Apparatus and method for intelligent operation of plasma equipment

Abstract

The present invention relates to an apparatus and a method for intelligent operation of plasma equipment, the apparatus comprising: a message queue for loading sensing data acquired from a plurality of sensors according to an input condition that is input into the plasma equipment during operation of the plasma equipment; a prediction module for predicting a plasma state value, a process state and a process result by applying the sensing data to a pre-learned artificial intelligence algorithm; a planning module, which applies the plasma state value, the process state, and the process result to the pre-learned artificial intelligence algorithm so as to plan a parameter value for controlling the plasma equipment; and a control module for controlling the plasma equipment by using the planned parameter value as the input condition, and other embodiments are also applicable.

Description

Device and method for intelligent operation of plasma equipment

[0001] The present invention relates to a device and method for intelligent operation of plasma equipment.

[0002] Plasma generated through high-frequency power is widely used in etching and deposition processes for wafers and display panel substrates. However, with the recent demand for higher performance in products, the amount of material required to be etched or deposited is decreasing. To etch or deposit these smaller amounts of material, it is necessary to accurately assess plasma state changes and control the plasma state accordingly.

[0003] Typically, assessing plasma state changes during product production requires direct contact with the plasma to measure its state. This creates the potential for plasma state changes during plasma state measurement. Therefore, plasma state changes are currently assessed after the product production process is complete, typically during the verification of process results. Alternatively, skilled engineers can verify plasma state and process state changes by reviewing plasma equipment component operation results and measured sensing data.

[0004] However, when engineers directly monitor plasma and process state changes, the amount of equipment they must manage increases. Furthermore, because the amount of material to be etched and deposited is reduced, data changes are less readily observed. Therefore, even with the most skilled engineers, human error can occur, reducing the accuracy of plasma state changes. Therefore, there is a growing need to develop technologies that can assist or even replace engineers to more accurately monitor plasma state changes.

[0005] Embodiments of the present invention for solving these conventional problems provide a device and method for intelligent operation of plasma equipment capable of monitoring plasma state changes and process state changes in real time to continuously produce products that meet established standards in a plasma process.

[0006] In addition, embodiments of the present invention provide a device and method for intelligent operation of plasma equipment, which learns an artificial intelligence algorithm using sensing data related to the operation of plasma equipment, monitors changes in plasma state and process state in real time through linkage with an artificial intelligence algorithm and a simulator for which learning has been completed, and can infer a plasma state value.

[0007] In addition, embodiments of the present invention provide a device and method for intelligent operation of plasma equipment capable of controlling a power device, a gas inlet device, or a vacuum pump installed in the plasma equipment in real time by setting parameter values ​​to change the plasma state to a state that can achieve a desired process result by checking when a change in the plasma state occurs and the influence of the change in the plasma state on the process.

[0008] An apparatus for intelligent operation of plasma equipment according to an embodiment of the present invention is characterized by including a message queue for loading sensing data acquired from a plurality of sensors according to input conditions input to the plasma equipment when operating the plasma equipment, a prediction module for predicting a plasma state value, a process state, and a process result by applying the sensing data to a pre-learned artificial intelligence algorithm, a planning module for planning a parameter value for controlling the plasma equipment by applying the plasma state value, the process state, and the process result to a pre-learned artificial intelligence algorithm, and a control module for controlling the plasma equipment using the planned parameter value as the input condition.

[0009] In addition, it is characterized by including a collection module that collects sensing data and transmits it to the message queue.

[0010] In addition, the prediction module includes a plasma state prediction module that predicts the plasma state value, a process state prediction module that predicts the process state, and a process result prediction module that predicts the process result, and each of the prediction modules is characterized in that it uses a different artificial intelligence algorithm.

[0011] In addition, the message queue is characterized in that, if a plurality of modules included in the prediction module are registered as multiple recipients for receiving the sensing data, the sensing data is transmitted to each of the modules registered as multiple recipients.

[0012] In addition, the message queue is characterized in that it transmits the sensing data to the simulator when the simulator is registered as the plurality of recipients.

[0013] In addition, the planning module is characterized in that it performs the planning based on data received from the prediction module or the simulator through the message queue.

[0014] In addition, it is characterized by including a reverse inference module that reversely infers the final plasma state value, final sensing data, and final parameter value to derive the desired process result.

[0015] In addition, the reverse inference module is characterized in that it infers the final parameter value by applying the above-described achievement process result to an artificial intelligence algorithm for reverse inference.

[0016] In addition, the control module is characterized in that it controls the plasma equipment based on the final parameter value inferred by the reverse inference module.

[0017] In addition, the prediction module, the planning module, the control module, and the reverse inference module are characterized in that they load data derived from each module into the message queue and independently operate by collecting the data loaded into the message queue as needed.

[0018] In addition, a method for intelligent operation of plasma equipment according to an embodiment of the present invention is characterized by including a step of an electronic device collecting sensing data acquired from a plurality of sensors when the plasma equipment is operated and loading the data into a message queue, and a step of the electronic device controlling the plasma equipment by planning parameter values ​​for controlling the plasma equipment based on plasma state values, process states, and process results predicted by the sensing data according to selection of an operation mode, or controlling the plasma equipment by inferring a final plasma state value, final sensing data, and final parameter values ​​to derive a desired process result.

[0019] As described above, the device and method for intelligent operation of plasma equipment according to the present invention have the effect of monitoring plasma state changes and process state changes in real time to continuously produce products that meet standards set in a plasma process, thereby enabling more accurate confirmation of plasma state changes and process state changes during the process.

[0020] In addition, the device and method for intelligent operation of plasma equipment according to an embodiment of the present invention learns an artificial intelligence algorithm using sensing data related to the operation of the plasma equipment, monitors changes in plasma state and process state in real time through linkage with the artificial intelligence algorithm and simulator for which learning has been completed, and infers plasma state values, thereby assisting and replacing engineers to more accurately confirm plasma state changes.

[0021] In addition, the device and method for intelligent operation of plasma equipment according to an embodiment of the present invention can control a power device, a gas inlet device, or a vacuum pump installed in the plasma equipment in real time by setting parameter values ​​to change the plasma state to a state that can achieve a desired process result by checking the case where a change in the plasma state occurs and the influence of the change in the plasma state on the process, thereby adjusting the parameters so that the plasma state, the process state, and the process result conform to the set range, thereby minimizing process by-products and process defects due to component damage that may occur as the process time accumulates, and improving product productivity.

[0022] FIG. 1 is a drawing showing a system for intelligent operation of plasma equipment according to an embodiment of the present invention.

[0023] FIG. 2 is a drawing showing the main configuration of an electronic device for intelligent operation of plasma equipment according to an embodiment of the present invention.

[0024] Figure 3 is a flowchart for explaining an intelligent operation method of plasma equipment according to an embodiment of the present invention.

[0025] FIG. 4 is a detailed flowchart for explaining a learning execution method for intelligent operation of plasma equipment according to an embodiment of the present invention.

[0026] FIG. 5 is a detailed flowchart for explaining an intelligent operation method of plasma equipment in a forward direction according to an embodiment of the present invention.

[0027] FIG. 6 is a detailed flowchart for explaining an intelligent operation method of plasma equipment in reverse according to an embodiment of the present invention.

[0028] FIG. 7 is a diagram showing the point in time at which data is processed during intelligent operation of plasma equipment using a message queue according to an embodiment of the present invention.

[0029] FIG. 8 is a diagram showing the point in time at which data of a plasma device is processed when a message queue is not used according to a general embodiment.

[0030] FIG. 9 is a screen example showing a change in a plasma state value or process state according to a change in sensing data according to an embodiment of the present invention.

[0031] FIG. 10 is a screen example showing the change in plasma state value or process state over time according to an embodiment of the present invention.

[0032] FIG. 11 and FIG. 12 are screen examples showing sensing data, plasma status values, and process status in normal and abnormal processes according to an embodiment of the present invention.

[0033] FIG. 13 is a screen example showing a method of operating a plasma device in forward mode and reverse mode according to an embodiment of the present invention.

[0034] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. The detailed description set forth below, together with the accompanying drawings, is intended to explain exemplary embodiments of the present invention and is not intended to represent the only embodiments in which the present invention may be practiced. In the drawings, portions irrelevant to the description may be omitted for clarity in describing the present invention, and the same reference numerals may be used throughout the specification for identical or similar components.

[0035] FIG. 1 is a drawing showing a system for intelligent operation of plasma equipment according to an embodiment of the present invention.

[0036] Referring to FIG. 1, a system (10) according to the present invention may include plasma equipment (100), an electronic device (200), and a simulator (300).

[0037] The plasma equipment (100) is equipment that generates plasma through high-frequency power to perform etching and deposition processes on wafers and display panel substrates, and may be equipped with a plurality of sensors (not shown) to obtain sensing data such as spectrum data and electrical characteristic data related to light generated during the plasma process. The plurality of sensors may communicate with the electronic device (200) on a 1:1 basis or may be connected in an N:1 manner via a LAN method to perform communication and transmit sensing data to the electronic device (200).

[0038] In addition, the plasma equipment (100) may be equipped with a plurality of plasma diagnostic devices (not shown), and the plurality of plasma diagnostic devices may collect electron density, electron temperature, gas flow rate values, etc. obtained during the plasma process and transmit them to the electronic device (200).

[0039] The electronic device (200) collects sensing data obtained from the plasma equipment (100) and performs intelligent operation of the plasma equipment (100) based on the sensing data. At this time, the main operation of the electronic device (200) will be described in more detail using FIG. 2 below.

[0040] The simulator (300) may be a plasma simulator that simulates plasma equipment. Such a simulator (300) is equipped with a number of theoretical modules, which are theoretical models that elucidate the basic theory of various physical / chemical changes inside the plasma equipment (100), and can simulate the plasma equipment (100) by numerically analyzing the theoretical models of the theoretical modules. At this time, the theoretical models may include a model that elucidates the physical / chemical phenomena of charged components such as electrons and ions, a model that elucidates the electromagnetic field formed in the plasma, a model that elucidates the reaction and flow of uncharged chemical species, a model that elucidates the generation and growth of fine particles in the plasma, a model that elucidates the flow of fine particles, and a model that elucidates the thin film formation and etching phenomenon by chemical species and ions on the wafer surface.

[0041] The simulator (300) can collect sensing data obtained from the plasma equipment (100) through communication with the electronic device (200), simulate plasma state changes, process states, and process results generated by the sensing data, and transmit the results to the electronic device (200). To this end, the simulator (300) can be connected to the electronic device (200) via wireless or wired communication to perform communication.

[0042] FIG. 2 is a drawing showing the main configuration of an electronic device for intelligent operation of plasma equipment according to an embodiment of the present invention.

[0043] Referring to FIG. 2, an electronic device (200) according to the present invention may include a collection module (210), a message queue (220), a prediction module (230), a planning module (240), a control module (250), a reverse inference module (260), a storage module (270), and a display module (280). The electronic device (200) may be a digital twin device capable of simultaneously performing forward prediction and backward prediction for a process performed in a plasma equipment (100).

[0044] The collection module (210) collects sensing data in real time through communication with multiple sensors equipped in the plasma equipment (100). To this end, the collection module (210) may perform 1:1 communication with the sensors or 1:N communication via LAN. For example, the collected sensing data may include OES sensing data, VI sensing data, and MFC sensing data.

[0045] The collection module (210) performs tasks such as normalizing or removing noise from the collected sensing data through preprocessing of the collected sensing data, and removing covariance between the sensing data through correlation analysis in the preprocessed sensing data, thereby extracting the characteristics of the sensing data. For example, when etching SiO2 using the CxFy series, wavelengths related to F neutral species or radicals and wavelengths of the SiFy series can be extracted as characteristics from the OES sensing data, and harmonic components affected by chamber impedance generated by attachment of etching byproducts can be extracted as characteristics from the VI sensing data.

[0046] The collection module (210) can transmit data on characteristics extracted from sensing data (hereinafter collectively referred to as sensing data) to a message queue (220) and load the data into the message queue (220). At this time, the sensing data loaded into the message queue (220) can be loaded in real time or periodically.

[0047] The message queue (220) can be configured with message broker software such as Rabbit MQ. The message queue (220) can load the sensing data provided from the collection module (210). In addition, the message queue (220) can receive and load the result values ​​derived from each module through communication with the prediction module (230), the planning module (240), the control module (250), and the reverse inference module (260), and can directly provide the data required by each module from among the loaded data to the corresponding module. In addition, the message queue (220) can receive and load the result values ​​derived from the simulator (300) through communication with the simulator (300), and can provide the data required by the simulator (300) from among the loaded data to the simulator. In addition, the message queue (220) can provide the loaded data to the storage module (270) and the display module (280).

[0048] The prediction module (230) may include a plasma state prediction module that predicts a plasma state value, a process state prediction module that predicts a process state, and a process result prediction module that predicts a process result, and each module may use different artificial intelligence algorithms.

[0049] The prediction module (230) applies the sensing data provided from the message queue (220) to the artificial intelligence algorithm included in each module to predict the plasma state value, process state, and process result, and performs learning of the artificial intelligence algorithm by comparing the predicted result with the acquired value actually acquired in the plasma diagnostic device.

[0050] More specifically, the plasma state prediction module merges OES sensing data and VI sensing data and applies them as input data to an artificial intelligence algorithm to predict electron density, and performs learning of the artificial intelligence algorithm by comparing the predicted electron density with the electron density acquired from a cutoff probe, which is a plasma diagnostic device, at the time when the OES sensing data and VI sensing data were acquired. In addition, the plasma state prediction module merges OES sensing data and MFC sensing data and applies them as input data to an artificial intelligence algorithm to predict electron temperature, and performs learning of the artificial intelligence algorithm by comparing the predicted electron temperature with the electron temperature acquired from a Langmuir probe, which is a plasma diagnostic device, at the time when the OES sensing data and MFC sensing data were acquired. At this time, the artificial intelligence algorithms for predicting electron density and electron temperature may be different algorithms.

[0051] The plasma state prediction module can provide the plasma state value to the message queue (220) in real time so that the message queue (220) can load the plasma state value predicted by the artificial intelligence algorithm.

[0052] The process state prediction module can predict the process state using only the sensing data loaded in the message queue (220), can predict the process state using the sensing data loaded in the message queue (220) and the plasma state value, and can predict the process state using the sensing data loaded in the message queue (220) and the plasma state value predicted in the simulator (300).

[0053] In this way, when the process state prediction module predicts the process state using sensing data or predicts the process state using the sensing data and the plasma state value predicted by the simulator (300), the process state prediction module can operate in parallel with the plasma state prediction module.

[0054] The process state prediction module predicts the etching process by merging OES sensing data and VI sensing data and applying them as input data to an artificial intelligence algorithm, and performs learning of the artificial intelligence algorithm by comparing the predicted etching process with the etching process acquired from the SEM, which is a process inspection device, at the time when the OES sensing data and VI sensing data were acquired. In addition, the process state prediction module predicts the etching process by applying the OES sensing data and VI sensing data and the electron temperature and electron density predicted by the simulator (300) and loaded into the message queue (220) to the artificial intelligence algorithm, and performs learning of the artificial intelligence algorithm by comparing the predicted etching process with the etching process acquired from the SEM, which is a process inspection device, at the time when the OES sensing data and VI sensing data were acquired.

[0055] The process status prediction module can provide the process status to the message queue (220) in real time so that the message queue (220) can load the process status predicted by the artificial intelligence algorithm.

[0056] The process result prediction module can predict the process result using the sensing data, plasma state value, and process state loaded into the message queue (220). At this time, the plasma state value and process state can be predicted by the simulator (300) and loaded into the message queue (220). Therefore, the process result prediction module can operate independently from the plasma state prediction module and the process state prediction module.

[0057] The process result prediction module applies OES sensing data and VI sensing data to an artificial intelligence algorithm to predict an etching profile, and performs learning of the artificial intelligence algorithm by comparing the predicted etching profile with the etching profile confirmed after the process is completed.

[0058] The process result prediction module can provide the process result to the message queue (220) in real time so that the message queue (220) can load the process result predicted by the artificial intelligence algorithm.

[0059] The planning module (240) plans parameter values ​​to be controlled during the process in the plasma equipment based on the process result prediction. For example, if a change or abnormality is identified in the plasma state or process state based on the process result prediction, the planning module (240) can perform parameter planning, control planning, product prediction, and tuning knob planning. At this time, parameter planning is to identify the target parameter that causes the identified change or abnormality, control planning is to determine which tuning knob is for controlling the target parameter and in which combination to control the tuning knob, product prediction is to predict the product result and determine whether to continue or stop tuning based on the predicted product result, and tuning knob planning may be to determine the adjustment value and order of the tuning knob.

[0060] More specifically, the planning module (240) is called when the plasma state value predicted by the plasma state prediction module has a difference greater than or equal to a threshold from a preset range, when the process state predicted by the process state prediction module has a difference greater than or equal to a threshold from a preset process state range, and when the process result predicted by the process result prediction module has a difference greater than or equal to a threshold from a range of process results measured after the actual process is completed.

[0061] More specifically, when a difference occurs between the plasma state value predicted by the plasma state prediction module and a preset range and a threshold value or more, the planning module (240) can receive input from the user of change items, change amounts, process parameters, process number of processes, process execution time, etc. that have changed in the plasma state.

[0062] If the process state predicted by the process state prediction module differs by a threshold or more from the range of the preset process state, the planning module (240) can receive from the user an input of a process influence related to a change item in which the plasma state has changed.

[0063] The planning module (240) can use an artificial intelligence algorithm to plan which parameters of the plasma equipment should be adjusted to which values ​​when a difference occurs between the process results predicted by the process result prediction module and the range of the actual measured process results that exceeds a threshold value.

[0064] The planning module (240) can generate a control combination based on an adjustable control factor in a parameter control combination condition having information on an adjustable item in at least one process among a process in which a predicted plasma state value deviates from a range of preset plasma state values ​​by a threshold or more, a process in which a predicted process state deviates from a range of preset process states by a threshold or more, and a process in which a predicted process result deviates from a range of an actually measured process result by a threshold or more.

[0065] The process control combination can perform learning by applying the sensing data acquired from the performed process and the predicted plasma state value, process state, and process result corresponding to the sensing data to the reinforcement learning algorithm, which is an artificial intelligence algorithm.

[0066] The planning module (240) performs a process impact assessment on the predicted control combination after the prediction of the process control combination is completed using a reinforcement learning algorithm. The process impact assessment determines whether the control combination exists within the range defined by the preset parameter control constraints, and if it is within the preset range, the recommended parameter control combination and the control information on the parameter adjustment amount are stored to be used as learning data for more stable control of the plasma equipment, and this can be loaded into the message queue (220) as a learning performance result.

[0067] The control module (250) performs control of parameter values ​​based on the process control combination set in the planning module (240), and the control module (250) can check whether the process result derived according to the control of the parameter values ​​is normal or abnormal and add it to the learning performance result and load it into the message queue (220).

[0068] Additionally, the parameter values ​​of the plasma equipment (100) can be controlled using the final equipment parameter values ​​selected by the reverse inference module (260) through the control module (250).

[0069] The reverse inference module (260) receives a desired process result (hereinafter, "achieved process result") from the user. For example, the desired process result input by the user may be an etching profile to be confirmed as a result of the process after the process is completed.

[0070] The reverse inference module (260) can exchange data with the message queue (220), prediction module (230), planning module (240), and control module (250) as needed to infer equipment parameters.

[0071] The reverse inference module (260) predicts the final plasma state value for deriving the achievement process result through data transmission and reception with the message queue (220). More specifically, the reverse inference module (260) applies the achievement process result to a regression-based artificial intelligence algorithm stored in the reverse inference module (260). The reverse inference module (260) applies the achievement process result to the artificial intelligence algorithm and predicts the plasma state value for deriving the achievement process result based on the plasma state value loaded in the message queue (220).

[0072] The reverse inference module (260) predicts the final sensing data for deriving the final plasma state value. More specifically, the reverse inference module (260) applies the final plasma state value to a regression-based artificial intelligence algorithm. The reverse inference module (260) applies the final plasma value and predicts the final sensing data for deriving the final plasma state value based on the sensing data loaded into the message queue (220).

[0073] The reverse inference module (260) predicts the final equipment parameter values ​​for deriving the final sensing data. More specifically, the reverse inference module (260) applies the final sensing data to a regression-based artificial intelligence algorithm. The reverse inference module (260) applies the final sensing data to predict the final parameter values ​​for deriving the final sensing data based on the parameter values ​​loaded in the message queue (220), and may provide the predicted final parameter values ​​directly to the control module (250) or provide the predicted final parameter values ​​to the control module (250) via the message queue (220).

[0074] The storage module (270) stores data provided from the message queue (220).

[0075] The display module (280) visualizes and displays data provided from the message queue (220).

[0076] Figure 3 is a flowchart for explaining an intelligent operation method of plasma equipment according to an embodiment of the present invention.

[0077] Referring to FIG. 3, if it is confirmed in step 301 that a start signal for intelligent operation has been received from the outside, the electronic device (200) performs step 303, and if the start signal is not received, it waits for reception of the start signal. In step 303, the collection module (210) of the electronic device (200) collects sensing data acquired from a plurality of sensors when the plasma equipment (100) is operated. For example, the collected sensing data may include OES sensing data, VI sensing data, and MFC sensing data.

[0078] The collection module (210) performs tasks such as normalizing or removing noise from the collected sensing data through preprocessing of the collected sensing data, and removing covariance between the sensing data through correlation analysis in the preprocessed sensing data, thereby extracting the characteristics of the sensing data. For example, when etching SiO2 using the CxFy series, wavelengths related to F neutral species or radicals and wavelengths of the SiFy series can be extracted as characteristics from the OES sensing data, and harmonic components affected by chamber impedance generated by attachment of etching byproducts can be extracted as characteristics from the VI sensing data.

[0079] In step 305, the collection module (210) can transmit data on characteristics extracted from the sensing data (hereinafter collectively referred to as sensing data) to the message queue (220) and load the data into the message queue (220). At this time, the sensing data loaded into the message queue (220) can be loaded in real time or periodically.

[0080] In step 307, the electronic device (200) checks whether the learning of each artificial intelligence algorithm operating in the prediction module (230), the planning module (240), and the reverse inference module (260) included in the electronic device (200) is complete. At this time, whether the learning is complete is determined by determining whether an artificial intelligence algorithm capable of predicting plasma state changes and process state changes based on the sensing data acquired according to the operation of the plasma equipment has been learned in advance. If, as a result of the check in step 307, each artificial intelligence algorithm stored in the prediction module (230) and the planning module (240) is not in a pre-learned state, the electronic device (200) can perform learning of the artificial intelligence algorithm based on the sensing data by performing step 309. At this time, the operation of learning each artificial intelligence algorithm stored in the prediction module (230) and the planning module (240) will be described in more detail using FIG. 4 below.

[0081] Conversely, if the result of the verification in step 307 indicates that each artificial intelligence algorithm operating in the prediction module (230), the planning module (240), and the reverse inference module (260) has been pre-learned, the electronic device (200) performs step 311. In step 311, if the operation mode of the plasma equipment (100) is selected as the forward mode from the outside, the electronic device (200) performs step 313. The forward mode means predicting the progress status of the process performed in the plasma equipment (100).

[0082] In step 313, the electronic device (200) performs forward prediction to operate the plasma equipment (100) in forward mode and performs step 319. At this time, step 313 will be described in more detail using FIG. 5 below.

[0083] Conversely, if the forward mode is not selected in step 311, the electronic device (200) performs step 315. If the operating mode of the plasma equipment (100) is selected as the reverse mode from the outside in step 315, the electronic device (200) performs step 317. The reverse mode means that the equipment parameters that need to be set in the plasma equipment (100) are inferred in reverse based on the forward prediction data that is updated in real time, thereby determining whether the plasma state and process state generated in the plasma equipment (100) are interpolation or extrapolation within the control range of the plasma equipment (100), and thereby confirming whether the plasma equipment (100) is abnormal and whether control is necessary and possible.

[0084] In step 317, the electronic device (200) performs reverse prediction to operate the plasma equipment (100) in reverse mode and performs step 319. At this time, step 317 will be described in more detail using FIG. 6 below.

[0085] In step 319, if an end signal for terminating intelligent operation is received from the outside, the electronic device (200) terminates the process, and if no end signal is received, the process returns to step 303 and steps 303 to 317 can be re-performed.

[0086] FIG. 4 is a detailed flowchart for explaining a learning execution method for intelligent operation of plasma equipment according to an embodiment of the present invention.

[0087] Referring to FIG. 4, in step 401, the message queue (220) provides the loaded sensing data to the prediction module (230). Although not shown, the prediction module (230) may include a plasma state prediction module that predicts a plasma state value, a process state prediction module that predicts a process state, and a process result prediction module that predicts a process result.

[0088] In step 403, the prediction module (230) can predict the plasma state value, process state, and process result using the sensing data. More specifically, the plasma state prediction module can predict the electron density and electron temperature using the sensing data. The process state prediction module can predict the etching process, etc., using the sensing data. The process result prediction module can predict the process result value, such as the etching profile after the process is completed, using the sensing data, the plasma state value, and the process state. In addition, the plasma state prediction module, the process state prediction module, and the process result prediction module can each use different artificial intelligence algorithms, and the plasma state prediction module and the process state prediction module can operate simultaneously.

[0089] The plasma state prediction module merges OES sensing data and VI sensing data and applies them as input data to an artificial intelligence algorithm to predict electron density, and performs learning of the artificial intelligence algorithm by comparing the predicted electron density with the electron density acquired from the cutoff probe, which is a plasma diagnostic device, at the time when the OES sensing data and VI sensing data were acquired. In addition, the plasma state prediction module merges OES sensing data and MFC sensing data and applies them as input data to an artificial intelligence algorithm to predict electron temperature, and performs learning of the artificial intelligence algorithm by comparing the predicted electron temperature with the electron temperature acquired from the Langmuir probe, which is a plasma diagnostic device, at the time when the OES sensing data and MFC sensing data were acquired. At this time, the artificial intelligence algorithms for predicting electron density and electron temperature may be different algorithms.

[0090] The plasma state prediction module can provide the plasma state value to the message queue (220) in real time so that the message queue (220) can load the plasma state value predicted by the artificial intelligence algorithm.

[0091] In addition, it is clarified that the combination of data applied to the artificial intelligence algorithm included in the plasma state prediction module is not limited to the above-described embodiment, and can be expanded and applied depending on the type of sensor and the type of plasma diagnostic device.

[0092] The process state prediction module can predict the process state using the sensing data loaded into the message queue (220), can predict the process state using the sensing data loaded into the message queue (220) and the plasma state value, and can predict the process state using the sensing data loaded into the message queue (220) and the plasma state value predicted by the simulator (300). In this way, when the process state prediction module predicts the process state using the sensing data or predicts the process state using the sensing data and the plasma state value predicted by the simulator (300), the process state prediction module can operate independently from the plasma state prediction module.

[0093] The process state prediction module predicts the etching process by merging OES sensing data and VI sensing data and applying them as input data to an artificial intelligence algorithm, and performs learning of the artificial intelligence algorithm by comparing the predicted etching process with the etching process acquired from the SEM, which is a process inspection device, at the time when the OES sensing data and VI sensing data were acquired. In addition, the process state prediction module predicts the etching process by applying the OES sensing data and VI sensing data and the electron temperature and electron density predicted by the simulator (300) and loaded into the message queue (220) to the artificial intelligence algorithm, and performs learning of the artificial intelligence algorithm by comparing the predicted etching process with the etching process acquired from the SEM, which is a process inspection device, at the time when the OES sensing data and VI sensing data were acquired.

[0094] The process status prediction module can provide the process status to the message queue (220) in real time so that the message queue (220) can load the process status predicted by the artificial intelligence algorithm.

[0095] In addition, it is clarified that the combination of data applied to the artificial intelligence algorithm included in the process status prediction module is not limited to the above-described embodiment, and can be expanded and applied depending on the type of sensor and the type of process inspection device.

[0096] The process result prediction module can predict the process result using the sensing data, plasma state value, and process state loaded into the message queue (220). At this time, the plasma state value and process state can be predicted by the simulator (300) and loaded into the message queue (220). Therefore, the process result prediction module can operate independently from the plasma state prediction module and the process state prediction module.

[0097] The process result prediction module applies OES sensing data and VI sensing data to an artificial intelligence algorithm to predict an etching profile, and performs learning of the artificial intelligence algorithm by comparing the predicted etching profile with the etching profile confirmed after the process is completed.

[0098] The process result prediction module can provide the process result to the message queue (220) in real time so that the message queue (220) can load the process result predicted by the artificial intelligence algorithm.

[0099] In step 405, the electronic device (200) performs planning by calling a planning module (240) to plan parameter values ​​to be controlled when the process is performed in the plasma equipment based on the process result prediction.

[0100] More specifically, the planning module (240) is called when the plasma state value predicted by the plasma state prediction module has a difference greater than or equal to a threshold from a preset range, when the process state predicted by the process state prediction module has a difference greater than or equal to a threshold from a preset process state range, and when the process result predicted by the process result prediction module has a difference greater than or equal to a threshold from a range of process results measured after the actual process is completed.

[0101] More specifically, if the predicted plasma state value from the plasma state prediction module differs by a threshold or more from a preset range, the planning module (240) can receive input from the user regarding the change items, change amounts, process parameters, number of processes, process execution time, etc. that have changed in the plasma state. Through this, the electronic device (200) can use the data input from the user as basic data for determining the timing of inspection of the plasma equipment and whether the plasma equipment is in operation.

[0102] If the process state predicted by the process state prediction module differs by a threshold or more from the range of the preset process state, the planning module (240) can receive from the user an input of a process influence related to a change item in which the plasma state has changed.

[0103] The planning module (240) can use an artificial intelligence algorithm to plan which parameters of the plasma equipment should be adjusted to which values ​​when a difference occurs between the process results predicted by the process result prediction module and the range of the actual measured process results that exceeds a threshold value.

[0104] The planning module (240) can generate a control combination based on an adjustable control factor in a parameter control combination condition having information on an adjustable item in at least one process among a process in which a predicted plasma state value deviates from a range of preset plasma state values ​​by a threshold or more, a process in which a predicted process state deviates from a range of preset process states by a threshold or more, and a process in which a predicted process result deviates from a range of an actually measured process result by a threshold or more.

[0105] The process control combination can perform learning by applying the sensing data acquired from the performed process and the predicted plasma state value, process state, and process result corresponding to the sensing data to the reinforcement learning algorithm, which is an artificial intelligence algorithm.

[0106] The planning module (240) performs a process impact assessment on the predicted control combination after the prediction of the process control combination is completed using a reinforcement learning algorithm. The process impact assessment determines whether the control combination exists within the range defined by the preset parameter control constraints, and if it is within the preset range, the recommended parameter control combination and the control information on the parameter adjustment amount are stored to be used as learning data for more stable control of the plasma equipment, and this can be loaded into the message queue (220) as a learning performance result.

[0107] For example, when performing process control when the electron density state changes at 2024.02.16. 14:10:20 and the process influence is determined to be 1.1%, the planning module (240) can adjust the Power item to an importance of 10, the Pressure item to an importance of 3, the MFC value position to an importance of 3, and the Pumping Speed ​​to an importance of 2 in the parameter control recommendation combination. In addition, the planning module (240) can set the control to increase the Power by 5 Watts according to the importance determination, and store the corresponding value of the parameter adjustment amount.

[0108] In step 407, the control module (250) of the electronic device (200) performs control of parameter values ​​based on the process control combination set in the planning module (240) and returns to step 319 of FIG. 3. For example, if the planning module (240) is set to perform control in the direction of increasing the power to 5 Watts, the control module (250) performs control to increase the power to 5 Watts according to the setting of the planning module (240). In addition, the control module (250) can add the process result to the learning performance result and load it into the message queue (220).

[0109] FIG. 5 is a detailed flowchart for explaining an intelligent operation method of plasma equipment in a forward direction according to an embodiment of the present invention.

[0110] Referring to FIG. 5, in step 501, the message queue (220) confirms the recipient for the sensing data loaded in the message queue (220) or the data derived from each module (hereinafter, referred to as “data”). At this time, the recipient may directly register a module requiring data in the message queue (220) through a user’s input, and the recipient may be at least one of a prediction module (230), a planning module (240), a control module (250), a reverse inference module (260), a display module (280), and a simulator (300).

[0111] Based on the confirmation in step 501, the message queue (220) copies the data required by the module or simulator (300) registered as the recipient and performs step 505. In step 505, the message queue (220) transmits the copied data to the recipient. For example, the registered recipient may be the prediction module (230) and the simulator (300).

[0112] In step 507, the electronic device (200) performs forward prediction using the prediction module (230) and the simulator (300). More specifically, the message queue (220) can simultaneously transmit the sensing data loaded in step 305 to the prediction module (230) and the simulator (300). Since the artificial intelligence algorithm stored in the prediction module (230) has completed pre-learning as in step 309 of FIG. 3, the prediction module (230) can apply the sensing message as input data to the artificial intelligence algorithm to predict the plasma state value, process state, and process result, and load the predicted value into the message queue (220).

[0113] More specifically, the plasma state prediction module can apply the sensing data as input data to an artificial intelligence algorithm to predict the electron density or electron temperature and load the predicted electron density or electron temperature into a message queue (220). The process state prediction module can apply the sensing data as input data to an artificial intelligence algorithm to predict the etching process and load the predicted etching process into a message queue (220). The process result prediction module can apply the sensing data as input data to an artificial intelligence algorithm to predict the process result, for example, the etching profile, and load the predicted process result into a message queue (220).

[0114] Additionally, the simulator (300) can predict process results, such as an etching profile, based on sensing data and load the predicted etching profile into the message queue (220).

[0115] In this way, the plasma prediction module, process state prediction module, process result prediction module, and simulator (300) included in the prediction module (230) can operate in parallel.

[0116] In step 509, the message queue (220) transmits the forward prediction result to the storage module (270) for storage or transmits the forward prediction result to the display module (280) for visualization and display of the forward prediction result on the display module (280), and performs step 511. At this time, the message queue (220) can confirm that the recipient of the predicted value predicted by the prediction module (230) and the etching profile predicted by the simulator (300) is the storage module (270) or the display module (280), and can transmit the corresponding data to each module. Through this, the electronic device (200) can perform storage and display of the corresponding data simultaneously, and can perform only at least one operation among storage or display.

[0117] In step 511, when the electronic device (200) receives a control signal for controlling the plasma equipment (100) from the user, the electronic device proceeds to step 513, performs control of the plasma equipment (100) based on the planned result in the planning module (240), and returns to step 319 of FIG. 3. At this time, the planning method in the planning module (240) has been specifically described in FIG. 4, so a detailed description thereof will be omitted.

[0118] Figure 6 is a detailed flowchart illustrating a method for intelligently operating plasma equipment in reverse according to an embodiment of the present invention. The method for intelligently operating plasma equipment in reverse refers to a method for operating plasma equipment by inferring equipment parameters that must be set in the plasma equipment to derive a desired process result.

[0119] Referring to FIG. 6, in step 601, the reverse inference module (260) receives a process result to be achieved (hereinafter, referred to as the achieved process result) from the user. For example, the achieved process result input by the user may be an etching profile to be confirmed as a result of the process after the process is completed. The reverse inference module (260) may exchange data with the message queue (220), the prediction module (230), the planning module (240), and the control module (250) as needed to infer equipment parameters.

[0120] In step 603, the reverse inference module (260) predicts the final plasma state value for deriving the achievement process result through data transmission and reception with the message queue (220). More specifically, the reverse inference module (260) applies the achievement process result to a regression-based artificial intelligence algorithm stored in the reverse inference module (260). The reverse inference module (260) applies the achievement process result to the artificial intelligence algorithm and predicts the plasma state value for deriving the achievement process result based on the plasma state value loaded in the message queue (220).

[0121] In step 605, the reverse inference module (260) predicts the final sensing data for deriving the final plasma state value. More specifically, the reverse inference module (260) applies the final plasma state value to a regression-based artificial intelligence algorithm. The reverse inference module (260) applies the final plasma value and predicts the final sensing data for deriving the final plasma state value based on the sensing data loaded into the message queue (220).

[0122] In step 607, the reverse inference module (260) predicts the final equipment parameter values ​​for deriving the final sensing data. More specifically, the reverse inference module (260) applies the final sensing data to a regression-based artificial intelligence algorithm. The reverse inference module (260) applies the final sensing data and predicts the final parameter values ​​for deriving the final sensing data based on the parameter values ​​loaded in the message queue (220).

[0123] In step 609, the reverse inference module (260) calls the control module (250) to provide the final equipment parameter values ​​selected by the control module (250), and the control module (250) controls the parameter values ​​of the plasma equipment using the final equipment parameter values. Through this, the present invention has the effect of being able to control in real time a power device, a gas inlet device, or a vacuum pump installed in the plasma equipment for the process result to be achieved by reverse inferring the equipment parameter values ​​that must be set in the plasma equipment to derive the process result to be achieved.

[0124] FIG. 7 is a diagram showing the point in time at which data is processed during intelligent operation of plasma equipment using a message queue according to an embodiment of the present invention.

[0125] Referring to FIG. 7, T1 to T6 represent the points in time when sensing data is collected and the points in time when the sensing data is actually processed. If the destinations of the sensing data loaded into the message queue (220) are the storage module (270), the prediction module (230), and the display module (280), the sensing data collected at point in time T1 are transmitted to the storage module (270), the prediction module (230), and the display module (280), respectively.

[0126] Accordingly, the storage module (270), prediction module (230), and display module (280) can store sensing data at time T1, perform plasma state prediction, process state prediction, and visualization of sensing data. More specifically, the electron density prediction, electron temperature prediction, and etching process prediction illustrated in FIG. 7 may be values ​​predicted by an artificial intelligence algorithm stored in the prediction module (230).

[0127] The prediction module (230) loads the predicted value into the message queue (220). If the recipients of the predicted value are the simulator (300) and the display module (280), the predicted value is transmitted to the simulator (300) and the display module (280) at time T1.

[0128] Accordingly, the simulator (300) can predict an etching profile based on the predicted value, and the display module (280) can visualize the predicted value. More specifically, the predicted data illustrated in FIG. 7 may refer to the predicted value.

[0129] The prediction module (230) loads the predicted etching profile into the message queue (220). If the message queue (220) receives the predicted etching profile from the display module (280), the predicted etching profile is transmitted to the display module (280) at time T1, and the display module (280) can perform visualization of the predicted etching profile.

[0130] Even if a problem occurs in the process of predicting electron density through data transmission based on a message queue (220), the process of predicting electron temperature can operate normally, and thus other processes that require predicted electron temperature values ​​can also operate normally. Accordingly, the present invention has the advantage of efficiently transmitting the data required by each module by using a message queue (220) to transmit and receive data, and since each module can operate independently, the speed of data processing is increased.

[0131] FIG. 8 is a diagram showing the point in time at which data of a plasma device is processed when a message queue is not used according to a general embodiment.

[0132] Referring to FIG. 8, T1 to T6 represent the points in time when sensing data is collected and the points in time when the sensing data is actually processed. When the message queue (220) is not used, the collected sensing data is stored in the storage module (270) and then transmitted to the display module (280), so that the sensing data can be visualized in the display module (280).

[0133] Next, the visualized sensing data is transmitted to the prediction module (230) to sequentially predict electron density, electron temperature, and etching process, predict an etching profile using the predicted values, and visualize the predicted values.

[0134] And the predicted etching profile is transmitted to the display module (280) so that the display module (280) can visualize the etching profile.

[0135] In this way, if a method of processing data by directly transferring it between modules without using a message queue (220) is used, it can be confirmed that the time required to predict the etching profile using the sensing data collected at time T1 is longer than that in Fig. 7. In addition, in the case of Fig. 8, if a problem occurs in the electron density prediction process, a problem occurs in that all operations performed after the electron density prediction process cannot be processed.

[0136] FIG. 9 is a screen example showing a change in a plasma state value or process state according to a change in sensing data according to an embodiment of the present invention.

[0137] Referring to FIG. 9, sensing data acquired according to the operation of the plasma equipment (100) and the plasma state, process state, and process results predicted based on the sensing data are provided in the form of a graph.

[0138] When a user executes a program for controlling intelligent operation through an electronic device (200) for intelligent operation of plasma equipment (100), the display module (280) can display a screen as in Fig. 9. At this time, reference numeral 901 represents sensing data acquired from a plurality of sensors provided in the plasma equipment (100) when the plasma equipment (100) is in operation, reference numeral 903 represents a plasma state value and a process state predicted through a plasma state prediction module and a process state prediction module included in the prediction module (230), and reference numeral 905 represents a process result predicted through a process result prediction module or simulator (300) included in the prediction module (230).

[0139] FIG. 10 is a screen example showing the change in plasma state value or process state over time according to an embodiment of the present invention.

[0140] Referring to FIG. 10, the electronic device (200) displays the accumulated results in the form of a graph according to the set time range or the set number of processes for the plasma state and the process state. At this time, the drawing reference numeral 1001 is a graph showing the changing aspect of the process result according to the accumulation of the process time and number of processes, and the drawing reference numeral 1003 is a graph providing standard deviation, moving average information, etc. for the plasma state value and the process state that change according to the accumulation of the process time and number of processes.

[0141] FIG. 11 and FIG. 12 are screen examples showing sensing data, plasma status values, and process status in normal and abnormal processes according to an embodiment of the present invention.

[0142] Referring to FIGS. 11 and 12, FIG. 11 is a screen that displays a list of cases in which a plasma process is performed normally and cases in which it is performed abnormally, and FIG. 12 is a screen that displays related data in a graph for each item so that the sensing data measured when performing a plasma process at a time related to the selected item in the list of FIG. 11 can be compared with the predicted plasma state and process state.

[0143] FIG. 13 is a screen example showing a method of operating a plasma device in forward mode and reverse mode according to an embodiment of the present invention.

[0144] Referring to FIG. 13, FIG. 13a shows a screen when the plasma equipment is to be operated in forward mode, and FIG. 13b shows a screen when the plasma equipment is to be operated in reverse mode.

[0145] Figure 13a shows conditions input to generate plasma in the plasma equipment (100) in the order of the images, such as equipment parameter values ​​such as power and pressure, and sensing data obtained from the plasma equipment (100) when operating the plasma equipment (100) with the equipment parameter values, such as OES sensing data and electron density by OES sensing data.

[0146] In this way, when a user inputs plasma generation conditions, the electronic device (200) can derive and display sensing data and electron density, which is a plasma state value, based on the plasma generation conditions, as shown in Fig. 13 a. At this time, the user can confirm in real time that the plasma state value changes as the plasma generation conditions are changed.

[0147] Figure 13b shows the final plasma state value to be obtained for the desired process result to be achieved in the order of the images, the plasma optical diagnostic data generation result corresponding to the plasma state value, e.g., the final sensing data, the equipment power parameter and equipment pressure parameter to be input for the final sensing data, e.g., the final equipment parameter value.

[0148] In this way, when a user inputs the achieved process result, the electronic device (200) can derive and display the final plasma state value, final sensing data, and final equipment parameter value based on the achieved process result, as shown in FIG. 13b. At this time, the user can check in real time that the sensing data and equipment parameter values ​​change while adjusting the final plasma state value.

[0149] Through this, the present invention has the effect of controlling a power device, a gas inlet device, or a vacuum pump installed in a plasma device in real time to achieve a process result to be achieved by reversely inferring the equipment parameter values ​​that must be set in the plasma device to derive the process result to be achieved.

[0150] The embodiments of the present invention disclosed in this specification and drawings are merely specific examples intended to facilitate understanding and easily explain the technical content of the present invention, and are not intended to limit the scope of the present invention. Therefore, the scope of the present invention should be interpreted to include all modifications or variations derived based on the technical concept of the present invention, in addition to the embodiments disclosed herein.

Claims

1. A message queue that loads sensing data acquired from multiple sensors according to input conditions input to the plasma equipment during operation of the plasma equipment; A prediction module that applies the above sensing data to a pre-learned artificial intelligence algorithm to predict plasma state values, process states, and process results; A planning module that plans parameter values ​​for controlling the plasma equipment by applying the plasma state value, the process state, and the process result to a pre-learned artificial intelligence algorithm; and A control module that controls the plasma equipment using the above-mentioned planned parameter values ​​as the input conditions; An electronic device for intelligent operation, characterized by including:

2. In paragraph 1, A collection module that collects the above sensing data and transmits it to the message queue; An electronic device for intelligent operation, characterized by including:

3. In paragraph 2, The above prediction module, It includes a plasma state prediction module that predicts the plasma state value, a process state prediction module that predicts the process state, and a process result prediction module that predicts the process result. An electronic device for intelligent operation, wherein each of the above prediction modules uses a different artificial intelligence algorithm.

4. In paragraph 3, The above message queue is, An electronic device for intelligent operation, characterized in that, if a plurality of modules included in the above prediction module are registered as multiple receivers for receiving the sensing data, the sensing data is transmitted to each of the modules registered as the multiple receivers.

5. In paragraph 4, The above message queue is, An electronic device for intelligent operation, characterized in that the sensing data is transmitted to the simulator when the simulator is registered as the plurality of receivers.

6. In paragraph 5, The above planning module is, An electronic device for intelligent operation, characterized in that the planning is performed based on data received from the prediction module or the simulator through the message queue.

7. In paragraph 2, A reverse inference module that reversely infers the final plasma state value, final sensing data, and final parameter value to derive the desired process result; An electronic device for intelligent operation, characterized by including:

8. In paragraph 7, The above reverse inference module is, An electronic device for intelligent operation characterized in that the final parameter value is inferred by applying the above-described achievement process result to an artificial intelligence algorithm for reverse inference.

9. In paragraph 8, The above control module, An electronic device for intelligent operation, characterized in that it controls the plasma equipment based on the final parameter value inferred from the reverse inference module.

10. In paragraph 7, The above prediction module, the planning module, the control module and the reverse inference module, An electronic device for intelligent operation, characterized in that it loads data derived from each module into the message queue and independently operates by collecting the data loaded into the message queue as needed.

11. A step of collecting sensing data obtained from multiple sensors when the electronic device operates the plasma equipment and loading the data into a message queue; and The electronic device controls the plasma equipment by planning parameter values ​​for controlling the plasma equipment based on the plasma state value, process state, and process result predicted by the sensing data according to the selection of the operating mode, or A step of controlling the plasma equipment by reversely inferring the final plasma state value, final sensing data, and final parameter value to derive the desired process result; A method for intelligent operation, characterized by including:

Images