Method for optimizing control parameters of an oxidation furnace for chip manufacturing and related device

Abstract

The application discloses a chip manufacturing oxidation furnace control parameter optimization method and related equipment, and relates to the technical field of semiconductor production and manufacturing. The chip manufacturing oxidation furnace control parameter optimization method comprises the following steps: obtaining sensor parameters in a wafer oxidation process; inputting the sensor parameters into a preset oxidation prediction model to determine a wafer oxidation prediction result, wherein the oxidation prediction model is obtained by training a preset to-be-trained model by using historical sensor data samples and wafer oxidation result labels; and dynamically adjusting oxidation furnace control parameters according to the wafer oxidation prediction result. The application can reduce defects caused by uneven oxidation and improve the pass rate of wafers by adjusting the control parameters. The oxidation prediction model can help avoid defects such as over-oxidation or insufficient oxidation and improve the quality of wafers. Through accurate prediction and dynamic adjustment, the oxidation process of wafers can be effectively controlled, thereby improving the yield of the oxidation process of wafers.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of semiconductor production and manufacturing, and particularly relates to an oxidation furnace control parameter optimization method for chip manufacturing and related equipment. BACKGROUND

[0002] In the field of semiconductor manufacturing, the quality and characteristics of the oxide layer have a decisive influence on the performance of the wafer. The oxidation furnace is a key device for forming an oxide layer on the surface of the wafer. How to oxidize the wafer by the oxidation furnace with high quality has become a hot issue in the field of semiconductors.

[0003] The conventional oxidation furnace usually adopts fixed parameter settings such as temperature, time, oxygen flow, etc. The settings of these parameters are often based on experience, lack of adaptability to wafer characteristics and environmental changes, and cannot adapt to the slight differences of different batches of wafers and the changes of environmental conditions, making it difficult to achieve precise quality control and resulting in a low yield of production wafers. SUMMARY

[0004] The main purpose of the present application is to provide an oxidation furnace control parameter optimization method for chip manufacturing and related equipment, aiming to solve the technical problem of low yield of production wafers.

[0005] To achieve the above-mentioned purpose, the present application provides an oxidation furnace control parameter optimization method for chip manufacturing, which comprises the following steps:

[0006] Obtaining sensor parameters in the wafer oxidation process;

[0007] Inputting the sensor parameters into a preset oxidation prediction model to determine a wafer oxidation prediction result, wherein the oxidation prediction model is obtained by training a preset to-be-trained model by historical sensor data samples and wafer oxidation result labels;

[0008] According to the wafer oxidation prediction result, dynamically adjusting the oxidation furnace control parameters.

[0009] In an embodiment, the step of obtaining sensor parameters in the wafer oxidation process comprises the following steps:

[0010] Obtaining historical sensor data samples and wafer oxidation result labels;

[0011] Inputting the historical sensor data samples into a preset to-be-trained model;

[0012] Predicting the wafer oxidation process by the to-be-trained model to determine a wafer oxidation prediction result;

[0013] Judging the difference between the wafer oxidation prediction result and the wafer oxidation result label to obtain a prediction loss value;

[0014] If the predicted loss value does not meet the preset loss standard, the step of predicting wafer oxidation by the to-be-trained model to determine a wafer oxidation prediction result is returned until the training is stopped after the predicted loss value meets the loss standard, and an oxidation prediction model meeting the loss standard is obtained.

[0015] In an embodiment, the step of dynamically adjusting the oxidation furnace control parameter comprises:

[0016] According to the deviation between the wafer oxidation prediction result and a preset oxidation target value, a corresponding control parameter adjustment deviation is calculated.

[0017] According to the sensitivity of the control parameter in different stages of the wafer oxidation process, a preset control parameter adjustment strategy is adjusted.

[0018] According to the change trend of the wafer oxidation prediction result, the control parameter adjustment strategy is adjusted.

[0019] According to the real-time monitored process environment factors, the control parameter adjustment strategy is corrected.

[0020] According to the corrected control parameter adjustment strategy, the control parameter adjustment deviation is converted into an actual control parameter adjustment value.

[0021] According to the actual control parameter adjustment value, the oxidation furnace control parameter is dynamically adjusted.

[0022] In an embodiment, the step of adjusting the preset control parameter adjustment strategy according to the sensitivity of the control parameter in different stages of the wafer oxidation process comprises:

[0023] The current wafer oxidation stage is determined.

[0024] If the wafer oxidation stage is the heating stage, the preset control parameter adjustment strategy is adjusted to temperature deviation, and a large-amplitude adjustment strategy is adopted for rapid correction.

[0025] If the wafer oxidation stage is the oxidation stage, the control parameter adjustment strategy is adjusted to focus on controlling temperature and oxygen partial pressure, and the deviation of temperature and oxygen partial pressure is accurately adjusted according to different periods of the oxidation stage.

[0026] If the wafer oxidation stage is the cooling stage, the control parameter adjustment strategy is adjusted to focus on controlling the temperature cooling rate, and a small-amplitude adjustment strategy is adopted for slow correction.

[0027] In an embodiment, the step of adjusting the control parameter adjustment strategy according to the change trend of the wafer oxidation prediction result comprises:

[0028] If the wafer oxidation prediction result is that the oxidation rate is too fast, according to the remaining time of oxidation expectation, the control parameter adjustment strategy is adjusted to use a step-by-step decreasing adjustment strategy;

[0029] If the wafer oxidation prediction result is that the oxidation rate is too slow, according to the remaining time of oxidation expectation, the control parameter adjustment strategy is adjusted to use a step-by-step increasing adjustment strategy;

[0030] If the wafer oxidation prediction result is that the oxidation process is stable operation, the control parameter adjustment strategy is adjusted to maintain the existing control parameter adjustment value.

[0031] In an embodiment, the step of correcting the control parameter adjustment strategy according to the real-time monitored process environment factors comprises:

[0032] When there is an unstable phenomenon in the furnace airflow, according to the airflow deviation and the stable compensation time, the control parameter adjustment strategy is corrected to use a step-by-step correction adjustment strategy control parameter;

[0033] When the temperature distribution is uneven, according to the temperature deviation and the oxidation sensitivity of each region, the control parameter adjustment strategy is corrected to accurately adjust the control parameter;

[0034] When the pressure fluctuates, according to the pressure change and the adjustment strategy of the pressure compensation time, the control parameter adjustment strategy is corrected to adjust the control parameter in real time.

[0035] In addition, in order to achieve the above-mentioned purpose, the present application also provides an oxidation furnace control parameter optimization device for chip manufacturing, which comprises:

[0036] An acquisition module is configured to acquire sensor parameters in a wafer oxidation process;

[0037] A prediction module is configured to input the sensor parameters into a preset oxidation prediction model to determine a wafer oxidation prediction result, wherein the oxidation prediction model is obtained by training a preset to-be-trained model by historical sensor data samples and wafer oxidation result labels;

[0038] An adjustment module is configured to dynamically adjust oxidation furnace control parameters according to the wafer oxidation prediction result.

[0039] In addition, in order to achieve the above-mentioned purpose, the present application also provides an oxidation furnace control parameter optimization device for chip manufacturing, which comprises: a memory, a processor, and a computer program stored on the memory and executable on the processor, the computer program being configured to implement the steps of the oxidation furnace control parameter optimization method for chip manufacturing as described above.

[0040] In addition, to achieve the above object, the present application also provides a storage medium, which is a computer readable storage medium, and a computer program is stored on the storage medium, and the computer program realizes the steps of the chip manufacturing oxidation furnace control parameter optimization method when executed by a processor.

[0041] In addition, to achieve the above object, the present application also provides a computer program product, which comprises a computer program, and the computer program realizes the steps of the chip manufacturing oxidation furnace control parameter optimization method when executed by a processor.

[0042] The one or more technical solutions provided by the present application have at least the following technical effects:

[0043] Compared with the prior art, the traditional oxidation furnace usually adopts fixed parameter settings such as temperature, time, oxygen flow, etc. The setting of these parameters is often based on experience, lacks adaptability to wafer characteristics and environmental changes, cannot adapt to the slight differences of different batches of wafers and the changes of environmental conditions, and is difficult to realize accurate quality control, resulting in low yield of production wafers. Compared with the prior art, the present application obtains sensor parameters in the wafer oxidation process; inputs the sensor parameters into a preset oxidation prediction model to determine a wafer oxidation prediction result, wherein the oxidation prediction model is obtained by training a preset to-be-trained model by historical sensor data samples and wafer oxidation result labels; and dynamically adjusts the oxidation furnace control parameters according to the wafer oxidation prediction result. It can be understood that, by adjusting the control parameters, the present application can ensure that the oxidation degree of each wafer is consistent, reduce defects caused by uneven oxidation, and improve the pass rate of wafers. The oxidation prediction model can help avoid over-oxidation or insufficient oxidation and other defects, and improve the quality of wafers. Optimizing the oxidation process can reduce material waste and production time, and reduce production cost. Through accurate prediction and dynamic adjustment, the oxidation process of the wafer is effectively controlled, thereby improving the yield of the wafer oxidation process. BRIEF DESCRIPTION OF DRAWINGS

[0044] The accompanying drawings, which are incorporated into and form part of the specification, illustrate embodiments consistent with the present application and, together with the specification, serve to explain the principles of the application.

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0046] Figure 1 The flowchart provided by the present application for the chip manufacturing oxidation furnace control parameter optimization method embodiment one;

[0047] Figure 2 The flowchart provided for the second embodiment of the chip manufacturing oxidation furnace control parameter optimization method of the present application is shown in the figure.

[0048] Figure 3 The module structure diagram of the chip manufacturing oxidation furnace control parameter optimization device of the embodiment of the present application is shown in the figure.

[0049] Figure 4 The device structure diagram of the hardware operating environment involved in the chip manufacturing oxidation furnace control parameter optimization method in the embodiment of the present application is shown in the figure.

[0050] The purpose realization, functional features and advantages of the present application will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0051] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and do not limit the present application.

[0052] In order to better understand the technical solutions of the present application, the specific embodiments will be described in detail below in conjunction with the drawings in the specification.

[0053] The main solution of the embodiment of the present application is:

[0054] Obtain sensor parameters in the wafer oxidation process;

[0055] Input the sensor parameters into a preset oxidation prediction model to determine a wafer oxidation prediction result, wherein the oxidation prediction model is obtained by training a preset to-be-trained model with historical sensor data samples and wafer oxidation result labels;

[0056] According to the wafer oxidation prediction result, dynamically adjust the oxidation furnace control parameters.

[0057] In the present embodiment, the chip manufacturing oxidation furnace control parameter optimization device is taken as the execution subject, and for the sake of description, the following will be specifically described in the form of a simple device.

[0058] Since the traditional oxidation furnace usually adopts fixed parameter settings such as temperature, time, oxygen flow, etc., the setting of these parameters is often based on experience, and lacks adaptability to wafer characteristics and environmental changes, cannot adapt to the slight differences of different batches of wafers and the changes of environmental conditions, and is difficult to realize accurate quality control, resulting in low yield of production wafers.

[0059] The application provides a solution that can ensure uniform oxidation of each wafer by adjusting control parameters, reduce defects caused by uneven oxidation, and improve the yield of wafers. The oxidation prediction model can help avoid over-oxidation or insufficient oxidation and improve the quality of wafers. Optimizing the oxidation process can reduce material waste and production time, and reduce production costs. By accurately predicting and dynamically adjusting, the oxidation process of the wafer can be effectively controlled, thereby improving the yield of the oxidation process of the wafer.

[0060] Based on this, the embodiment of the application provides an oxidation furnace control parameter optimization method for chip manufacturing. Referring to Figure 1 , Figure 1 The flowchart of the first embodiment of the oxidation furnace control parameter optimization method for chip manufacturing of the application is shown in the figure.

[0061] In this embodiment, the oxidation furnace control parameter optimization method for chip manufacturing includes steps S10-S30:

[0062] Step S10, obtaining sensor parameters in the wafer oxidation process;

[0063] It should be noted that the wafer oxidation process refers to the process of heat treating silicon wafers by an oxidation furnace to form an oxide film on the surface in semiconductor manufacturing. This process is crucial for manufacturing integrated circuits and semiconductor devices, as it can form insulating layers, protective layers, or serve as the basis for other process steps. Sensor parameters refer to the data of physical quantities and operating conditions monitored and recorded by various sensors in the wafer oxidation process. These parameters usually include temperature, pressure, gas flow, humidity, oxidation time, etc., which are key factors in controlling the oxidation process and evaluating the quality of oxidation.

[0064] It can be understood that in the field of semiconductor manufacturing, the wafer oxidation process is a basic step for forming key insulating layers and other functional layers in integrated circuits and semiconductor devices. This process involves exposing silicon wafers to high temperatures and specific gas environments to promote the occurrence of oxidation reactions. In order to ensure the quality and uniformity of the oxide layer, it is crucial to accurately control various parameters in the oxidation process. In this application, a method is proposed to optimize the control parameters of the oxidation furnace by obtaining sensor parameters in the wafer oxidation process. The core of this method is to monitor and analyze key physical quantities in the oxidation process in real time, such as temperature, pressure, and gas flow, which have a direct impact on the formation and properties of the oxide layer.

[0065] Step S20, inputting the sensor parameters into a preset oxidation prediction model to determine a wafer oxidation prediction result, wherein the oxidation prediction model is obtained by training a preset to-be-trained model with historical sensor data samples and wafer oxidation result labels;

[0066] It is noted that the sensor parameters refer to the physical quantities and operating conditions monitored and recorded by various sensors during the wafer oxidation process, including but not limited to temperature, pressure, gas flow, humidity, and oxidation time, among other key parameters. The oxidation prediction model is a model constructed using mathematical and statistical methods to predict the outcome of the wafer oxidation process. This model can predict the characteristics of the oxide layer, such as thickness, uniformity, and quality, based on the input sensor parameters. The historical sensor data samples refer to the sensor parameter data collected in the past during the wafer oxidation process, which is used to train the oxidation prediction model to learn the relationship between the oxidation process and the oxidation outcome. The wafer oxidation outcome labels refer to the records of the wafer oxidation outcomes corresponding to the historical sensor data samples, which are usually determined by a quality inspection step and used as the "ground truth" in model training to evaluate the accuracy of the model predictions. The pre-set model to be trained refers to the model framework set before the training begins, which has not been trained and needs to be optimized by learning from historical data to improve the accuracy of predicting the wafer oxidation outcome.

[0067] It can be understood that in the field of semiconductor manufacturing, wafer oxidation is a critical step that directly affects the performance and reliability of devices. In order to ensure the quality of the oxide layer, it is necessary to accurately control multiple parameters in the oxidation process. Traditional control methods often rely on fixed parameter settings, lack of adaptability to different wafer characteristics and environmental changes, resulting in unstable oxide layer quality. The method described in this application determines the predicted outcome of wafer oxidation by inputting real-time collected sensor parameters into a pre-set oxidation prediction model. The core of this method is to use historical sensor data samples and wafer oxidation outcome labels to train a model to be trained, so that it can predict future oxidation outcomes. In this way, the oxidation prediction model can learn the relationship between oxidation parameters and oxide layer quality and make predictions based on these relationships. The training of the oxidation prediction model is an iterative process, in which the model continuously adjusts its internal parameters until the difference between the predicted outcome and the actual outcome (i.e., the prediction loss value) meets the pre-set loss standard. Once the model is trained, it can be used for real-time prediction to provide predicted outcomes of wafer oxidation based on input sensor parameters.

[0068] The data-driven approach provides a new way to optimize the oxidation process, as it can dynamically adjust control parameters based on real-time data to achieve higher quality oxide layers. This method not only improves the uniformity and quality of the oxide layer, but also reduces the dependence on physical measurements, reduces production costs, and enhances the controllability and predictability of the semiconductor manufacturing process. In this way, this application helps to promote the progress of semiconductor manufacturing technology to meet the growing demand for high-performance integrated circuits.

[0069] Step S30, dynamically adjusting the oxidation furnace control parameters according to the wafer oxidation prediction results.

[0070] It should be noted that the wafer oxidation prediction results refer to the prediction data of the expected characteristics and quality of the wafer oxide layer obtained by the oxidation prediction model based on real-time sensor parameter analysis. These results usually include key indicators such as the thickness, uniformity, and defect rate of the oxide layer, which are used to evaluate and optimize the oxidation process. The oxidation furnace is a device used in the semiconductor manufacturing process to oxidize the wafer under high temperature conditions. This device can generate an oxide layer in a strictly controlled environment and is an indispensable part of semiconductor device manufacturing. The control parameters refer to the parameters used to accurately control the performance of the oxidation furnace and the oxidation process during the operation of the oxidation furnace. These parameters include but are not limited to temperature, pressure, gas flow, oxidation time, etc., which directly affect the formation and characteristics of the oxide layer. Dynamic adjustment refers to real-time, automatic adjustment of the control parameters of a system or process based on real-time feedback or prediction results. In the context of this application, dynamic adjustment means automatically optimizing the control parameters of the oxidation furnace according to the changes in the wafer oxidation prediction results to achieve more accurate process control.

[0071] It can be understood that in the field of semiconductor manufacturing, accurate control of the wafer oxidation process is crucial to ensure the performance and reliability of the final product. The quality and characteristics of the oxide layer, such as thickness and uniformity, directly affect the electrical performance and stability of the device. Therefore, real-time monitoring and optimization of the oxidation process is key to improving the efficiency and quality of semiconductor manufacturing. The method described in this application involves dynamically adjusting the control parameters of the oxidation furnace according to the wafer oxidation prediction results. This method uses advanced prediction models combined with real-time sensor data to predict the results of the oxidation process and adjust the operating parameters of the oxidation furnace accordingly. This dynamic adjustment strategy allows the oxidation process to adapt to real-time changes in conditions such as changes in wafer characteristics, environmental fluctuations, etc., thereby achieving higher precision and consistency of the oxide layer.

[0072] In one possible implementation, step S30 can include:

[0073] According to the deviation between the wafer oxidation prediction results and the preset oxidation target value, the corresponding control parameter adjustment deviation is calculated;

[0074] According to the sensitivity of the control parameters in different stages of the wafer oxidation process, the preset control parameter adjustment strategy is adjusted;

[0075] According to the trend of the wafer oxidation prediction results, the control parameter adjustment strategy is adjusted;

[0076] According to the real-time monitored process environmental factors, the control parameter adjustment strategy is corrected;

[0077] According to the modified control parameter adjustment strategy, the control parameter adjustment deviation is converted into an actual control parameter adjustment value;

[0078] According to the actual control parameter adjustment value, the oxidation furnace control parameter is dynamically adjusted.

[0079] It should be noted that the wafer oxidation prediction result refers to the prediction data of the expected characteristics and quality of the wafer oxide layer obtained by the oxidation prediction model based on real-time sensor parameters. These results usually include key indicators such as the thickness, uniformity, and defect rate of the oxide layer, which are used to evaluate and optimize the oxidation process. The preset oxidation target value refers to the ideal value of the oxide layer characteristics pre-set according to process requirements and product quality standards during the wafer oxidation process. These target values are used to evaluate the accuracy and consistency of the oxidation process, and serve as a benchmark for adjusting control parameters. The control parameter adjustment deviation refers to the amount of adjustment required for the oxidation furnace control parameter calculated based on the deviation between the wafer oxidation prediction result and the preset oxidation target value. This deviation is the basis for control parameter adjustment. The control parameter adjustment strategy refers to a series of rules and methods for guiding how to adjust the oxidation furnace control parameter based on the deviation between the wafer oxidation prediction result and the preset target value. The strategy may include the magnitude, direction, and timing of the adjustment. The different stages of the wafer oxidation process refer to different time periods in the wafer oxidation process, each stage may have different sensitivity to specific control parameters, such as the heating stage, oxidation stage, and cooling stage. The process environmental factors refer to various environmental conditions that may affect the quality and efficiency of the oxidation process, such as temperature fluctuations, pressure changes, gas purity, and humidity. The control parameter adjustment value refers to the specific value calculated based on the control parameter adjustment deviation and the adjustment strategy, which is used to actually adjust the oxidation furnace control parameter.

[0080] It can be understood that in semiconductor manufacturing, wafer oxidation is a critical step that directly affects the performance and reliability of the device. Traditional oxidation process control methods often rely on fixed parameter settings, which cannot adapt to different wafer characteristics and environmental changes, resulting in unstable oxide layer quality. The method proposed in this application achieves optimization of the oxidation process through the following steps:

[0081] Calculate the control parameter adjustment deviation: by comparing the wafer oxidation prediction result with the preset oxidation target value, calculate the control parameter adjustment deviation required. This step is the basis for subsequent adjustment, ensuring the accuracy of the adjustment direction and magnitude.

[0082] Adjust the control parameter adjustment strategy: adjust the preset control parameter adjustment strategy according to the sensitivity of the control parameter in different stages of the wafer oxidation process. This step takes into account the dynamic characteristics of the oxidation process, ensuring effective control in critical stages.

[0083] Adjusting the strategy according to the trend: According to the trend of the wafer oxidation prediction results, further adjust the control parameter adjustment strategy. This step makes the control strategy adapt to the real-time changes of the oxidation process, improves the response speed and adjustment accuracy.

[0084] According to the process environment factor correction strategy: According to the real-time monitoring of the process environment factor, the control parameter adjustment strategy is corrected. This step ensures that the control strategy can still maintain effectiveness when the environmental conditions change.

[0085] Convert the adjustment deviation into actual value: According to the corrected control parameter adjustment strategy, convert the control parameter adjustment deviation into the actual control parameter adjustment value. This step is the key to realize the dynamic adjustment of the control parameter, and ensures the implementation of the adjustment value.

[0086] Dynamic adjustment of control parameters: According to the actual control parameter adjustment value, dynamically adjust the oxidation furnace control parameters. This step realizes the real-time optimization of the oxidation process, improves the quality and consistency of the oxide layer.

[0087] In summary, the method described in this application realizes intelligent and dynamic adjustment of the oxidation furnace control parameters by comprehensively considering the prediction results, target deviation, process stage, trend and environmental factors, and provides an innovative solution for the semiconductor manufacturing field to improve the quality and efficiency of wafer oxidation process.

[0088] In a feasible implementation, the step of adjusting the preset control parameter adjustment strategy according to the sensitivity of the control parameters in different stages of the wafer oxidation process includes:

[0089] Determine the current wafer oxidation stage;

[0090] If the wafer oxidation stage is the heating stage, adjust the preset control parameter adjustment strategy to temperature deviation, and use the large amplitude adjustment strategy for fast correction;

[0091] If the wafer oxidation stage is the oxidation stage, adjust the control parameter adjustment strategy to focus on controlling temperature and oxygen partial pressure, and accurately adjust the deviation of temperature and oxygen partial pressure according to different periods of the oxidation stage;

[0092] If the wafer oxidation stage is the cooling stage, adjust the control parameter adjustment strategy to focus on controlling the temperature cooling rate, and use the small amplitude adjustment strategy for slow correction.

[0093] It should be noted that the wafer oxidation stage refers to different periods in the wafer oxidation process, including the heating stage, the oxidation stage and the cooling stage, and each stage has different requirements for the adjustment strategy of the control parameters. The heating stage is the initial stage of the wafer oxidation process, and the main purpose is to heat the wafer to the required temperature for oxidation. The oxidation stage is the core stage of the wafer oxidation process, and in this stage, the wafer surface reacts with the oxidizing agent to form an oxide layer. The cooling stage is the last stage of the wafer oxidation process, and the wafer is slowly cooled from the oxidation temperature to room temperature. The control parameter adjustment strategy refers to the adjustment strategy of the control parameters (such as temperature, pressure, etc.) of the oxidation furnace according to the different stages of the wafer oxidation process. The temperature deviation refers to the difference between the actual temperature and the preset temperature target value, which is used to guide the rapid temperature adjustment in the heating stage. The oxygen partial pressure refers to the partial pressure of oxygen in the oxidation process, which is one of the key parameters that need to be accurately controlled in the oxidation stage. The temperature cooling rate refers to the cooling speed of the wafer in the cooling stage, which needs to be accurately controlled to avoid thermal stress and oxide layer quality problems.

[0094] It can be understood that the present application describes a method of dynamically adjusting control parameters according to different stages of wafer oxidation process. The method accurately identifies the current oxidation stage, and adjusts the control parameters according to the characteristics of each stage to achieve better oxide layer quality and process control.

[0095] In the heating stage, since the wafer needs to be quickly heated to the set oxidation temperature, a large amplitude temperature adjustment strategy is adopted to achieve rapid correction. This strategy helps to reduce the time delay in the heating process and ensures the timely start of the oxidation process.

[0096] The oxidation stage is the key period for forming the oxide layer, and the control of temperature and oxygen partial pressure is particularly important. The method of the present application accurately adjusts the deviation of temperature and oxygen partial pressure, and fine-tunes according to the characteristics of different periods to optimize the growth rate and quality of the oxide layer.

[0097] The cooling stage needs to slowly reduce the temperature of the wafer to avoid thermal stress, and the method of the present application accurately controls the cooling rate through a small amplitude adjustment strategy to protect the structural integrity of the oxide layer and prevent device damage caused by rapid temperature change.

[0098] Overall, the present application dynamically adjusts the control parameters according to the characteristics and needs of different oxidation stages to achieve accurate control of the wafer oxidation process. This method not only improves the quality and uniformity of the oxide layer, but also enhances the flexibility and adaptability of the production process, which is of great significance for improving the performance and reliability of semiconductor devices.

[0099] In a feasible implementation manner, the step of adjusting the control parameter adjustment strategy according to the change trend of the wafer oxidation prediction result comprises:

[0100] If the wafer oxidation prediction result is that the oxidation rate is too fast, adjust the control parameter adjustment strategy to use a step-by-step decreasing adjustment strategy based on the expected remaining time for oxidation.

[0101] If the wafer oxidation prediction result is that the oxidation rate is too slow, adjust the control parameter adjustment strategy to use a step-by-step increasing adjustment strategy based on the expected remaining time for oxidation.

[0102] If the wafer oxidation prediction result is that the oxidation process is stable operation, adjust the control parameter adjustment strategy to maintain the existing control parameter adjustment value.

[0103] It should be noted that the wafer oxidation prediction result refers to the prediction data of the expected characteristics and quality of the wafer oxidation layer obtained by the oxidation prediction model based on real-time sensor parameter analysis, including key indicators such as oxidation rate. Oxidation rate refers to the speed of forming an oxide layer on the wafer surface, which is an important parameter in the wafer oxidation process, affecting the thickness and uniformity of the oxide layer. The control parameter adjustment strategy refers to a series of rules and methods for guiding how to adjust the oxidation furnace control parameters based on the wafer oxidation prediction result. The step-by-step decreasing adjustment strategy is a control parameter adjustment method that gradually reduces relevant control parameters (such as temperature or gas flow) to slow down the oxidation rate when the prediction result shows that the oxidation rate is too fast. The step-by-step increasing adjustment strategy is a control parameter adjustment method that gradually increases relevant control parameters to speed up the oxidation rate when the prediction result shows that the oxidation rate is too slow. The expected remaining time for oxidation refers to the time remaining to complete the oxidation process based on the current oxidation progress and the planned total oxidation time. The stable operation of the oxidation process refers to the normal operation of the wafer oxidation process within the expected parameter range, with the oxidation rate and oxide layer quality meeting the preset standards.

[0104] It can be understood that the present application describes a method of dynamically adjusting control parameters based on wafer oxidation prediction results, aiming to optimize the oxidation process of the oxidation furnace. This method intelligently adjusts control parameters to maintain the stability and efficiency of the oxidation process by monitoring the oxidation rate in real time and predicting its trend.

[0105] When the prediction result shows that the oxidation rate is too fast, the method adjusts the control parameters through a step-by-step decreasing adjustment strategy to slow down the oxidation rate and prevent the oxide layer from being too thick or having quality problems. This strategy helps to avoid device performance degradation caused by too fast oxidation rate.

[0106] On the contrary, when the prediction result shows that the oxidation rate is too slow, the method adjusts the control parameters through a step-by-step increasing adjustment strategy to speed up the oxidation rate and ensure that the oxidation process proceeds as planned. This strategy helps to improve production efficiency and reduce production delays caused by too slow oxidation rate.

[0107] When the prediction result shows that the oxidation process is operating stably, the method adjusts the control parameter adjustment strategy to maintain the existing control parameter adjustment value, so as to maintain the continuity and stability of the oxidation process. This strategy helps to reduce unnecessary adjustments and improve the efficiency and reliability of the oxidation process.

[0108] Overall, the present application realizes precise control of the oxidation process of the oxidation furnace by intelligently adjusting the control parameters. This method not only improves the quality and uniformity of the oxide layer, but also enhances the flexibility and adaptability of the production process, which is of great significance for improving the performance and reliability of semiconductor devices. Through this dynamic adjustment strategy, the uncertainty in the production process can be better addressed, and the overall efficiency and quality of semiconductor manufacturing can be improved.

[0109] In a feasible implementation, the step of correcting the control parameter adjustment strategy according to the real-time monitored process environment factors includes:

[0110] When there is instability in the airflow in the furnace, according to the airflow deviation and the stability compensation time, the control parameter adjustment strategy is corrected to use a step-by-step correction adjustment strategy to control the parameters;

[0111] When the temperature distribution is uneven, according to the temperature deviation and the oxidation sensitivity of each region, the control parameter adjustment strategy is corrected to accurately adjust the control parameters;

[0112] When there is fluctuation in the pressure, according to the pressure change and the adjustment strategy of the pressure compensation time, the control parameter adjustment strategy is corrected to adjust the control parameters in real time.

[0113] It is necessary to note that the instability of the gas flow in the furnace refers to the abnormal or deviation of the gas flow inside the oxidation furnace from the predetermined pattern, which may affect the uniformity and efficiency of the oxidation process. The gas flow deviation refers to the degree of deviation between the actual gas flow and the predetermined gas flow path, which is used to evaluate the stability of the gas flow in the furnace. The stable compensation time refers to the time needed to restore the stability of the gas flow, which is used to adjust the control parameters to compensate for the deviation of the gas flow. The step-by-step correction adjustment strategy is a control parameter adjustment method that optimizes the gas flow by gradually adjusting rather than drastically changing when the gas flow is unstable, reducing the impact on the oxidation process. The temperature distribution is uneven, which means that the temperature in different areas inside the oxidation furnace is different, which may cause the quality of the oxidation layer to be inconsistent. The temperature deviation refers to the difference between the actual temperature and the set target temperature, which is used to evaluate and adjust the temperature control in the oxidation process. The oxidation sensitivity refers to the degree of response of different regions to temperature changes, which affects the oxidation rate and the quality of the oxidation layer. Precise adjustment of the control parameters is a control parameter adjustment method that optimizes the oxidation process by precisely adjusting parameters such as temperature when the temperature distribution is uneven. Pressure fluctuations refer to irregular changes in the pressure inside the oxidation furnace, which may affect the stability of the oxidation atmosphere. The pressure change refers to the difference between the actual pressure and the set pressure, which is used to evaluate and adjust the pressure control in the oxidation process. The pressure compensation time refers to the time needed to restore the stability of the pressure, which is used to adjust the control parameters to compensate for the change in pressure. Real-time adjustment of the control parameters is a control parameter adjustment method that maintains the stability of the oxidation process by adjusting parameters such as pressure in real time when the pressure fluctuates.

[0114] It can be understood that the present application describes a method of dynamically adjusting control parameters based on changes in the furnace environment in an oxidation furnace. The method intelligently adjusts control parameters to optimize the oxidation process by monitoring changes in gas flow, temperature and pressure in the furnace in real time.

[0115] When the instability of the gas flow in the furnace is detected, the device will use the step-by-step correction adjustment strategy to correct the control parameters according to the gas flow deviation and the stable compensation time. This method helps to smooth the gas flow changes and reduce the disturbance to the oxidation process, thereby improving the uniformity and quality of the oxidation layer.

[0116] During the oxidation process, if the temperature distribution is found to be uneven, the device will accurately adjust the control parameters according to the temperature deviation and the oxidation sensitivity of each region. This precise adjustment helps to ensure the consistency of the oxidation rate on the entire wafer surface and avoids quality problems of the oxidation layer caused by temperature differences.

[0117] When the pressure in the furnace fluctuates, the device will adjust the control parameters in real time according to the pressure change and the pressure compensation time. This real-time adjustment helps to maintain the stability of the oxidation atmosphere and ensure the continuity and uniformity of the oxidation process.

[0118] Overall, the present application realizes intelligent and dynamic adjustment of oxidation furnace control parameters by comprehensively considering the changes of gas flow, temperature and pressure in the furnace. This method not only improves the quality and uniformity of the oxidation layer, but also enhances the flexibility and adaptability of the production process, which is of great significance for improving the performance and reliability of semiconductor devices. Through this dynamic adjustment strategy, the uncertainty in the production process can be better addressed, and the overall efficiency and quality of semiconductor manufacturing can be improved.

[0119] The embodiment provides an oxidation furnace control parameter optimization method for chip manufacturing. By adjusting the control parameters, the oxidation degree of each wafer can be ensured to be consistent, the defects caused by uneven oxidation can be reduced, and the yield of the wafer can be improved. The oxidation prediction model can help avoid defects such as over-oxidation or insufficient oxidation, and improve the quality of the wafer. Optimizing the oxidation process can reduce material waste and production time, and reduce production cost. Through precise prediction and dynamic adjustment, the oxidation process of the wafer can be effectively controlled, thereby improving the yield of the oxidation process of the wafer.

[0120] In a feasible implementation manner, the step of obtaining the sensor parameters in the wafer oxidation process comprises the following steps:

[0121] Obtain historical sensor data samples and wafer oxidation result labels;

[0122] Input the historical sensor data samples into a preset to-be-trained model;

[0123] Predict the wafer oxidation process by using the to-be-trained model to determine the wafer oxidation prediction result;

[0124] Judge the difference between the wafer oxidation prediction result and the wafer oxidation result label to obtain a prediction loss value;

[0125] If the prediction loss value does not meet a preset loss standard, return to the step of predicting the wafer oxidation by using the to-be-trained model to determine the wafer oxidation prediction result until the training is stopped after the prediction loss value meets the loss standard, and an oxidation prediction model meeting the loss standard is obtained.

[0126] It is important to note that historical sensor data samples refer to the sensor data collected during past wafer oxidation processes, which record various parameters such as temperature, pressure, gas flow, etc. during the oxidation process, used to train the prediction model. Wafer oxidation result labels are labels of wafer oxidation results corresponding to historical sensor data samples, which are usually determined by a quality detection step, indicating whether the actual oxidation result meets the expected standard. The model to be trained refers to the model framework set before training begins, which has not yet learned historical data and needs to be optimized through the training process to improve its parameters and prediction ability. Wafer oxidation prediction results are expected wafer oxidation results analyzed by the model to be trained based on input historical sensor data samples, used for comparison with actual result labels. Prediction loss value refers to the difference between wafer oxidation prediction results and wafer oxidation result labels, used to evaluate the accuracy of model prediction. The pre-set loss standard refers to a threshold set during model training, when the prediction loss value is lower than this standard, it is considered that the prediction accuracy of the model has reached an acceptable level. Oxidation prediction model refers to the model that has been trained and meets the pre-set loss standard, which can accurately predict the results of wafer oxidation process.

[0127] It can be understood that the present application describes an oxidation prediction model trained based on historical data, used for predicting and optimizing wafer oxidation process. The method involves the following key steps:

[0128] Data collection: Collect historical sensor data samples and wafer oxidation result labels, which are the basis for training the prediction model.

[0129] Model training: Input historical sensor data samples into the model to be trained, and use these data to train the model, so that it can learn the relationship between parameters and results in the oxidation process.

[0130] Prediction and evaluation: Use the trained model to predict wafer oxidation process, and compare its prediction results with actual wafer oxidation result labels to calculate prediction loss value.

[0131] Model optimization: If the prediction loss value does not meet the pre-set loss standard, it means that the prediction accuracy of the model needs to be improved, and the model needs to be returned and continued to be trained until the prediction loss value meets the standard.

[0132] Model application: Once the model meets the pre-set loss standard, it can be used to predict wafer oxidation process in real time, guiding the adjustment of control parameters of oxidation furnace to optimize the quality of oxidation layer.

[0133] The advantage of this method is that it can improve the understanding and control of the oxidation process through machine learning technology, reduce the dependence on experience, improve production efficiency and product quality. By continuously optimizing the model, it can better cope with the variables in the production process, achieve more accurate process control, and ultimately improve the performance and reliability of semiconductor devices.

[0134] For example, to facilitate understanding of the implementation process of the chip manufacturing oxidation furnace control parameter optimization method obtained after combining the above-embodiment one, please refer to Figure 2 , Figure 2 A brief flowchart of a chip manufacturing oxidation furnace control parameter optimization method is provided, specifically:

[0135] Collect sensor data and corresponding oxidation result labels during wafer oxidation process for training oxidation prediction model. Input the collected data into a pre-set model which has not been trained. Use the to-be-trained model to predict the wafer oxidation process and obtain the prediction result. Compare the difference between the model prediction result and the actual oxidation result label, and calculate the prediction loss value. If the prediction loss value does not meet the pre-set standard, the step of using the to-be-trained model to predict the wafer oxidation process and obtaining the prediction result needs to be returned until the prediction loss value of the model meets the standard, at which time the obtained model can be used for actual wafer oxidation prediction. In the actual wafer oxidation process, real-time sensor parameters are obtained. Input the real-time sensor parameters obtained into the trained oxidation prediction model. According to the input sensor parameters, the model predicts the result of wafer oxidation. According to the prediction result, the control parameters of the oxidation furnace are dynamically adjusted. Calculate the deviation between the prediction result and the target value, and calculate the deviation of the control parameters to be adjusted accordingly. According to different stages of wafer oxidation (heating, oxidation, and cooling), adjust the adjustment strategy of the control parameters. According to the change trend of the prediction result (oxidation rate is too fast, too slow or stable), adjust the adjustment strategy of the control parameters. According to the real-time monitored process environment factors (such as unstable airflow, uneven temperature distribution, and pressure fluctuation), correct the control parameter adjustment strategy. Convert the calculated control parameter deviation into actual control parameter adjustment value. According to the actual control parameter adjustment value, dynamically adjust the control parameters of the oxidation furnace.

[0136] It should be noted that the above examples are only for understanding the present application and do not constitute a limitation on the chip manufacturing oxidation furnace control parameter optimization method of the present application. More simple transformations based on this technical concept are within the scope of protection of the present application.

[0137] The present application also provides a chip manufacturing oxidation furnace control parameter optimization device, please refer to Figure 3 , the chip manufacturing oxidation furnace control parameter optimization device comprises:

[0138] an acquisition module 10, configured to acquire a sensor parameter in a wafer oxidation process;

[0139] a prediction module 20, configured to input the sensor parameter into a preset oxidation prediction model to determine a wafer oxidation prediction result, wherein the oxidation prediction model is obtained by training a preset to-be-trained model by historical sensor data samples and wafer oxidation result labels;

[0140] an adjustment module 30, configured to dynamically adjust an oxidation furnace control parameter according to the wafer oxidation prediction result.

[0141] And / or, the oxidation furnace control parameter optimization device for chip manufacturing comprises:

[0142] a first acquisition module, configured to acquire historical sensor data samples and wafer oxidation result labels;

[0143] a first input module, configured to input the historical sensor data samples into a preset to-be-trained model;

[0144] a first determination module, configured to determine a wafer oxidation prediction result by predicting a wafer oxidation process through the to-be-trained model;

[0145] a first judgment module, configured to judge a difference between the wafer oxidation prediction result and the wafer oxidation result label to obtain a prediction loss value;

[0146] a first loop module, configured to return to the step of determining a wafer oxidation prediction result by predicting a wafer oxidation through the to-be-trained model if the prediction loss value does not meet a preset loss standard, and stop training until the prediction loss value meets the loss standard to obtain an oxidation prediction model meeting the loss standard.

[0147] And / or, the adjustment module 30 comprises:

[0148] a first calculation module, configured to calculate a corresponding control parameter adjustment deviation according to a deviation between the wafer oxidation prediction result and a preset oxidation target value;

[0149] a first adjustment module, configured to adjust a preset control parameter adjustment strategy according to a sensitivity of a control parameter in different stages of a wafer oxidation process;

[0150] a second adjustment module, configured to adjust the control parameter adjustment strategy according to a change trend of the wafer oxidation prediction result;

[0151] a first correction module, configured to correct the control parameter adjustment strategy according to a real-time monitored process environment factor;

[0152] a third adjusting module, configured to convert the control parameter adjustment deviation into an actual control parameter adjustment value according to the corrected control parameter adjustment strategy;

[0153] a fourth adjusting module, configured to dynamically adjust the oxidation furnace control parameter according to the actual control parameter adjustment value.

[0154] and / or the first adjusting module comprises:

[0155] a second determining module, configured to determine a current wafer oxidation stage;

[0156] a first correcting module, configured to adjust the preset control parameter adjustment strategy to a temperature deviation if the wafer oxidation stage is a heating stage, and to adopt a large-amplitude adjustment strategy for fast correction;

[0157] a second correcting module, configured to adjust the control parameter adjustment strategy to focus on control temperature and oxygen partial pressure if the wafer oxidation stage is an oxidation stage, and to accurately adjust the deviations of temperature and oxygen partial pressure according to different periods of the oxidation stage;

[0158] a third correcting module, configured to adjust the control parameter adjustment strategy to focus on control temperature and oxygen partial pressure if the wafer oxidation stage is a cooling stage, and to adopt a small-amplitude adjustment strategy for slow correction.

[0159] and / or the second adjusting module comprises:

[0160] a fourth correcting module, configured to adjust the control parameter adjustment strategy to adopt a gradually decreasing adjustment strategy according to an expected remaining time of oxidation if the wafer oxidation prediction result is that the oxidation rate is too fast;

[0161] a fifth correcting module, configured to adjust the control parameter adjustment strategy to adopt a gradually increasing adjustment strategy according to an expected remaining time of oxidation if the wafer oxidation prediction result is that the oxidation rate is too slow;

[0162] a sixth correcting module, configured to adjust the control parameter adjustment strategy to maintain the existing control parameter adjustment value if the wafer oxidation prediction result is that the oxidation process is in stable operation.

[0163] and / or the first correcting module comprises:

[0164] a seventh correcting module, configured to correct the control parameter adjustment strategy to adopt a gradually correcting adjustment strategy according to a gas flow deviation and a stable compensation time if there is an unstable phenomenon in the gas flow in the furnace;

[0165] An eighth correction module is configured to correct the control parameter adjustment strategy to accurate adjustment of the control parameter according to the temperature deviation and the oxidation sensitivity of each region when the temperature distribution is uneven.

[0166] A ninth correction module is configured to correct the control parameter adjustment strategy to real-time adjustment of the control parameter according to the pressure change and the adjustment strategy of the pressure compensation time when the pressure fluctuates.

[0167] The chip manufacturing oxidation furnace control parameter optimization device provided by the present application adopts the chip manufacturing oxidation furnace control parameter optimization method in the above embodiment, and can solve the technical problem of low yield of the produced wafer. Compared with the prior art, the chip manufacturing oxidation furnace control parameter optimization device provided by the present application has the same beneficial effects as the chip manufacturing oxidation furnace control parameter optimization method provided by the above embodiment, and other technical features in the chip manufacturing oxidation furnace control parameter optimization device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0168] The present application provides a chip manufacturing oxidation furnace control parameter optimization device, which comprises at least one processor and a memory in communication connection with the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the chip manufacturing oxidation furnace control parameter optimization method in the above embodiment one.

[0169] Reference will now be made to Figure 4 which shows a structural schematic diagram of a chip manufacturing oxidation furnace control parameter optimization device suitable for implementing the embodiments of the present application. The chip manufacturing oxidation furnace control parameter optimization device in the embodiments of the present application can include but is not limited to mobile terminals such as mobile phones, tablet computers, notebook computers, digital broadcast receivers, PDAs (Personal Digital Assistant), PMPs (Portable Media Player), vehicle terminals (such as vehicle navigation terminals), and the like, and fixed terminals such as digital TVs, desktop computers, and the like. Figure 4 The chip manufacturing oxidation furnace control parameter optimization device shown is only an example, and should not bring any limitation to the functions and use range of the embodiments of the present application.

[0170] As Figure 4As shown, the chip manufacturing oxidation furnace control parameter optimization device can include a processing device 1001 (for example, a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM) 1002 or programs loaded from a storage device 1003 to a random access memory (RAM) 1004. In the RAM 1004, various programs and data required for the operation of the chip manufacturing oxidation furnace control parameter optimization device are also stored. The processing device 1001, the ROM 1002, and the RAM 1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; the storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the chip manufacturing oxidation furnace control parameter optimization device to communicate with other devices wirelessly or by wire to exchange data. Although the chip manufacturing oxidation furnace control parameter optimization device with various systems is shown in the figure, it should be understood that all the systems shown are not required to be implemented or possessed. More or fewer systems can be alternatively implemented or possessed.

[0171] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product comprising a computer program carried on a computer readable medium, the computer program containing program codes for executing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network through a communication device, or installed from the storage device 1003, or installed from the ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the methods of the embodiments disclosed in the present application are performed.

[0172] The chip manufacturing oxidation furnace control parameter optimization device provided by the present application adopts the chip manufacturing oxidation furnace control parameter optimization method in the above-mentioned embodiments, and can solve the technical problem of low yield of wafer production. Compared with the prior art, the chip manufacturing oxidation furnace control parameter optimization device provided by the present application has the same beneficial effects as the chip manufacturing oxidation furnace control parameter optimization method provided by the above-mentioned embodiments, and other technical features in the chip manufacturing oxidation furnace control parameter optimization device are the same as the features disclosed in the previous embodiment method, which will not be repeated here.

[0173] It is to be understood that the various parts of the present application can be realized by hardware, software, firmware or any combination thereof. In the description of the foregoing embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0174] The above description is merely illustrative of the application and is not intended to limit the scope of the application. Any variations and modifications that can be made by any person skilled in the art within the spirit and scope of the application are intended to be encompassed by the claims of the application. The scope of the application is defined by the claims.

[0175] The present application provides a computer readable storage medium having stored thereon computer readable program instructions (i.e., a computer program) for performing the chip manufacturing oxidation furnace control parameter optimization method in the above embodiments.

[0176] The computer readable storage medium provided by the present application may, for example, be a U disk, but is not limited to an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, system, or device, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more conductive wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present embodiment, the computer readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer readable storage medium can be transmitted in any suitable medium, including but not limited to electrical wire, optical cable, RF (Radio Frequency), etc., or any suitable combination of the above.

[0177] The above computer readable storage medium can be included in the chip manufacturing oxidation furnace control parameter optimization device; or can exist separately and not be assembled into the chip manufacturing oxidation furnace control parameter optimization device.

[0178] The computer readable storage medium described above carries one or more programs, when the one or more programs are executed by the chip manufacturing oxidation furnace control parameter optimization device, the chip manufacturing oxidation furnace control parameter optimization device is caused to: acquire sensor parameters in a wafer oxidation process;

[0179] input the sensor parameters into a preset oxidation prediction model to determine a wafer oxidation prediction result, wherein the oxidation prediction model is obtained by training a preset to-be-trained model by historical sensor data samples and wafer oxidation result labels;

[0180] According to the wafer oxidation prediction result, dynamically adjust the oxidation furnace control parameters.

[0181] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object-oriented programming language such as Java, Smalltalk, C++, or conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0182] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present application. In this regard, each block in the flowcharts or block diagrams can represent a module, a program segment, or a portion of code that contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur in different orders than that shown in the accompanying drawings. For example, two blocks that are shown in succession can actually be executed substantially in parallel, or they can be executed in reverse order, depending on the involved functions. It should also be noted that each block in the block diagrams and / or flowcharts, and the combination of blocks in the block diagrams and / or flowcharts, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0183] The modules described in the embodiments of the present application can be implemented in the form of software or in the form of hardware. In some cases, the names of the modules do not constitute a limitation on the modules themselves.

[0184] The readable storage medium provided by the present application is a computer readable storage medium, which stores computer readable program instructions (i.e., a computer program) for executing the chip manufacturing oxidation furnace control parameter optimization method described above, and can solve the technical problem of low yield of produced wafers. Compared with the prior art, the computer readable storage medium provided by the present application has the same beneficial effects as the chip manufacturing oxidation furnace control parameter optimization method provided by the above-mentioned embodiments, and will not be described here.

[0185] The present application also provides a computer program product comprising a computer program, which, when executed by a processor, implements the steps of the chip manufacturing oxidation furnace control parameter optimization method as described above.

[0186] The computer program product provided by the present application can solve the technical problem of low yield of produced wafers. Compared with the prior art, the computer program product provided by the present application has the same beneficial effects as the chip manufacturing oxidation furnace control parameter optimization method provided by the above-mentioned embodiments, and will not be described here.

[0187] The above only describes some embodiments of the present application, and does not limit the patent scope of the present application. Any equivalent structural transformation, direct / indirect application in other related technical fields based on the technical concept of the present application, and the content of the present application specification and drawings are included in the patent protection scope of the present application.

Claims

1. A method for optimizing control parameters of an oxidation furnace for chip manufacturing, characterized by, The method comprises: obtaining sensor parameters in a wafer oxidation process; inputting the sensor parameters into a preset oxidation prediction model to determine a wafer oxidation prediction result, wherein the oxidation prediction model is obtained by training a preset to-be-trained model by historical sensor data samples and wafer oxidation result labels, the oxidation prediction model can learn the relationship between oxidation parameters and oxide layer quality, and can make a prediction according to the relationship between the oxidation parameters and the oxide layer quality; dynamically adjusting oxidation furnace control parameters according to the wafer oxidation prediction result; the step of dynamically adjusting the oxidation furnace control parameters comprises: calculating a corresponding control parameter adjustment deviation according to the deviation between the wafer oxidation prediction result and a preset oxidation target value; adjusting a preset control parameter adjustment strategy according to the sensitivity of the control parameters in different stages of the wafer oxidation process; adjusting the control parameter adjustment strategy according to the change trend of the wafer oxidation prediction result; correcting the control parameter adjustment strategy according to real-time monitored process environmental factors; the step of correcting the control parameter adjustment strategy according to real-time monitored process environmental factors comprises: when detecting that the gas flow in the oxidation furnace is abnormal or deviates from the predetermined mode, correcting the control parameter adjustment strategy to a step-by-step correction adjustment strategy control parameter according to the deviation between the actual gas flow and the predetermined gas flow path and the time required to restore the gas flow to be stable, and optimizing the gas flow by step-by-step adjustment instead of sharp change; when the temperature distribution is uneven, correcting the control parameter adjustment strategy to an accurate adjustment control parameter according to the temperature deviation and the response degree of each region to temperature change; when the pressure fluctuates, correcting the control parameter adjustment strategy to a real-time adjustment control parameter according to the pressure change and the pressure compensation time adjustment strategy; converting the control parameter adjustment deviation into an actual control parameter adjustment value according to the corrected control parameter adjustment strategy; dynamically adjusting the oxidation furnace control parameters according to the actual control parameter adjustment value.

2. The method of claim 1, wherein, the step of obtaining sensor parameters in a wafer oxidation process comprises: obtaining historical sensor data samples and wafer oxidation result labels; inputting the historical sensor data samples into a preset to-be-trained model; determining a wafer oxidation prediction result by predicting the wafer oxidation process through the to-be-trained model; judging the difference between the wafer oxidation prediction result and the wafer oxidation result label to obtain a prediction loss value; if the prediction loss value does not meet a preset loss standard, returning to the step of determining a wafer oxidation prediction result by predicting the wafer oxidation through the to-be-trained model until the training is stopped after the prediction loss value meets the loss standard to obtain an oxidation prediction model meeting the loss standard.

3. The method of claim 1, wherein, the step of adjusting a preset control parameter adjustment strategy according to the sensitivity of the control parameters in different stages of the wafer oxidation process comprises: determining the current wafer oxidation stage; if the wafer oxidation stage is a heating stage, adjusting the preset control parameter adjustment strategy to a temperature deviation and adopting a large-amplitude adjustment strategy for rapid correction; If the wafer oxidation stage is an oxidation stage, the control parameter adjustment strategy is adjusted to focus on controlling temperature and oxygen partial pressure, and the deviation of temperature and oxygen partial pressure is accurately adjusted according to different periods of the oxidation stage; If the wafer oxidation stage is a cooling stage, the control parameter adjustment strategy is adjusted to focus on the cooling rate of the controlled temperature, and a small adjustment strategy is adopted for slow correction.

4. The method of claim 1, wherein, The step of adjusting the control parameter adjustment strategy according to the change trend of the wafer oxidation prediction result comprises: If the wafer oxidation prediction result is that the oxidation rate is too fast, the control parameter adjustment strategy is adjusted to be a gradually decreasing adjustment strategy according to the expected remaining time of oxidation; If the wafer oxidation prediction result is that the oxidation rate is too slow, the control parameter adjustment strategy is adjusted to be a gradually increasing adjustment strategy according to the expected remaining time of oxidation; If the wafer oxidation prediction result is that the oxidation process is stable, the control parameter adjustment strategy is adjusted to maintain the existing control parameter adjustment value.

5. A device for optimizing control parameters of an oxidation furnace used in chip manufacturing, characterized in that, The device comprises: an acquisition module for acquiring sensor parameters in a wafer oxidation process; a prediction module for inputting the sensor parameters into a preset oxidation prediction model to determine a wafer oxidation prediction result, wherein the oxidation prediction model is obtained by training a preset to-be-trained model with historical sensor data samples and wafer oxidation result labels, the oxidation prediction model can learn the relationship between oxidation parameters and oxidation layer quality, and can make predictions according to the relationship between oxidation parameters and oxidation layer quality; an adjustment module for dynamically adjusting oxidation furnace control parameters according to the wafer oxidation prediction result; The step of dynamically adjusting the oxidation furnace control parameters comprises: calculating a corresponding control parameter adjustment deviation according to the deviation between the wafer oxidation prediction result and a preset oxidation target value; adjusting a preset control parameter adjustment strategy according to the sensitivity of control parameters in different stages of the wafer oxidation process; adjusting the control parameter adjustment strategy according to the change trend of the wafer oxidation prediction result; correcting the control parameter adjustment strategy according to real-time monitored process environmental factors; The step of correcting the control parameter adjustment strategy according to real-time monitored process environmental factors comprises: when detecting that the gas flow in the oxidation furnace is abnormal or deviates from the predetermined mode, correcting the control parameter adjustment strategy to be a gradually correcting adjustment strategy for controlling parameters by gradually adjusting rather than sharply changing to optimize the gas flow according to the deviation between the actual gas flow and the predetermined gas flow path and the time required to restore the gas flow to be stable; when the temperature distribution is uneven, correcting the control parameter adjustment strategy to be an accurate adjustment of the control parameters according to the temperature deviation and the response degree of each region to the temperature change; when the pressure fluctuates, correcting the control parameter adjustment strategy to be a real-time adjustment of the control parameters according to the pressure change and the adjustment strategy of the pressure compensation time; converting the control parameter adjustment deviation into an actual control parameter adjustment value according to the corrected control parameter adjustment strategy; dynamically adjusting the oxidation furnace control parameters according to the actual control parameter adjustment value.

6. A device for optimizing control parameters of an oxidation furnace used in chip manufacturing, characterized in that, The device comprises a memory, a processor, and a computer program stored on the memory and executable on the processor, and the computer program is configured to implement the steps of the control parameter optimization method for the oxidation furnace used in chip manufacturing according to any one of claims 1 to 4.

7. A storage medium, characterized by The storage medium is a computer readable storage medium, and the storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the control parameter optimization method for the oxidation furnace used in chip manufacturing according to any one of claims 1 to 4.

8. A computer program product, characterised in that, The computer program product comprises a computer program, and the computer program is executed by a processor to implement the steps of the control parameter optimization method for the oxidation furnace used in chip manufacturing according to any one of claims 1 to 4.

Images