An equal initial power recommendation method, device and storage medium

Abstract

The application provides an equal-diameter initial power recommendation method, device and storage medium, the equal-diameter initial power recommendation method comprises the following steps: screening reference bar in the history bar of the current heat; when there is reference bar, the target pulling speed curve is constructed; the average pulling speed deviation of the equal-diameter actual average pulling speed curve determined by the equal-diameter actual average pulling speed of the reference bar in the specified single crystal length interval of the equal-diameter head and the target pulling speed curve is calculated, and the first type main power correction value is determined according to the average pulling speed deviation; the equal-diameter initial power of the current bar is calculated based on the equal-diameter initial power of the reference bar and the first type main power correction value; when there is no reference bar, the equal-diameter initial power of the current bar is calculated based on the seeding power of the current bar, the total drop of the default shoulder power and the artificial adjustment power drop factor of the current bar shoulder stage. The application realizes the intelligent recommendation of the equal-diameter initial power of the current bar.

Description

Technical Field

[0001] This invention relates to the field of Czochralski method for preparing single-crystal silicon, and in particular to a method, apparatus and storage medium for recommending initial power of constant diameter. Background Technology

[0002] Czochralski (CZ) single-crystal silicon growth is one of the most important manufacturing processes in the production of photovoltaic and semiconductor silicon materials. During single-crystal silicon growth, the constant-diameter stage is a critical step that determines product quality, production stability, and yield per unit volume. The constant-diameter stage requires precise adjustment of the main heating power, pulling speed, and thermal field conditions to maintain a stable crystal diameter, thereby obtaining a single-crystal product that meets the process requirements.

[0003] In the Czochralski process of growing single-crystal silicon, the equal-diameter stage is a critical step that determines crystal quality, production stability, and yield. Setting the initial equal-diameter power too high can easily lead to reduced pulling speed, low yield, and excessively hot thermal field; conversely, setting it too low can easily lead to excessively high pulling speed, single-crystal instability, or even wire breakage. Therefore, the appropriate setting of the initial equal-diameter power directly affects the growth stability and crystallization rate in the early stages of equal-diameter growth. How to accurately recommend the initial equal-diameter power based on the current process conditions, reduce reliance on manual experience, and improve the control stability of the equal-diameter head stage has become an important technical problem in the field of automatic control for single-crystal silicon.

[0004] Currently, the industry mainly uses the following methods to set the initial power for equal diameter.

[0005] The first method is the manual experience-based setting method.

[0006] Operators manually input the initial power parameters for the equal-diameter head based on product specifications, remaining material quantity, furnace status, and historical production experience. This method is simple to implement and widely applicable, but it heavily relies on the operator's experience level. Significant differences exist between operators, and even the same operator may exhibit judgment biases at different production stages, resulting in poor stability of the equal-diameter head and making standardized control difficult.

[0007] The second method is the fixed formula setting method.

[0008] Standard process formulas are pre-defined for specific product specifications, and preset power values ​​are directly applied when entering the equal diameter stage. This method can reduce the degree of manual intervention, but because it cannot reflect factors such as changes in residual material, thermal field conditions, and individual furnace differences during actual production, the recommended power often does not match the actual operating conditions; this can easily lead to unreasonable power settings, thus affecting the stability of the equal diameter head.

[0009] In summary, the existing technology has at least the following shortcomings: it relies too much on human experience, has a low degree of automation, the fixed process formula cannot adapt to changes in real-time production status, it only uses historical power values ​​for reference, and lacks quantitative analysis of pulling speed deviation.

[0010] Furthermore, during continuous production within the same furnace cycle, the thermal state inside the furnace changes constantly as the furnace running time increases. Even with the same process conditions, the drawing speed of the equal-diameter head of different bars may still vary significantly. Existing technologies typically only consider the influence of some process parameters and fail to fully account for the impact of changes in thermal insulation capacity on power requirements, which can easily lead to overly high recommended power. Summary of the Invention

[0011] To address at least one technical problem in the prior art, embodiments of the present invention provide a method, device, and storage medium for recommending initial power for equal-diameter bars. By selecting reference bars, constructing a target casting speed curve, analyzing the average casting speed deviation between the actual average casting speed curve of the reference bars and the target casting speed curve, a power compensation and correction mechanism is established. This is combined with the furnace's cumulative operating time, the equal-diameter length of the reference bars, and process parameter data from the current bar's shoulder-forming stage for comprehensive correction, thereby achieving intelligent recommendation of the initial power for the current bar's equal-diameter bars. This effectively reduces reliance on manual experience, improves the stability of thermal field control during the equal-diameter head stage, and reduces wire breakage caused by casting speed fluctuations. To achieve the above technical objectives, the technical solution adopted in embodiments of the present invention is as follows: In a first aspect, embodiments of the present invention provide a method for recommending initial power of constant diameter, comprising: Obtain the first type of process parameter data of historical and current rods during the current single crystal production process; based on the comparison of the remaining material in the first type of process parameter data of the current rod and historical rods in the current furnace, select reference rods from the historical rods of the current furnace. When a reference batch exists, acquire the second type of process parameter data of multiple historical furnaces and multiple historical bars in each historical furnace during the historical production process of the same product's sample single crystal, and construct a target pulling speed curve accordingly; calculate the average pulling speed deviation between the actual average pulling speed curve of the reference batch within a specified single crystal length range at the equal diameter head and the target pulling speed curve, and determine the first type of main applied power correction value based on the average pulling speed deviation; calculate the initial equal diameter power of the current batch based at least on the initial equal diameter power of the reference batch and the first type of main applied power correction value; When there is no reference order, the initial equal-diameter power of the current order is calculated based on the current order's lead-in power, the default total power reduction during shoulder formation, and the manually adjusted power reduction factor during the current order's shoulder formation stage.

[0012] Further, the first type of process parameter data includes furnace batch, bar batch, remaining material amount, actual average pulling speed for equal diameter, set average pulling speed for equal diameter, initial power for equal diameter, and crystal pulling power; the step of comparing the remaining material amount in the first type of process parameter data of the current bar batch and historical bar batches in the current furnace batch, and selecting reference bar batches from the historical bar batches of the current furnace batch, includes: The first type of process parameter data of the historical rod batches in the current single crystal production process are sorted according to the temperature adjustment end time to construct the production sequence of the current batch. The remaining material quantity of the current batch of bars in the current furnace is compared with the remaining material quantity of each historical batch to obtain the remaining material quantity deviation; If there are historical batches in the current furnace production sequence that meet the condition that the remaining material deviation is less than or equal to the set material deviation threshold, the historical batch that is closest in time to the current batch and has entered the equal diameter stage is selected as the reference batch; if there are no batches in the current furnace production sequence that meet the condition that the remaining material deviation is less than or equal to the set material deviation threshold, then it is determined that there are no reference batches.

[0013] Further, the second type of process parameter data includes: single crystal length, actual average pulling speed for constant diameter; the step of obtaining the second type of process parameter data from multiple historical furnaces and multiple historical bars in each historical furnace during the historical production process of sample single crystals of the same product, and constructing a target pulling speed curve accordingly, includes: Data cleaning was performed on the second type of process parameter data for each historical bar in each historical furnace of the sample single crystal, and the sample single crystals whose single crystal length was greater than the specified single crystal length threshold of the equal diameter head after the equal diameter stage were retained as valid sample single crystals. Align the effective sample single crystals of the same process according to the single crystal length of the equal diameter stage; for each single crystal length position, calculate the average of the actual average pulling speed of all effective sample single crystals at each single crystal length position to generate the target pulling speed trajectory; wherein, the single crystal length in the target pulling speed trajectory is less than or equal to the single crystal length threshold specified at the equal diameter head. The target pulling speed trajectory is spliced ​​with the equal diameter set average pulling speed trajectory determined by the current bar's equal diameter set average pulling speed to obtain the target pulling speed curve, and Gaussian smoothing algorithm is used to smooth the splicing point; wherein, the single crystal length in the equal diameter set average pulling speed trajectory is greater than the single crystal length threshold specified by the equal diameter head.

[0014] Furthermore, the specified single crystal length threshold for the equal-diameter head is 900mm to 1100mm.

[0015] Further, the calculation of the average pulling speed deviation between the actual average pulling speed curve of the reference bar within the specified single crystal length range of the equal diameter head and the target pulling speed curve, and the determination of the first type of main applied power correction value based on the average pulling speed deviation, includes: Obtain the actual average pulling speed of reference bars within the specified single crystal length range of the equal diameter head and construct the actual average pulling speed curve of the equal diameter head; the upper limit of the specified single crystal length range of the equal diameter head is greater than the specified single crystal length threshold of the equal diameter head. Calculate the average pulling rate deviation between the actual average pulling rate curve and the target pulling rate curve for reference bars within a specified single crystal length range at the constant diameter head. ; The average pulling speed deviation With the set speed fluctuation range Compare; If the average pulling speed deviation Exceeding the upper limit of the speed fluctuation range This indicates that the initial power of the reference rod with equal diameter is too low. Therefore, the first type of main applied power correction value should be set to the same level as... Positive values ​​with linear correlation are applied, and power adjustment and limiting are performed. If the average pulling speed deviation Below the lower limit of the pulling speed fluctuation range This indicates that the initial power of the reference rod with equal diameter is too high. Therefore, the first type of main applied power correction value should be set to the same level as... Negative values ​​of linear correlation are applied, and power adjustment limiting is performed. If the average pulling speed deviation Within the range of speed fluctuation If the value is within the range, it indicates that the initial power of the reference rod is appropriate, and the first type of main power correction value is set to zero.

[0016] Furthermore, the upper limit of the specified single crystal length range for the equal diameter head is 1500mm to 1700mm; The set pulling speed fluctuation range is [-2kw, 2kw].

[0017] Furthermore, the method for recommending the initial equal-diameter power also includes: when a reference bar number exists, the calculation of the initial equal-diameter power of the current bar number based at least on the initial equal-diameter power of the reference bar number and the first type of main applied power correction value further includes: When the cumulative running time of the furnace platform in the current furnace batch exceeds the set running time threshold, the second type of main power correction value is calculated by multiplying the constant diameter length of the reference rod batch and the negative power correction coefficient; the constant diameter initial power of the current rod batch is calculated based on the constant diameter initial power of the reference rod batch, the first type of main power correction value, and the second type of main power correction value.

[0018] Furthermore, the power reduction factor manually adjusted during the current bar-shoulder-setting stage is obtained through the following method: Obtain the power reduction of several manual adjustments during the current bar shoulder formation stage; sum the power reduction of several manual adjustments and multiply it by a shoulder formation power compensation coefficient to obtain the power reduction factor of the current bar shoulder formation stage; and limit the power reduction factor of the manual adjustments.

[0019] In a second aspect, embodiments of the present invention provide an electronic device, comprising: Memory, which stores computer programs; A processor is configured to run the computer program, which, when executed, performs the steps of the constant-diameter initial power recommendation method as described above.

[0020] Thirdly, embodiments of the present invention provide a storage medium storing a computer program configured to execute the steps of the equal-diameter initial power recommendation method described above.

[0021] The beneficial effects of the technical solution provided by the embodiments of the present invention are as follows: 1) A reference bar selection mechanism based on the amount of remaining material; This invention utilizes the remaining material quantity as a reference for selecting rod batches. By comparing the remaining material quantity of the current rod batch with that of historical rod batches in the current furnace, reference rod batches are selected from the historical rod batches of the current furnace. Since the remaining material quantity directly reflects the amount of melt and the thermal state in the crucible, compared to simply using the rod batch order or time order for reference rod selection, this invention can more accurately select historical rod batches that match the current production state, thereby improving the accuracy of power recommendation.

[0022] 2) A dual-mode recommendation mechanism based on the existence of a reference bar; When a reference bar is available, power correction is performed based on the average pulling speed deviation between the actual average pulling speed curve of the reference bar within the specified single crystal length range of the equal diameter head and the target pulling speed curve. When there is no reference order, the initial equal-diameter power of the current order is calculated based on the current order's lead-in power, the default total power reduction during shoulder formation, and the manually adjusted power reduction factor during the current order's shoulder formation stage.

[0023] 3) Construct the target pulling speed trajectory based on historical valid sample single crystals; The effective sample single crystals are aligned according to the single crystal length of the equal diameter stage; for each single crystal length position, the average value of the actual average pulling speed of all effective sample single crystals at each single crystal length position is calculated to generate the target pulling speed trajectory; the target pulling speed trajectory can reflect the ideal pulling speed change law under stable production conditions and can be used as an important reference benchmark for evaluating the current production status.

[0024] 4) In the process of constructing the target pulling speed curve, the target pulling speed trajectory is spliced ​​with the average pulling speed trajectory determined by the average pulling speed of the current bar to obtain the target pulling speed curve, and the splicing point is processed by Gaussian smoothing; to eliminate the abrupt phenomenon at the trajectory splicing point, so that the target pulling speed curve is more in line with the continuous change characteristics of the actual production process.

[0025] 5) Power correction mechanism based on pulling speed deviation By calculating the average pulling speed deviation between the actual average pulling speed curve of the reference bar and the target pulling speed curve within the specified single crystal length range of the equal diameter head, it is determined whether the initial power of the reference bar is too high or too low; and the first type of main power correction value is automatically calculated based on the average pulling speed deviation; and a limiting strategy is adopted to prevent over-adjustment.

[0026] 6) Insulation compensation mechanism based on cumulative furnace operating time and reference bar diameter length This invention takes into account the issue of continuously increasing thermal insulation capacity during the continuous operation of a single crystal furnace, and introduces a second type of main power correction value. When the cumulative operating time of the furnace exceeds the set operating time threshold, the second type of main power correction value is calculated by multiplying the equal diameter length of the reference bar and the negative power correction coefficient to correct the recommended power. This can take into account the problem of thermal field overheating caused by heat accumulation during continuous production, and improve the long-term stability of the recommended results.

[0027] In summary, the embodiments of the present invention integrate the selection of reference bars, construction of target pulling speed trajectory, Gaussian smoothing when splicing to form target pulling speed curve, average pulling speed deviation analysis, automatic calculation of the first type of main power correction value, thermal insulation compensation and power limiting control, forming a complete equal-diameter initial power recommendation process, realizing intelligent optimization control of the equal-diameter initial power of the current bar. Attached Figure Description

[0028] Figure 1 This is a flowchart of the equal-diameter initial power recommendation method in an embodiment of the present invention.

[0029] Figure 2 This is a flowchart illustrating the process of selecting reference bars from the historical bars of the current furnace in an embodiment of the present invention.

[0030] Figure 3This is a flowchart illustrating the construction of the target pulling speed curve in an embodiment of the present invention.

[0031] Figure 4 This is a flowchart illustrating the calculation of the average pulling speed deviation and the determination of the first type of main power correction value based on the average pulling speed deviation in this embodiment of the invention.

[0032] Figure 5 This is a schematic diagram of an electronic device in an embodiment of the present invention. Detailed Implementation

[0033] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0034] In the description of the embodiments of the present invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the present invention. In addition, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0035] In the description of the embodiments of the present invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can also refer to the internal connection of two components; and they can refer to a wireless connection or a wired connection. Those skilled in the art can understand the specific meaning of the above terms in the present invention based on the specific circumstances.

[0036] Furthermore, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0037] The current Czochralski method for preparing single-crystal silicon includes stages such as temperature control, crystal pulling, shoulder formation, shoulder rotation, diameter equalization, and finishing. In this embodiment of the invention, the single-crystal silicon (single crystal rod) produced by the Czochralski method is simply referred to as single crystal. Each batch typically produces multiple batches (e.g., 10-20 batches) of single crystals. Each batch involves stages such as temperature control, crystal pulling, shoulder formation, shoulder rotation, diameter equalization, and finishing. The power in the following embodiments refers to the main applied power in the Czochralski method, i.e., the heating power used to heat the polycrystalline silicon raw material in the crucible. Polycrystalline silicon raw material needs to be added back to the crucible before each batch begins. The remaining material amount is the weight of the polycrystalline silicon raw material in the crucible at the beginning of a batch.

[0038] like Figure 1 As shown in the figure, an embodiment of the present invention proposes a method for recommending initial power of constant diameter, including: Step S10: Obtain the first type of process parameter data of historical and current single crystal production processes in the current furnace batch; based on the comparison of the remaining material in the first type of process parameter data of the current and historical single crystal production processes in the current furnace batch, select reference single crystals from the historical single crystal production processes of the current furnace batch.

[0039] The first type of process parameter data includes furnace batch, bar batch, remaining material amount, actual average pulling speed for equal diameter, set average pulling speed for equal diameter, initial power for equal diameter, and crystal pulling power; step S10 compares the remaining material amount in the first type of process parameter data of the current bar batch and historical bar batches in the current furnace batch, and selects reference bar batches from the historical bar batches of the current furnace batch, such as... Figure 2 As shown, it specifically includes: Step S101: Sort the first type of process parameter data of the historical rod batches in the current single crystal production process according to the temperature adjustment end time to construct the production sequence of the current furnace batch.

[0040] Step S102: Compare the remaining material quantity of the current batch of bars in the current furnace with the remaining material quantity of each historical batch to obtain the remaining material quantity deviation; For example, the deviation of remaining material quantity is defined as: ; in This represents the remaining material quantity for the current bar order; This refers to the remaining material quantity from historical batches.

[0041] Step S103: If there is a historical bar in the current furnace production sequence that satisfies the remaining material deviation being less than or equal to the set material deviation threshold, select the historical bar that is closest in time to the current bar and has entered the equal diameter stage as a reference bar; if there is no bar in the current furnace production sequence that satisfies the remaining material deviation being less than or equal to the set material deviation threshold, then it is determined that there is no reference bar.

[0042] For example, when the deviation of the remaining material quantity meets the following: hour, To set a material quantity deviation threshold, it is necessary to indicate that the thermal environment of these historical bars is similar to that of the current bar. Historical bars that are closest in time to the current bar and have entered the constant diameter stage are selected as reference bars.

[0043] Step S20: When a reference bar exists, acquire the second type of process parameter data of multiple historical furnaces and multiple historical bars in each historical furnace during the historical production process of the same product sample single crystal, and construct a target pulling speed curve accordingly; calculate the average pulling speed deviation between the actual average pulling speed curve of the reference bar within the specified single crystal length range of the equal diameter head and the target pulling speed curve, and determine the first type of main applied power correction value based on the average pulling speed deviation; calculate the initial equal diameter power of the current bar based at least on the initial equal diameter power of the reference bar and the first type of main applied power correction value.

[0044] Step S20 specifically includes the following steps: Step S201: Obtain the second type of process parameter data for multiple historical furnace batches and multiple historical bar batches within each historical furnace batch of the same product's sample single crystal production process, and construct the target pulling speed curve accordingly, such as... Figure 3 As shown, it specifically includes: The second category of process parameter data includes: single crystal length and actual average pulling speed for constant diameter crystals.

[0045] Step S2011: Perform data cleaning on the second type of process parameter data for each historical bar in each historical furnace batch of the sample single crystal, and retain the sample single crystals whose single crystal length is greater than the specified single crystal length threshold of the equal diameter head after the equal diameter stage as valid sample single crystals; data cleaning includes outlier removal, missing value processing, etc.; the specified single crystal length threshold of the equal diameter head can usually be 900mm to 1100mm, preferably, for example, 1000mm.

[0046] Step S2012: Align the effective sample single crystals of the same process according to the single crystal length of the equal diameter stage; for each single crystal length position, calculate the average of the actual average pulling speed of all effective sample single crystals at each single crystal length position to generate the target pulling speed trajectory; wherein, the single crystal length in the target pulling speed trajectory is less than or equal to the single crystal length threshold specified at the equal diameter head; for example, the target pulling speed trajectory can be represented as: ; in, This indicates the position of the kth effective sample single crystal within the single crystal length. The actual average pulling speed at the constant diameter point; N is the number of effective sample single crystals (i.e., the number of effective historical bar passes). This indicates the position of all valid sample single crystals along the single crystal length. The average actual average tension rate at the constant diameter location, where, ; This indicates the threshold value for the single crystal length of the equal-diameter head.

[0047] For example, when the threshold value for the single crystal length of a constant-diameter head is 1000 mm, a single crystal length position can be taken every 1 mm. Calculations are performed to obtain the target pulling speed trajectory.

[0048] Step S2013: The target pulling speed trajectory is spliced ​​with the equal diameter set average pulling speed trajectory determined by the equal diameter set average pulling speed of the current bar to obtain the target pulling speed curve, and Gaussian smoothing algorithm is used to smooth the splicing point to eliminate the abrupt change phenomenon at the splicing point of the target pulling speed trajectory and the equal diameter set average pulling speed trajectory of the current bar; wherein, the single crystal length in the equal diameter set average pulling speed trajectory is greater than the single crystal length threshold specified by the equal diameter head.

[0049] The current bar's constant diameter average pulling speed is set according to the single crystal length, meaning that each single crystal length position corresponds to a corresponding constant diameter average pulling speed, thus forming the constant diameter average pulling speed trajectory; the target pulling speed trajectory ( The trajectory of the average pulling speed set by the current bar diameter setting () By splicing the curves together, the target pulling speed curve can be obtained.

[0050] Step S202: Calculate the average pulling speed deviation between the actual average pulling speed curve determined by the reference rod's actual average pulling speed within the specified single crystal length range at the equal diameter head and the target pulling speed curve, and determine the first type of main applied power correction value based on the average pulling speed deviation; such as... Figure 4 As shown, it specifically includes: Step S2021: Obtain the actual average pulling speed of the reference rod within the specified single crystal length range of the equal diameter head and construct the actual average pulling speed curve of the equal diameter head; the upper limit of the specified single crystal length range of the equal diameter head is greater than the specified single crystal length threshold of the equal diameter head.

[0051] The actual average pulling speed of the reference bar is obtained according to the single crystal length, that is, each single crystal length position corresponds to a corresponding actual average pulling speed of the reference bar.

[0052] For example, the upper limit of the single crystal length range specified for the equal diameter head is 1500mm to 1700mm, preferably, for example, 1600mm.

[0053] Step S2022: Calculate the average pulling rate deviation between the actual average pulling rate curve of the reference rod and the target pulling rate curve within the specified single crystal length range of the equal-diameter head. ; In this step, for example, within the specified single crystal length range [0, 1600 mm] at the constant diameter head, at each single crystal length position... The difference between the actual average pulling speed curve of the reference bar and the target pulling speed curve is calculated, and the average of all pulling speed differences is taken as the average pulling speed deviation. .

[0054] Step S2023, the average tension deviation With the set speed fluctuation range Compare; If the average pulling speed deviation Exceeding the upper limit of the speed fluctuation range This indicates that the initial power of the reference rod with equal diameter is too low. Therefore, the first type of main applied power correction value should be set to the same level as... Positive values ​​with linear correlation are applied, and power adjustment and limiting are performed; specifically including: when > ,but ; in, This is the first type of main applied power correction value. This is the power correction factor. This is the power adjustment limiting factor; If the average pulling speed deviation Below the range of pulling speed fluctuations This indicates that the initial power of the reference rod with equal diameter is too high. Therefore, the first type of main applied power correction value should be set to the same level as... Negative values ​​of linear correlation are applied, and power adjustment and limiting are performed; specifically including: when ,but ; in, This is the first type of main applied power correction value. This is the power correction factor. This is the power adjustment limiting factor; In a practical example, the set pulling speed fluctuation range is [-2kW, 2kW]; If the average pulling speed deviation Within the range of speed fluctuation If the initial power of the reference rod is within the range of equal diameter, it indicates that the initial power is appropriate, and the first type of main applied power correction value is set to zero; including: when ,but ; in, This is the first type of main applied power correction value.

[0055] Step S203: Calculate the initial power of the current rod based at least on the initial power of the reference rod and the first type of main applied power correction value. Typically, the initial power of the current rod with constant diameter is calculated using the following formula: ; in, The initial power of the current rod with constant diameter. For the reference bar, the initial power of the constant diameter bar, This is the first type of main applied power correction value.

[0056] Step S20 may further include: Step S204: When the cumulative running time of the furnace platform in the current furnace batch is greater than the set running time threshold, the second type of main power correction value is calculated by multiplying the constant diameter length of the reference rod batch and the negative power correction coefficient; the constant diameter initial power of the current rod batch is calculated based on the constant diameter initial power of the reference rod batch, the first type of main power correction value and the second type of main power correction value.

[0057] As the single crystals continue to grow in this batch, the heat preservation capacity inside the furnace will increase. If the main power correction value for subsequent batches continues to use the first type of main power correction value... (Power correction factor) If it is fixed, then the thermal field inside the furnace platform will be too hot; therefore, it is necessary to calculate the insulation compensation amount, i.e., the correction value of the second type of main power. The second type of main applied power correction value is calculated using the following formula: ; in The reference bar's equal diameter length (single crystal length after the equal diameter stage is completed); This is the negative power correction factor.

[0058] Typically, the second type of main power correction value is calculated several times after the current furnace cycle (e.g., after three or four cycles).

[0059] The initial power of the current rod with constant diameter is calculated using the following formula: ; in, The initial power of the current rod with constant diameter. For the reference bar, the initial power of the constant diameter bar, This is the first type of main applied power correction value. This is the second type of main applied power correction value.

[0060] Step S30: When there is no reference order, calculate the initial equal-diameter power of the current order based on the current order's lead-in power, the default total power reduction during shoulder formation, and the manually adjusted power reduction factor during the current order's shoulder formation stage; specifically including: Step S301: Obtain the power reduction amplitude of several manual adjustments during the current bar shoulder formation stage; sum the power reduction amplitudes of these manual adjustments and multiply them by a shoulder formation power compensation coefficient to obtain the power reduction amplitude factor of the current bar shoulder formation stage; for example: ; Where M represents the number of times the power reduction is manually adjusted during the current bar-shoulder-setting phase. The manual adjustment power reduction during the current bar-shoulder-setting phase; This is the shoulder power compensation coefficient; The power reduction factor is manually adjusted during the current bar-shoulder release phase.

[0061] Step S302: Limit the manual adjustment of the power reduction factor during the current bar shoulder-laying stage to prevent excessive adjustment from causing wire breakage during the equal-diameter stage; for example: ; Step S303: Calculate the initial power of the current rod with constant diameter; for example: ; in, The initial power for the current rod with constant diameter; This represents the crystal-driving power for the current order of crystals. The power reduction factor is manually adjusted during the current bar-shoulder-setting stage; This represents the default total power reduction for shoulder mounting.

[0062] The above methods enable continuous transfer of thermal field status, identification of pulling speed deviation, and adaptive optimization of initial power for equal diameter, effectively reducing the breakage rate of equal diameter head and improving the stability and product consistency of the monocrystalline silicon growth process.

[0063] like Figure 5As shown, this embodiment of the invention also provides an electronic device, including: a processor and a memory; the processor and the memory communicate with each other, for example, by being connected and communicating with each other via a communication bus; the memory stores a computer program; the processor is used to run the computer program, and the computer program executes the steps of the equal-diameter initial power recommendation method described above when it runs; the processor may be a CPU, or other general-purpose processor, digital signal processor (DSP), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or combinations of the above-mentioned chips, devices, or circuits; the memory may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk, or solid-state drive; the memory may also include combinations of the above-mentioned types of memory.

[0064] This invention also proposes a storage medium storing a computer program configured to execute the steps of the equal-diameter initial power recommendation method described above during runtime. The storage medium may be a magnetic disk, optical disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk drive (HDD), or solid-state drive (SSD), etc.; the storage medium may also include combinations of the above types of memory.

[0065] Finally, it should be noted that the above specific 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 embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for recommending initial power for constant-diameter applications, characterized in that, include: Obtain the first type of process parameter data for the historical and current single crystal production processes during the current batch; Based on the comparison of the remaining material in the first type of process parameter data of the current bar batch and the historical bar batch, reference bar batches are selected from the historical bar batches of the current furnace batch. When a reference batch exists, acquire the second type of process parameter data of multiple historical furnaces and multiple historical bars in each historical furnace during the historical production process of the same product's sample single crystal, and construct a target pulling speed curve accordingly; calculate the average pulling speed deviation between the actual average pulling speed curve of the reference batch within a specified single crystal length range at the equal diameter head and the target pulling speed curve, and determine the first type of main applied power correction value based on the average pulling speed deviation; calculate the initial equal diameter power of the current batch based at least on the initial equal diameter power of the reference batch and the first type of main applied power correction value; When there is no reference order, the initial equal-diameter power of the current order is calculated based on the current order's lead-in power, the default total power reduction during shoulder formation, and the manually adjusted power reduction factor during the current order's shoulder formation stage.

2. The method for recommending initial power of constant diameter as described in claim 1, characterized in that, The first type of process parameter data includes furnace batch, bar batch, remaining material amount, actual average pulling speed for equal diameter, set average pulling speed for equal diameter, initial power for equal diameter, and crystal pulling power; the step of comparing the remaining material amount in the first type of process parameter data of the current bar batch and historical bar batches in the current furnace batch, and selecting reference bar batches from the historical bar batches of the current furnace batch, includes: The first type of process parameter data of the historical rod batches in the current single crystal production process are sorted according to the temperature adjustment end time to construct the production sequence of the current batch. The remaining material quantity of the current batch of bars in the current furnace is compared with the remaining material quantity of each historical batch to obtain the remaining material quantity deviation; If there are historical batches in the current furnace production sequence that meet the condition that the remaining material deviation is less than or equal to the set material deviation threshold, the historical batch that is closest in time to the current batch and has entered the equal diameter stage is selected as the reference batch; if there are no batches in the current furnace production sequence that meet the condition that the remaining material deviation is less than or equal to the set material deviation threshold, then it is determined that there are no reference batches.

3. The method for recommending initial power of constant diameter as described in claim 2, characterized in that, The second type of process parameter data includes: single crystal length and actual average pulling speed for constant diameter; the acquisition of second type of process parameter data from multiple historical furnaces and multiple historical bars within each historical furnace during the historical production process of sample single crystals of the same product, and the construction of a target pulling speed curve based on this data, includes: Data cleaning was performed on the second type of process parameter data for each historical bar in each historical furnace of the sample single crystal, and the sample single crystals whose single crystal length was greater than the specified single crystal length threshold of the equal diameter head after the equal diameter stage were retained as valid sample single crystals. Align the effective sample single crystals of the same process according to the single crystal length of the equal diameter stage; for each single crystal length position, calculate the average of the actual average pulling speed of all effective sample single crystals at each single crystal length position to generate the target pulling speed trajectory; wherein, the single crystal length in the target pulling speed trajectory is less than or equal to the single crystal length threshold specified at the equal diameter head. The target pulling speed trajectory is spliced ​​with the equal diameter set average pulling speed trajectory determined by the current bar's equal diameter set average pulling speed to obtain the target pulling speed curve, and Gaussian smoothing algorithm is used to smooth the splicing point; wherein, the single crystal length in the equal diameter set average pulling speed trajectory is greater than the single crystal length threshold specified by the equal diameter head.

4. The method for recommending initial power of constant diameter as described in claim 3, characterized in that, The specified single crystal length threshold for the equal diameter head is 900mm to 1100mm.

5. The method for recommending initial power of constant diameter as described in claim 3, characterized in that, The calculation of the average pulling speed deviation between the actual average pulling speed curve of the reference bar within a specified single crystal length range at the equal diameter head and the target pulling speed curve, and the determination of the first type of main applied power correction value based on the average pulling speed deviation, includes: Obtain the actual average pulling speed of reference bars within the specified single crystal length range of the equal diameter head and construct the actual average pulling speed curve of the equal diameter head; the upper limit of the specified single crystal length range of the equal diameter head is greater than the specified single crystal length threshold of the equal diameter head. Calculate the average pulling rate deviation between the actual average pulling rate curve and the target pulling rate curve for reference bars within a specified single crystal length range at the constant diameter head. ; The average pulling speed deviation With the set speed fluctuation range Compare; If the average pulling speed deviation Exceeding the upper limit of the speed fluctuation range This indicates that the initial power of the reference rod with equal diameter is too low. Therefore, the first type of main applied power correction value should be set to the same level as... Positive values ​​with linear correlation are applied, and power adjustment and limiting are performed. If the average pulling speed deviation Below the lower limit of the pulling speed fluctuation range This indicates that the initial power of the reference rod with equal diameter is too high. Therefore, the first type of main applied power correction value should be set to the same level as... Negative values ​​of linear correlation are applied, and power adjustment limiting is performed. If the average pulling speed deviation Within the range of speed fluctuation If the value is within the range, it indicates that the initial power of the reference rod is appropriate, and the first type of main power correction value is set to zero.

6. The method for recommending initial power of constant diameter as described in claim 5, characterized in that, The upper limit of the specified single crystal length range for the equal diameter head is 1500mm to 1700mm; The set pulling speed fluctuation range is [-2kw, 2kw].

7. The method for recommending initial power of constant diameter as described in any one of claims 1 to 6, characterized in that, The method for recommending the initial power of equal diameter bars further includes: when a reference bar number exists, the calculation of the initial power of equal diameter bars for the current bar number based at least on the initial power of equal diameter bars based on the reference bar number and the first type of main applied power correction value further includes: When the cumulative running time of the furnace platform in the current furnace batch exceeds the set running time threshold, the second type of main power correction value is calculated by multiplying the constant diameter length of the reference rod batch and the negative power correction coefficient; the constant diameter initial power of the current rod batch is calculated based on the constant diameter initial power of the reference rod batch, the first type of main power correction value, and the second type of main power correction value.

8. The method for recommending initial power of constant diameter as described in any one of claims 1 to 6, characterized in that, The manually adjusted power reduction factor for the current bar shoulder formation stage is obtained through the following method: Obtain the power reduction of several manual adjustments during the current bar shoulder formation stage; sum the power reduction of several manual adjustments and multiply it by a shoulder formation power compensation coefficient to obtain the power reduction factor of the current bar shoulder formation stage; and limit the power reduction factor of the manual adjustments.

9. An electronic device, characterized in that, include: Memory, which stores computer programs; A processor for running the computer program, which, when running, performs the steps of the constant-diameter initial power recommendation method as described in any one of claims 1 to 8.

10. A storage medium, characterized in that, The storage medium stores a computer program configured to execute the steps of the constant-diameter initial power recommendation method as described in any one of claims 1 to 8 when running.

Images