Method, apparatus, device, medium and program for controlling semiconductor process equipment

Abstract

The application relates to the field of semiconductor manufacturing process control, and provides a control method, device, equipment, medium and program of a semiconductor process equipment, the method comprising the following steps: acquiring a configuration file and a diagnosis time period of a target machine, and extracting N process variables; aggregating the process variables into M control groups (M < N) by specifying process variable correlation; identifying basic process variables in each control group, extracting corresponding basic process variables in a database, and determining a sub time period in the diagnosis time period; extracting non-basic process variables of non-basic process variables based on each sub time period; analyzing whether the non-basic process variables meet boundary conditions in the configuration file, and updating a closed-loop control model; and feeding back proportional gain and bias items output by the model to a control module, so as to optimize subsequent control response. The method fully utilizes configuration file information, realizes flexible adjustment of a closed-loop control strategy, and improves the control precision and stability of the semiconductor process equipment.

Description

Technical Field

[0001] The present invention relates to the field of semiconductor manufacturing process control, and in particular to a control method, device, equipment, medium and program for semiconductor process equipment. Background Art

[0002] In semiconductor manufacturing, the operational stability and control accuracy of process equipment significantly impact product yield and production efficiency. As process nodes continue to shrink, the demand for more sophisticated equipment control becomes increasingly stringent. Typically, semiconductor process equipment incorporates multiple sensors, actuators, and control system modules, each of which interacts in complex ways, forming a multivariable control system.

[0003] Currently, most equipment control systems employ closed-loop control strategies based on fixed parameters, relying on preset target values ​​and feedback signals for regulation. However, in actual operation, due to variations in equipment status, environmental factors, and process conditions, traditional control methods struggle to maintain consistently high levels of control accuracy and responsiveness. Furthermore, the vast amount of process data generated during equipment operation is often underutilized, lacking systematic analytical tools to support the optimization and adjustment of control models.

[0004] Therefore, there is an urgent need for a control method, device, equipment, medium and program for semiconductor process equipment to improve the above problems. Summary of the Invention

[0005] The present invention provides a control method, apparatus, device, medium and program for semiconductor process equipment. The method makes full use of a configuration file to flexibly adjust a closed-loop control strategy to optimize the closed-loop control of the semiconductor process equipment.

[0006] According to a first aspect of an embodiment of the present invention, a control method for semiconductor process equipment is provided, comprising: obtaining a configuration file and a diagnostic time period of a target machine; the configuration file includes data types and a closed-loop control model of process variables of the target machine; the closed-loop control model has preset boundary conditions; extracting N process variables to be analyzed based on the configuration file and the diagnostic time period, where N is a positive integer; specifying the correlation between the process variables based on a control relationship pre-set in the configuration file; aggregating the relevant process variables in a single closed-loop control loop to generate M control groups, where M is a positive integer less than N; identifying the data types of basic process variables in each control group based on the data types, extracting the basic process variables from a database, and determining the time period in which the basic process variables change based on the time period in which the basic process variables change. The sub-time period to be analyzed in the diagnostic time period; the basic process variable is the target value or set value of the closed-loop control loop of the target machine; based on each sub-time period, the non-basic process variables in the database are extracted; the non-basic process variables are analyzed to see whether they meet the boundary conditions to obtain the judgment results; the non-basic process variables include the actual value and output value of the closed-loop control model, and the average value of the actual value and output value of each non-basic variable in each sub-time period when the boundary conditions are met is counted, and then regression analysis is performed based on the average value to update the closed-loop control model; the proportional gain and bias term of the output of the updated closed-loop control model are fed back to the control module so that the control module calculates the output value based on the proportional gain and bias term when the input value is obtained next time; the input value is the basic process variable, and the output value is the control parameter of the target machine.

[0007] Optionally, the method further includes: outputting a report file to a host computer, where the report file includes the determination results of each sub-time period.

[0008] Optionally, the method further includes controlling the host computer to generate visualization interface data according to the content of the report file; and outputting the visualization interface data to the human-computer interaction module so that the human-computer interaction module displays the determination results of each sub-time period.

[0009] Optionally, the basic process variables include process setpoints over the complete diagnostic time period.

[0010] Optionally, non-basic process variables include the actual process value, process approach time and steady-state fluctuation range corresponding to each sub-time period; the process approach time is calculated based on the time required for the actual process value to reach the target fluctuation range of the process set value; the steady-state fluctuation range is calculated based on the difference between the process set value and the actual process value.

[0011] Optionally, the boundary conditions include a set value of the process approach time, a maximum alarm time for reaching a steady-state fluctuation range, and a maximum fluctuation range allowed for the actual process value.

[0012] Optionally, the determination result includes a result of a comprehensive determination of whether a plurality of preset control conditions are satisfied, including a determination result of whether the process approach time satisfies a set time and a determination result of whether the process approach time satisfies a steady-state fluctuation range.

[0013] Optionally, the output value Y and the input value X of the control module satisfy the following linear relationship; Y=PX+b; where P is the proportional gain and b is the bias term; the input value X is any one of a temperature value, a pressure value, and a flow value; and the output value Y is any one of a power value of the heater, an opening of the proportional valve, and an opening of the needle valve.

[0014] According to a second aspect of an embodiment of the present invention, a control device for semiconductor process equipment is provided, which is used for any method of the first aspect, including: an analysis module for obtaining a configuration file and a diagnostic time period of a target machine; the configuration file includes the data type and a closed-loop control model of the process variables of the target machine; the closed-loop control model has preset boundary conditions; N process variables to be analyzed are extracted based on the configuration file and the diagnostic time period, where N is a positive integer; the correlation between the process variables is specified based on the control relationship pre-set in the configuration file; the related process variables in a single closed-loop control loop are aggregated to generate M control groups, where M is a positive integer less than N; a database module is used to identify the data type of the basic process variables in each control group based on the data type, extract the basic process variables in the database, and calculate the correlation between the process variables based on the basic process relationship. The time period in which the variable changes determines the sub-time period to be analyzed in the diagnosis time period; the basic process variable is the target value or set value of the closed-loop control loop of the target machine; the diagnosis module is used to extract the non-basic process variables in the database based on each sub-time period; analyze whether the non-basic process variables meet the boundary conditions and obtain the judgment results; the non-basic process variables include the actual value and output value of the closed-loop control model, and the average value of the actual value and output value of each non-basic variable in each sub-time period when the boundary conditions are met is counted, and then regression analysis is performed based on the average value to update the closed-loop control model; the control module is used to obtain the proportional gain and bias term of the output of the updated closed-loop control model, and the control module calculates the output value based on the proportional gain and bias term when the input value is obtained next time; the input value is the basic process variable, and the output value is the control parameter of the target machine.

[0015] According to a third aspect of an embodiment of the present invention, an electronic device is provided, comprising a memory and a processor, wherein the memory is used to store a computer program executable by the processor; and the processor is used to execute the computer program in the memory to implement any method as in the first aspect.

[0016] According to a fourth aspect of an embodiment of the present invention, a computer-readable storage medium is provided, on which a computer program is stored. When the executable computer program in the storage medium is executed by a processor, the method as described in any one of the first aspects can be implemented.

[0017] According to a fifth aspect of the embodiments of the present application, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the method according to any one of the first aspect.

[0018] Compared with the prior art, the present application has the beneficial effects that: the redundant information is reduced by process variable aggregation, the key control path is focused, and the execution efficiency of the control system is improved. The basic process variables in the database are used as the basis for dividing the sub-time periods, the non-basic process variables are used as the basis for diagnosing whether the boundary conditions are met, and a closed-loop control model with adaptive capability is constructed. Through sub-time period division and data analysis, abnormal fluctuations are found in time, and the consistency of the process is ensured. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 is a flow chart of a data processing method of a semiconductor device according to an exemplary embodiment.

[0020] Figure 2 is a structural schematic diagram of a data processing device of a semiconductor device according to an exemplary embodiment.

[0021] Figure 3 is a block diagram of an electronic device according to another exemplary embodiment.

[0022] EXPLANATION OF REFERENCE NUMERALS IN DRAWINGS

[0023] 10, host computer; 20, control module; 30, sensing assembly; 40, execution assembly; 50, human-computer interaction module;

[0024] 11, analysis module; 12, database module; 13, diagnosis module. DETAILED DESCRIPTION

[0025] Unless otherwise defined, technical or scientific terms used in this specification shall have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. In the following description, specific embodiments of the application are described in connection with the appended drawings, in which it is noted that, for purposes of clarity, not every component can be shown in every drawing. It is intended that modifications and substitutions be made herein which fall within the scope of the underlying application. Accordingly, although specific embodiments have been illustrated and described herein, it should be appreciated that any arrangement which is calculated to achieve the same results could be substituted for the specific embodiments shown.

[0026] As Figure 1As shown, the first embodiment of the present invention provides a control method for semiconductor process equipment, including: S1, obtaining a configuration file and a diagnostic time period of a target machine; the configuration file includes the data type and closed-loop control model of the process variables of the target machine; the closed-loop control model has preset boundary conditions; S2, extracting N process variables to be analyzed based on the configuration file and the diagnostic time period, where N is a positive integer; specifying the correlation between each process variable based on the control relationship pre-set in the configuration file; aggregating the related process variables in a single closed-loop control loop to generate M control groups, where M is a positive integer less than N; S3, identifying the data type of the basic process variables in each control group based on the data type, extracting the basic process variables in the database, and determining the time period in the diagnostic time period based on the time period in which the basic process variables change. The sub-time period to be analyzed; the basic process variable is the target value or set value of the closed-loop control loop of the target machine; S4, based on each of the sub-time periods, extract the non-basic process variables in the database; analyze whether the non-basic process variables meet the boundary conditions to obtain a judgment result; the non-basic process variables include the actual value and output value of the closed-loop control model, and the average value of the actual value and output value of each non-basic variable in each sub-time period when the boundary conditions are met is counted, and then regression analysis is performed based on the average value to update the closed-loop control model; S5, the proportional gain and bias term of the output of the updated closed-loop control model are fed back to the control module, so that the control module calculates the output value according to the proportional gain and bias term when the input value is obtained next time; the input value is the basic process variable, and the output value is the control parameter of the target machine.

[0027] In some specific embodiments, N is set to 6, and the single closed-loop circuit includes a set temperature value, a set flow value, a set pressure value, an actual temperature value, an actual flow value, and an actual pressure value.

[0028] In some examples, M is set to 3, and the set temperature value and the actual temperature value are aggregated into a temperature control group; the set flow value and the actual flow value are aggregated into a flow control group; and the set pressure value and the actual pressure value are aggregated into a pressure control group.

[0029] In other examples, M is set to 2, and the set temperature, actual temperature, set pressure, and actual pressure values ​​are aggregated into a thermodynamic control group; the set flow rate and actual flow rate are aggregated into a flow control group. In this example, temperature and pressure have a strong physical correlation in the plasma chamber, and their changing trends are highly consistent. Aggregating the set temperature, actual temperature, set pressure, and actual pressure values ​​into a single thermodynamic control group can more accurately reflect the system's thermodynamic state and avoid the lag or misjudgment caused by single-variable control.

[0030] In some other examples, set temperature values, set flow values, and set pressure values ​​are grouped together into a set value control group. Actual temperature values, actual flow values, and actual pressure values ​​are grouped together into an actual value control group. This example simplifies the data processing process and reduces complexity by grouping all set values ​​into one category and all actual values ​​into another.

[0031] In some embodiments, the non-basic process variables are analyzed to see whether they meet the boundary conditions in the configuration file, and are also used to generate a judgment result; the method further includes: outputting a report file to a host computer, the report file including the judgment results of each sub-time period; the judgment result is input into a closed-loop control model as a feedback signal, and is used to update the proportional gain and bias term.

[0032] In some specific embodiments, the determination results include normal results, slightly deviated results, and severely deviated results.

[0033] In some examples, a normal result indicates that all non-basic process variables meet the boundary conditions, a slight deviation result indicates that some parameters of the non-basic process variables do not meet the boundary conditions in some time periods, and a severe deviation result indicates that some parameters of the non-basic process variables do not meet the boundary conditions in all time periods.

[0034] For normal results, the current proportional gain and bias items can continue to be used; there is no need to adjust the control parameters, only update the report records for subsequent tracing.

[0035] For slightly deviated results, the entire diagnostic time period is delayed to skip the time window where abnormal fluctuations occur; data is collected again in the new diagnostic cycle and the judgment results are generated; if the subsequent judgment is normal, there is no need to modify the closed-loop control model.

[0036] For severely deviated results, the configuration file needs to be re-obtained. The latest configuration file should be obtained to adapt to the current equipment status or changes in process requirements. Based on the new configuration file, process variables are re-extracted, control groups are divided, and a new closed-loop control model is generated. The proportional gain and bias terms are updated to adapt to the new control requirements.

[0037] In some embodiments, the method further includes controlling the host computer to generate visualization interface data according to the content of the report file; and outputting the visualization interface data to a human-computer interaction module so that the human-computer interaction module displays the determination results of each sub-time period.

[0038] In some embodiments, the basic process variables include process setting values ​​within a complete diagnostic time period. It is worth noting that the process setting values ​​vary over time in a closed-loop control loop.

[0039] In some specific embodiments, the basic process variables are not only used for dividing the sub-time periods, but also serve as a reference benchmark for establishing a closed-loop control model.

[0040] In some examples, the set value includes at least one of a temperature set value, a flow set value, and a pressure set value.

[0041] It is worth noting that by converting the content of the report file into visual interface data and outputting it to human-computer interaction modules such as touch screens, industrial computers or remote monitoring terminals, operators can obtain the operating status and process execution status of the target machine more intuitively and efficiently, thereby improving operational efficiency.

[0042] In some embodiments, the non-basic process variables include the actual process value, process approach time and steady-state fluctuation range corresponding to each of the sub-time periods; the process approach time is calculated based on the time required for the actual process value to reach the target fluctuation range of the process setting value; the steady-state fluctuation range is calculated based on the difference between the process setting value and the actual process value.

[0043] In some examples, the process set point is at least one of a temperature set point, a flow set point, and a pressure set point.

[0044] In other examples, the temperature setting value is a fixed value, and the steady-state fluctuation range of the temperature setting value is an upper and lower floating range based on the temperature setting value. The flow setting value is a fixed value, and the steady-state fluctuation range of the flow setting value is an upper and lower floating range based on the flow setting value. The pressure setting value is a fixed value, and the steady-state fluctuation range of the pressure setting value is an upper and lower floating range based on the pressure setting value.

[0045] In some embodiments, the boundary conditions include a set value of the process approach time, a maximum alarm time for reaching a steady-state fluctuation range, and a maximum fluctuation range allowed for the actual process value.

[0046] Taking temperature process control as an example, the process approach time setting specifies how long, starting from the set temperature value, the system expects the actual temperature to reach a reasonable range close to the steady-state value. For example, if the target temperature is 50°C and the set temperature approach time is 300 seconds, the system expects the actual temperature to reach a stable range (e.g., 50 ± 2°C) within 300 seconds. The process approach time setting is used to assess whether the temperature control system's response speed meets process requirements and prevent slow heating and cooling processes from impacting production cycle time.

[0047] In some examples, when the temperature setpoint rises from 20°C to 50°C, the system detects the impending temperature increase and gradually increases the heating power, rather than waiting until the temperature falls below the setpoint before significantly increasing the temperature. This effectively prevents temperature overshoot. This example uses targeted pre-judgment to intervene in closed-loop control. When the controlled variable is adjusted in advance, overshoot is avoided and the controlled variable is adjusted quickly.

[0048] The maximum alarm time to reach the steady-state fluctuation range specifies the maximum duration the system allows the actual temperature to fluctuate after entering the steady-state fluctuation range. If the temperature remains within the 50±2°C range after this time, an alarm will be triggered. For example, if the maximum alarm time is set to 120 seconds, if the actual temperature still fails to stabilize within the 50±2°C range after 120 seconds, an abnormality will be detected and an alarm will be triggered. This prevents long-term temperature instability caused by control system imbalance or external disturbances, ensuring consistent product quality.

[0049] The maximum allowable fluctuation range of the actual process value refers to the maximum range within which the temperature is allowed to deviate from the set value under steady-state conditions. Exceeding this range is considered out of control. For example, if the set temperature value has a maximum fluctuation range of 50±3°C, the system will consider the actual temperature to be outside this range if it exceeds 53°C or falls below 47°C. This setting ensures that the temperature remains within the acceptable range for process safety and quality, preventing product defects or equipment damage caused by excessive fluctuations.

[0050] Taking flow process control as an example, the process approach time setting specifies how long, starting from the set flow rate, the system expects the actual flow rate to reach a reasonable range close to the steady-state value. For example, if the target flow rate is 4900 sccm and the flow approach time is set to 200 seconds, the system expects the actual flow rate to enter a stable range (e.g., 4900 ± 2 sccm) within 200 seconds. The process approach time setting is used to assess whether the temperature control system's response speed meets process requirements and prevent slow heating and cooling processes from impacting production cycle time.

[0051] The maximum alarm time to reach the steady-state fluctuation range specifies the maximum duration the system allows the actual flow rate to fluctuate after entering the steady-state fluctuation range. If the flow rate remains unstable after this time, an alarm will be triggered. For example, if the maximum alarm time is set to 60 seconds, if the flow rate fluctuates within the range of 4900 ± 2 sccm for more than 60 seconds and remains unstable, an abnormality alarm will be triggered. This prevents long-term flow instability caused by control system imbalance or external disturbances, ensuring consistent product quality.

[0052] The maximum allowable fluctuation range of the actual process value refers to the maximum range within which the flow rate can deviate from the set value under steady-state conditions; any deviation beyond this range is considered out of control. For example, if the set flow value has a maximum fluctuation range of 4900 ± 3 sccm, the system will deem the fluctuation to be outside this maximum range if the actual flow rate exceeds 4903 sccm or falls below 4897 sccm. This setting ensures that the flow rate remains within the acceptable range for process safety and quality, preventing product defects or equipment damage caused by excessive fluctuations.

[0053] Taking pressure process control as an example, the process approach time setting specifies how long the system expects the actual pressure to reach a reasonable range close to the steady-state value, starting from the set pressure value. For example, if the target pressure is 235 kPa and the pressure approach time is set to 30 seconds, the system expects the actual pressure to reach a stable range (e.g., 235 ± 2 kPa) within 30 seconds. The process approach time setting is used to assess whether the temperature control system's response speed meets process requirements and prevent slow heating and cooling processes from impacting production cycle time.

[0054] The maximum alarm time to reach the steady-state fluctuation range specifies the maximum duration the system allows the actual pressure to fluctuate after entering the steady-state fluctuation range. If the pressure remains unstable after this time, an alarm will be triggered. For example, if the maximum alarm time is set to 10 seconds, if the actual pressure still fails to stabilize within the range of 235 ± 2 kPa after 10 seconds, an abnormality will be detected and an alarm will be triggered. This prevents long-term unstable pressure due to control system imbalance or external disturbances, ensuring consistent product quality.

[0055] The maximum allowable fluctuation range of the actual process value refers to the maximum range within which the pressure can deviate from the set value under steady-state conditions. Any deviation beyond this range is considered out of control. For example, if the maximum allowable fluctuation range of the set pressure value is 235 ± 3 kPa, if the actual pressure value is above 238 kPa or below 232 kPa, the system will determine that it exceeds the maximum fluctuation range. This setting ensures that the pressure remains within the process safety and quality standards, preventing product defects or equipment damage caused by excessive fluctuations.

[0056] In some embodiments, the determination result includes a comprehensive determination of whether a plurality of preset control conditions are satisfied, including a determination of whether the process approach time satisfies a set time and a determination of whether the process approach time satisfies a steady-state fluctuation range.

[0057] In some specific embodiments, taking temperature control as an example, the temperature approach time is set to 300 seconds. When the actual temperature approach time is 260 seconds, it is determined that the temperature approach time meets the set time; when the actual temperature approach time is 330 seconds, it is determined that the temperature approach time does not meet the set time.

[0058] In some embodiments, the temperature control is taken as an example, and the steady-state fluctuation range is set to be 100±2℃. When the actual temperature value is 100℃, it is determined that the actual temperature value satisfies the steady-state fluctuation range. When the actual temperature value is 96℃, it is determined that the actual temperature value does not satisfy the steady-state fluctuation range.

[0059] In some embodiments, the output value Y of the control module satisfies a linear relationship with the input value X; Y=PX+b; wherein P is the proportional gain, and b is the bias term; the input value X is any one of a temperature value, a pressure value, and a flow value; and the output value Y is any one of a power value of a heater, an opening degree of a proportional valve, and an opening degree of a needle valve.

[0060] In some embodiments, the output value Y and the input value X also have a partial nonlinear relationship, the definition domain of the input value X has subintervals X1, X2, …, Xn, and the proportional gain P has a modified mapping relationship with respect to the definition domain, that is, the subinterval X1 maps a modified proportional gain P1, the subinterval X2 maps a modified proportional gain P2, …, and the subinterval Xn maps a modified proportional gain Pn, n being a positive integer. By introducing a segmented control strategy and setting different proportional gains Pi for each subinterval, the control precision and system stability are improved, and the response speed and flexibility of the system are enhanced, so that the process equipment control system can maintain efficient and stable operation under complex working conditions.

[0061] It is worth noting that in some other embodiments, the bias term b also has a modified mapping relationship with respect to the definition domain, for example, the subinterval Xn maps a modified proportional gain Pn and a modified bias term bn, which will not be described here.

[0062] It is worth noting that the target machine is set to be a semiconductor device, such as a chemical cleaning machine. The number of target machines can be any positive integer.

[0063] In some embodiments, the target machine is set to include a front station machine and a rear station machine. For example, the front station machine is a delivery pump, and the control variable, that is, the input value, is: a front station outlet temperature Tin (℃), a front station outlet flow Fin (sccm), and a front station outlet pressure Pin (kPa). The rear station machine is a chemical cleaning machine, and the controlled variable, that is, the output value, is: a rear station inlet temperature Tout (℃), a rear station inlet flow Fout (sccm), and a rear station inlet pressure Pout (kPa). A multi-output regression model from (Tin, Fin, Pin) to (Tout, Fout, Pout) is trained using historical operation data.

[0064] For example, if the feed pump settings for a given cleaning operation are Tin = 60°C, Fin = 5000 sccm, and Pin = 250 kPa, the data model predicts that the chemical cleaning machine settings are Tout = 58.5°C, Fout = 4900 sccm, and Pout = 235 kPa.

[0065] like Figure 2 As shown, the second embodiment provides a control device for semiconductor process equipment, which is used for the method in the above embodiment, including: an analysis module 11, which is used to obtain a configuration file and a diagnostic time period of a target machine; the configuration file includes the data type and closed-loop control model of the process variables of the target machine; the closed-loop control model has preset boundary conditions; N process variables to be analyzed are extracted based on the configuration file and the diagnostic time period, where N is a positive integer; the correlation between each process variable is specified based on the control relationship pre-set in the configuration file; the related process variables in a single closed-loop control loop are aggregated to generate M control groups, where M is a positive integer less than N; a database module 12, which is used to identify the data type of the basic process variables in each control group based on the data type, extract the basic process variables in the database, and determine the sub-time period to be analyzed in the diagnostic time period based on the time period when the basic process variables change; The basic process variables are the target values ​​or set values ​​of the closed-loop control loop of the target machine; the diagnostic module 13 is used to extract the non-basic process variables in the database based on each of the sub-time periods; analyze whether the non-basic process variables meet the boundary conditions to obtain a judgment result; the non-basic process variables include the actual value and output value of the closed-loop control model, and the average value of the actual value and output value of each non-basic variable in each sub-time period when the boundary conditions are met is calculated, and then regression analysis is performed based on the average value to update the closed-loop control model; the control module 20 is used to obtain the proportional gain and bias term of the output of the closed-loop control model, and the control module calculates the output value based on the proportional gain and bias term when the input value is obtained next time; the input value is the basic process variable, and the output value is the control parameter of the target machine.

[0066] In some specific embodiments, the analysis module 11, the database module 12, and the diagnosis module 13 are integrated into the same host computer 10. The host computer 10 can be deployed in a local control cabinet or a remote server, and has unified data collection, storage, analysis, and diagnosis capabilities, which is conducive to improving the system's integration and response efficiency.

[0067] In some examples, the analysis module 11 is configured as an Equipment Data Analysis platform (EQDA) for data collection and storage, real-time data and historical data monitoring, and preliminary analysis.

[0068] The database module 12 is configured as a time series database (eg, InfluxDB), which is used as a basic system for analysis software to add, delete, modify, and query data, and is suitable for continuous monitoring data.

[0069] The diagnostic module is implemented by diagnostic software, which calls a database module based on a configuration file to output proportional gain and offset terms. The proportional gain and offset terms are transmitted to the analysis module 11 via a first Ethernet protocol. In some examples, the first Ethernet protocol is set to the Automation Device Specification (ADS) protocol.

[0070] In other specific embodiments, the analysis module 11, database module 12, and diagnostic module 13 are deployed in different computing units or devices, forming a distributed architecture. For example, the analysis module 11 runs on a local controller or edge computing device; the database module 12 is deployed on a cloud server or a local database server; and the diagnostic module 13 can be executed in an independent high-performance computing process variable, supporting simultaneous diagnostic analysis of multiple devices.

[0071] In some embodiments, the control module 20 is configured as a control chip or a programmable logic controller (PLC). The actual values ​​acquired by the sensor assembly are transmitted to the analysis module via the PLC using a second Ethernet communication protocol. In some examples, the second Ethernet communication protocol is configured as the User Datagram Protocol (UDP). It is worth noting that this embodiment achieves continuous optimization of the control strategy by configuring a spreadsheet file, quickly specifying the analysis target, analyzing historical data, and automatically generating reports for feedback to the PLC.

[0072] In some embodiments, the device further includes a sensor component 30, and the sensor component 30 is used to obtain the actual value of the process.

[0073] In some examples, the sensing assembly 30 includes a temperature sensor, a flow sensor, and a pressure sensor. The temperature sensor is used to obtain an actual temperature value, the flow sensor is used to obtain an actual flow value, and the pressure sensor is used to obtain an actual pressure value.

[0074] In some embodiments, the apparatus further includes an execution component 40, and the execution component 40 is used to adjust the actual process value.

[0075] In some examples, the actuator 40 includes a heater, a needle valve, and a proportional valve. The heater is configured to adjust the temperature of the chamber according to a temperature setting value. The needle valve is configured to adjust the gas flow rate through the chamber according to the flow setting value. The proportional valve is configured to adjust the pressure of the chamber according to the pressure setting value.

[0076] In some embodiments, the device further includes a human-computer interaction module 50, and the human-computer interaction module 50 is used to set the basic process variables, and the basic process variables include a set temperature value, a set flow value, and a set pressure value.

[0077] In some specific embodiments, the human machine interface (HMI) module 50 may be a touch screen, an industrial computer, or a remote monitoring terminal.

[0078] In other specific embodiments, operators can view the current status and perform manual intervention through the HMI interface. Interactive operations include manually triggering diagnostic processes, viewing historical reports and graphs, confirming or ignoring system suggestions, and exporting report files for archiving or sharing. Compared to existing touchscreen-PLC communication methods, this embodiment offers enhanced historical data analysis capabilities and allows for horizontal comparison of differences between different devices, resulting in more intuitive and convenient data presentation.

[0079] According to a third aspect of an embodiment of the present invention, an electronic device is provided, comprising a memory and a processor, wherein the memory is used to store a computer program executable by the processor; and the processor is used to execute the computer program in the memory to implement a method as described in any one of the first aspects.

[0080] Figure 3 9 is a block diagram of an electronic device according to an exemplary embodiment. For example, the electronic device 900 can be provided as a server. Figure 3 The electronic device 900 includes a processing component 922, which further includes one or more processors, and a memory resource represented by a memory 932 for storing instructions executable by the processing component 922, such as an application. The application stored in the memory 932 may include one or more modules, each corresponding to a set of instructions. In addition, the processing component 922 is configured to execute the instructions to perform the above-described method.

[0081] The electronic device 900 can further include a power supply component 926 configured to perform power management of the electronic device 900, a wired or wireless network interface 950 configured to connect the electronic device 900 to a network, and an input / output (I / O) interface 958. The electronic device 900 can operate based on an operating system stored in the memory 932, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, or the like.

[0082] In an exemplary embodiment, a non-transitory computer readable storage medium including instructions, such as the memory 932 including instructions, is also provided, which when executed by the processing component 922 of the electronic device 900, performs the above-described method. For example, the non-transitory computer readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disc, and an optical data storage device.

[0083] A fourth embodiment provides a readable storage medium, wherein a program is stored in the readable storage medium, and the program is executed to implement the method of any one of the above-described embodiments.

[0084] A fifth embodiment provides a computer program product, which includes a computer program, and the computer program is executed to implement the method of any one of the above-described embodiments.

[0085] In the present disclosure, the terms "first", "second" are used only for the purpose of description, and should not be understood as indicating or implying relative importance. The term "a plurality of" means two or more, unless otherwise explicitly limited.

[0086] The above description of the embodiments is provided to enable those with ordinary skill in the art to understand and apply the present disclosure. Those skilled in the art will readily understand and appreciate that various modifications can be made to the embodiments without departing from the scope and spirit of the present disclosure. Therefore, the present disclosure is not limited to the embodiments described herein, and improvements and modifications made by those skilled in the art based on the disclosure provided herein without departing from the scope and spirit of the present disclosure are within the scope of the present disclosure.

Claims

1. A control method for semiconductor process equipment, characterized in that: include: Obtaining a configuration file and a diagnostic time period of a target machine; the configuration file includes a data type and a closed-loop control model of a process variable of the target machine; The closed-loop control model has preset boundary conditions; Extracting N process variables to be analyzed based on the configuration file and the diagnostic time period, where N is a positive integer; specifying correlations between the process variables based on control relationships pre-set in the configuration file; Aggregate the relevant process variables in a single closed-loop control loop to generate M control groups, where M is a positive integer less than N; Identifying the data type of a basic process variable in each control group based on the data type, extracting the basic process variable from a database, and determining a sub-time period to be analyzed within the diagnostic time period based on a time period in which the basic process variable changes; the basic process variable being a target value or set value of a closed-loop control loop of the target machine; Extracting non-basic process variables from the database based on each of the sub-time periods; analyzing whether the non-basic process variables meet the boundary conditions to obtain a determination result; The non-basic process variables include the actual value and output value of the closed-loop control model, and the average value of the actual value and output value of each non-basic variable in each sub-time period when the boundary conditions are met is calculated, and then regression analysis is performed based on the average value to update the closed-loop control model; Feeding back the updated proportional gain and bias term output by the closed-loop control model to the control module, so that the control module calculates the output value according to the proportional gain and bias term when obtaining the input value next time; The input value is the basic process variable, and the output value is the control parameter of the target machine.

2. The method according to claim 1, characterized in that The basic process variables include process setting values ​​within the complete diagnosis period.

3. The method according to claim 2, characterized in that The non-basic process variables include the process actual value, process approach time and steady-state fluctuation range corresponding to each of the sub-time periods; The process approach time is calculated based on the time required for the process actual value to reach the target fluctuation range of the process setting value; The steady-state fluctuation range is calculated based on the difference between the process set value and the process actual value.

4. The method according to claim 1, wherein The boundary conditions include a set value of the process approach time, a maximum alarm time for reaching a steady-state fluctuation range, and a maximum fluctuation range allowed for the process actual value.

5. The method according to claim 1, wherein The determination result includes a result of a comprehensive determination of the satisfaction of multiple preset control conditions, including a determination result of whether the process approach time satisfies the set time and a determination result of whether the process approach time satisfies the steady-state fluctuation range.

6. The method according to claim 1, characterized in that The output value Y and input value X of the control module satisfy the following linear relationship: Y=PX+b; Wherein, P is the proportional gain, b is the bias term; The input value X is at least one of the basic process variables; The output value Y is at least one of the non-basic process variables.

7. The method according to any one of claims 1 to 6, characterized in that The method further includes: outputting a report file to a host computer, wherein the report file includes the determination results of each sub-time period.

8. The method according to claim 7, characterized in that The method further includes controlling the host computer to generate visualization interface data according to the content of the report file; and outputting the visualization interface data to a human-computer interaction module so that the human-computer interaction module displays the determination results of each sub-time period.

9. A control device for semiconductor process equipment, used in the method according to any one of claims 1 to 8, characterized in that: include: an analysis module configured to obtain a configuration file and a diagnostic time period of a target machine; the configuration file including data types of process variables of the target machine and a closed-loop control model; the closed-loop control model having preset boundary conditions; extract N process variables to be analyzed based on the configuration file and the diagnostic time period, where N is a positive integer; and specify correlations between the process variables based on control relationships pre-defined in the configuration file; Aggregate the relevant process variables in a single closed-loop control loop to generate M control groups, where M is a positive integer less than N; a database module, configured to identify the data type of a basic process variable in each control group based on the data type, extract the basic process variable from the database, and determine a sub-time period to be analyzed within the diagnostic time period based on a time period in which the basic process variable changes; the basic process variable being a target value or set value of a closed-loop control loop of the target machine; a diagnosis module, configured to extract non-basic process variables from the database based on each of the sub-time periods; analyze whether the non-basic process variables meet the boundary conditions, and obtain a determination result; The non-basic process variables include the actual value and output value of the closed-loop control model, and the average value of the actual value and output value of each non-basic variable in each sub-time period when the boundary conditions are met is calculated, and then regression analysis is performed based on the average value to update the closed-loop control model; A control module, configured to obtain a proportional gain and a bias term output by the updated closed-loop control model, and calculate an output value based on the proportional gain and the bias term when obtaining an input value next time; The input value is the basic process variable, and the output value is the control parameter of the target machine.

10. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory is used to store a computer program executable by the processor; and the processor is used to execute the computer program in the memory to implement the method according to any one of claims 1 to 8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the executable computer program in the storage medium is executed by a processor, the method according to any one of claims 1 to 8 can be implemented.

12. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images