A method and system for automatically adjusting PID parameters in a process

Abstract

A method and system for automatically adjusting PID parameters in a process, belonging to the field of semiconductor processing technology. The method and system cyclically simulate wafer transfer and process flow under certain PID parameters, obtain process indicators such as the temperature fluctuation of the debug wafer in the process, the average temperature in the process, power fluctuation, and power output situation, automatically adjust the PID parameter values according to the performance of the process indicators, and conduct further cyclic tests on the debug wafer based on the adjusted parameters. Iteratively optimize in this way until the optimal PID parameters are automatically found. This method can quickly and automatically find the optimal PID parameters, significantly reducing the workload of manual parameter adjustment.

Description

Technical Field

[0001] The present invention relates to the field of semiconductor processing technology, and particularly relates to a method and system for automatically adjusting PID parameters during a process. Background Art

[0002] Semiconductor processing equipment includes semiconductor thin film deposition equipment, plasma stripping equipment, plasma etching equipment, furnace tube equipment, etc.

[0003] The temperature control in the process of existing semiconductor processing equipment mainly adopts PID (proportion integration differentiation) control. Some processes have strict requirements for temperature fluctuations, and tiny temperature changes have a great impact on the process results; when some parameters in the process change (such as time, chamber pressure, gas flow rate, etc.), it will affect the temperature fluctuation situation. Therefore, for a certain process menu, how to automatically find the most suitable PID to make the temperature fluctuation situation in the process meet the process requirements is particularly important. Summary of the Invention

[0004] The present disclosure provides a method and system for automatically adjusting PID parameters during a process. By automatically adjusting the PID parameters and cyclically testing the wafers for debugging, the PID parameters most suitable for a certain process condition can be quickly and automatically found, so that each system parameter meets the process requirements.

[0005] The method for automatically adjusting PID parameters during a process provided by the present disclosure mainly includes the following steps:

[0006] S1, Remove the main reaction gas in the process and replace it with an inert gas with the same flow rate to ensure that no reaction occurs on the surface of the wafer for debugging.

[0007] S2, Place a group of wafers for debugging on the machine table and set the initial PID parameters.

[0008] S3, Based on the set parameters, simulate at least one round of wafer transfer and process flow during machine production, and obtain the process indexes of each wafer for debugging in the process, including: temperature fluctuation situation, average temperature during the process, power fluctuation, and power output situation.

[0009] S4, Determine whether the process index data obtained based on the current PID parameters meets the process requirements, including:

[0010] Whether the temperature fluctuation situation of each wafer meets the set standard requirements;

[0011] Whether the deviation between the average temperature during the process of each wafer and the set value is within the set standard;

[0012] Whether the power fluctuation is within the set standard;

[0013] Whether there is no limit state of 100% or 0% power output in the power output;

[0014] S5. If one of the above indicators does not meet the requirements, enter the PID automatic adjustment process to automatically adjust the PID parameters. After each parameter adjustment, repeat step S3 for simulation testing;

[0015] Thus, the PID parameter values that meet the process requirements for each indicator are obtained.

[0016] Furthermore, the automatic adjustment process in step S5 specifically includes:

[0017] First, gradually adjust the value of one of the P, I, and D parameters within the set adjustment range, and set the other two parameters as fixed values;

[0018] After each parameter adjustment, repeat step S3;

[0019] Select the value that makes the process indicators of each debug wafer meet the process requirements;

[0020] Then, for the other two types of parameters in the P, I, and D parameters, also use the same method respectively to obtain the values that make the process indicators of the debug wafer meet the process requirements; among the two parameters set as fixed values, for the parameter whose value has been determined in the previous test, take that parameter value;

[0021] Through the above cyclic and iterative tests, the PID parameter values that meet the process requirements for each process indicator are obtained.

[0022] Furthermore, in step S5, the gradual adjustment method includes any one of the following: gradual increase or decrease, and the dichotomy method.

[0023] Furthermore, when using the dichotomy method for automatic parameter adjustment in step S5, the specific process includes:

[0024] ① Assume that the current value of the parameter to be adjusted is a, then take values a1 and a2 respectively, where a1 < a < a2, and repeat step S3 for further testing;

[0025] ② Assume that the process indicator is better at a1 than at a, and worse at a2 than at a, then continue to take a3 and a4 for testing, where a4 < a l < a3 < a, a3 = (a l + a) / 2, a4 = 2a1 - a3;

[0026] ③If it is better at a4, then continue to search in a smaller direction than a4 in the same method and trend;

[0027] If it performs better at a1, then the optimal value is between a4 and a3. Select a5 < a1 < a6 < a3, where a5 = (a4 + a1) / 2 and a6 = (a1 + a3) / 2;

[0028] If it is better at a3, then the optimal value is between a1 and a. Select a5 < a3 < a6, where a5 = (a1 + a3) / 2 and a6 = (a3 + a) / 2;

[0029] ④And so on until the parameter value that meets the process requirements is found.

[0030] Furthermore, in the step S5:

[0031] After each change of the PID parameters, it is necessary to idle for 10 minutes. After the temperature fluctuation in the idle state is < ±0.3°C, the debug wafer can be used for cyclic testing;

[0032] If the temperature fluctuation in the idle state > ±0.3°C after idling for 10 minutes, then the current PID is not applicable and the debug wafer cyclic testing will not continue.

[0033] A system for automatically adjusting PID parameters in the process of applying the above method mainly includes:

[0034] A debug wafer cyclic testing device, which is used to simulate the wafer transfer and process flow during machine production. During testing, the main reaction gas in the process is replaced by an equal-flow inert gas;

[0035] A PID parameter automatic adjustment module: used to automatically adjust the value of the PID parameters within the set range according to the above method, so that the adjusted process indicators meet the set process requirements.

[0036] Compared with the prior art, the beneficial effects of the present disclosure are:

[0037] (1) Replace the main reaction gas in the process with an equal-flow inert gas to ensure that no reaction occurs on the surface of the debug wafer and realize the cyclic utilization of the wafer; (2) Use the adjustment trend chart and then combine the dichotomy method to automatically find the most suitable PID parameters, so that the temperature fluctuation situation in the process meets the process requirements; (3) Through a logical judgment system that judges whether the automatically adjusted PID value meets the requirements and whether further adjustment is needed, realize the progressive screening of the PID parameter values, and finally obtain the optimal parameter values; (4) The method and system have low implementation difficulty and good engineering applicability. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The above and other objects, features, and advantages of the present disclosure will become more apparent by describing the exemplary embodiments of the present disclosure in more detail with reference to the accompanying drawings, wherein, in the exemplary embodiments of the present disclosure, the same reference numerals generally represent the same components.

[0039] Figure 1 It is a flowchart for automatically adjusting the PID parameter value according to the present disclosure;

[0040] Figure 2 It is an example of temperature and power index data when testing parameter P;

[0041] Figure 3 It is an example of temperature and power index data when testing parameter D. Detailed implementation manners

[0042] The preferred embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the preferred embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to make the present disclosure more thorough and complete, and to fully convey the scope of the present disclosure to those skilled in the art.

[0043] The present disclosure provides a method and system for automatically adjusting PID parameters during a process. Through the cyclic testing of debug wafers, the PID parameters most suitable for a certain process condition can be quickly and automatically found, so that the temperature fluctuation situation meets the process requirements.

[0044] Test conditions: Simulate the wafer transfer and process flow during machine production. The main reaction gas in the process can be removed and replaced with an inert gas of equal flow rate to ensure that no reaction (such as depositing a thin film or being etched) occurs on the surface of the debug wafer, so as to achieve recyclability.

[0045] In an exemplary embodiment, the process of automatically adjusting the PID parameter value according to the temperature fluctuation is as follows:

[0046] Step1: Place 3 - 25pcs of prepared debug wafers on the machine;

[0047] Step2: Set the PID parameter I to a default value such as 80, and set the D value to a default value such as 30;

[0048] Step3: Adjust parameter P, select an adjustment range such as 10 - 50, and take values at intervals of 5 (P takes 10 - 15 - 20 - 25 - 30 - 35 - 40 - 45 - 50). At this time, the D value can be default set to 30; Conduct cyclic testing on the debug wafers. After each run, check the temperature fluctuation situation of each wafer in the data record, record the change trend of the temperature fluctuation with the P value, and select the P value with the smallest temperature fluctuation;

[0049] Step 4: Parameter D adjustment. Select an adjustment range, such as 10 - 50, with an interval of 5 (D takes 10 - 15 - 20 - 25 - 30 - 35 - 40 - 45 - 50). At this time, the P value is the P value determined in Step 3. Use the debug wafer for cyclic testing. After each run, check the temperature fluctuation of each wafer in the data record, record the change trend of the temperature fluctuation with the D value, and select the D value with the smallest temperature fluctuation.

[0050] Step 5: Parameter I adjustment. Select an adjustment range, such as 20 - 200, with an interval of 20 (I takes 20 - 40 - 60 - 80 - 100 - 120 - 140 - 160 - 200). At this time, the P and D values are the P and D values determined in Step 4. Use the prepared debug wafer for cyclic testing. After each run, check the temperature fluctuation of each wafer in the data record, record the change trend of the temperature fluctuation with the D value, and select the D value with the smallest temperature fluctuation.

[0051] Step 6: If higher requirements for temperature fluctuation are required, use the bisection method to adjust for better parameters near the PID determined above. For example, if the best P / D value determined above is 30, then take 28 and 33 for testing (25 - 28 - 30 - 33 - 35). If 28 is better than 30 and the temperature fluctuation of 33 is greater than that of 30, then continue to take 26 / 27 and 29 for testing until the best PID value is found.

[0052] In this embodiment, the adjustment targets are defined as follows: the temperature fluctuation < ±1.5°C; the average temperature during the process deviates from the set value < 1.5°C.

[0053] It should be noted that after changing the PID parameters, it is necessary to idle for about 10 minutes. After the temperature fluctuates stably < ±0.3°C during idle, the debug wafer can be used for cyclic testing. If the temperature fluctuation during idle is > ±0.3°C after 10 minutes of idle, then the current PID is not applicable and there is no need to do the debug wafer cyclic testing anymore.

[0054] Based on the above method, continuously improving the parameter adjustment standard can find the optimal PID parameters. As shown in the appendix Figure 1 shown, the automatic adjustment of PID needs to judge in turn:

[0055] Whether the temperature fluctuation is within the defined standard,

[0056] Whether the deviation between the average temperature and the set value during the process is within the defined standard,

[0057] Whether the power fluctuation is within the defined standard,

[0058] And whether there is an extreme state of 100% power output or 0% power output in the power output;

[0059] When the parameters of the PID meet the current standard, adjustment is made for the next standard on this basis.

[0060] In this embodiment, the temperature and power index data examples when testing parameters P and D respectively are shown in the appendix Figure 2 and 3 as shown.

[0061] In different processing scenarios, users have different concerns about process indicators. Some require the best temperature fluctuation and the temperature average within a certain range; some require the smallest deviation of the temperature average and the temperature fluctuation within a certain range. When comparing which parameter makes the process indicators better, it is determined according to the user's concern about process indicators in the current processing scenario.

[0062] A system for automatically adjusting PID parameters in the process of applying the above method mainly includes:

[0063] A debug wafer cycle test device, which is used to simulate the wafer transfer and process flow during machine production. Among them, the main reaction gas in the process is removed and replaced with an inert gas with the same flow rate to ensure that no reaction occurs on the surface of the debug wafer, so as to achieve recyclability;

[0064] A PID parameter adjustment module: used to adjust the values of parameters P, I, and D at set intervals within a set range. The adjusted parameters are used for the cycle test of the debug wafer; collect the system process indicators during the cycle test of the debug wafer, and select the PID parameter values that meet the set standards. The process indicators include one or more of the temperature fluctuation situation in the process, the deviation amount between the average temperature and the set value, the power fluctuation situation, and the power output situation.

[0065] This system provides a recipe, that is, a process menu, and the processing process can be freely selected.

[0066] The above technical solutions are only exemplary embodiments of the present invention. For those skilled in the art, on the basis of the disclosed application methods and principles of the present invention, it is very easy to make various types of improvements or deformations, not limited to the methods described in the above specific embodiments of the present invention. Therefore, the above-described manner is only preferred and does not have a restrictive meaning.

Claims

1. A method for automatically adjusting PID parameters in a technological process, characterized in that, It includes the following steps: S1. Remove the main reaction gas in the process and replace it with an inert gas of equal flow rate to ensure that no reaction occurs on the surface of the wafers for debugging. S2. Place a set of wafers for debugging on the machine and set the initial PID parameters. S3. Based on the set parameters, simulate at least one round of wafer transfer and process flow during machine production, and obtain the process indicators of each wafer for debugging in the process, including: temperature fluctuation, average temperature during the process, power fluctuation, and power output. S4. Determine whether the process indicator data obtained based on the current PID parameters meets the process requirements, including: Whether the temperature fluctuation of each wafer meets the set standard requirements; Whether the deviation between the average temperature during the process of each wafer and the set value is within the set standard; Whether the power fluctuation is within the set standard; Whether there is no limit state of 100% or 0% power output in the power output; S5. If one of the indicators does not meet the requirements, enter the PID automatic adjustment process to automatically adjust the PID parameters. After each parameter adjustment, repeat step S3 for simulation testing. Thus, the PID parameter values that meet the process requirements for all indicators are obtained.

2. The method according to claim 1, wherein The automatic adjustment process in step S5 specifically includes: First, gradually adjust the value of one of the P, I, and D parameters within the set adjustment range, and set the other two parameters to fixed values. After each parameter adjustment, repeat step S3. Select the value that makes the process indicators of each wafer for debugging meet the process requirements. Then, for the other two types of parameters in the P, I, and D parameters, also use the same method to obtain the values that make the process indicators of the wafers for debugging meet the process requirements. Among the two parameters set to fixed values, for the parameter whose value has been determined in the previous test, take that parameter value. Through the above loop and iterative testing, the PID parameter values that meet the process requirements for all process indicators are obtained.

3. The method according to claim 2, characterized in that, In step S5, the gradual adjustment method includes any one of the following: gradual increase or decrease, and the dichotomy method.

4. The method according to claim 3, wherein In step S5, when using the dichotomy method for automatic parameter adjustment, the specific process includes: Assume that the current value of the parameter to be adjusted is a, then take values a1 and a2 respectively, where a1 < a < a2, and repeat step S3 for further testing. Suppose the process index at a1 is better than that at a, and the one at a2 is worse than that at a. Then continue to take tests at a3 and a4, where a4 < a l < a3 < a, a3 = (a l + a) / 2, a4 = 2a1 - a3; If it is better at a4, then continue to search in a smaller direction than a4 in the same method and trend. If it performs better at a1, then the optimal value is between a4 and a3. Select a5 < a1 < a6 < a3, where a5 = (a4 + a1) / 2 and a6 = (a1 + a3) / 2. If it is better at a3, then the optimal value is between a1 and a. Select a5 < a3 < a6, where a5 = (a1 + a3) / 2 and a6 = (a3 + a) / 2. And so on until the parameter value that makes the process indicators meet the requirements is found.

5. The method according to any one of claims 1-4, characterized in that, In step S5: After each change of the PID parameters, it is necessary to idle for 10 minutes. After the temperature fluctuation in the idle state < ±0.3°C, the wafer cycle test for debugging can be carried out. If the temperature fluctuation in the idle state is > ±0.3 °C after 10 minutes of idling, the current PID is not applicable and the debugging wafer cycle test will no longer be continued using it.

6. A system for automatically adjusting PID parameters in a process using any of the methods described in claims 1-5, characterized in that, Including: A debugging wafer cycle test device, which is used to simulate the wafer transfer and process flow during machine production. During the test, the main reaction gas in the process is replaced by an equal-flow inert gas; A PID parameter automatic adjustment module: used to automatically adjust the value of the PID parameter within the set range according to any one of the methods described in claims 1-5, so that the adjusted process indicators meet the set process requirements.

Images