Hexachlorodisilane automated production control system and method

Abstract

The present invention relates to the technical field of chemical production automatic control, and in particular to an automated production control system and method for hexachlorodisilane. The present invention corrects data through a temperature compensation model to avoid side reactions caused by inaccurate temperature, and utilizes an adaptive temperature and pressure control model for negative feedback regulation of the temperature and pressure of a reactor, dynamically adjusting a gain function according to reaction time to improve control accuracy. With the help of online component analysis, a mathematical regression analysis model is established, coefficients are determined using machine learning, a production parameter optimization module calculates the opening of a reflux valve, and distillation tower parameters are optimized to reduce by-products and ensure the quality of hexachlorodisilane. The present invention automatically carries out automatic safety inspections based on data analysis results, quantitatively assesses explosion risks, detects presumed leaks using a leak location algorithm, matches hierarchical protection signals, and executes safety measures to ensure production safety.

Description

Technical Field

[0001] The present invention relates to the technical field of chemical production automatic control, in particular to a hexachlorodisilane automatic production control system and method. Background Art

[0002] Chemical production automation technology is a key enabler for driving the chemical industry toward safe, efficient, and green development. This development stems from the challenges inherent in traditional manual operations, such as delayed response times, large parameter fluctuations, and high safety risks, coupled with the industry's urgent need for consistent product quality, reduced energy consumption, and environmental compliance. Currently, the use of high-precision sensors to collect multi-parameter data in real time, combined with advanced algorithms for dynamic process optimization and the integration of safety interlock systems and leak detection technologies, has effectively improved the stability, safety, and economic efficiency of chemical production, becoming a crucial component of the core competitiveness of modern chemical enterprises.

[0003] However, there are still many problems in the existing automated production process of hexachlorodisilane.

[0004] First, traditional temperature sensors are susceptible to environmental interference, which can cause temperature drift in the collected data and a large deviation between the actual reaction temperature and the set value.

[0005] Secondly, most existing systems use fixed-gain PID control, which is difficult to adapt to the different requirements in the early stage of the reaction (rapid temperature and pressure increase) and the late stage (fine-tuning and stabilization are required), and is prone to overshoot or under-adjustment.

[0006] Finally, the traditional hexachlorodisilane production process relies on empirical parameter adjustment and lacks quantitative analysis of the relationship between operating parameters and by-product formation.

[0007] In view of the above problems, it is necessary to propose an automated production control system and method for hexachlorodisilane. Summary of the Invention

[0008] The purpose of the present invention is to solve the problems existing in the background technology and to provide an automated production control system and method for hexachlorodisilane.

[0009] The purpose of the present invention can be achieved through the following technical solutions:

[0010] The invention discloses a hexachlorodisilane automated production control system and method, including a hexachlorodisilane automated production control system and a hexachlorodisilane automated production control system method.

[0011] In a first aspect, the present invention provides an automated production control system for hexachlorodisilane, comprising

[0012] The automated production control system for hexachlorodisilane includes an intelligent temperature and pressure control module for the reactor, a purity control module for the multi-stage distillation tower, a production parameter optimization module, and a safety interlock and disaster suppression module.

[0013] The reactor intelligent temperature and pressure control module monitors the reactor's temperature, pressure, and silane concentration data in real time. It corrects the collected temperature data through temperature compensation, calculates the predicted reaction rate based on the corrected temperature data, and sends this predicted reaction rate to the production parameter optimization module. It monitors temperature and pressure fluctuations in the reactor, provides negative feedback control of the internal temperature and pressure, and outputs pressure or temperature anomaly signals to the safety interlock and disaster suppression module based on analysis of fluctuation anomalies.

[0014] The internal temperature and pressure of the reactor and the motor's stirring power are collected in real time. The preset temperature drift coefficient, stirring thermal effect coefficient, and motor's baseline stirring power are obtained and input into the temperature compensation model to correct the collected reactor temperature and obtain the corrected temperature.

[0015] As a preferred embodiment of the present invention, the reaction rate prediction is performed, and the specific process is as follows:

[0016] Based on the corrected temperature, the real-time silane concentration and activation energy of the main reaction in the reactor are obtained and input into a preset formula to calculate the predicted reaction rate. The predicted reaction rate is then sent to the production parameter optimization module.

[0017] As a preferred embodiment of the present invention, the temperature fluctuation and pressure fluctuation of the reactor are monitored, and negative feedback regulation is performed on the temperature and pressure inside the reactor. The specific process is as follows:

[0018] Obtaining the target temperature and target pressure preset by the user, calculating the difference between the internal temperature of the reactor and the target temperature and its integral part through the adaptive temperature and pressure control model, and calculating the temperature control input through the preset first and second temperature adaptive gain functions;

[0019] The difference between the internal pressure of the reactor and the target pressure and its integral are calculated using the adaptive temperature and pressure control model, and the pressure control input is calculated using the preset first and second pressure adaptive gain functions;

[0020] Among them, the first and second temperature adaptive gain functions and the first and second pressure adaptive gain functions are all power functions that converge over time, so that the error changes of temperature and pressure can be quickly adjusted in the early stage of the reaction, and the temperature and pressure errors can be fine-tuned in a small range at the end of the reaction. When the temperature and pressure are close to the target values, the value of the adaptive gain function is reduced to avoid over-adjustment.

[0021] The temperature control input and pressure control input are sent to the temperature control system and booster valve of the reactor as output power adjustment values ​​to adjust the temperature and pressure of the reactor.

[0022] As a preferred embodiment of the present invention, abnormality identification is performed based on temperature control input and pressure control input. The specific process is as follows:

[0023] If the temperature control input is greater than the first preset threshold, the system determines that the temperature deviation is too large, preventing the temperature control system from achieving temperature regulation, and generates a temperature anomaly signal. If the pressure control input is greater than the second preset threshold, the system determines that the pressure deviation is too large, preventing the boost valve from achieving pressure regulation, and generates a pressure anomaly signal. The generated temperature anomaly and pressure anomaly signals are sent to the safety interlock and disaster suppression module.

[0024] The multi-stage distillation column purity control module accesses the gas chromatograph and differential pressure level gauge in the multi-stage distillation column to perform online component analysis and obtain the target component and by-product qualities. It deeply mines real-time and historical data on temperature gradients, pressure gradients, steam control valve opening, reflux ratio, target component quality, light component by-product quality, and heavy component by-product quality during the distillation process, analyzing the linkages and underlying patterns between these data points to develop a regression prediction model for control decisions.

[0025] Obtain the temperature, pressure and distillate quality of each column section;

[0026] Obtain the steam regulating valve opening and reflux valve opening of the distillation tower;

[0027] As a preferred method of the present invention, the mass of the target components and by-products is calculated by scanning the distillate composition of each tower section using a gas chromatograph, locating the tower section identifier where the final product hexachlorodisilane is located, and obtaining the mass of the final product distillate. The mass of the by-products in the distillation tower is obtained, including the mass of the light components and the mass of the heavy components.

[0028] The light component index number is greater than all distillates in the column section where the hexachlorodisilane is located;

[0029] The heavy component refers to all distillates in the column section where the hexachlorodisilane is located.

[0030] As a preferred embodiment of the present invention, a mathematical regression analysis model is established based on historical data to describe the effects of the temperature, pressure, steam regulating valve opening, and reflux valve opening of each tower section on the mass of the light component, the mass of the heavy component, and the mass ratio of the final product. The specific process is as follows:

[0031] The ratio of the mass of the light component to the mass of the final product is defined as the first characteristic efficiency;

[0032] The ratio of the mass of the heavy component to the mass of the final product is defined as the second characteristic efficiency.

[0033] Establishing a regression prediction model for the first characteristic efficiency and the second characteristic efficiency and the temperature, pressure, steam control valve opening, and reflux valve opening of each tower section, and assigning regression coefficients to the temperature, pressure, steam control valve opening, and reflux valve opening of each tower section, including a regression coefficient for the first characteristic efficiency and a regression coefficient for the second characteristic efficiency;

[0034] The first characteristic efficiency is fitted by calculating the sum of the product of the temperature, pressure, steam regulating valve opening and reflux valve opening of each tower section and their corresponding regression coefficients and the sum of the first error term;

[0035] The second characteristic efficiency is fitted by calculating the sum of the product of the temperature, pressure, steam regulating valve opening and reflux valve opening of each tower section and their corresponding regression coefficients and the sum of the second error term;

[0036] As a preferred embodiment of the present invention, historical data is used to train a machine learning model, including a support vector machine and a neural network, to determine the specific values ​​of all regression coefficients and the first and second error terms in the regression prediction model, and substitute them for the regression prediction model.

[0037] The production parameter optimization module uses the predicted reaction rate to estimate the mass of the reactor product and the mass percentage of hexachlorodisilane, and determines the opening range of the steam control valve from the reactor to the distillation column. Based on the regression prediction model and the steam control valve opening range, the production parameter optimization calculation is performed to calculate the reflux valve opening that minimizes byproducts.

[0038] Obtain the predicted reaction rate and determine the opening range of the steam regulating valve for the reaction product to enter the distillation tower from the reactor. The specific process is as follows:

[0039] The volume of the liquid product in the reactor and the pressure differential across the steam control valve between the reactor and the distillation tower are calculated. Combined with the predicted reaction rate, these values ​​are used to calculate the reference opening value for the steam control valve. The maximum opening of the steam control valve is determined to be the maximum value between the reference opening value and the preset reference value. The minimum opening of the steam control valve is determined to be 0.

[0040] As a preferred embodiment of the present invention, production parameter optimization is performed based on the adjustment range of the steam control valve opening and the regression prediction model. The operating parameters of the distillation tower are adjusted by predicting the target component quality and the by-product quality. The tower section temperature, tower section pressure, steam control valve opening, and reflux valve opening that minimize the first characteristic efficiency and the second characteristic efficiency under the equipment constraints are calculated, thereby providing decision support for the optimal operation of the distillation tower. The specific process is as follows:

[0041] In the regression prediction model, the range of the steam control valve opening is set to 0 to the maximum value between the opening reference value and the preset reference value, and the temperature and pressure of each tower section are set to the preset value.

[0042] Find the return valve opening curve that satisfies the condition: minimizing the sum of the first characteristic efficiency and the second characteristic efficiency, and record it as the return valve opening control curve.

[0043] The return valve opening control curve is sent to the return valve control unit, and the return valve opening is controlled by the PID algorithm to fluctuate around the return valve opening control curve obtained by the solution.

[0044] After receiving an abnormal temperature signal or an abnormal pressure signal, the safety interlock and disaster suppression module conducts automatic safety inspections, accesses equipment instruments and flame detectors, obtains combustible gas concentration, equipment vibration acceleration and fire water pressure, and conducts a quantitative assessment of the explosion risk. It also detects leaks and infers the leak location through a leak location algorithm, matches the corresponding graded protection signal, and executes corresponding safety protection measures, including ESD emergency shutdown, opening of the nitrogen inerting valve, and directional opening of the explosion vent.

[0045] When an abnormal temperature signal is received, the equipment instrument is accessed to obtain the equipment vibration acceleration and the real-time pressure in the reactor.

[0046] When the vibration acceleration of the equipment or the real-time pressure in the reactor is detected to be greater than the preset threshold, a first-level pressure protection signal is generated;

[0047] When it is detected that the vibration acceleration of the equipment and the real-time pressure in the reactor are both greater than the preset threshold, a secondary pressure protection signal is generated.

[0048] When the abnormal pressure signal is received, the flame detector is accessed to obtain the combustible gas concentration and the real-time temperature of the reactor.

[0049] When the combustible gas concentration or the real-time temperature of the reactor is detected to be greater than the preset threshold, a first-level temperature protection signal is output;

[0050] When it is simultaneously identified that the combustible gas concentration and the real-time temperature of the reactor are greater than the preset threshold, a secondary temperature protection signal is output.

[0051] Execute corresponding safety protection measures according to the generated protection signal. The specific process is as follows:

[0052] When the first-level pressure protection signal is received, the directional opening part of the explosion vent is activated and the specific value of the abnormal pressure value is transmitted to the central monitoring equipment;

[0053] When the secondary pressure protection signal is received, the emergency shutdown procedure and the directional opening part of the explosion vent are started, and the nitrogen inerting valve is started to inject nitrogen into the reactor and distillation tower to reduce the concentration of combustible gas.

[0054] When receiving the first-level temperature protection signal, the emergency cooling device is activated to enhance the heat exchange capacity of the equipment;

[0055] When the secondary temperature protection signal is received, the emergency shutdown program, nitrogen inerting valve and emergency cooling device are activated, and the specific value of the temperature abnormality is transmitted to the central monitoring equipment.

[0056] In a second aspect, the present invention provides a method for controlling the automated production of hexachlorodisilane, comprising the following steps:

[0057] Step 1: Reactor temperature and pressure control and reaction rate prediction;

[0058] The temperature, pressure and silane concentration of the reactor are monitored in real time and temperature compensation correction is performed.

[0059] Collect temperature, pressure, and silane concentration data within the reactor. Correct the temperature data to compensate for temperature drift and stirring heat effects. Calculate the corrected temperature to ensure precise temperature control during the reaction and avoid side reactions.

[0060] The reaction rate is predicted based on the corrected temperature and the predicted value is sent to the production parameter optimization module. Temperature and pressure fluctuations are monitored in real time, negative feedback is adjusted, and abnormal signals are output to the safety interlock and disaster suppression module.

[0061] Step 2: Reaction rate prediction and reaction control;

[0062] The reaction rate is predicted by the silane concentration in the reactor, which helps optimize the production process and calculate the predicted reaction rate value. The calculation is performed using formulas such as silane concentration and preset reaction activation energy.

[0063] The predicted reaction rate is sent to the production parameter optimization module to optimize subsequent operations.

[0064] Step 3: Identification of temperature and pressure anomalies and safety interlock;

[0065] Adjust the temperature and pressure control inputs based on the deviation between the actual temperature and pressure in the reactor and the target values. Obtain the user's preset target temperature and target pressure, calculate the temperature and pressure control inputs using the adaptive temperature and pressure control model, and send these as output power adjustment values ​​to the reactor's temperature control system and booster valve to adjust the reactor's temperature and pressure.

[0066] Step 4: Purity control of multi-stage distillation tower;

[0067] Control the operation of the distillation tower to ensure that the final product quality of hexachlorodisilane meets standards and minimize byproducts. Monitor distillation tower operating parameters such as tower section temperature, pressure, and steam control valve opening in real time. Gas chromatograph scans are used to determine the component masses of each tower section, locate the final product section, and determine the quality of the final product distillate. A mathematical regression analysis model is developed to predict the impact of tower section operating parameters on the quality of light components, heavy components, and the target product. Based on the regression model, optimize the distillation tower operating parameters to minimize byproducts.

[0068] Step 5: Optimize production parameters;

[0069] Based on the reaction rate prediction, the steam control valve opening range is calculated and steam flow control is performed. Based on the regression model and optimization algorithm, the optimal control curve of the reflux valve opening is calculated to ensure the minimization of by-products and optimize the distillation process.

[0070] Compared with the prior art, the present invention has the following beneficial effects:

[0071] 1. The present invention uses a temperature compensation model to correct the collected temperature data, which can obtain more accurate temperature data for subsequent control, avoid data inaccuracies caused by temperature drift and stirring thermal effects, and prevent side reactions caused by temperature inaccuracies. At the same time, the adaptive temperature and pressure control model is used to perform negative feedback regulation on the temperature and pressure inside the reactor, and the gain function is dynamically adjusted according to the reaction time to achieve rapid adjustment in the early stage of the reaction and small-scale fine-tuning in the late stage, avoiding over-adjustment and improving the accuracy of temperature and pressure control.

[0072] 2. The present invention obtains target components and by-product qualities through online component analysis, establishes a mathematical regression analysis model based on historical data, describes the impact of various operating parameters on product quality, and uses a machine learning model to determine the regression coefficient. The production parameter optimization module calculates the reflux valve opening that minimizes by-products based on the predicted reaction rate and the regression prediction model. By optimizing the distillation column operating parameters, by-products are minimized to ensure that the final hexachlorodisilane product quality meets the standards.

[0073] 3. The present invention carries out automatic safety inspections, conducts quantitative assessments of explosion risks, detects and infers leaks through leak location algorithms, matches graded protection signals and implements corresponding safety protection measures, such as ESD emergency shutdown, nitrogen inerting valve opening, and directional opening of explosion vents, effectively ensuring the safety of the production process. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings:

[0075] Figure 1is a system block diagram of the present invention;

[0076] Figure 2 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0077] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0078] See also Figure 1 As shown, the hexachlorodisilane automated production control system includes a reactor intelligent temperature and pressure control module, a multi-stage distillation tower purity control module, a production parameter optimization module, and a safety interlock and disaster suppression module.

[0079] The reactor intelligent temperature and pressure control module monitors the reactor's temperature, pressure, and silane concentration data in real time. It corrects the collected temperature data through temperature compensation, calculates the predicted reaction rate based on the corrected temperature data, and sends this predicted reaction rate to the production parameter optimization module. It monitors temperature and pressure fluctuations in the reactor, provides negative feedback control of the internal temperature and pressure, and outputs pressure or temperature anomaly signals to the safety interlock and disaster suppression module based on analysis of fluctuation anomalies.

[0080] Real-time collection of reactor internal temperature T (t), internal pressure P (t) and motor stirring power , where t is the data collection time. The collected reactor temperature is corrected by the temperature compensation model using the preset formula:

[0081]

[0082] Calculate corrected temperature , where k1 is the temperature drift coefficient, k2 is the stirring thermal effect coefficient, It is the benchmark stirring power of the motor.

[0083] It should be noted that in the actual production of hexachlorodisilane, temperature drift and motor stirring can cause thermal effects, which can lead to inaccurate collected temperature data. Therefore, a temperature compensation model is used to correct the collected reactor temperature using the temperature drift coefficient k1, the stirring thermal effect coefficient k2, and the motor's reference stirring power to obtain more accurate temperature data for subsequent control. Furthermore, the hexachlorodisilane synthesis reaction is very sensitive to temperature, and temperature drift can cause the actual temperature to exceed the sensor reading, triggering side reactions.

[0084] Furthermore, the reaction rate prediction is performed, and the specific process is as follows:

[0085] Obtain real-time silane concentration in the reactor , by the preset formula:

[0086]

[0087] Calculate the predicted reaction rate r(t). Where A is the preset pre-factor, is the preset reaction activation energy, where α is the preset pressure correction coefficient.

[0088] The predicted reaction rate r(t) is sent to the production parameter optimization module.

[0089] Furthermore, the temperature and pressure fluctuations of the reactor are monitored, and negative feedback regulation is performed on the temperature and pressure inside the reactor. The specific process is as follows:

[0090] Get the target temperature preset by the user and target pressure , through the adaptive temperature and pressure control model:

[0091]

[0092] Calculating temperature control input and pressure control input , among which and The first and second temperature adaptive gain functions dynamically adjust the difference between the internal temperature of the reactor and the target temperature as the reaction time in the reactor increases. and its integral part The impact weight of

[0093] Among them and The first and second pressure adaptive gain functions dynamically adjust the difference between the internal pressure of the reactor and the target pressure as the reaction time in the reactor increases. and its integral part The impact weight of

[0094] Among them, the first and second temperature adaptive gain functions and the first and second pressure adaptive gain functions are all power functions that converge with time t, so that the error changes of temperature and pressure can be quickly adjusted at the beginning of the reaction, and the temperature and pressure errors can be fine-tuned in a small range at the end of the reaction. When the temperature and pressure are close to the target values, the value of the adaptive gain function is reduced to avoid over-adjustment.

[0095] Enter the temperature control and pressure control input The output power adjustment value is sent to the temperature control system and booster valve of the reactor to adjust the temperature and pressure of the reactor.

[0096] Furthermore, based on the temperature control input and pressure control input Perform abnormality identification. The specific process is as follows:

[0097] When the temperature control input is recognized If the temperature deviation is greater than the first preset threshold, it is determined that the temperature deviation is too large and the temperature control system cannot complete the temperature adjustment, and a temperature abnormality signal is generated; when the pressure control input is identified If the pressure deviation is greater than the second preset threshold, it is determined that the pressure deviation is too large and the pressure regulation cannot be completed by the boost valve, and a pressure abnormality signal is generated. The generated temperature abnormality signal and pressure abnormality signal are sent to the safety interlock and disaster suppression module.

[0098] The multi-stage distillation column purity control module accesses the gas chromatograph and differential pressure level gauge in the multi-stage distillation column to perform online component analysis and obtain the target component and by-product qualities. It deeply mines real-time and historical data on temperature gradients, pressure gradients, steam control valve opening, reflux ratio, target component quality, light component by-product quality, and heavy component by-product quality during the distillation process, analyzing the linkages and underlying patterns between these data points to develop a regression prediction model for control decisions.

[0099] Each section of the distillation column is numbered with the symbol i, where i=1, 2, ..., n, and n is the total number of sections in the distillation column. The symbol i=1 represents the bottom section of the distillation column, and the symbol i=n represents the top section of the distillation column.

[0100] Real-time collection of distillation column data, including:

[0101] The temperature Ti(t), pressure Pi(t) and distillate mass mi(t) of each column section;

[0102] The steam regulating valve opening θ(t) and the reflux valve opening L(t) of the distillation tower;

[0103] Where t is the time of data collection.

[0104] Furthermore, the target component and byproduct mass statistics were calculated. The distillate composition of each tower section was scanned using a gas chromatograph. The tower section number i0, where the final product hexachlorodisilane resides, was located, and the mass m0(t) of the distillate from that section was obtained. The byproduct mass in the distillation tower was obtained, including the mass of the light component m1(t) and the mass of the heavy component m2(t).

[0105] The light components refer to all distillates in the tower sections whose numbers are greater than i0;

[0106] The heavy components refer to all distillates in the column sections whose numbers are less than i0.

[0107] It should be noted that during the distillation process, light components with lower boiling points are taken away at the top of the distillation tower and cannot be completely recovered by the reflux liquid in the tower; heavy components with higher boiling points are collected at the top of the distillation tower.

[0108] Furthermore, based on historical data, a mathematical regression analysis model was established to describe the effects of temperature, pressure, steam control valve opening, and reflux valve opening of each tower section on the quality of the light component, the quality of the heavy component, and the quality ratio of the final product. The specific process is as follows:

[0109] Define the ratio of the mass of the light component to the mass of the final product: η1(t)=m1(t) / m0(t), and record it as the first characteristic efficiency;

[0110] Define the ratio of the mass of the heavy component to the mass of the final product: η2(t)=m2(t) / m0(t), and record it as the second characteristic efficiency.

[0111] A regression prediction model is established for the first characteristic efficiency, the second characteristic efficiency, the temperature, pressure, steam regulating valve opening, and reflux valve opening of each tower section:

[0112]

[0113] Among them and are the regression coefficients of temperature gradient and pressure gradient, respectively, representing the influence of temperature and pressure of each tower section i on the first characteristic efficiency;

[0114] Among them and is the regression coefficient of the steam control valve opening and the return valve opening, representing the influence of the steam control valve opening and the return valve opening on the first characteristic efficiency; b1 is the first error term, which simulates the error of the first characteristic efficiency caused by the noise part and the unmodeled factors.

[0115] Among them and are the regression coefficients of temperature gradient and pressure gradient, respectively, representing the influence of temperature and pressure of each tower section i on the second characteristic efficiency;

[0116] Among them and is the regression coefficient of the steam control valve opening and the return valve opening, representing the influence of the steam control valve opening and the return valve opening on the second characteristic efficiency; b2 is the second error term, which simulates the error of the second characteristic efficiency caused by the noise part and the unmodeled factors.

[0117] Furthermore, historical data is used to train machine learning models, including support vector machines and neural networks, to determine the specific values ​​of all regression coefficients and the first and second error terms in the regression prediction model, and then substitute them into the regression prediction model.

[0118] The production parameter optimization module uses the predicted reaction rate to estimate the mass of the reactor product and the mass percentage of hexachlorodisilane, and determines the opening range of the steam control valve from the reactor to the distillation column. Based on the regression prediction model and the steam control valve opening range, the production parameter optimization calculation is performed to calculate the reflux valve opening that minimizes byproducts.

[0119] Obtain the predicted reaction rate r(t) and determine the opening range of the steam regulating valve for the reaction product to enter the distillation tower from the reactor. The specific process is as follows:

[0120] Get the volume of liquid product in the reactor product , obtain the pressure difference before and after the steam regulating valve between the reactor and the distillation tower , substitute it into the preset formula:

[0121]

[0122] Calculate the opening reference value of the steam control valve , where K1 is the system structure constant, is the molar mass of hexachlorodisilane, is the density of hexachlorodisilane, where Cv is the preset valve flow coefficient.

[0123] Determine the maximum opening of the steam control valve as the opening reference value Compared with the preset reference value The maximum value between , }; Determine that the minimum opening of the steam control valve is 0.

[0124] Determine the opening range of the steam control valve as (0, Max{ , }).

[0125] Furthermore, production parameters are optimized based on the adjustment range of the steam control valve opening and the regression prediction model. The operating parameters of the distillation tower are adjusted by predicting the quality of the target component and by-product. The tower section temperature, tower section pressure, steam control valve opening, and reflux valve opening that minimize the first characteristic efficiency and the second characteristic efficiency under equipment constraints are calculated, thereby providing decision support for the optimal operation of the distillation tower. The specific process is as follows:

[0126] In the regression prediction model, the range of the steam control valve opening θ(t) is (0, Max{ , }), the temperature Ti(t) and pressure Pi(t) of each tower section are preset values.

[0127] Solve the return valve opening L(t) curve that satisfies the condition: minimize the sum of the first characteristic efficiency η1(t) and the second characteristic efficiency η2(t), and record it as the return valve opening control curve.

[0128] The return valve opening control curve is sent to the return valve control unit, and the return valve opening is controlled by the PID algorithm to fluctuate around the return valve opening control curve obtained by the solution.

[0129] After receiving an abnormal temperature signal or an abnormal pressure signal, the safety interlock and disaster suppression module conducts automatic safety inspections, accesses equipment instruments and flame detectors, obtains combustible gas concentration, equipment vibration acceleration and fire water pressure, and conducts a quantitative assessment of the explosion risk. It also detects leaks and infers the leak location through a leak location algorithm, matches the corresponding graded protection signal, and executes corresponding safety protection measures, including ESD emergency shutdown, opening of the nitrogen inerting valve, and directional opening of the explosion vent.

[0130] When an abnormal temperature signal is received, the equipment instrument is accessed to obtain the equipment vibration acceleration and the real-time pressure in the reactor.

[0131] When the vibration acceleration of the equipment or the real-time pressure in the reactor is detected to be greater than the preset threshold, a first-level pressure protection signal is generated;

[0132] When it is detected that the vibration acceleration of the equipment and the real-time pressure in the reactor are both greater than the preset threshold, a secondary pressure protection signal is generated.

[0133] When the abnormal pressure signal is received, the flame detector is accessed to obtain the combustible gas concentration and the real-time temperature of the reactor.

[0134] When the combustible gas concentration or the real-time temperature of the reactor is detected to be greater than the preset threshold, a first-level temperature protection signal is output;

[0135] When it is simultaneously identified that the combustible gas concentration and the real-time temperature of the reactor are greater than the preset threshold, a secondary temperature protection signal is output.

[0136] Execute corresponding safety protection measures according to the generated protection signal. The specific process is as follows:

[0137] When the first-level pressure protection signal is received, the directional opening part of the explosion vent is activated and the specific value of the abnormal pressure value is transmitted to the central monitoring equipment;

[0138] When the secondary pressure protection signal is received, the emergency shutdown procedure and the directional opening part of the explosion vent are started, and the nitrogen inerting valve is started to inject nitrogen into the reactor and distillation tower to reduce the concentration of combustible gas.

[0139] When receiving the first-level temperature protection signal, the emergency cooling device is activated to enhance the heat exchange capacity of the equipment;

[0140] When the secondary temperature protection signal is received, the emergency shutdown program, nitrogen inerting valve and emergency cooling device are activated, and the specific value of the temperature abnormality is transmitted to the central monitoring equipment.

[0141] See also Figure 2 As shown, the hexachlorodisilane automated production control method comprises the following steps:

[0142] Step 1: Reactor temperature and pressure control and reaction rate prediction;

[0143] The temperature, pressure and silane concentration of the reactor are monitored in real time and temperature compensation correction is performed.

[0144] Collect temperature, pressure, and silane concentration data within the reactor. Correct the temperature data to compensate for temperature drift and stirring heat effects. Calculate the corrected temperature to ensure precise temperature control during the reaction and avoid side reactions.

[0145] The reaction rate is predicted based on the corrected temperature and the predicted value is sent to the production parameter optimization module. Temperature and pressure fluctuations are monitored in real time, negative feedback is adjusted, and abnormal signals are output to the safety interlock and disaster suppression module.

[0146] Step 2: Reaction rate prediction and reaction control;

[0147] The reaction rate is predicted by the silane concentration in the reactor, which helps optimize the production process and calculate the predicted reaction rate value. The calculation is performed using formulas such as silane concentration and preset reaction activation energy.

[0148] The predicted reaction rate is sent to the production parameter optimization module to optimize subsequent operations.

[0149] Step 3: Identification of temperature and pressure anomalies and safety interlock;

[0150] Adjust the temperature and pressure control inputs based on the deviation between the actual temperature and pressure in the reactor and the target values. Obtain the user's preset target temperature and target pressure, calculate the temperature and pressure control inputs using the adaptive temperature and pressure control model, and send these as output power adjustment values ​​to the reactor's temperature control system and booster valve to adjust the reactor's temperature and pressure.

[0151] Step 4: Purity control of multi-stage distillation tower;

[0152] Control the operation of the distillation tower to ensure that the final product quality of hexachlorodisilane meets standards and minimize byproducts. Monitor distillation tower operating parameters such as tower section temperature, pressure, and steam control valve opening in real time. Gas chromatograph scans are used to determine the component masses of each tower section, locate the final product section, and determine the quality of the final product distillate. A mathematical regression analysis model is developed to predict the impact of tower section operating parameters on the quality of light components, heavy components, and the target product. Based on the regression model, optimize the distillation tower operating parameters to minimize byproducts.

[0153] Step 5: Optimize production parameters;

[0154] Based on the reaction rate prediction, the steam control valve opening range is calculated and steam flow control is performed. Based on the regression model and optimization algorithm, the optimal control curve of the reflux valve opening is calculated to ensure the minimization of by-products and optimize the distillation process.

[0155] It should be understood that the terms “include” and “comprising” used in the specification and claims of the present disclosure indicate the presence of described features, integers, steps, operations, elements and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.

[0156] It should also be understood that the terms used in this disclosure are for the purpose of describing particular embodiments only and are not intended to limit the disclosure. As used in this disclosure and the claims, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should also be further understood that the term "and / or" used in this disclosure and the claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations;

[0157] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to specific embodiments. Obviously, many modifications and variations are possible based on the contents of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. The automated production control system for hexachlorodisilane includes an intelligent temperature and pressure control module for the reactor, a purity control module for the multi-stage distillation tower, and a production parameter optimization module. It is characterized by: The intelligent temperature and pressure control module of the reactor monitors the temperature, pressure and silane concentration data of the reactor in real time; it corrects the collected temperature data through temperature compensation, calculates the predicted reaction speed based on the corrected temperature data, and sends the predicted reaction speed to the production parameter optimization module; Monitor the temperature and pressure fluctuations of the reactor, perform negative feedback regulation on the internal temperature and pressure of the reactor, and output abnormal pressure or temperature signals to the safety interlock and disaster suppression module based on abnormal fluctuation analysis; The multi-stage distillation tower purity control module accesses the gas chromatograph and differential pressure level gauge in the multi-stage distillation tower to perform online component analysis and obtain the quality of the target component and by-products. It deeply mines real-time and historical data on the temperature gradient, pressure gradient, steam control valve opening, reflux ratio, target component quality, light component by-product quality, and heavy component by-product quality during the distillation process, analyzes the linkage effects and potential patterns among various data, and provides a regression prediction model for control decisions. The production parameter optimization module obtains the predicted reaction rate to estimate the mass of the reactor product and the mass ratio of hexachlorodisilane, and determines the opening range of the steam control valve entering the distillation tower from the reactor. Based on the regression prediction model and the opening range of the steam control valve, the production parameter optimization calculation is carried out to calculate the reflux valve opening when the by-product is minimized.

2. The hexachlorodisilane automated production control system according to claim 1, characterized in that: Also includes safety interlock and disaster suppression modules: After receiving abnormal temperature or pressure signals, the safety interlock and disaster suppression module conducts automatic safety inspections, accesses equipment instruments and flame detectors, obtains combustible gas concentration, equipment vibration acceleration, and fire water pressure, and conducts a quantitative assessment of the explosion risk. It also detects leaks and estimates the leak location through a leak location algorithm, matches the corresponding graded protection signal, and executes corresponding safety protection measures.

3. The hexachlorodisilane automated production control system according to claim 1, characterized in that: The specific process of correcting the collected temperature data through temperature compensation is as follows: The internal temperature and pressure of the reactor and the stirring power of the motor are collected in real time; the preset temperature drift coefficient, stirring thermal effect coefficient and reference stirring power of the motor are obtained, and the obtained temperature of the reactor is corrected by inputting them into the temperature compensation model to obtain the corrected temperature; To predict the reaction rate, the specific process is as follows: Based on the corrected temperature, the real-time silane concentration and main reaction activation energy data in the reactor are obtained, and input into the preset formula to calculate the predicted reaction rate; the predicted reaction rate is sent to the production parameter optimization module.

4. The hexachlorodisilane automated production control system according to claim 1, characterized in that: The specific process of negative feedback regulation of the temperature and pressure inside the reactor is as follows: Obtaining the target temperature and target pressure preset by the user, calculating the difference between the internal temperature of the reactor and the target temperature and its integral part through the adaptive temperature and pressure control model, and calculating the temperature control input through the preset first and second temperature adaptive gain functions; The difference between the internal pressure of the reactor and the target pressure and its integral are calculated using the adaptive temperature and pressure control model, and the pressure control input is calculated using the preset first and second pressure adaptive gain functions; The first and second temperature adaptive gain functions, as well as the first and second pressure adaptive gain functions, are all power functions that converge over time, allowing for rapid adjustments to temperature and pressure errors at the beginning of the reaction, and small adjustments to the temperature and pressure errors at the end of the reaction. When the temperature and pressure approach the target values, the adaptive gain function values ​​are reduced to avoid over-adjustment. The temperature control input and pressure control input are sent to the temperature control system and booster valve of the reactor as output power adjustment values ​​to adjust the temperature and pressure of the reactor.

5. The hexachlorodisilane automated production control system according to claim 1, characterized in that: The specific process of outputting abnormal pressure or abnormal temperature signals to the safety interlock and disaster suppression module based on abnormal fluctuation analysis is as follows: Identify anomalies based on temperature control input and pressure control input; When it is identified that the temperature control input is greater than the first preset threshold, it is determined that the temperature deviation is too large, and the temperature adjustment cannot be completed through the temperature control system, and a temperature abnormality signal is generated; when it is identified that the pressure control input is greater than the second preset threshold, it is determined that the pressure deviation is too large, and the pressure adjustment cannot be completed through the boost valve, and a pressure abnormality signal is generated; the generated temperature abnormality signal and pressure abnormality signal are sent to the safety interlock and disaster suppression module.

6. The hexachlorodisilane automated production control system according to claim 1, characterized in that: The specific process of deep mining of real-time data and historical data is as follows: Obtain the temperature, pressure and distillate quality of each column section; Obtain the steam regulating valve opening and reflux valve opening of the distillation tower; Perform target component and by-product mass statistics, scan the distillate composition of each tower section using a gas chromatograph, locate the tower section number where the final product hexachlorodisilane is located, and obtain the mass of the final product distillate; obtain the mass of by-products in the distillation tower, including the mass of light components and heavy components; The light component index number is greater than all distillates in the column section where the hexachlorodisilane is located; The heavy component index number is smaller than all distillates in the column section where the hexachlorodisilane is located; Based on historical data, a mathematical regression analysis model was established to describe the effects of temperature, pressure, steam control valve opening, and reflux valve opening of each tower section on the quality of the light component, the quality of the heavy component, and the quality ratio of the final product. The specific process is as follows: The ratio of the mass of the light component to the mass of the final product is defined as the first characteristic efficiency; The ratio of the mass of the heavy component to the mass of the final product is defined as the second characteristic efficiency; Establishing a regression prediction model for the first characteristic efficiency and the second characteristic efficiency and the temperature, pressure, steam control valve opening, and reflux valve opening of each tower section, and assigning regression coefficients to the temperature, pressure, steam control valve opening, and reflux valve opening of each tower section, including a regression coefficient for the first characteristic efficiency and a regression coefficient for the second characteristic efficiency; The first characteristic efficiency is fitted by calculating the sum of the product of the temperature, pressure, steam regulating valve opening and reflux valve opening of each tower section and their corresponding regression coefficients and the sum of the first error term; The second characteristic efficiency is fitted by calculating the sum of the product of the temperature, pressure, steam regulating valve opening and reflux valve opening of each tower section and their corresponding regression coefficients and the sum of the second error term; Use historical data to train machine learning models, including support vector machines and neural networks, to determine the specific values ​​of all regression coefficients and the first and second error terms in the regression prediction model, and substitute them for the regression prediction model.

7. The hexachlorodisilane automated production control system according to claim 1, characterized in that: The specific process of optimizing production parameters based on the regression prediction model and the steam control valve opening range is as follows: Obtain the predicted reaction rate and determine the opening range of the steam control valve for the reaction product entering the distillation tower from the reactor. Obtain the volume of the liquid product in the reactor product and the pressure difference before and after the steam control valve between the reactor and the distillation tower, perform comprehensive calculations on these and the predicted reaction rate, and calculate the opening reference value of the steam control valve. Determine the maximum opening of the steam control valve as the maximum value between the opening reference value and the preset reference value. Determine the minimum opening of the steam control valve as 0. Production parameters are optimized based on the adjustment range of the steam control valve opening and the regression prediction model. The operating parameters of the distillation tower are adjusted by predicting the quality of the target components and by-products. The tower section temperature, tower section pressure, steam control valve opening, and reflux valve opening that minimize the first and second characteristic efficiencies under equipment constraints are calculated, thereby providing decision support for the optimized operation of the distillation tower.

8. The hexachlorodisilane automated production control system according to claim 7, characterized in that: The specific process of adjusting the operating parameters of the distillation column by predicting the quality of the target component and the by-product is as follows: In the regression prediction model, the steam control valve opening range is set to 0 to the maximum value between the opening reference value and the preset reference value, and the temperature and pressure of each tower section are set to the preset value; Find the return valve opening curve that satisfies the condition: the sum of the first characteristic efficiency and the second characteristic efficiency is minimized, and record it as the return valve opening control curve; The return valve opening control curve is sent to the return valve control unit, and the return valve opening is controlled by the PID algorithm to fluctuate around the return valve opening control curve obtained by the solution.

9. The hexachlorodisilane automated production control system according to claim 2, characterized in that: The specific process of matching the corresponding hierarchical protection signal and executing the corresponding security protection measures is as follows: When receiving a temperature anomaly signal, access the equipment instrument to obtain the equipment vibration acceleration and real-time pressure in the reactor; When the vibration acceleration of the equipment or the real-time pressure in the reactor is detected to be greater than the preset threshold, a first-level pressure protection signal is generated; When it is detected that the vibration acceleration of the equipment and the real-time pressure in the reactor are both greater than the preset threshold, a secondary pressure protection signal is generated; When receiving the abnormal pressure signal, the flame detector is accessed to obtain the combustible gas concentration and the real-time temperature of the reactor; When the combustible gas concentration or the real-time temperature of the reactor is detected to be greater than the preset threshold, a first-level temperature protection signal is output; When it is simultaneously identified that the combustible gas concentration and the real-time temperature of the reactor are greater than the preset threshold, a secondary temperature protection signal is output; Execute corresponding safety protection measures according to the generated protection signal. The specific process is as follows: When the first-level pressure protection signal is received, the directional opening part of the explosion vent is activated and the specific value of the abnormal pressure value is transmitted to the central monitoring equipment; When the secondary pressure protection signal is received, the emergency shutdown procedure and the directional opening part of the explosion vent are started, and the nitrogen inerting valve is started to inject nitrogen into the reactor and distillation tower to reduce the concentration of combustible gas. When receiving the first-level temperature protection signal, the emergency cooling device is activated to enhance the heat exchange capacity of the equipment; When the secondary temperature protection signal is received, the emergency shutdown program, nitrogen inerting valve and emergency cooling device are activated, and the specific value of the temperature abnormality is transmitted to the central monitoring equipment.

10. The automated production control method of hexachlorodisilane is characterized in that: The following steps are involved: Step 1: Reactor temperature and pressure control and reaction rate prediction; Real-time monitoring of the reactor's temperature, pressure and silane concentration, and temperature compensation correction; Collect temperature, pressure, and silane concentration data within the reactor; correct the temperature data to compensate for temperature drift and stirring thermal effects; calculate the corrected temperature to ensure precise temperature control during the reaction and avoid side reactions; The reaction rate is predicted based on the corrected temperature and the predicted value is sent to the production parameter optimization module; temperature and pressure fluctuations are monitored in real time, negative feedback is adjusted, and abnormal signals are output to the safety interlock and disaster suppression module; Step 2: Reaction rate prediction and reaction control; The reaction rate is predicted by the silane concentration in the reactor, which helps optimize the production process and calculate the reaction rate prediction value. The calculation is performed through the formula of silane concentration and preset reaction activation energy. Send the predicted reaction rate to the production parameter optimization module to optimize subsequent operations; Step 3: Identification of temperature and pressure anomalies and safety interlock; Adjust the temperature and pressure control inputs based on the deviation between the actual temperature and pressure in the reactor and the target values; obtain the target temperature and target pressure preset by the user, calculate the temperature control input and pressure control input through the adaptive temperature and pressure control model, and send the temperature control input and pressure control input as the output power adjustment value to the temperature control system and booster valve of the reactor to adjust the temperature and pressure of the reactor; Step 4: Purity control of multi-stage distillation tower; Control the operation of the distillation tower to ensure that the final product quality of hexachlorodisilane meets standards and reduces by-products; monitor the tower section temperature, pressure, steam control valve opening and other operating parameters in real time; obtain the component mass of each tower section through gas chromatography scanning, locate the final product tower section and obtain the quality of the distillate; establish a mathematical regression analysis model to predict the impact of tower section operating parameters on the quality of light components, heavy components and target products; and optimize the various distillation tower operating parameters based on the regression model to minimize by-products. Step 5: Optimize production parameters; Based on the reaction rate prediction, the opening range of the steam control valve is calculated and steam flow control is performed; based on the regression model and optimization algorithm, the optimal control curve of the reflux valve opening is calculated to ensure that by-products are minimized and the distillation process is optimized.

Images