Non-contact measurement and closed-loop control system for melt level of single crystal furnace based on internet of things

Abstract

The application provides a single crystal furnace melt liquid level non-contact measurement and closed loop control system based on the Internet of Things, and relates to the technical field of single crystal furnaces.The application comprises a data acquisition module, a radar measurement module and a closed loop control module.The data acquisition module is used for acquiring single crystal furnace data of a single crystal furnace melt, and the single crystal furnace data comprises environmental data, equipment state data and control feedback data.The radar measurement module is used for measuring the liquid level of the single crystal furnace melt by using a radar to obtain an initial liquid level.The liquid level correction module is used for constructing a liquid level correction model, inputting the initial liquid level into the correction model and obtaining a corrected liquid level.The closed loop control module sets a target liquid level range, calculates the deviation value of the corrected liquid level data and the target liquid level in real time, and judges whether the deviation value exceeds a preset threshold value.If yes, the single crystal furnace parameters are adjusted.Through the installation of a multi-directional radar, the processing of signals by a wavelet algorithm and a liquid level correction model and other means, the influence of these interference factors on the measurement accuracy is effectively reduced, and the accuracy of liquid level measurement is improved.

Description

Technical Field

[0001] This invention relates to the field of single crystal furnace technology, and in particular to a non-contact measurement and closed-loop control system for melt level in a single crystal furnace based on the Internet of Things. Background Technology

[0002] Single crystal furnaces are key equipment for the growth of single crystals in semiconductor and photovoltaic materials. The stable control of the melt level directly affects the growth quality and production efficiency of the single crystal. Traditional melt level measurement methods have problems such as easy contamination of the melt by contact measurement and significant interference from high-temperature environments by non-contact measurement. In addition, the linkage between measurement data and the control system is poor, making it difficult to achieve real-time closed-loop control.

[0003] Traditional non-contact methods for measuring melt level in single crystal furnaces are often susceptible to the effects of high temperatures, complex environments, and external interference, leading to inaccurate radar signals and significant level measurement errors. Previously, relying on manual observation of the melt level and adjustment of furnace parameters resulted in problems such as untimely information acquisition, operational delays, and human error, making it difficult to ensure the melt level remained accurately and stably within the appropriate range. Summary of the Invention

[0004] This invention provides an IoT-based non-contact measurement and closed-loop control system for the melt level of a single crystal furnace, which solves the shortcomings of existing non-contact measurement methods, such as high temperatures inside the furnace, untimely information acquisition, and delayed operation.

[0005] On one hand, the present invention provides a non-contact measurement and closed-loop control system for melt level in a single crystal furnace based on the Internet of Things, comprising:

[0006] The data acquisition module is used to collect single crystal furnace data of the melt in the single crystal furnace. The single crystal furnace data includes: environmental data, equipment status data, and control feedback data.

[0007] The radar measurement module is used to combine data from the single crystal furnace with radar to measure the liquid level of the melt in the single crystal furnace and obtain the initial liquid level.

[0008] The liquid level correction module is used to build a liquid level correction model. The initial liquid level is input into the correction model to obtain the corrected liquid level.

[0009] The closed-loop control module sets the target liquid level range, calculates the deviation value ΔL between the corrected liquid level data and the target liquid level in real time, and determines whether the deviation value exceeds the preset threshold. If so, it adjusts the parameters of the single crystal furnace.

[0010] The IoT-based non-contact measurement and closed-loop control system for melt level in a single crystal furnace provided by this invention includes environmental data such as high-temperature environmental data inside and outside the furnace. Equipment status data includes single crystal furnace process parameter data and measurement and control system status data. The single crystal furnace process parameter data includes heating data and crucible parameter data.

[0011] According to the IoT-based non-contact measurement and closed-loop control system for single-crystal furnace melt level provided by the present invention, the specific steps for measuring the melt level of the single-crystal furnace melt using radar are as follows:

[0012] Radar equipment is installed on different sides and at different locations on the top of the furnace.

[0013] Turn on the radar power and initialize the radar settings.

[0014] In the radar's operating interface, input the current location coordinates, measurement range, and radar wave transmission frequency.

[0015] The radar continuously transmits radar waves at a set frequency and receives echo data from the surface of the melt.

[0016] The echo data is processed using the wavelet algorithm to obtain wavelet echo data.

[0017] The real-time value of the melt level was calculated using wavelet echo data.

[0018] The specific steps for processing echo data using wavelet algorithm in the IoT-based non-contact measurement and closed-loop control system for single-crystal furnace melt level provided by this invention are as follows:

[0019] Preprocessing of the echo data yields preprocessed echo data.

[0020] The number of wavelet decomposition layers N is determined based on the frequency range of the preprocessed echo data.

[0021] Using Symlets wavelet functions, discrete wavelet transforms were performed on most preprocessed echo data according to the number of wavelet decomposition levels, decomposing most preprocessed echo data into low-frequency approximation coefficients cA at different scales. j and high-frequency detail coefficients cD j .

[0022] Using high-frequency detail coefficients cD j Calculate the corresponding threshold T j And perform threshold processing.

[0023] The low-frequency approximation coefficients cA after thresholding are... N and high-frequency detail coefficients of each layer (cD) j As input, wavelet echo data x is obtained using inverse discrete wavelet transform. new(n).

[0024] According to the IoT-based non-contact measurement and closed-loop control system for single crystal furnace melt level provided by the present invention, threshold processing includes: when At that time, the high-frequency detail coefficients cD j Updated to ,when At the same time, the high-frequency detail coefficients cD j Set to 0.

[0025] According to the IoT-based non-contact measurement and closed-loop control system for single-crystal furnace melt level provided by the present invention, the formula for calculating the solution position is expressed as follows:

[0026] ;

[0027] In the formula, H eff For the effective installation height actually used to calculate the liquid level, c eff Let t be the actual effective propagation speed of radar waves inside the furnace, and t be the actual time difference.

[0028] According to the IoT-based non-contact measurement and closed-loop control system for single-crystal furnace melt level provided by the present invention, the calculation formula for correcting the melt level data in the melt level correction module is as follows:

[0029] ;

[0030] In the formula, Lcorrected is the corrected liquid level, Lraw is the original liquid level data, k is the temperature influence coefficient, and ΔT is the difference between the actual temperature and the standard temperature.

[0031] According to the IoT-based non-contact measurement and closed-loop control system for single crystal furnace melt level provided by the present invention, the parameters for adjusting the single crystal furnace parameters include: adjustment when the liquid level is higher than a threshold, and adjustment when the liquid level is lower than a preset value.

[0032] According to the IoT-based non-contact measurement and closed-loop control system for melt level in a single crystal furnace provided by this invention, adjusting the melt level when it exceeds a threshold includes: gradually increasing the casting speed; gradually decreasing the heating power; reducing the feeding speed by a certain proportion in the feeding control system; activating the crucible lowering function in the crucible lifting control system of the single crystal furnace; and adjusting the flow rate according to the gas type in the gas flow control system.

[0033] According to the IoT-based non-contact measurement and closed-loop control system for single-crystal furnace melt level provided by the present invention, the adjustment of the melt level below a threshold includes: increasing the heating power; increasing the feeding speed in the feeding control system; raising the height of the crucible in the control system; and gradually increasing the gas flow rate to adjust the melt level when the melt level is below the threshold.

[0034] The present invention provides an IoT-based non-contact measurement and closed-loop control system for the molten liquid level in a single crystal furnace. This system utilizes multi-directional radar installation, wavelet algorithm signal processing, and a liquid level correction model for non-contact measurement of the molten liquid level in the single crystal furnace. The beneficial effects achieved are as follows:

[0035] Installing radar equipment on different sides and at different locations on the top of the furnace body overcomes the signal blind zone problem that may exist in traditional single-angle measurements. Transmitting and receiving radar waves from multiple directions allows for the acquisition of more comprehensive information on the molten metal surface reflection, avoiding measurement errors caused by factors such as localized surface fluctuations, uneven furnace structure, or impurity distribution. This invention effectively reduces the impact of these interference factors on measurement accuracy and improves the accuracy of level measurement through multi-directional radar installation, wavelet algorithm signal processing, and a level correction model. The closed-loop control mechanism of this system enables automatic, real-time parameter adjustment, responding promptly based on accurate level measurement data, overcoming the shortcomings of manual control, and ensuring the timeliness and accuracy of level control.

[0036] Based on IoT technology, the system can collect data from various aspects of the single crystal furnace in real time and quickly compare the liquid level data obtained by the radar measurement module with the set target liquid level range to calculate the deviation value in real time. Once the deviation value exceeds the preset threshold, the closed-loop control module can automatically make a decision based on the pre-set control strategy without manual intervention, determining whether to raise or lower the liquid level, and which single crystal furnace parameters need to be adjusted and by what extent. This real-time automatic decision-making capability greatly shortens the time interval from detecting abnormal liquid level to taking measures, avoiding the problem of prolonged deviation of the liquid level from the normal range due to untimely observation or inaccurate judgment by humans, and ensuring the stability of the liquid level during the operation of the single crystal furnace. Traditional single crystal furnace liquid level control relies on frequent manual observation of the liquid level gauge reading and manual operation of various devices to adjust parameters. Operators are not only prone to visual fatigue and misjudgment, but differences in the experience and operating habits of different operators can also lead to inconsistent control effects. This system achieves automated control, eliminating the need for manual monitoring and adjustment of the liquid level. This significantly reduces the risk of errors caused by manual operation and also lowers the workload of operators, allowing them to focus more on monitoring the entire production process and managing other important aspects, thereby improving production efficiency and management level.

[0037] The closed-loop control module continuously compares the actual liquid level with the target liquid level, dynamically adjusts the single crystal furnace parameters based on the deviation, and then acquires new liquid level data for the next round of comparison and adjustment, forming a continuous closed-loop feedback mechanism. During production, regardless of changes in process parameters or external environmental factors, the system can rely on this closed-loop feedback mechanism to promptly detect changes in the liquid level and automatically adjust parameters to adapt to new conditions, ensuring that the liquid level remains stable within a reasonable target range. This adaptive capability allows the system to maintain good performance not only under routine production conditions but also under various complex and dynamically changing production conditions, improving the reliability and flexibility of the entire single crystal furnace production system and helping to ensure the consistency and stability of product quality. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0039] Figure 1 This is a schematic diagram of the module of the IoT-based non-contact measurement and closed-loop control system for melt level in a single crystal furnace provided in an embodiment of the present invention.

[0040] Figure 2 This is a flowchart illustrating the steps of measuring the liquid level of a single-crystal furnace melt using radar, as provided in an embodiment of the present invention.

[0041] Figure 3 This is a schematic diagram of the process of using wavelet algorithm to process echo data according to an embodiment of the present invention. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0043] The following is combined Figures 1-3 This invention describes an Internet of Things-based non-contact measurement and closed-loop control system for melt level in a single crystal furnace.

[0044] like Figures 1-3 As shown in the embodiment of the present invention, the non-contact measurement and closed-loop control system for melt level in a single crystal furnace based on the Internet of Things includes:

[0045] The data acquisition module is used to collect single crystal furnace data of the melt in the single crystal furnace.

[0046] The data for the single crystal furnace includes: environmental data, equipment status data, and control feedback data.

[0047] Environmental data includes: high-temperature environmental data inside the furnace and high-temperature environmental data outside the furnace. The high-temperature environmental data inside the furnace is acquired through high-temperature sensors installed inside the furnace to monitor the real-time temperature. The high-temperature environmental data outside the furnace is acquired through temperature sensors installed near the exterior of the furnace body to understand the potential impact of the external environment on the furnace temperature.

[0048] Equipment status data includes: single crystal furnace process parameter data, measurement and control system status data, and the measured time difference t. m The installation height deviation of the radar sensor relative to the nominal total height H of the bottom of the single crystal furnace is △H.

[0049] Equipment status data includes heating data such as heating power P, heating current I, and voltage V. These data are acquired in real time through current transformers and voltage sensors installed on the heating elements, reflecting the energy input status of the melt. Pulling data includes seed crystal pulling speed and rotation speed ω. These data are acquired through encoders or sensors of the pulling and rotating mechanisms, directly affecting the melt consumption rate and surface stability.

[0050] The single crystal furnace process parameters include heating data and crucible parameter data. The crucible parameter data includes the crucible rotation speed and crucible position. These data are acquired through encoders or displacement sensors of the crucible rotation mechanism and position sensors, and are used to analyze the coupling relationship between liquid level changes and crucible movement.

[0051] Heating data: Heating power and heating current / voltage reflect the energy input state of the melt and are directly related to changes in the liquid level. Pulling data: Seed crystal pulling speed and rotation speed affect the melt consumption rate and surface stability. Crucible parameter data: Crucible rotation speed and crucible position are used to correlate the liquid level changes with the crucible movement.

[0052] Control feedback data includes: liquid level deviation data, control effect data, and abnormal status data.

[0053] Liquid level deviation data: The deviation between the real-time liquid level and the target liquid level serves as the core input for closed-loop control, triggering adjustment actions such as power adjustment and lifting speed correction. The deviation change rate is the amount of change in deviation per unit time, used to determine the liquid level change trend and avoid over-adjustment.

[0054] Control effect data: The actual change in liquid level after the control command is executed. Control efficiency is the ratio of the reduction in deviation to the control amplitude, used to optimize control parameters.

[0055] Abnormal status data: Exceeding limits: liquid level fluctuation > 1mm / 2s, sensor communication interruption > 200ms, heating power exceeding the safe range, etc., triggering an alarm and recording an abnormal snapshot.

[0056] The specific steps for using radar to detect the melt level in a single crystal furnace are as follows:

[0057] Radar equipment is installed on different sides and top of the furnace body to ensure accurate detection of the molten metal level from multiple angles. The radar is undamaged and all interfaces are intact. The radar is securely installed in the predetermined position using appropriate mounting brackets, clamps, and other fixing devices to ensure that the radar does not loosen or shift during operation.

[0058] Turn on the radar power and initialize its settings. In the radar's operating interface, input the relevant basic parameters, such as current location coordinates, measurement range, and radar wave transmission frequency, to ensure the radar can start operating normally according to the set requirements.

[0059] The radar continuously emits radar waves at a set frequency and receives echoes from the surface of the melt. Based on principles such as the time difference between the radar wave's round trip, it automatically calculates the real-time melt level and transmits this value to a connected control system or data acquisition device. Operators can monitor the changes in the melt level in real time through the control system's display screen or the host computer software interface.

[0060] The echo data is processed using the wavelet algorithm to obtain wavelet echo data, including:

[0061] The echo signal is preprocessed to obtain preprocessed echo data. Preprocessing includes: removing noise spikes.

[0062] Noise spike removal is used to examine the acquired raw signal data, identify and remove obvious abnormal noise spikes caused by accidental factors. This can be achieved by setting a threshold, where a signal amplitude exceeding the normal range by a certain multiple is identified as a noise spike and removed.

[0063] Normalization: Normalize the amplitude of the signal to a specific interval, such as [0,1] or [-1,1], to facilitate subsequent wavelet transform calculations and comparative analysis between data of different magnitudes.

[0064] The number of wavelet decomposition levels N is determined based on the signal's frequency range and the level of detail required for analysis. First, observe the coefficients of each level after decomposition at the lowest level. If the low-frequency components still contain significant information requiring further subdivision, or if noise in the high-frequency components is not effectively separated, the number of decomposition levels is appropriately increased. Analysis of the signal spectrum and prior knowledge can help determine the appropriate number of decomposition levels.

[0065] Using Symlets wavelet functions, discrete wavelet transform is performed on the preprocessed original signal x(n) according to a determined number of decomposition levels, decomposing the signal into low-frequency approximation coefficients cA at different scales. j and high-frequency detail coefficients cD j , j=1,2,…,N, where j represents the decomposition level. In a specific embodiment, for the first decomposition, a low-frequency component, containing the main contour information cA1 of the signal, and a high-frequency component, containing details, abrupt changes, and other information of the signal cD1, are obtained.

[0066] For the high-frequency detail coefficients cD of each layer obtained by decomposition j The corresponding threshold T is calculated according to the selected threshold determination method. j Then, thresholding is performed. In a specific embodiment, in hard thresholding, when When, keep the value of this coefficient unchanged, when When the threshold value is 0, the coefficient value is set to 0. In soft thresholding, when... When, update the coefficient value to ,when Similarly, the coefficient values ​​are set to 0, k represents the coefficient index, and sgn is the sign function. Thresholding removes most of the noise components from the high-frequency detail coefficients, preserving the effective detail information of the signal.

[0067] High-frequency detail factor cD j The corresponding threshold T j The calculation formula is expressed as follows:

[0068] ;

[0069] The low-frequency approximation coefficients cA after thresholding are... N The high-frequency detail coefficients cDj (j=1,2,…,N) after each layer of processing are used as input. The inverse discrete wavelet transform, which corresponds to wavelet decomposition, is used to perform reconstruction operations in reverse order to gradually recover the processed signal x. new (n). The reconstruction process recombines the information contained in the coefficients of each layer to restore the filtered, enhanced and restored signal.

[0070] The reconstructed signal is evaluated by comparing it with the original signal, observing its time-domain waveform characteristics, and analyzing its spectral properties to determine if the processing effect meets expectations. If the effect is unsatisfactory, it may be necessary to readjust parameters such as the wavelet basis function, the number of decomposition levels, and the thresholding method, and perform wavelet transform processing again until satisfactory filtering, enhancement, and restoration effects are obtained.

[0071] The real-time value of the melt level was calculated using wavelet echo data.

[0072] The formula for calculating the solution level is expressed as:

[0073] ;

[0074] In the formula, H eff For the effective installation height actually used to calculate the liquid level, c eff Let t be the actual effective propagation speed of radar waves inside the furnace, and t be the actual time difference.

[0075] ;

[0076] In the formula, H is the nominal total height of the radar sensor installation position relative to the bottom of the single crystal furnace, and △H is the installation height deviation.

[0077] The actual effective propagation speed of radar waves inside the furnace, c eff It can be represented as:

[0078] ;

[0079] In the formula, c is the ideal speed of light, and k v This is the speed correction factor caused by the furnace environment.

[0080] The formula for calculating the true time difference is as follows:

[0081] ;

[0082] In the formula, t m For the measured time difference, k t This is the error correction factor.

[0083] The liquid level correction module is used to build a liquid level correction model. The initial liquid level is input into the correction model to obtain the corrected liquid level. The correction formula is expressed as: Lcorrected=Lraw×(1+k×ΔT).

[0084] Where Lcorrected is the corrected liquid level, Lraw is the original liquid level data, k is the temperature influence coefficient, and ΔT is the difference between the actual temperature and the standard temperature.

[0085] The closed-loop control module sets the target liquid level range, calculates the deviation value ΔL between the corrected liquid level data and the target liquid level in real time, and determines whether the deviation value exceeds the preset threshold. If so, it adjusts the parameters of the single crystal furnace.

[0086] When the melt level is too high, closely observe the extent of the excess in the single crystal furnace, and comprehensively determine the appropriate pulling speed adjustment range by considering the current crystal growth stage, diameter, and other relevant parameters. Generally, the melt level value is accurately read using the melt level monitoring device on the furnace body and compared with the preset standard melt level value to determine the specific amount of excess. For example, if the melt level is 5 mm higher than the standard melt level, and the crystal is in the constant diameter growth stage with a stable diameter, the next pulling speed adjustment operation can be considered.

[0087] Pulling speed adjustment: Following the equipment operating procedures, gradually increase the pulling speed in the control system, with each adjustment recommended to be controlled within 0.1-0.5 mm / min. In a specific embodiment, if the initial pulling speed is 1 mm / min, it can be increased to 1.2 mm / min first, and then the liquid level change can be continuously observed. Usually, within a few minutes after adjusting the pulling speed, the liquid level will begin to decrease, because as the pulling speed increases, more material is pulled out of the melt to form crystals, and the total amount of melt will decrease accordingly, thus causing the liquid level to drop. After adjusting the pulling speed, check the liquid level value, crystal growth diameter, appearance quality, etc., every 1-2 minutes. If the liquid level drops too quickly and approaches the standard liquid level, the pulling speed should be adjusted appropriately to avoid the liquid level becoming too low. If the drop rate is too slow, the pulling speed can be increased slightly again until the liquid level stabilizes and returns to the standard liquid level. At the same time, it is necessary to ensure that no quality problems such as broken edges or dislocations occur during crystal growth. If any abnormalities occur, the pulling speed and other parameters need to be adjusted in a coordinated manner.

[0088] Adjusting the heating power: Check the current heating power setting of the single crystal furnace, and determine the power adjustment range by combining factors such as melt level, temperature monitoring data, and the amount of raw material remaining to be melted. If the melt level is too high and the furnace temperature is also slightly higher than the upper limit of the normal growth temperature range, and there is still a lot of unmelted raw material, it indicates that the heating power is too high.

[0089] Based on the equipment's power adjustment function, the heating power is gradually reduced, typically by 1%-5% of the total power each time. In a specific embodiment, for a single crystal furnace with a total heating power of 100 kW, the power can be reduced by 3 kW initially. During operation, precise settings are made on the control cabinet according to the corresponding operation interface prompts. As the heating power decreases, the melting rate of the raw materials slows down. Due to the reduced heat supply, losses such as volatilization in the melt are relatively reduced, and the crystal growth process continuously consumes melt, causing the melt level to gradually decrease.

[0090] Adjusting the raw material feeding rate: In the feeding control system, reduce the feeding rate by a certain percentage. The reduction amount can be set according to the actual feeding capacity and process requirements. For example, if the initial feeding rate is 5 grams / minute, it can be reduced to 3 grams / minute first. During operation, simply input the new feeding rate parameter in the corresponding operating interface. If you decide to pause feeding, simply click the pause feeding button in the control system to stop the feeding device from conveying raw materials into the furnace. In this way, as the existing crystal growth is consumed, the melt level will gradually drop back to the standard state.

[0091] Adjusting the crucible's lifting or lowering: In the crucible lifting and lowering control system of the single crystal furnace, carefully activate the crucible lowering function according to the prescribed operating procedures. The distance of each descent should be 1-5 mm. In a specific embodiment, first lower the crucible by 3 mm, so that the relative position of the melt inside the crucible moves downward, which is equivalent to artificially lowering the visual height of the liquid level. This also helps the melt to be more evenly distributed inside the crucible, which is beneficial for the consumption of melt during subsequent crystal growth, thereby gradually bringing the melt level closer to the standard liquid level.

[0092] Adjusting gas flow rate: In the gas flow control system, the flow rate is adjusted according to the type of gas, generally by reducing the flow rate by 10%-30% each time. In a specific embodiment, for argon gas with an initial flow rate of 10 liters / minute, it can be reduced to 7 liters / minute by adjusting the opening of the corresponding gas valve or by inputting new flow parameters in the flow control interface. As the gas flow rate decreases, the evaporation of the melt surface will be improved to some extent, and the melt level will gradually decrease due to the consumption of crystal growth, etc.

[0093] When the liquid level is low, carefully reduce the pulling speed in the control system, with each reduction being 0.05-0.2 mm / min. For example, if the original pulling speed is 1.2 mm / min, it can be reduced to 1.1 mm / min first, and then the liquid level change can be observed. As the pulling speed decreases, the crystal growth rate slows down, and the melt consumption slows down. With the subsequent melting and replenishment of raw materials, the melt level will gradually rise.

[0094] Adjusting the heating power: Within the safe range allowed by the equipment, gradually increase the heating power, with each increase recommended to be controlled at around 2%-6%. For example, for an 80 kW single crystal furnace, you can first increase the power by 4 kW, precisely adjusting the input power value through the control cabinet. As the heating power increases, the raw materials will melt faster and replenish the melt, thereby increasing the melt level.

[0095] Adjusting the raw material feeding speed: Increase the feeding speed in the feeding control system. The amount of increase each time can be determined according to the feeding accuracy of the equipment and actual needs, generally 1-3 grams per minute each time. For example, if the original feeding speed is 2 grams per minute, it can be increased to 4 grams per minute. After accurately inputting the new feeding speed value through the operation interface, the feeding device will deliver raw materials into the furnace at the new speed. More raw materials enter the furnace and melt in time, replenishing the melt and causing the liquid level to rise.

[0096] Adjusting the crucible lifting: Operate the crucible lifting in the control system. Each lifting height is controlled within the range of 1-3 mm. For example, first lift the crucible by 2 mm to make the melt move relatively upward in the crucible, which increases the amount of melt in the upper part. Visually, the liquid level will rise, and it can improve the conditions for the melt to supply the growing crystal, causing the liquid level to approach the standard liquid level.

[0097] Adjusting the gas flow rate: In the gas flow control system, gradually increase the gas flow rate, with each increase ideally between 15% and 40%. For example, if the initial flow rate of a reactive gas is 5 liters per minute, it can be increased to 7 liters per minute by precisely adjusting the gas valve or inputting the new flow rate value into the flow control interface. As the gas flow rate increases, the melting conditions of the raw materials may improve, the melt distribution may become more reasonable, and the liquid level will gradually rise due to the replenishment of raw materials.

[0098] In summary, this embodiment provides an IoT-based non-contact measurement and closed-loop control system for the melt level of a single-crystal furnace. It achieves non-contact measurement of the melt level through multi-directional radar installation, wavelet algorithm signal processing, and a level correction model. The beneficial effects obtained are as follows:

[0099] Based on IoT technology, the system can collect data from various aspects of the single crystal furnace in real time and quickly compare the liquid level data obtained by the radar measurement module with the set target liquid level range to calculate the deviation value in real time. Once the deviation value exceeds the preset threshold, the closed-loop control module can automatically make a decision based on the pre-set control strategy without manual intervention, determining whether to raise or lower the liquid level, and which single crystal furnace parameters need to be adjusted and by what extent. This real-time automatic decision-making capability greatly shortens the time interval from detecting abnormal liquid level to taking measures, avoiding the problem of prolonged deviation of the liquid level from the normal range due to untimely observation or inaccurate judgment by humans, and ensuring the stability of the liquid level during the operation of the single crystal furnace. Traditional single crystal furnace liquid level control relies on frequent manual observation of the liquid level gauge reading and manual operation of various devices to adjust parameters. Operators are not only prone to visual fatigue and misjudgment, but differences in the experience and operating habits of different operators can also lead to inconsistent control effects. This system achieves automated control, eliminating the need for manual monitoring and adjustment of the liquid level. This significantly reduces the risk of errors caused by manual operation and also lowers the workload of operators, allowing them to focus more on monitoring the entire production process and managing other important aspects, thereby improving production efficiency and management level.

[0100] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0101] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. These modifications or substitutions do not cause the essence of the corresponding technical solutions to depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A non-contact measurement and closed-loop control system for melt level in a single crystal furnace based on the Internet of Things, characterized in that, include: The data acquisition module is used to collect single crystal furnace data of the melt in the single crystal furnace. The single crystal furnace data includes: environmental data, equipment status data, and control feedback data. The radar measurement module is used to combine the data from the single crystal furnace and use radar to measure the liquid level of the melt in the single crystal furnace to obtain the initial liquid level. Radar equipment was installed on different sides and at different locations on the top of the furnace body. Turn on the radar power and initialize the radar settings; In the radar's operating interface, input the current location coordinates, measurement range, and radar wave transmission frequency; The radar continuously transmits radar waves at a set frequency and receives echo data from the surface of the melt. The echo data is processed using a wavelet algorithm to obtain wavelet echo data; The echo data is preprocessed to obtain preprocessed echo data; The number of wavelet decomposition layers N is determined based on the frequency range of the preprocessed echo data; Using Symlets wavelet functions, discrete wavelet transforms were performed on most preprocessed echo data according to the number of wavelet decomposition levels, decomposing most preprocessed echo data into low-frequency approximation coefficients cA at different scales. j and high-frequency detail coefficients cD j ; Using high-frequency detail coefficients cD j Calculate the corresponding threshold T j And perform threshold processing; The low-frequency approximation coefficients cA after thresholding are... N and high-frequency detail coefficients of each layer (cD) j As input, wavelet echo data x is obtained using inverse discrete wavelet transform. new (n); The real-time value of the melt level is calculated using the wavelet echo data. The liquid level correction module is used to construct a liquid level correction model. The initial liquid level is input into the correction model to obtain the corrected liquid level. The closed-loop control module sets the target liquid level range, calculates the deviation value ΔL between the corrected liquid level data and the target liquid level in real time, and determines whether the deviation value exceeds the preset threshold. If so, the parameters of the single crystal furnace are adjusted.

2. The IoT-based non-contact measurement and closed-loop control system for single-crystal furnace melt level according to claim 1, characterized in that, The environmental data includes high-temperature environmental data inside the furnace and high-temperature environmental data outside the furnace; the equipment status data includes single crystal furnace process parameter data and measurement and control system status data; the single crystal furnace process parameter data includes heating data and crucible parameter data.

3. The IoT-based non-contact measurement and closed-loop control system for single-crystal furnace melt level according to claim 1, characterized in that, The threshold processing includes: when At that time, the high-frequency detail coefficients cD j Updated to ,when At the same time, the high-frequency detail coefficients cD j Set to 0.

4. The IoT-based non-contact measurement and closed-loop control system for single-crystal furnace melt level according to claim 1, characterized in that, The formula for calculating the solution position is expressed as: In the formula, H eff For the effective installation height actually used to calculate the liquid level, c eff Let t be the actual effective propagation speed of radar waves inside the furnace, and t be the actual time difference.

5. The IoT-based non-contact measurement and closed-loop control system for single-crystal furnace melt level according to claim 1, characterized in that, In the liquid level correction module, the formula for calculating the corrected liquid level data is: In the formula, Lcorrected is the corrected liquid level, L raw The original liquid level data is given, k is the temperature influence coefficient, and ΔT is the difference between the actual temperature and the standard temperature.

6. The IoT-based non-contact measurement and closed-loop control system for single-crystal furnace melt level according to claim 1, characterized in that, The parameters that can be adjusted for the single crystal furnace include: adjusting the liquid level when it is above a threshold and adjusting the liquid level when it is below a preset value.

7. The IoT-based non-contact measurement and closed-loop control system for single-crystal furnace melt level according to claim 6, characterized in that, Adjustments for liquid levels exceeding a threshold include: gradually increasing the pulling speed; gradually decreasing the heating power; reducing the feeding speed by a certain proportion in the feeding control system; activating the crucible lowering function in the crucible lifting control system of the single crystal furnace; and adjusting the flow rate according to the gas type in the gas flow control system.

8. The IoT-based non-contact measurement and closed-loop control system for melt level in a single crystal furnace according to claim 6, characterized in that, Adjustments for liquid levels below the threshold include: increasing heating power; increasing the feeding speed in the feeding control system; raising the height of the crucible in the control system; and gradually increasing the gas flow rate to adjust the liquid level when the liquid level is below the threshold.

Images