When the polysilicon deposition rate Stability index of the reaction system Conversion rate with trichlorosilane When all three indicators are within the central range of the preset range, it is judged to be in the optimal state, indicating that the matching degree of various parameters in the current growth process is high and the quality consistency of polycrystalline silicon products is good.

[0038] S1.3 Determining Normal State: When all three indicators are within the preset range but close to the boundary, or when a single indicator briefly exceeds the boundary but returns to the range within 30 seconds, it is determined to be a normal state, indicating that there are slight fluctuations in the current growth process but it does not affect the basic quality of the product.

[0039] S1.4 Determining Deviation Status: When a single indicator exceeds the preset range by less than 5% and the duration is less than 5 minutes, or when two indicators are close to the boundary at the same time, it is determined to be a slight deviation; when a single indicator exceeds the preset range by 5% to 10% and the duration is 5 to 15 minutes, or when two indicators exceed the preset range, it is determined to be a serious deviation; when any indicator exceeds the preset range by more than 10% and the duration is ≥ 15 minutes, or when three indicators exceed the preset range at the same time, it is determined to be an abnormal status.

[0040] Preferably, the online process parameter control module automatically generates targeted adjustment strategies based on the output results of the growth status assessment module, and adjusts the process parameters online. The adjustment content includes:

[0041] (1) Gas flow rate adjustment: based on polysilicon deposition rate and trichlorosilane conversion rate Adjust the flow rates of trichlorosilane and hydrogen;

[0042] (2) Adjustment of reaction temperature: based on the polysilicon deposition rate and the stability index of the reaction system Adjust the temperature of the reactor;

[0043] (3) Reaction pressure adjustment: based on the trichlorosilane conversion rate and the stability index of the reaction system Adjust the pressure of the reactor.

[0044] Compared with the prior art, the present invention provides a trichlorosilane hydrogen reduction polycrystalline silicon growth monitoring system that integrates multi-source data, which has the following beneficial effects:

[0045] 1. This invention improves the polycrystalline silicon deposition rate. The calculations are used to evaluate the polysilicon growth efficiency and production progress, comparing them with a preset range, when the polysilicon deposition rate... If the temperature is below the lower limit of the range, it indicates that the reaction temperature or the feed rate needs to be increased; when the polysilicon deposition rate... Exceeding the upper limit of the range may cause a decrease in deposition uniformity; when the polysilicon deposition rate... By maintaining the current process parameters within the optimal range, the goal is to achieve precise control of the production pace and ensure the quality of polysilicon deposition.

[0046] 2. This invention utilizes the stability index of the reaction system. The calculation is used to assess the degree of fluctuation in temperature and pressure field inside the reactor, when the stability index of the reaction system... A stability index within the range of 0.85-0.95 or 0.80-0.90 indicates a stable reaction state; when the stability index of the reaction system is within the range of 0.85-0.95 or 0.80-0.90, it indicates a stable reaction state. When the value is below 0.8 or 0.85, the process parameter adjustment mechanism is automatically triggered. By fine-tuning the heating power and the opening of the exhaust valve, fluctuations are suppressed, thereby reducing polysilicon defects caused by reaction instability.

[0047] 3. This invention improves the conversion rate of trichlorosilane. The calculations are used to evaluate the raw material utilization rate and reaction efficiency, when the trichlorosilane conversion rate... Maintain the current hydrogen-to-raw material ratio when the conversion rate is within the range of 85%–92% or 80%–88%; when the trichlorosilane conversion rate... When the concentration is below 80% or 85%, increasing the hydrogen flow rate or optimizing the temperature distribution can promote feedstock conversion, thereby reducing feedstock consumption and tail gas treatment costs. Attached Figure Description

[0048] Figure 1 This is a system flowchart of the present invention. Detailed Implementation

[0049] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0050] Please see Figure 1 A trichlorosilane hydrogen reduction polycrystalline silicon growth monitoring system integrating multi-source data includes a multi-sensor monitoring module, a multi-sensor data acquisition module, a multi-source data preprocessing module, a growth process data analysis module, an online growth status assessment module, an online process parameter control module, a control result feedback module, and a system collaborative management module.

[0051] The multi-sensor monitoring module monitors the growth process of trichlorosilane hydrogen reduction polycrystalline silicon through various sensors installed in the reduction furnace and its auxiliary systems;

[0052] The multi-sensor data acquisition module is responsible for acquiring raw data from various sensors in real time;

[0053] The multi-source data preprocessing module is responsible for preprocessing the raw data from the multi-sensor data acquisition module by cleaning, filtering, timestamp alignment and unit unification, and using weighted average or Kalman filter algorithms to perform preliminary fusion of multi-source heterogeneous data.

[0054] The growth process data analysis module receives the preprocessed data and performs polysilicon deposition rate analysis. Stability index of the reaction system Conversion rate with trichlorosilane Calculation;

[0055] The online growth status evaluation module performs an online comprehensive evaluation of the current polysilicon growth process based on the calculation results and the preset parameter range, and outputs the evaluation results.

[0056] Based on the output of the growth status assessment module, the online process parameter control module automatically generates targeted adjustment strategies and controls the process parameters online.

[0057] The control result feedback module implements specific control commands to the reduction furnace and its auxiliary systems based on the adjustment strategy generated by the process parameter adjustment module, and then feeds back the adjusted results.

[0058] The system collaborative management module records and stores system parameters, optimizes the system, and handles anomalies based on feedback information.

[0059] Through the aforementioned multi-module collaborative end-to-end monitoring architecture, deep fusion technology of multi-source sensor data, quantitative calculation model of core process indicators, and dynamic closed-loop control mechanism, real-time perception, precise analysis, and intelligent control of the polysilicon growth process can be achieved, ultimately resulting in the beneficial effects of improving the quality stability of polysilicon products, increasing raw material conversion rate, reducing production energy consumption, and reducing manual intervention costs.

[0060] The multi-sensor monitoring module is equipped with temperature sensors, pressure sensors, flow sensors, composition sensors, and optical sensors. The specific monitoring targets are:

[0061] Temperature sensors monitor the real-time temperature (T, unit: °C) of different deposition areas (such as silicon core surface and gas phase space) inside the reactor online.

[0062] Pressure sensors monitor the real-time pressure of the hydrogen-trichlorosilane mixed gas phase inside the reactor online. (Unit: MPa)

[0063] Online monitoring of trichlorosilane feed flow rate using flow sensors ( (Unit: L / min), hydrogen feed rate ( (Unit: L / min)

[0064] The component sensor monitors the concentration of unreacted trichlorosilane in the reaction tail gas online. (Unit: mol / L), hydrogen concentration ( (Unit: mol / L), concentration of hydrogen chloride byproducts () (Unit: mol / L)

[0065] Optical sensors monitor the thickness of polycrystalline silicon deposits online. ,unit: ) and surface flatness ( (Unit: nm)

[0066] The multi-sensor data acquisition module includes a temperature-pressure data acquisition unit, a gas-liquid flow rate data acquisition unit, an exhaust gas composition data acquisition unit, and a sedimentation state data acquisition unit.

[0067] The temperature-pressure data acquisition unit synchronously acquires real-time data from each temperature sensor (3-5, distributed in different areas of the reactor) and pressure sensor (2, monitoring the pressure at the top and bottom of the reactor respectively), with a sampling frequency of 1Hz. The output data format is "timestamp-sensor number-temperature value / pressure value".

[0068] The gas-liquid flow data acquisition unit acquires real-time data from the trichlorosilane feed flow sensor and the hydrogen feed flow sensor, and simultaneously records the feed valve opening (V, unit: %). The sampling frequency is 0.5Hz, and the output data format is "timestamp-medium type-flow value-valve opening".

[0069] The exhaust gas composition data acquisition unit interfaces with a composition sensor (using gas chromatography) to collect data on the composition of exhaust gas. , , The real-time concentration data is sampled at a frequency of 0.1Hz (the component detection response is relatively slow), and the output data format is "timestamp-component name-concentration value".

[0070] The deposition state data acquisition unit acquires the polycrystalline silicon deposition thickness monitored by the optical sensor. ), surface flatness ( The data is sampled at a frequency of 0.2Hz, and the output data format is "timestamp-monitoring indicator-value".

[0071] The advantages are: by designing the above-mentioned unit-specific collection of temperature-pressure, gas-liquid flow rate, exhaust gas composition and deposition state data, and by matching the response characteristics of each parameter with differentiated sampling frequency, the integrity and timeliness of the original data are ensured, thereby achieving comprehensive coverage and accurate capture of key parameters for polycrystalline silicon growth, so as to provide a high-quality data source for subsequent data processing and analysis.

[0072] The growth process data analysis module receives the preprocessed data and performs polysilicon deposition rate analysis. The calculation formula is as follows:

[0073]

[0074] In the formula, This indicates the polycrystalline silicon deposition rate, expressed in micrometers per minute. This indicates the thickness of polysilicon deposited per unit time. The rate constant represents the reaction rate and is related to reaction conditions and catalysts, etc. It is usually obtained through experimental fitting. The molar flow rate of trichlorosilane (TCS) is expressed in moles per minute (mol / min), representing the number of moles of TCS entering the reaction zone per unit time. This represents the absolute pressure of the system, measured in Pascals (Pa). The absolute pressure within the reaction system affects the reaction rate. This indicates the temperature of the dominant reaction zone, in Kelvin (K). The temperature of the reaction zone has a significant impact on the reaction rate. The activation energy of a reaction is expressed in joules per mole (J / mol). It is the minimum energy required for a reaction to occur. The value represents the ideal gas constant, with units of joules per (mol·K). a and b represent empirical indices of the molar flow rate of trichlorosilane and the absolute pressure of the system, respectively, obtained by fitting historical data, and are used to describe the degree of influence of each parameter on the deposition rate.

[0075] The advantage is that, through the above-mentioned polycrystalline silicon deposition rate The calculations are used to evaluate the polycrystalline silicon growth efficiency and production progress, comparing them with preset ranges (5-8 nm / s for electronic grade and 4-7 nm / s for photovoltaic grade) when the polycrystalline silicon deposition rate... If the temperature is below the lower limit of the range, it indicates that the reaction temperature or the feed rate needs to be increased; when the polysilicon deposition rate... Exceeding the upper limit of the range may cause a decrease in deposition uniformity; when the polysilicon deposition rate... By maintaining the current process parameters within the optimal range, the goal is to achieve precise control of the production pace and ensure the quality of polysilicon deposition.

[0076] The growth process data analysis module is responsible for performing reaction system stability index analysis on the preprocessed data. The calculation formula is as follows:

[0077]

[0078] In the formula, Indicates the stability index of the reaction system. The standard deviation of the average temperature inside the reactor after pretreatment (unit: °C, calculated from data from multiple temperature sensors). This indicates the reaction temperature setpoint (unit: °C, process baseline parameter). This represents the standard deviation of the pressure inside the reactor after pretreatment (unit: MPa, calculated from pressure sensor data). This indicates the reaction pressure setpoint (unit: MPa, process reference parameter). , These represent the weights of temperature fluctuations and pressure fluctuations, respectively. + =1, based on process experience, pressure fluctuations have a greater impact on reaction stability than temperature fluctuations, therefore, a higher weight is assigned to the reaction system stability index. The value range is [0,1], and the stability index of the reaction system is... The closer the value is to 1, the more stable the reaction system is.

[0079] The advantage is that, based on the stability index of the above reaction system... The calculation is used to assess the degree of fluctuation in temperature and pressure field inside the reactor, when the stability index of the reaction system... A stability index within the range of 0.85-0.95 (electronic grade) or 0.80-0.90 (photovoltaic grade) indicates a stable reaction state; when the stability index of the reaction system is within the range of 0.85-0.95 (electronic grade), it indicates a stable reaction state. When the value is below 0.8 (photovoltaic grade) or 0.85 (electronic grade), the process parameter control mechanism is automatically triggered. By fine-tuning the heating power and the opening of the exhaust valve, fluctuations are suppressed, thereby reducing polycrystalline silicon defects (such as impurity doping and loose structure) caused by reaction instability.

[0080] The growth process data analysis module is responsible for calculating the trichlorosilane conversion rate from the preprocessed data. The calculation formula is as follows:

[0081]

[0082] In the formula, Indicates the trichlorosilane conversion rate. This indicates the pre-treated trichlorosilane feed flow rate (unit: L / min, from the flow acquisition unit). This indicates the initial concentration of trichlorosilane raw material (unit: mol / L, known process parameters, determined by batch testing of raw materials). This indicates the equivalent flow rate of unreacted trichlorosilane in the pretreated exhaust gas (unit: L / min, determined by the exhaust gas flow rate (default is the same as the total feed flow rate) and... (derived from conversion) This indicates the concentration of unreacted trichlorosilane in the exhaust gas after pretreatment (unit: mol / L, from the exhaust gas component acquisition unit).

[0083] The advantage is that, through the above-mentioned trichlorosilane conversion rate... The calculations are used to evaluate the raw material utilization rate and reaction efficiency, when the trichlorosilane conversion rate... Maintain the current hydrogen-to-raw material ratio within the range of 85%–92% (electronic grade) or 80%–88% (photovoltaic grade); when the trichlorosilane conversion rate... When the content is below 80% (photovoltaic grade) or 85% (electronic grade), the conversion of raw materials can be promoted by increasing the hydrogen flow rate (increasing the driving force of the reduction reaction) or optimizing the temperature distribution, thereby achieving the beneficial effects of reducing raw material consumption and reducing tail gas treatment costs.

[0084] The online growth status evaluation module performs a comprehensive online evaluation of the current polysilicon growth process based on the calculation results and a preset parameter range, and outputs the evaluation results. The specific process is as follows:

[0085] S1.1 Setting Preset Range: Based on the polysilicon product grade (electronic grade / photovoltaic grade), set the benchmark threshold and fluctuation range for core indicators, specifically including:

[0086] (1) Polycrystalline silicon deposition rate Electronic grade preset range 5-8nm / s, photovoltaic grade preset range 4-7nm / s;

[0087] (2) Stability index of the reaction system The preset range for electronic grade is 0.85-0.95, and the preset range for photovoltaic grade is 0.80-0.90.

[0088] (3) Trichlorosilane conversion rate The preset range for electronic grade is 85%–92%, and for photovoltaic grade it is 80%–88%.

[0089] S1.2 Determining the optimal state: When all three indicators are within the central range of the preset range (polysilicon deposition rate) The median of the range 0.5 nm / s, stability index of the reaction system The median of the range 0.02, trichlorosilane conversion rate The median of the range When the growth rate reaches 1%, it is considered the optimal state, indicating that the parameters of the current growth process are highly matched and the polysilicon product quality is consistent.

[0090] S1.3 Determining Normal State: When all three indicators are within the preset range but close to the boundary (polysilicon deposition rate) ≤1nm / s from the boundary of the reaction system stability index ≤0.03 from the range boundary, trichlorosilane conversion rate When the deviation from the boundary is ≤2%, or when a single indicator briefly exceeds the boundary but returns to the range within 30 seconds, it is judged as a normal state, indicating that there are slight fluctuations in the current growth process but it does not affect the basic quality of the product.

[0091] S1.4 Determine the deviation status:

[0092] (1) Slight deviation: When a single indicator exceeds the preset range by less than 5% (such as electronic grade conversion rate of 83% to 85%) and the duration is less than 5 minutes, or when two indicators are close to the boundary at the same time (≤0.5nm / s, 0.02, 1% from the boundary), it is judged as a slight deviation, indicating that there is a need for fine adjustment of the current raw material ratio or temperature field distribution;

[0093] (2) Serious deviation: When a single indicator exceeds the preset range by 5% to 10% (e.g., electronic conversion rate of 80% to 83%) and lasts for 5 to 15 minutes, or when two indicators exceed the preset range, it is judged as a serious deviation, indicating that the reaction system has significant fluctuations (e.g., unstable hydrogen flow rate, abnormal local temperature).

[0094] S1.5 Determining Abnormal Status: When any indicator exceeds the preset range by more than 10% (e.g., electronic conversion rate < 80%) and the duration is ≥ 15 minutes, or when all three indicators exceed the preset range simultaneously, it is determined to be an abnormal status, indicating that there may be serious problems such as sensor failure, raw material purity not meeting standards, or reactor seal failure.

[0095] The above calculation formula can be used to locate potential problems (such as "high temperature leading to decreased uniformity" or "low raw material conversion rate").

[0096] The advantages are: by using the above-mentioned graded preset parameter range and multi-index collaborative judgment logic, a refined assessment of the growth state of polycrystalline silicon can be achieved. Combined with the intelligent positioning function of potential problems, it provides a clear direction for subsequent regulation and avoids over-regulation caused by misjudgment of a single index, thereby achieving the beneficial effect of improving the accuracy of state assessment and guiding the targeted regulation.

[0097] Based on the output of the growth status assessment module, the online process parameter control module automatically generates targeted adjustment strategies and adjusts the process parameters online. The adjustments include:

[0098] (1) Gas flow rate adjustment: based on polysilicon deposition rate and trichlorosilane conversion rate When the polysilicon deposition rate When the conversion rate of trichlorosilane is more than 10% below the preset lower limit, increase the feed flow rate of trichlorosilane proportionally (by 5% to 10%). When the flow rate is more than 5% below the preset lower limit, increase the flow ratio of hydrogen to trichlorosilane (from 3:1 to 4:1) to enhance the driving force of the reduction reaction.

[0099] (2) Adjustment of reaction temperature: based on the polysilicon deposition rate and the stability index of the reaction system When the trichlorosilane conversion rate When the value remains consistently below the median of the optimal interval, Gradually increase the temperature of the dominant reaction zone within a 5℃ range, until the stability index of the reaction system is... When the temperature drops below 0.8 (photovoltaic level) due to temperature fluctuations, reduce the heating power adjustment range (from...). 2℃ / min decreased 1℃ / min);

[0100] (3) Reaction pressure adjustment: based on the trichlorosilane conversion rate and the stability index of the reaction system When the trichlorosilane conversion rate When the pressure is below the preset range and the pressure fluctuation is small, increase the system pressure by 0.02-0.05 MPa; when the stability index of the reaction system... When the pressure drops due to pressure fluctuations, the pressure fluctuations can be controlled by fine-tuning the opening of the exhaust valve. Within 0.01MPa;

[0101] The control result feedback module, based on the adjustment strategy generated by the process parameter adjustment module, implements specific control commands to the reduction furnace and its auxiliary systems, and provides feedback on the adjusted results. The feedback content includes:

[0102] (1) Adjusted parameter values: Record and report the adjusted gas flow rate, reaction temperature and reaction pressure;

[0103] (2) Evaluation of adjustment effect: Feedback on the adjusted deposition rate, stability index and conversion rate to facilitate further optimization;

[0104] (3) Implement control commands: Convert the adjustment strategy into specific control commands and send them to the reduction furnace and its auxiliary systems to ensure that the adjustment strategy is implemented;

[0105] The system collaborative management module records and stores system parameters, optimizes the system, and handles anomalies based on feedback information. Its management functions include:

[0106] (1) Parameter recording and storage: Record the parameter values ​​and evaluation results after each adjustment for easy subsequent analysis;

[0107] (2) System optimization: Based on feedback information, optimize system configuration and parameter adjustment strategies to improve system stability and efficiency;

[0108] (3) Abnormal handling: Handle abnormal situations during system operation to ensure stable system operation. When an abnormal state is detected, trigger an alarm and take emergency measures (such as suspending production, checking sensors, etc.).

[0109] (4) User interface management: Provides a user interface to facilitate operators to monitor system status, view evaluation results and adjust parameters in real time;

[0110] (5) Data backup and recovery: Regularly back up system data to ensure data security and support data recovery function.

[0111] The advantages are: through the above-mentioned full-process data recording and storage, dynamic optimization and control strategies, hierarchical anomaly handling mechanisms and user interface interaction functions, intelligent management and continuous improvement of the monitoring system can be achieved. Combined with data backup and recovery functions, production continuity can be ensured, thereby improving the reliability of system operation, extending the service life of equipment, and providing data support for process optimization.

[0112] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A monitoring system for the growth of trichlorosilane hydrogen-reduction polycrystalline silicon integrating multi-source data, characterized in that, It includes a multi-sensor monitoring module, a multi-sensor data acquisition module, a multi-source data preprocessing module, a growth process data analysis module, an online growth status assessment module, an online process parameter control module, a control result feedback module, and a system collaborative management module; The multi-sensor monitoring module monitors the growth process of trichlorosilane hydrogen reduction polycrystalline silicon through various sensors installed in the reduction furnace and its auxiliary systems. The multi-sensor data acquisition module is responsible for acquiring raw data from various sensors in real time; The multi-source data preprocessing module is responsible for preprocessing the raw data from the multi-sensor data acquisition module. The growth process data analysis module receives the preprocessed data and performs polysilicon deposition rate analysis. Stability index of the reaction system Conversion rate with trichlorosilane Calculation; The online growth status evaluation module performs an online comprehensive evaluation of the current polysilicon growth process based on the calculation results and a preset parameter range, and outputs the evaluation results. The online process parameter control module automatically generates targeted adjustment strategies based on the output results of the growth status assessment module, and controls the process parameters online. The control result feedback module implements specific control commands to the reduction furnace and its auxiliary systems based on the adjustment strategy generated by the process parameter adjustment module, and then feeds back the adjusted results. The system collaborative management module records and stores system parameters, optimizes the system, and handles anomalies based on feedback information.

2. The trichlorosilane hydrogen reduction polycrystalline silicon growth monitoring system integrating multi-source data according to claim 1, characterized in that: The multi-sensor monitoring module is equipped with temperature sensors, pressure sensors, flow sensors, component sensors, and optical sensors. The specific monitoring objects are: The temperature sensor monitors the real-time temperature of different deposition zones within the reactor online. The pressure sensor monitors the real-time pressure of the hydrogen-trichlorosilane mixed gas phase inside the reactor online. The flow sensor monitors the feed flow rate of trichlorosilane and the feed flow rate of hydrogen online. The component sensor monitors the concentrations of unreacted trichlorosilane and hydrogen chloride byproducts in the reaction tail gas online. The optical sensor monitors the thickness and surface flatness of the polycrystalline silicon deposited layer online.

3. The trichlorosilane hydrogen reduction polycrystalline silicon growth monitoring system integrating multi-source data according to claim 1, characterized in that: The multi-sensor data acquisition module includes a temperature-pressure data acquisition unit, a gas-liquid flow rate data acquisition unit, an exhaust gas composition data acquisition unit, and a deposition state data acquisition unit.

4. The trichlorosilane hydrogen reduction polycrystalline silicon growth monitoring system integrating multi-source data according to claim 3, characterized in that: The temperature-pressure data acquisition unit synchronously acquires real-time data from each temperature sensor and pressure sensor. The gas-liquid flow data acquisition unit acquires real-time data from the trichlorosilane feed flow sensor and the hydrogen feed flow sensor, and simultaneously records the opening degree of the feed valve.

5. The trichlorosilane hydrogen reduction polycrystalline silicon growth monitoring system integrating multi-source data according to claim 3, characterized in that: The exhaust gas composition data acquisition unit is connected to the composition sensor to collect real-time concentration data of C1, C2, and C3 in the exhaust gas; The deposition state data acquisition unit acquires data on the polycrystalline silicon deposition thickness and surface flatness monitored by the optical sensor.

6. The trichlorosilane hydrogen reduction polycrystalline silicon growth monitoring system integrating multi-source data according to claim 1, characterized in that: The growth process data analysis module receives the preprocessed data and performs polysilicon deposition rate analysis. The calculation formula is as follows: ； In the formula, Indicates the polysilicon deposition rate. This indicates the thickness of polysilicon deposited per unit time. Represents the reaction rate constant. This indicates the molar flow rate of trichlorosilane. This represents the absolute pressure of the system, expressed in Pascals. Indicates the temperature of the dominant reaction zone. Indicates the activation energy of the reaction. denoted by , where a and b represent the ideal gas constant, and respectively, the empirical indices of the molar flow rate of trichlorosilane and the absolute pressure of the system.

7. The trichlorosilane hydrogen reduction polycrystalline silicon growth monitoring system integrating multi-source data according to claim 1, characterized in that: The growth process data analysis module is responsible for performing a reaction system stability index analysis on the preprocessed data. The calculation formula is as follows: ； In the formula, Indicates the stability index of the reaction system. This represents the standard deviation of the average temperature inside the reactor after pretreatment. This indicates the reaction temperature setpoint. This represents the standard deviation of the pressure inside the reactor after pretreatment. This indicates the reaction pressure setpoint. , These represent the weights of temperature fluctuations and pressure fluctuations, respectively. + =1.

8. The trichlorosilane hydrogen reduction polycrystalline silicon growth monitoring system integrating multi-source data according to claim 1, characterized in that: The growth process data analysis module is responsible for calculating the trichlorosilane conversion rate from the preprocessed data. The calculation formula is as follows: ； In the formula, Indicates the trichlorosilane conversion rate. This indicates the feed flow rate of trichlorosilane after pretreatment. This indicates the initial concentration of the trichlorosilane raw material. This represents the equivalent flow rate of unreacted trichlorosilane in the pretreated exhaust gas. This indicates the concentration of unreacted trichlorosilane in the exhaust gas after pretreatment.

9. The trichlorosilane hydrogen reduction polycrystalline silicon growth monitoring system integrating multi-source data according to claim 1, characterized in that: The online growth status evaluation module performs an online comprehensive evaluation of the current polysilicon growth process based on the calculation results and a preset parameter range, and outputs the evaluation results. The specific process is as follows: S1.1 Setting preset range: Set the benchmark threshold and fluctuation range of core indicators according to the polysilicon product grade; S1.2 Determining the optimal state: When the polysilicon deposition rate Stability index of the reaction system Conversion rate with trichlorosilane When all three indicators are within the central range of the preset range, it is judged to be in the optimal state, indicating that the matching degree of various parameters in the current growth process is high and the quality consistency of polycrystalline silicon products is good. S1.3 Determining Normal State: When all three indicators are within the preset range but close to the boundary, or when a single indicator briefly exceeds the boundary but returns to the range within 30 seconds, it is determined to be a normal state, indicating that there are slight fluctuations in the current growth process but it does not affect the basic quality of the product. S1.4 Determining Deviation Status: When a single indicator exceeds the preset range by less than 5% and the duration is less than 5 minutes, or when two indicators are close to the boundary at the same time, it is determined to be a slight deviation; when a single indicator exceeds the preset range by 5% to 10% and the duration is 5 to 15 minutes, or when two indicators exceed the preset range, it is determined to be a serious deviation; when any indicator exceeds the preset range by more than 10% and the duration is ≥ 15 minutes, or when three indicators exceed the preset range at the same time, it is determined to be an abnormal status.

10. The trichlorosilane hydrogen reduction polycrystalline silicon growth monitoring system integrating multi-source data according to claim 1, characterized in that: The online process parameter control module automatically generates targeted adjustment strategies based on the output of the growth status assessment module, and adjusts the process parameters online. The adjustments include: (1) Gas flow rate adjustment: based on polysilicon deposition rate and trichlorosilane conversion rate Adjust the flow rates of trichlorosilane and hydrogen; (2) Adjustment of reaction temperature: based on the polysilicon deposition rate and the stability index of the reaction system Adjust the temperature of the reactor; (3) Reaction pressure adjustment: based on the trichlorosilane conversion rate and the stability index of the reaction system Adjust the pressure of the reactor.

Images