Intelligent temperature field regulation and control method and system based on HVPE equipment

Abstract

The invention discloses an intelligent temperature field regulation and control method and system based on HVPE equipment, and relates to the field of temperature field control, and the method comprises the steps: S1, carrying out the multi-dimensional collection and time-space fusion of multi-source heterogeneous data; s2, temperature field balance degree quantitative evaluation and distributed cooperative control are carried out, the temperature field balance degree is calculated through data collected in the S1, temperature field unbalance judgment and cooperative control of a heater are completed through a six-dimensional fuzzy PID controller and a consistency algorithm, and the uniformity and stability of the temperature field are improved; s3, heat, flow, deposition and pressure four-field coupling dynamic compensation; and S4, dynamic learning optimization based on digital twinning. According to the intelligent temperature field regulation and control method and system based on the HVPE equipment, comprehensive improvement of temperature field control precision, complex working condition adaptability and equipment operation and maintenance efficiency is achieved based on multi-source heterogeneous data acquisition, temperature field balance control, four-field coupling compensation and digital twinning optimization.

Description

Technical Field

[0001] The present invention relates to the field of temperature field control, and in particular to an intelligent temperature field control method and system based on HVPE equipment. Background Art

[0002] HVPE equipment is a key device for growing high-quality semiconductor single crystal films, and plays a core role in the preparation of wide-bandgap semiconductors such as gallium nitride and zinc oxide.

[0003] Traditional HVPE equipment relies solely on single-point thermocouple temperature measurement, which is unable to capture micro-area temperature differences, easily leading to temperature field evaluation distortion and control command lag. At the same time, most of them use single-zone PID control, without considering the thermal conduction coupling of the sample zone, transition zone, and exhaust zone in the HVPE equipment reaction chamber, resulting in cross-zone interference causing temperature field oscillation. In addition, through temperature feedback adjustment, it is easy to ignore the cross-influence of flow field turbulence, deposition by-products, and pressure fluctuations on the temperature field, resulting in a decrease in control accuracy under complex working conditions.

[0004] Therefore, it is necessary to propose an intelligent temperature field control method and system based on HVPE equipment to solve the above problems. Summary of the Invention

[0005] The main purpose of the present invention is to provide an intelligent temperature field control method and system based on HVPE equipment, which can effectively solve the problems in the background technology.

[0006] To achieve the above object, the technical solution adopted by the present invention is: An intelligent temperature field control method based on HVPE equipment includes the following steps: S1: Multi-source heterogeneous data multi-dimensional acquisition and spatiotemporal fusion. Data acquisition equipment is deployed in different areas of the HVPE equipment to collect data, which is then processed through a spatiotemporal fusion model to complete the integration of multi-source heterogeneous data, providing an accurate data foundation for subsequent temperature field control. S2: Quantitative evaluation of temperature field balance and distributed collaborative control. The data collected by S1 is used to calculate the temperature field balance. Through a six-dimensional fuzzy PID controller and consistency algorithm, temperature field imbalance judgment and heater collaborative control are completed to improve temperature field uniformity and stability. S3: Dynamic compensation of the four-field coupling of heat, flow, deposition, and pressure. Based on the four-field coupling model, the compensation amount is calculated to control the heater power and carrier gas flow rate, completing the dynamic compensation of the HVPE equipment to eliminate the interference of multiple factors on the temperature field; S4: Dynamic learning optimization based on digital twins, using LSTM and PPO algorithms, combined with digital twin simulation, completes temperature field trend prediction and control parameter optimization to achieve autonomous optimization of HVPE equipment control strategies.

[0007] Preferably, the S1 specifically includes the following steps: S101: Multi-dimensional acquisition of multi-source heterogeneous data. The reaction chamber of the HVPE equipment is divided into a sample area, a transition area, and an exhaust area. 25 infrared thermocouples are arranged in the sample area, and the maximum temperature difference allowed between the 25 infrared thermocouples is 2°C. 24 fiber optic temperature sensors are arranged in the transition zone, where the transition zone is the reaction chamber wall of the reaction chamber; A contact thermocouple is arranged in the exhaust area to collect temperature data; Data collection of overall pressure fluctuations is performed using a laser particle imager, a first CCD camera, and a pressure sensor; Data collection of deposited surface morphology and roughness using an X-ray diffractometer and a second CCD camera; S102: Verification of multi-source heterogeneous data. When the HVPE device starts, the devices used in S101 start synchronously and begin data collection. When the measured value of any device used in S1 changes, that is, temperature > ±5°C, flow rate > ±2m / s, and pressure > ±10Pa, a triple verification mechanism is triggered: first, two repeated samples are taken within 100ms. If the standard deviation of the three measured values ​​exceeds twice that of the device, the data point is deemed invalid and marked as abnormal, and a warning signal is issued. S103: Spatiotemporal fusion of multi-source heterogeneous data. The verified multi-source heterogeneous data is stored in the local SSD and pre-processed using the spatiotemporal fusion model. The formula is: ; in The real-time temperature value after integrating the temperature data collected by S101; is the average temperature collected by the infrared thermocouple; It is the weighted average value of the temperature field collected by the optical fiber temperature sensor; Collect temperature data for contact thermocouples; is the epitaxial layer deposition rate, which is measured by X-ray diffractometer The scanning mode is calculated; It is the difference between the real-time pressure in the reaction chamber of the HVPE equipment and the process set pressure; 、 、 、 is the model weight coefficient, and its initial values ​​are 0.4, 0.3, 0.1, 0.15, and 0.05 respectively. .

[0008] Preferably, the S2 specifically includes the following steps: S201: The real-time temperature data collected by 25 infrared thermocouples in step S101 is ; Standard deviation of flow field velocity measured by laser particle imaging instrument ;Temperature difference of infrared thermocouple Flow field influencing factors , and the dynamic temperature field balance calculation formula is used to complete the quantitative evaluation of temperature field uniformity. The calculation period is 100ms per sampling period. The formula is: ; in is the temperature field balance; It is one of 25 infrared thermocouples; is the real-time average temperature of the sample area; Set the velocity for the flow field; for The standard deviation of the flow field velocity measured by the laser particle imager at each moment; S202: Divide the sample area, transition area, and exhaust area into three independent temperature control areas, and deploy the following equipment to achieve distributed control: Sample area: A radio frequency heater is used to directly heat the substrate tray placed in the sample area through a PWM signal; Transition zone: A resistance heater is used and installed around the middle and upper part of the outer wall of the transition zone; Exhaust area: Use a resistance heater and install it 100mm outside the exhaust pipe inlet; S203: When S201 calculates When the threshold is preset, the temperature control is coordinated by the six-dimensional fuzzy PID controller: Input variables include real-time temperature difference , temperature change rate , deposition thickness deviation , pressure deviation , it is divided into seven fuzzy subsets: negative large, negative medium, negative small, zero, positive small, positive medium, and positive large. Through 81 core fuzzy rules, the output variable is the PID parameter adjustment coefficient 、 、 The adjustment range is 50%-200% of the base value. The adjusted parameters directly act on the heaters in the sample area, transition area and tail gas area; S204: When the temperature difference between adjacent temperature zones exceeds ±5°C, the distributed cooperative control mechanism is triggered via the CAN bus: a consistency algorithm is used, and the formula is: ; in For the region The current power; , is the synergy coefficient; middle Indicates area The adjacent area number of ; For the region At the next discrete moment Heating power; For the region At the current discrete moment Heating power; For adjacent areas With the current area At discrete moments The power difference is calculated Send it to the corresponding area for power adjustment.

[0009] Preferably, in S202, the temperature control target of the sample zone is ±0.5°C; the temperature control target of the transition zone is ±1°C; and the temperature control target of the tail gas zone is ±2°C.

[0010] Preferably, the step S3 specifically includes the following steps: S301: Establish a coupling compensation model, where the compensation formula for thermal field-flow field-pressure coupling is: ; in It is the heating power compensation, which is used to correct the interference of flow field, deposition and pressure change on the temperature field; The real-time state of the flow field measured by the laser particle imager; is the coupling coefficient, where =0.8W·s / m; =50W·s / nm; =200W / Pa; The coordinated adjustment formula of process and thermal field is: ; in The carrier gas flow adjustment amount is used to balance the flow field fluctuation caused by thermal field adjustment; =0.1sccm / W, which is the power-flow conversion coefficient; =5sccm / (m / s), is the flow rate stability compensation coefficient, when Increase flow stability when increasing; is the standard deviation of the flow field velocity measured by the laser particle imager; S302: Dynamic supplement execution steps: Heating power adjustment, including sample area: according to Calculate the PWM duty cycle adjustment amount and output it in real time at a modulation frequency of 100Hz through the RF heater, with a power adjustment accuracy of ±0.5W; Transition zone: According to the cross-zone heat conduction compensation value required by the transition zone and Multiply as the power compensation value, and adjust through the resistance heater. The power change rate is ≤10W / s; exhaust area: only when When As an auxiliary adjustment amount, it is used to prevent drastic fluctuations in exhaust gas temperature and is adjusted by the heater; The carrier gas flow rate is adjusted by using a Brooks 5850S ​​mass flow controller to adjust the carrier gas flow rate of the required gas when the HVPE equipment is working according to the formula to maintain process stability. The formula is: ; in Set the target flow rate of the adjusted carrier gas; To adjust the original flow setting value of the front carrier gas.

[0011] Preferably, when any data fluctuation in S301 exceeds a self-set threshold, the data filtering algorithm is activated to eliminate the interference of high-frequency noise on the compensation calculation, and the adjustment range of the heating power of a single temperature zone does not exceed 30% of the current power to prevent current overload; the carrier gas flow rate does not exceed 90% of the mass flow controller range after adjustment to avoid exceeding the gas line pressure limit.

[0012] Preferably, in S3, when the following combined conditions are detected: 、 When the preset threshold and surface roughness are greater than 5nm, the following actions are triggered: A six-axis robotic arm with a silicon carbide scraper at the end was used to clean byproducts from the reaction chamber wall along a preset path at a speed of 5 mm / s. A 254 nm ultraviolet lamp was simultaneously turned on for 30 seconds to enhance the desorption of adsorbed atoms on the wall through photochemical action, thereby reducing particle redeposition.

[0013] Preferably, the S4 specifically includes the following steps: S401: The data collected by S1 is input into the LSTM neural network model. Using two layers of 128-dimensional LSTM units, the model captures the dynamic changes in the temperature field during HVPE equipment operation. Finally, the output layer predicts the operating parameters of the sample area, transition area, and exhaust area within the next 10 minutes. If the LSTM neural network model predicts that the temperature field fluctuation exceeds ±1°C, S2 is used to adjust the power of the HVPE equipment and issue an early warning. S402: The control strategy of the HVPE equipment is optimized through the PPO reinforcement learning model. Every hour, the PPO reinforcement learning model generates a set of optimization schemes including heating power, carrier gas flow, and fuzzy PID parameters based on the current temperature field balance, deposition thickness uniformity, and energy consumption. The scheme is first verified through 100 process cycle simulations in the digital twin simulation platform to ensure that there are no temperature overshoots or deposition anomalies. After encryption processing by the industrial host, it is sent to S2 and S3 via the CAN bus for execution of the control strategy.

[0014] An intelligent temperature field control system based on HVPE equipment includes a multi-source heterogeneous data acquisition module, a data processing and analysis module, a temperature field balance control module, a four-field coupling dynamic compensation module, and a digital twin optimization module. The multi-source heterogeneous data acquisition module collects temperature data through infrared thermocouples, fiber optic temperature sensors, and contact thermocouples placed in the sample area, transition area, and exhaust area of ​​the HVPE equipment; a laser particle imager and pressure sensor collect flow field velocity and pressure data; and an X-ray diffractometer and CCD camera collect deposition thickness and surface morphology data. The data processing and analysis module is used to transmit the data collected by the multi-source heterogeneous data acquisition module to the industrial host, pre-process the multi-source data through the spatiotemporal fusion model, and when the system starts, the acquisition equipment used by the multi-source heterogeneous data acquisition module synchronously collects data at a frequency of 10Hz; when the data is abnormal, three repeated sampling verifications are performed, invalid data is marked and an alarm is issued, and the fused data is used to calculate the temperature field balance, drive the four-field coupling compensation, and support the training of the digital twin model; The temperature field balance control module evaluates the uniformity of the temperature field through a balance calculation formula. When the balance is less than a preset threshold, the six-dimensional fuzzy PID controller is activated to adjust the heater power of the sample area, transition area and exhaust area; when the temperature difference between adjacent temperature zones exceeds ±0.5°C, a consistency algorithm is used to implement distributed collaborative control to maintain temperature field balance.

[0015] Preferably, the four-field coupling dynamic compensation module calculates the heating power compensation amount and the carrier gas flow adjustment amount based on the heat-flow-deposition-pressure coupling compensation formula, controls the heater power and mass flow controller of the sample area, transition area and tail gas area, and triggers the robot arm cleaning and ultraviolet light irradiation when the deposition rate is greater than 1nm / s and the temperature field balance is less than a preset threshold value to eliminate the influence of multi-field interference on the temperature field; The digital twin optimization module uses an LSTM model to predict the temperature field parameters for the next 10 minutes. When the preset threshold is exceeded, the pre-compensation parameters are sent to the temperature field balance control module. The PPO reinforcement learning algorithm generates an optimization strategy every hour. After verification by digital twin simulation, the model is updated according to the new data to achieve autonomous optimization of the control strategy.

[0016] Compared with the prior art, the present invention provides an intelligent temperature field control method and system based on HVPE equipment, which has the following beneficial effects: 1. This intelligent temperature field control method and system based on HVPE equipment can accurately perceive the three-dimensional temperature field of the HVPE equipment through the arrangement of multiple infrared thermocouples and fiber optic sensors, which can effectively improve the real-time temperature monitoring accuracy. Compared with single-point temperature measurement, the ability to identify micro-area temperature differences can be effectively improved, which can avoid the hidden problem of epitaxial layer defects caused by local overheating despite the overall temperature meeting the standard. At the same time, it can solve the problem of heat conduction lag caused by measuring only the temperature of the sample area. By collecting the flow field velocity distribution and identifying the turbulent area, it can avoid the temperature field blowing caused by high-speed airflow. Combined with the X-ray diffractometer and CCD camera, the deposition rate and surface morphology can be monitored in real time to solve the local thermal resistance changes caused by the accumulation of deposition by-products.

[0017] 2. The intelligent temperature field control method and system based on HVPE equipment divides the reaction chamber of the HVPE equipment into three temperature control areas: sample area, transition area, and exhaust area, deploys heating equipment of different precisions and builds a distributed collaborative control network: the sample area uses a high-frequency response radio frequency heater to achieve rapid and precise adjustment of the micro-area temperature, the transition area uses a resistance heater to buffer cross-zone heat conduction, and the exhaust area maintains overall thermal balance. Through quantitative evaluation of balance, the temperature uniformity and flow field stability parameters are integrated to dynamically judge the temperature field state. When the balance is not up to standard, the temperature difference, temperature difference change rate, deposition thickness deviation, and pressure deviation are converted into fuzzy language variables through the set six-dimensional fuzzy PID controller. The proportional, integral, and differential parameters are dynamically adjusted through 81 core rules, which can effectively solve the problem of insufficient control accuracy of traditional PID in multi-variable coupling scenarios. Based on the problem of excessive cross-zone temperature difference, the dynamic allocation of heating power to adjacent areas is achieved by adopting a consistency algorithm, forming a fast response, buffer compensation and global balance control system, which can effectively suppress the temperature field drift caused by heat conduction.

[0018] 3. This intelligent temperature field control method and system based on HVPE equipment constructs a four-field coupling model of heat-flow-deposition-pressure, and incorporates flow rate deviation, deposition rate, and pressure fluctuation into the compensation calculation of heating power and carrier gas flow rate. It can realize the active identification and dynamic offset of multi-physical field interference. When flow field turbulence or deposition anomaly is detected, the system automatically triggers targeted compensation measures: a silicon carbide scraper equipped with a robotic arm cleans the by-products on the reaction chamber wall along a preset path, and ultraviolet light irradiation is combined to enhance the desorption of adsorbed atoms on the wall, reducing the impact of particle redeposition on local thermal resistance; the carrier gas flow rate and heating power are adjusted in real time according to the coupling formula to balance the temperature field fluctuations caused by flow rate changes and the heat consumption changes caused by abnormal deposition rate. Strict hardware safety boundaries are set during the compensation process to avoid equipment overload caused by excessive adjustment of power and flow, ensuring the accuracy and safety of the execution action.

[0019] 4. This intelligent temperature field control method and system based on HVPE equipment can provide real-time operating condition information for balanced control and coupling compensation through the set multi-source data acquisition module. The temperature field balanced control module generates basic adjustment instructions based on the evaluation results. The four-field coupling compensation module can dynamically correct complex interference. The digital twin optimization module continuously improves the accuracy of the control strategy through historical data and real-time feedback.

[0020] 5. This intelligent temperature field control method and system based on HVPE equipment can effectively suppress temperature differences in micro-areas of the sample area through the deployment of high-density temperature sensors and distributed collaborative control, avoiding epitaxial layer defects caused by local overheating or overcooling; the multi-field coupling compensation mechanism can offset interference from flow field, deposition, etc. in real time, ensuring that the temperature field remains stable under complex working conditions and providing a uniform thermal environment for high-quality crystal growth.

[0021] 6. This intelligent temperature field control method and system based on HVPE equipment can quickly respond to harsh conditions such as flow rate fluctuations, pressure changes, and abnormal deposition rates through multi-field data fusion analysis and intelligent algorithms, automatically adjusting control parameters and executing actions to maintain temperature field balance within the ideal range. This can significantly improve the reliability of HPE equipment under extreme process conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0023] In order to make the technical means, creative features, objectives and effects achieved by the present invention easier to understand, the present invention is further described below in conjunction with specific implementation methods.

[0024] Example 1: like Figure 1 As shown, an intelligent temperature field control method based on HVPE equipment includes the following steps: S1: Multi-source heterogeneous data multi-dimensional acquisition and spatiotemporal fusion. Data acquisition equipment is deployed in different areas of the HVPE equipment to collect data. The data is processed through a spatiotemporal fusion model to complete the integration of multi-source heterogeneous data, providing an accurate data foundation for subsequent temperature field control. The specific operation steps include the following: S101: Multi-dimensional acquisition of multi-source heterogeneous data. The reaction chamber of the HVPE equipment is divided into a sample area, a transition area, and an exhaust area. 25 infrared thermocouples are arranged in the sample area, and the maximum temperature difference allowed between the 25 infrared thermocouples is 2°C. 24 fiber optic temperature sensors are arranged in the transition zone, where the transition zone is the reaction chamber wall of the reaction chamber; A contact thermocouple is arranged in the exhaust area to collect temperature data; Data collection of overall pressure fluctuations is performed using a laser particle imager, a first CCD camera, and a pressure sensor; Data collection of deposited surface morphology and roughness using an X-ray diffractometer and a second CCD camera; S102: Verification of multi-source heterogeneous data. When the HVPE equipment is started, the equipment used in S101 is started synchronously and data collection begins at the preset frequency: infrared thermocouple, fiber optic temperature sensor, contact thermocouple 10Hz, laser particle imager 10Hz, pressure sensor 20Hz, X-ray diffractometer 1 time / minute, first CCD camera and second CCD camera 15fps. When the measurement value of any device used in S1 changes suddenly, that is, temperature > ±5°C, flow rate > ±2m / s, pressure > ±10Pa, a triple verification mechanism is triggered: first, two repeated samples are taken within 100ms. If the standard deviation of the three measurement values ​​exceeds twice that of the device, the data point is deemed invalid and marked as abnormal, and a warning signal is issued. S103: Spatiotemporal fusion of multi-source heterogeneous data. The verified multi-source heterogeneous data is stored in the local SSD and pre-processed using the spatiotemporal fusion model. The formula is: ; in The real-time temperature value after integrating the temperature data collected by S101; is the average temperature collected by the infrared thermocouple; It is the weighted average value of the temperature field collected by the optical fiber temperature sensor; Collect temperature data for contact thermocouples; is the epitaxial layer deposition rate, which is measured by X-ray diffractometer The scanning mode is calculated; It is the difference between the real-time pressure in the reaction chamber of the HVPE equipment and the process set pressure; 、 、 、 is the model weight coefficient, and its initial values ​​are 0.4, 0.3, 0.1, 0.15, and 0.05 respectively. ; The particle swarm optimization algorithm is used to adjust the weight coefficient in real time. The objective function is to minimize the root mean square error between the fusion temperature and the measured temperature. The parameters of the particle swarm optimization algorithm are set as follows: the number of particles is fixed at 30, the number of iterations is 50, and the inertia weight is The acceleration factor decreases linearly from 0.8 to 0.4. The calibration cycle is every 10 minutes. During the calibration process, data acquisition and basic control functions remain running, and the optimized weight parameters take effect immediately after the iteration ends.

[0025] S2: Quantitative evaluation of temperature field balance and distributed collaborative control. The data collected by S1 is used to calculate the temperature field balance. Through the six-dimensional fuzzy PID controller and consistency algorithm, the temperature field imbalance judgment and the coordinated control of the heater are completed to improve the uniformity and stability of the temperature field. The specific operation steps include the following: S201: The real-time temperature data collected by 25 infrared thermocouples in step S101 is ; Standard deviation of flow field velocity measured by laser particle imaging instrument ;Temperature difference of infrared thermocouple Flow field influencing factors , and the dynamic temperature field balance calculation formula is used to complete the quantitative evaluation of temperature field uniformity. The calculation period is 100ms per sampling period. The formula is: ; in is the temperature field balance; It is one of 25 infrared thermocouples; is the real-time average temperature of the sample area; Set the velocity for the flow field; for The standard deviation of the flow field velocity measured by the laser particle imager at each moment; S202: Divide the sample area, transition area, and exhaust area into three independent temperature control areas, and deploy the following equipment to achieve distributed control: Sample area: A MKS247B RF heater (power range 0-500W, pulse width modulation frequency 100Hz, power adjustment accuracy ±0.5W) is used to directly heat the substrate tray placed in the sample area through a PWM signal (duty cycle 0-100%). Transition zone: Use Watlow F4T resistance heater (power range 0-1000W, equipped with thyristor power regulator, power adjustment resolution 1W), and install it around the middle and upper part of the outer wall of the transition zone; Exhaust area: Use OmegaCN7600 resistance heater (power range 0-500W, control accuracy ±2℃), and install it 100mm outside the exhaust pipe inlet; The temperature control target of the sample area is ±0.5°C; the temperature control target of the transition area is ±1°C; the control target of the exhaust area is ±2°C; S203: When S201 calculates When the threshold is preset, the temperature control is coordinated by the six-dimensional fuzzy PID controller: Input variables include real-time temperature difference , temperature change rate , deposition thickness deviation , pressure deviation , it is divided into seven fuzzy subsets: negative large, negative medium, negative small, zero, positive small, positive medium, and positive large. Through 81 core fuzzy rules, the output variable is the PID parameter adjustment coefficient 、 、 The adjustment range is 50%-200% of the base value. The adjusted parameters directly act on the heaters in the sample area, transition area and exhaust area. The six-dimensional fuzzy PID controller has four key input variables, namely the real-time temperature difference , temperature change rate , deposition thickness deviation , pressure deviation , where real-time temperature difference , temperature change rate , deposition thickness deviation , pressure deviation The fuzzy rule table contains 7×7×7×7=2401 rules, which are simplified to 81 core rules through orthogonal experiments in practical applications. Set the temperature for the process in the sample area; is the target thickness of the thin film deposition; The deposition thickness is measured in real time by an online ellipsometer; Set pressure for reaction chamber process; is the real-time measurement value of the pressure sensor, is the proportional coefficient adjustment amount; is the integral coefficient adjustment amount; is the differential coefficient adjustment amount, the core rules are as follows: when For negative, For Zhengda, For negative, When it is positive, the output PID parameter adjustment coefficient is Increase by 40%, Reduce by 20%, Increase by 15%. This rule is used to quickly increase the temperature while avoiding overshoot due to excessive deposition rate. like Zero, Zero, For small, If it is a negative hour, Reduce by 10%, A 5% increase is used to fine-tune the heating power to balance the effects of a slight increase in deposition thickness and a slight decrease in pressure on the temperature field; The 81 rules are based on multi-source data collected by S1 and the simulation results of the S3 four-field coupling model. They are a minimal and complete set formed by eliminating redundant and low-impact combinations to ensure coverage of more than 95% of typical operating conditions. S204: When the temperature difference between adjacent temperature zones exceeds ±5°C, the distributed cooperative control mechanism is triggered via the CAN bus: a consistency algorithm is used, and the formula is: ; in For the region The current power; , is the synergy coefficient; middle Indicates area The adjacent area number of ; For the region At the next discrete moment Heating power; For the region At the current discrete moment Heating power; For adjacent areas With the current area At discrete moments The power difference of each collaborative calculation cycle is 500ms, and the power adjustment step does not exceed 5% of the current power to avoid temperature overshoot caused by power mutation. The signal is sent to the corresponding area for power adjustment. The sample area controller adopts high-frequency PWM modulation to achieve fast response. The controllers of the transition area and exhaust area adopt smooth power adjustment to prevent thermal stress concentration. When it is detected that the actual temperature of any temperature zone deviates from the set value by more than 2 times the control target, that is, the sample area is greater than ±1℃, the transition area is greater than ±2℃, and the exhaust area is greater than ±4℃, the hardware redundancy protection is automatically triggered: the power supply of the heater in the corresponding area is cut off, and the liquid nitrogen cooling valve is started at the same time to limit the temperature fluctuation to a safe range: the sample area is ≤±1.5℃, the transition area is ≤±3℃, and the exhaust area is ≤±6℃.

[0026] S3: Dynamic compensation of the four-field coupling of heat, flow, deposition, and pressure. Based on the four-field coupling model, the compensation amount is calculated to control the heater power and carrier gas flow rate, complete the dynamic compensation of the HVPE equipment, and eliminate the interference of multiple factors on the temperature field. The specific operation steps include the following: S301: Establish a coupling compensation model, where the compensation formula for thermal field-flow field-pressure coupling is: ; in It is the heating power compensation, which is used to correct the interference of flow field, deposition and pressure change on the temperature field; The real-time state of the flow field measured by the laser particle imager; is the coupling coefficient, where =0.8W·s / m; =50W·s / nm; =200W / Pa; The coordinated adjustment formula of process and thermal field is: ; in The carrier gas flow adjustment amount is used to balance the flow field fluctuation caused by thermal field adjustment; =0.1sccm / W, which is the power-flow conversion coefficient; =5sccm / (m / s), is the flow rate stability compensation coefficient, when Increase flow stability when increasing; is the standard deviation of the flow field velocity measured by the laser particle imager; When any data fluctuation in S301 exceeds the self-set threshold, the data filtering algorithm is activated to eliminate the interference of high-frequency noise on the compensation calculation. The adjustment range of the heating power of a single temperature zone shall not exceed 30% of the current power to prevent current overload; the carrier gas flow rate shall not exceed 90% of the mass flow controller range after adjustment to avoid gas line pressure exceeding the limit; S302: Dynamic supplement execution steps: Heating power adjustment, including sample area: according to Calculate the PWM duty cycle adjustment amount and output it in real time with a modulation frequency of 100Hz through the MKS247B RF heater. The power adjustment accuracy is ±0.5W. Transition zone: According to the cross-zone heat conduction compensation value required by the transition zone and Multiplied as the power compensation value, the adjustment is performed through the WatlowF4T resistance heater, the power change rate is ≤10W / s; exhaust area: only when When As an auxiliary adjustment amount, it is used to prevent drastic fluctuations in exhaust gas temperature and is adjusted by the OmegaCN7600 heater; The carrier gas flow rate is adjusted by a Brooks 5850S ​​mass flow controller (adjustment step size 0.1 sccm, response time < 500 ms) according to the formula to adjust the carrier gas flow rate of the required gas when the HVPE equipment is working to maintain process stability. The formula is: ; in Set the target flow rate of the adjusted carrier gas; To adjust the original flow setting value of the front carrier gas When the following combined conditions are detected: 、 When the preset threshold and surface roughness are greater than 5nm, the following actions are triggered: A six-axis robotic arm with a silicon carbide scraper at the end cleans byproducts from the reaction chamber walls along a preset path at a speed of 5 mm / s. Simultaneously, a 254nm UV lamp with a power of 500 mW / cm² is turned on, covering 80% of the reaction chamber wall area for 30 seconds. This photochemical effect enhances the desorption of adsorbed atoms on the wall to reduce particle redeposition.

[0027] S4: Dynamic learning optimization based on digital twins, using LSTM and PPO algorithms, combined with digital twin simulation, completes temperature field trend prediction and control parameter optimization to achieve autonomous optimization of HVPE equipment control strategies. Specific steps include: S401: The data collected by S1 is input into the LSTM neural network model. Using two layers of 128-dimensional LSTM units, the model captures the dynamic changes in the temperature field during HVPE equipment operation. Finally, the output layer predicts the operating parameters of the sample area, transition area, and exhaust area within the next 10 minutes. If the LSTM neural network model predicts that the temperature field fluctuation exceeds ±1°C, S2 is used to adjust the power of the HVPE equipment and issue an early warning. S402: The control strategy of the HVPE equipment is optimized through the PPO reinforcement learning model. Every hour, the PPO reinforcement learning model generates a set of optimization solutions including heating power, carrier gas flow, and fuzzy PID parameters based on the current temperature field balance, deposition thickness uniformity, and energy consumption. The solutions are first verified through 100 process cycle simulations in the digital twin simulation platform to ensure that there are no temperature overshoots or deposition anomalies. After encryption processing by the industrial host, they are sent to S2 and S3 via the CAN bus for control strategy execution; If the parameters generated by the PPO reinforcement learning model for three consecutive times cause the temperature field balance of the HVPE equipment to deteriorate, the system will automatically revert to the parameters that were valid for the previous three times, mark the model training abnormality, and wait for engineer review and confirmation.

[0028] Example 2: An intelligent temperature field control system based on HVPE equipment includes a multi-source heterogeneous data acquisition module, a data processing and analysis module, a temperature field balance control module, a four-field coupling dynamic compensation module, and a digital twin optimization module. The multi-source heterogeneous data acquisition module collects temperature data through infrared thermocouples, fiber optic temperature sensors, and contact thermocouples placed in the sample area, transition area, and exhaust area of ​​the HVPE equipment; a laser particle imager and pressure sensor collect flow field velocity and pressure data; and an X-ray diffractometer and a CCD camera collect deposition thickness and surface morphology data. The data processing and analysis module is used to transmit the data collected by the multi-source heterogeneous data acquisition module to the industrial host. The multi-source data is pre-processed through a spatiotemporal fusion model. When the system starts, the acquisition equipment used by the multi-source heterogeneous data acquisition module synchronizes data at a frequency of 10Hz. When the data is abnormal, three repeated sampling verifications are performed, invalid data is marked and an alarm is issued. The fused data is used to calculate the temperature field balance, drive the four-field coupling compensation, and support the training of the digital twin model. The temperature field balance control module evaluates the uniformity of the temperature field through the balance calculation formula. When the balance is less than the preset threshold, the six-dimensional fuzzy PID controller is activated to adjust the heater power of the sample area, transition area and exhaust area; when the temperature difference between adjacent temperature zones exceeds ±0.5℃, the consistency algorithm is used to implement distributed collaborative control to maintain temperature field balance.

[0029] The four-field coupling dynamic compensation module calculates the heating power compensation and carrier gas flow adjustment based on the heat-flow-deposition-pressure coupling compensation formula, controls the heater power and mass flow controller in the sample area, transition area, and tail gas area. When the deposition rate is greater than 1nm / s and the temperature field balance is less than the preset threshold, the robot arm cleaning and UV irradiation are triggered to eliminate the impact of multi-field interference on the temperature field. The digital twin optimization module uses the LSTM model to predict the temperature field parameters in the next 10 minutes. When the preset threshold is exceeded, the pre-compensation parameters are sent to the temperature field balance control module. The PPO reinforcement learning algorithm generates an optimization strategy every hour. After verification by digital twin simulation, the model is updated according to the new data to achieve autonomous optimization of the control strategy.

[0030] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent temperature field control method based on HVPE equipment, characterized by: The following steps are included: S1: Multi-source heterogeneous data multi-dimensional acquisition and spatiotemporal fusion. Data acquisition equipment is deployed in different areas of the HVPE equipment to collect data, which is then processed through a spatiotemporal fusion model to complete the integration of multi-source heterogeneous data, providing an accurate data foundation for subsequent temperature field control. S2: Quantitative evaluation of temperature field balance and distributed collaborative control. This system uses the data collected by S1 to calculate the temperature field balance. Through a six-dimensional fuzzy PID controller and consistency algorithm, it completes the temperature field imbalance judgment and the coordinated control of the heaters to improve the temperature field uniformity and stability. S3: Dynamic compensation of the four-field coupling of heat, flow, deposition, and pressure. Based on the four-field coupling model, the compensation amount is calculated to control the heater power and carrier gas flow rate, completing the dynamic compensation of the HVPE equipment to eliminate the interference of multiple factors on the temperature field; S4: Dynamic learning optimization based on digital twins, using LSTM and PPO algorithms, combined with digital twin simulation, completes temperature field trend prediction and control parameter optimization, which is used to achieve autonomous optimization of HVPE equipment control strategies.

2. The intelligent temperature field control method based on HVPE equipment according to claim 1, characterized in that: Said S1 specifically includes the following steps: S101: Multi-dimensional acquisition of multi-source heterogeneous data. The reaction chamber of the HVPE equipment is divided into a sample area, a transition area, and an exhaust area. 25 infrared thermocouples are arranged in the sample area, and the maximum temperature difference allowed between the 25 infrared thermocouples is 2°C. 24 fiber optic temperature sensors are arranged in the transition zone, where the transition zone is the reaction chamber wall of the reaction chamber; A contact thermocouple is arranged in the exhaust area to collect temperature data; Data collection of overall pressure fluctuations is performed using a laser particle imager, a first CCD camera, and a pressure sensor; Data collection of deposited surface morphology and roughness using an X-ray diffractometer and a second CCD camera; S102: Verification of multi-source heterogeneous data. When the HVPE device starts, the devices used in S101 start synchronously and begin data collection. When the measured value of any device used in S1 changes suddenly, that is, temperature > ±5°C, flow rate > ±2m / s, and pressure > ±10Pa, a triple verification mechanism is triggered: first, two repeated samples are taken within 100ms. If the standard deviation of the three measured values ​​exceeds twice that of the device, the data point is deemed invalid and marked as abnormal, and a warning signal is issued. S103: Spatiotemporal fusion of multi-source heterogeneous data. The verified multi-source heterogeneous data is stored in the local SSD and pre-processed using the spatiotemporal fusion model. The formula is: ； in The real-time temperature value after integrating the temperature data collected by S101; is the average temperature collected by infrared thermocouple; It is the weighted average value of the temperature field collected by the optical fiber temperature sensor; Collect temperature data for contact thermocouples; is the epitaxial layer deposition rate, which is measured by X-ray diffractometer The scanning mode is calculated; It is the difference between the real-time pressure in the reaction chamber of the HVPE equipment and the process set pressure; 、 、 、 is the model weight coefficient, and its initial values ​​are 0.4, 0.3, 0.1, 0.15, and 0.05 respectively. .

3. The intelligent temperature field control method based on HVPE equipment according to claim 1, characterized in that: Said S2 specifically includes the following steps: S201: The real-time temperature data collected by 25 infrared thermocouples in step S101 is ; Standard deviation of flow field velocity measured by laser particle imaging instrument ;Temperature difference of infrared thermocouple Flow field influencing factors , and the dynamic temperature field balance calculation formula is used to complete the quantitative evaluation of temperature field uniformity. The calculation period is 100ms per sampling period. The formula is: ； in is the temperature field balance; It is one of 25 infrared thermocouples; is the real-time average temperature of the sample area; Set the velocity for the flow field; for The standard deviation of the flow field velocity measured by the laser particle imager at each moment; S202: Divide the sample area, transition area, and exhaust area into three independent temperature control areas, and deploy the following equipment to achieve distributed control: Sample area: A radio frequency heater is used to directly heat the substrate tray placed in the sample area through a PWM signal; Transition zone: A resistance heater is used and installed around the middle and upper part of the outer wall of the transition zone; Exhaust area: Use a resistance heater and install it 100mm outside the exhaust pipe inlet; S203: When S201 calculates When the threshold is preset, the temperature control is coordinated by the six-dimensional fuzzy PID controller: Input variables include real-time temperature difference , temperature change rate , deposition thickness deviation , pressure deviation , it is divided into seven fuzzy subsets: negative large, negative medium, negative small, zero, positive small, positive medium, and positive large. Through 81 core fuzzy rules, the output variable is the PID parameter adjustment coefficient 、 、 The adjustment range is 50%-200% of the base value. The adjusted parameters directly act on the heaters in the sample area, transition area and tail gas area; S204: When the temperature difference between adjacent temperature zones exceeds ±5°C, the distributed cooperative control mechanism is triggered via the CAN bus: a consistency algorithm is used, and the formula is: ； in For the region The current power; , is the synergy coefficient; middle Indicates area The adjacent area number of ; For the region At the next discrete moment Heating power; For the region At the current discrete moment Heating power; For adjacent areas With the current area At discrete moments The power difference is calculated Send it to the corresponding area for power adjustment.

4. The intelligent temperature field control method based on HVPE equipment according to claim 3, characterized in that: In the above S202, the temperature control target of the sample zone is ±0.5°C; the temperature control target of the transition zone is ±1°C; and the temperature control target of the tail gas zone is ±2°C.

5. The intelligent temperature field control method based on HVPE equipment according to claim 1, characterized in that: The S3 specifically includes the following steps: S301: Establish a coupling compensation model, where the compensation formula for thermal field-flow field-pressure coupling is: ； in It is the heating power compensation, which is used to correct the interference of flow field, deposition and pressure change on the temperature field; The real-time state of the flow field measured by the laser particle imager; is the coupling coefficient, where =0.8W·s / m; =50W·s / nm; =200W / Pa; The formula for coordinated adjustment of process and thermal field is: ； in The carrier gas flow adjustment amount is used to balance the flow field fluctuation caused by thermal field adjustment; =0.1sccm / W, which is the power-flow conversion coefficient; =5sccm / (m / s), is the flow rate stability compensation coefficient, when Increase flow stability when increasing; is the standard deviation of the flow field velocity measured by the laser particle imager; S302: Dynamic supplement execution steps: Heating power adjustment, including sample area: according to Calculate the PWM duty cycle adjustment and output it in real time through the RF heater at a modulation frequency of 100Hz, with a power adjustment accuracy of ±0.5W; Transition zone: According to the cross-zone heat conduction compensation value required by the transition zone Multiply as the power compensation value, and adjust through the resistance heater. The power change rate is ≤10W / s; exhaust area: only when When As an auxiliary adjustment amount, it is used to prevent the exhaust gas temperature from fluctuating violently, and the adjustment is performed by the heater; The carrier gas flow rate is adjusted by using a Brooks 5850S ​​mass flow controller to adjust the carrier gas flow rate of the required gas when the HVPE equipment is working according to the formula to maintain process stability. The formula is: ； in Set the target flow rate of the adjusted carrier gas; To adjust the original flow setting value of the front carrier gas.

6. The intelligent temperature field control method based on HVPE equipment according to claim 5, characterized in that: When any data fluctuation in S301 exceeds the self-set threshold, the data filtering algorithm is activated to eliminate the interference of high-frequency noise on the compensation calculation. The adjustment range of the heating power of a single temperature zone shall not exceed 30% of the current power to prevent current overload; the carrier gas flow rate shall not exceed 90% of the mass flow controller range after adjustment to avoid excessive gas line pressure.

7. The intelligent temperature field control method based on HVPE equipment according to claim 5, characterized in that: In S3, when the following combined conditions are detected: 、 When the preset threshold and surface roughness are greater than 5nm, the following actions are triggered: A six-axis robotic arm with a silicon carbide scraper at the end was used to clean byproducts from the reaction chamber wall along a preset path at a speed of 5 mm / s. A 254 nm ultraviolet lamp was simultaneously turned on for 30 seconds to enhance the desorption of adsorbed atoms on the wall through photochemical action, thereby reducing particle redeposition.

8. The intelligent temperature field control method based on HVPE equipment according to claim 1, characterized in that: The S4 specifically includes the following steps: S401: The data collected by S1 is input into the LSTM neural network model. Using two layers of 128-dimensional LSTM units, the model captures the dynamic changes in the temperature field during HVPE equipment operation. Finally, the output layer predicts the operating parameters of the sample area, transition area, and exhaust area within the next 10 minutes. If the LSTM neural network model predicts that the temperature field fluctuation exceeds ±1°C, S2 is used to adjust the power of the HVPE equipment and issue an early warning. S402: The control strategy of the HVPE equipment is optimized through the PPO reinforcement learning model. Every hour, the PPO reinforcement learning model generates a set of optimization schemes including heating power, carrier gas flow, and fuzzy PID parameters based on the current temperature field balance, deposition thickness uniformity, and energy consumption. The scheme is first verified through 100 process cycle simulations in the digital twin simulation platform to ensure that there are no temperature overshoots or deposition anomalies. After encryption processing by the industrial host, it is sent to S2 and S3 via the CAN bus for execution of the control strategy.

9. An intelligent temperature field control system based on HVPE equipment, using the intelligent temperature field control method based on HVPE equipment according to any one of claims 1 to 8, comprising a multi-source heterogeneous data acquisition module, a data processing and analysis module, a temperature field balance control module, a four-field coupling dynamic compensation module, and a digital twin optimization module, characterized in that: The multi-source heterogeneous data acquisition module collects temperature data through infrared thermocouples, fiber optic temperature sensors, and contact thermocouples placed in the sample area, transition area, and exhaust area of ​​the HVPE equipment; laser particle imagers and pressure sensors collect flow field velocity and pressure data; and X-ray diffractometers and CCD cameras collect deposition thickness and surface morphology data. The data processing and analysis module is used to transmit the data collected by the multi-source heterogeneous data acquisition module to the industrial host, pre-process the multi-source data through the spatiotemporal fusion model, and when the system starts, the acquisition equipment used by the multi-source heterogeneous data acquisition module synchronously collects data at a frequency of 10Hz; when the data is abnormal, three repeated sampling verifications are performed, invalid data is marked and an alarm is issued, and the fused data is used to calculate the temperature field balance, drive the four-field coupling compensation, and support the training of the digital twin model; The temperature field balance control module evaluates the uniformity of the temperature field through a balance calculation formula. When the balance is less than a preset threshold, the six-dimensional fuzzy PID controller is activated to adjust the heater power of the sample area, transition area and exhaust area; when the temperature difference between adjacent temperature zones exceeds ±0.5°C, a consistency algorithm is used to implement distributed collaborative control to maintain temperature field balance.

10. The intelligent temperature field control system based on HVPE equipment according to claim 9, characterized in that: The four-field coupling dynamic compensation module calculates the heating power compensation and carrier gas flow adjustment according to the heat-flow-deposition-pressure coupling compensation formula, controls the heater power and mass flow controller of the sample area, transition area and tail gas area, and triggers the robot arm cleaning and ultraviolet light irradiation when the deposition rate is greater than 1nm / s and the temperature field balance is less than the preset threshold to eliminate the influence of multi-field interference on the temperature field. The digital twin optimization module uses an LSTM model to predict the temperature field parameters for the next 10 minutes. When the preset threshold is exceeded, the pre-compensation parameters are sent to the temperature field balance control module. The PPO reinforcement learning algorithm generates an optimization strategy every hour. After verification by digital twin simulation, the model is updated according to the new data to achieve autonomous optimization of the control strategy.

Images