Learning device, information processing device, substrate processing device, substrate processing system, learning method, and processing condition determination method

Abstract

To perform machine learning of a condition that changes with the elapse of time in order to process a substrate.SOLUTION: A learning device 200 includes: an experimental data acquisition part 261 for acquiring a processing amount showing the difference of film thickness before and after processing a coating film formed on a substrate after a substrate processing device for processing the substrate by supplying process liquid to the substrate with the coating film formed is driven under a processing condition including a variation condition that changes with the lapse of time; a first compression part 263 for converting the variation condition so as to reduce the number of dimensions; and predictor generation part 265 for generating a learning model for performing machine learning of learning data including a conversion result obtained by converting the variation condition by the first compression part 263 and a processing amount corresponding to the processing condition and estimating a processing amount showing the difference of film thickness before and after processing the coating film formed on the substrate before being processed by the substrate processing device.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present invention relates to a learning device, an information processing device, a substrate processing device, a substrate processing system, a learning method, and a processing condition determination method, and relates to a learning device that generates a learning model that simulates processing according to processing conditions by a substrate processing device, an information processing device that determines processing conditions using the learning model, a substrate processing device equipped with the information processing device, a learning method executed by the learning device, and a processing condition determination method executed by the information processing device. [Background technology]

[0002] Semiconductor manufacturing processes include cleaning processes. In cleaning processes, the thickness of a coating formed on a substrate is adjusted by etching, which involves applying a chemical solution to the substrate. For this film thickness adjustment, it is important to perform the etching process so that the substrate surface is uniform, or to flatten the substrate surface through the etching process. When discharging etching solution from a nozzle onto a portion of the substrate, the nozzle must be moved radially relative to the substrate. However, the etching process is a complex process in which the amount of coating processed varies depending on the nozzle movement. Furthermore, the amount of coating processed by the etching process is determined after the substrate is processed. Therefore, setting the nozzle movement operation requires trial and error by engineers. Determining the optimal nozzle operation requires significant cost and time.

[0003] Japanese Patent Application Laid-Open No. 2021-108367 describes an apparatus that determines scan speed information from a target processing amount using a trained model that has been machine-learned using learning data in which the "input" is the processing amount (etching amount) and the "output" is scan speed information. According to this technology, one piece of scan speed information is determined from the target processing amount. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2021-108367 Summary of the Invention [Problem to be solved by the invention]

[0005] On the other hand, it is desirable to make the nozzle movement operation more complex. The nozzle movement operation is time-series data that indicates the position that changes over time. If the nozzle movement operation is made more complex, the sampling interval increases, and the number of dimensions of the time-series data increases. Generally, as the number of dimensions of the learning data increases, the amount of data required for machine learning increases exponentially. Therefore, as the number of dimensions of the learning data increases, it becomes difficult to optimize the learning model obtained by machine learning. Furthermore, because etching is a complex process, there is not necessarily one nozzle operation that is suitable for the target processing volume, but rather there may be multiple.

[0006] An object of the present invention is to provide a learning device and a substrate processing system suitable for machine learning of conditions that change over time for processing a substrate.

[0007] Another object of the present invention is to provide an information processing apparatus, a substrate processing apparatus, a substrate processing system, and a processing condition determination method that are capable of presenting multiple processing conditions for the processing results of a complex process for processing a substrate. [Means for solving the problem]

[0008] The system includes an experimental data acquisition unit that acquires a processing amount indicating the difference in film thickness between before and after processing of a coating formed on a substrate after the substrate processing apparatus, which processes the substrate by supplying a processing liquid to the substrate on which a coating is formed, is operated under processing conditions including variable conditions that vary over time; a compressor that converts the variable conditions to reduce the number of dimensions; and a model generation unit that performs machine learning on learning data including the conversion results of the variable conditions converted by the compressor and the processing amount corresponding to the processing conditions, to generate a learning model that estimates the processing amount indicating the difference in film thickness between before and after processing of a coating formed on a substrate before it is processed by the substrate processing apparatus.

[0009] According to this aspect, the learning data includes a transformation result obtained by transforming the data so as to reduce the number of dimensions of the variable conditions that change over time, and a processing amount. This reduces the number of dimensions of the learning data. As a result, a learning device suitable for machine learning of conditions that change over time for processing a substrate can be provided.

[0010] Preferably, the system further includes a variation condition generation unit that generates a plurality of variation conditions, and a compressor generation unit that generates a compressor by having an autoencoder learn the plurality of variation conditions generated by the variation condition generation unit.

[0011] According to this aspect, the compressor is generated by having the autoencoder learn the variation conditions, which makes it easier to generate the compressor.

[0012] Preferably, the processing conditions further include fixed conditions that do not vary over time, and the learning data includes the fixed conditions.

[0013] According to this aspect, it is possible to deal with processes with different fixed conditions, and there is no need to generate multiple learning models with different fixed conditions.

[0014] Preferably, the system further includes a performance acquisition unit that provides the learning model with the conversion results obtained by converting the tentative variable conditions by the compressor, and operates the substrate processing apparatus under processing conditions including the tentative variable conditions if the processing volume estimated by the learning model satisfies the tolerance conditions, thereby acquiring a processing volume indicating the difference in film thickness before and after processing of a coating formed on a substrate processed by the substrate processing apparatus, and an additional learning unit that trains the learning model using additional learning data including the conversion results obtained by converting the tentative variable conditions by the compressor and the processing volume acquired by the performance acquisition unit.

[0015] According to this aspect, the learning model is additionally trained, so that the performance of the learning model can be improved.

[0016] Preferably, the system further includes a distillation unit that provides the learning model with the conversion results obtained by converting the tentative variable conditions using the compressor, and generates a new learning model using distillation data including the conversion results and the processing amount estimated by the learning model if the processing amount estimated by the learning model satisfies the tolerance condition.

[0017] Following this approach makes it easier to prepare data for training a new learning model.

[0018] Preferably, the substrate processing apparatus includes a nozzle that supplies processing liquid to the substrate, and a moving unit that changes the relative position between the nozzle and the substrate, and the variable condition is the relative position between the nozzle and the substrate that is changed by the moving unit.

[0019] According to this aspect, a learning model is generated that estimates the amount of processing of a coating that is processed by supplying a processing liquid to a substrate from a nozzle while changing the relative position between the nozzle and the substrate, thereby enabling the generation of a learning model that estimates the amount of processing in an etching process.

[0020] According to another aspect of the present invention, an information processing apparatus is an information processing apparatus that manages a substrate processing apparatus, and the substrate processing apparatus processes a substrate by supplying a processing liquid to a substrate on which a coating has been formed under processing conditions including variable conditions that vary over time, and includes a second compressor that converts the variable conditions so as to reduce the number of dimensions, and a processing condition determination unit that determines processing conditions for driving the substrate processing apparatus using a learning model that estimates a processing amount indicating the difference in film thickness before and after processing of a coating formed on a substrate before being processed by the substrate processing apparatus, wherein the learning model is an inference model that has been machine-learned using learning data that includes the conversion results obtained by converting the variable conditions included in the processing conditions under which the substrate processing apparatus processed the substrate using a first compressor that is the same as the second compressor, and the processing amount indicating the difference in film thickness before and after processing of a coating formed on a substrate processed by the substrate processing apparatus, and the processing condition determination unit provides the conversion results obtained by converting the hypothetical variable conditions by the second compressor to the learning model, and determines the processing conditions including the hypothetical variable conditions as the processing conditions for driving the substrate processing apparatus if the processing amount estimated by the learning model satisfies the allowable condition.

[0021] According to this aspect, when the conversion result obtained by converting the tentative variable conditions that vary over time is provided to the learning model, and the processing conditions including the tentative variable conditions are determined as the processing conditions for driving the substrate processing apparatus when the processing amount estimated by the learning model satisfies the allowable condition, it is possible to determine a plurality of tentative variable conditions for the processing amount that satisfies the allowable condition. As a result, it is possible to provide an information processing apparatus that can present a plurality of processing conditions for the processing result of a complex process for processing substrates.

[0022] Preferably, the substrate processing apparatus includes the above-described information processing apparatus.

[0023] According to this aspect, it is possible to provide a substrate processing apparatus capable of presenting a plurality of processing conditions for the processing results of a complex process for processing a substrate.

[0024] Preferably, the substrate processing system is a substrate processing system that manages a substrate processing apparatus that processes substrates, and includes a learning device and an information processing device, wherein the substrate processing apparatus processes the substrate by supplying a processing liquid to the substrate on which a film is formed under processing conditions including variable conditions that vary over time, and the learning device includes an experimental data acquisition unit that acquires a processing amount indicating a difference in film thickness before and after processing of a film formed on the substrate after the substrate processing apparatus is operated under the processing conditions, a first compressor that converts the variable conditions so as to reduce the number of dimensions, and learning data including a conversion result obtained by converting the variable conditions by the first compressor and a processing amount corresponding to the processing conditions. and a model generation unit that performs machine learning to generate a learning model that estimates a processing amount indicating the difference in film thickness before and after processing of a coating formed on a substrate before processing by the substrate processing apparatus, and the information processing apparatus includes a second compressor that is the same as the first compressor, and a processing condition determination unit that determines processing conditions for operating the substrate processing apparatus using the learning model generated by the learning apparatus, and the processing condition determination unit provides the learning model with the conversion results obtained by converting the tentative variable conditions by the second compressor, and determines the processing conditions including the tentative variable conditions as the processing conditions for operating the substrate processing apparatus if the processing amount estimated by the learning model satisfies the allowable condition.

[0025] According to this aspect, it is possible to provide a substrate processing system that is suitable for machine learning of conditions that change over time for processing a substrate, and that is capable of presenting multiple processing conditions for the processing results of a complex process for processing a substrate.

[0026] According to yet another aspect of the present invention, the learning method causes a computer to perform the following processes after a substrate processing apparatus that processes substrates by supplying a processing liquid to a substrate on which a coating has been formed is operated under processing conditions including variable conditions that vary over time: a process of acquiring a processing amount indicating the difference in film thickness of a coating formed on a substrate before and after processing; a compression process of converting the variable conditions to reduce the number of dimensions; and a process of machine learning learning data that includes the conversion results of the variable conditions converted in the compression process and the processing amount corresponding to the processing conditions to generate a learning model that estimates the processing amount indicating the difference in film thickness of a coating formed on a substrate before it is processed by the substrate processing apparatus.

[0027] According to this aspect, it is possible to provide a learning method suitable for machine learning of conditions that change over time for processing a substrate.

[0028] According to yet another aspect of the present invention, a processing condition determination method is a processing condition determination method executed by a computer that manages a substrate processing apparatus, wherein the substrate processing apparatus processes a substrate by supplying a processing liquid to a substrate on which a coating has been formed under processing conditions including variable conditions that vary over time, and includes a process of converting the variable conditions so as to reduce the number of dimensions, and a process of determining processing conditions for driving the substrate processing apparatus using a learning model that estimates a processing amount indicating the difference in film thickness before and after processing of a coating formed on a substrate before processing by the substrate processing apparatus, wherein the learning model is an inference model that has been machine-learned to learn data including a conversion result obtained by converting variable conditions included in the processing conditions under which the substrate was processed by the substrate processing apparatus using the same process as the converting process, and a processing amount indicating the difference in film thickness before and after processing of a coating formed on a substrate processed by the substrate processing apparatus, and the process of determining processing conditions includes a process of providing the conversion result obtained by converting the hypothetical variable conditions by the converting process to the learning model, and determining the processing conditions including the hypothetical variable conditions as the processing conditions for driving the substrate processing apparatus if the processing amount estimated by the learning model satisfies the allowable condition.

[0029] According to this aspect, it is possible to provide a processing condition determination method capable of presenting a plurality of processing conditions for the processing result of a complex process for processing a substrate. [Effects of the Invention]

[0030] A learning device and a substrate processing system suitable for machine learning conditions that change over time for processing a substrate can be provided.

[0031] Furthermore, it is possible to provide an information processing apparatus, a substrate processing apparatus, a substrate processing system, and a processing condition determining method that are capable of presenting a plurality of processing conditions for the processing results of a complex process for processing a substrate. [Brief explanation of the drawings]

[0032] [Figure 1] 1 is a diagram illustrating a configuration of a substrate processing system according to an embodiment of the present invention. [Figure 2] FIG. 1 illustrates an example of a configuration of an information processing device. [Figure 3] FIG. 1 illustrates an example of the configuration of a learning device. [Figure 4] 1 is a diagram showing an example of a functional configuration of a substrate processing system according to one embodiment of the present invention. [Figure 5] FIG. 1 is a first diagram showing an example of a nozzle operation pattern. [Figure 6] FIG. 10 is a second diagram showing an example of a nozzle operation pattern. [Figure 7] FIG. 10 is a diagram showing an example of film thickness characteristics. [Figure 8] FIG. 10 is a diagram illustrating a predictor. [Figure 9] 10 is a flowchart illustrating an example of the flow of an autoencoder generation process. [Figure 10] 10 is a flowchart illustrating an example of the flow of a predictor generation process. [Figure 11] 10 is a flowchart showing an example of the flow of a processing condition determination process. [Figure 12]10 is a flowchart illustrating an example of the flow of an additional learning process. DETAILED DESCRIPTION OF THE INVENTION

[0033] A substrate processing system according to an embodiment of the present invention will be described in detail below with reference to the drawings. In the following description, the term "substrate" refers to a semiconductor substrate (semiconductor wafer), a substrate for an FPD (Flat Panel Display) such as a liquid crystal display device or an organic EL (Electro Luminescence) display device, a substrate for an optical disk, a substrate for a magnetic disk, a substrate for a magneto-optical disk, a substrate for a photomask, a ceramic substrate, a substrate for a solar cell, or the like.

[0034] 1. Overall configuration of the substrate processing system Fig. 1 is a diagram illustrating the configuration of a substrate processing system according to one embodiment of the present invention. The substrate processing system 1 in Fig. 1 includes an information processing device 100, a learning device 200, and a substrate processing device 300. The learning device 200 is, for example, a server, and the information processing device 100 is, for example, a personal computer.

[0035] The learning device 200 and the information processing device 100 are used to manage the substrate processing device 300. The number of substrate processing devices 300 managed by the learning device 200 and the information processing device 100 is not limited to one, and a plurality of substrate processing devices 300 may be managed.

[0036] In the substrate processing system 1 according to this embodiment, the information processing device 100, the learning device 200, and the substrate processing device 300 are connected to one another by a wired or wireless communication line or a communication network. The information processing device 100, the learning device 200, and the substrate processing device 300 are each connected to a network, and are capable of transmitting and receiving data to and from one another. The network may be, for example, a local area network (LAN) or a wide area network (WAN). The network may also be the Internet. The information processing device 100 and the substrate processing device 300 may also be connected by a dedicated communication network. The network connection may be a wired connection or a wireless connection.

[0037] The learning device 200 does not necessarily need to be connected to the substrate processing device 300 and the information processing device 100 via a communication line or a communication network. In this case, data generated by the substrate processing device 300 may be transferred to the learning device 200 via a recording medium. Also, data generated by the learning device 200 may be transferred to the information processing device 100 via a recording medium.

[0038] A display device, a sound output device, and an operation unit (not shown) are provided in the substrate processing apparatus 300. The substrate processing apparatus 300 is operated according to predetermined processing conditions (processing recipe) of the substrate processing apparatus 300.

[0039] 2. Overview of substrate processing equipment The substrate processing apparatus 300 includes a control device 10 and a plurality of substrate processing units WU. The control device 10 controls the plurality of substrate processing units WU. The plurality of substrate processing units WU processes the substrates by supplying a processing liquid to the substrates W having a coating formed thereon. The processing liquid includes an etching liquid, and the substrate processing units WU perform an etching process. The etching liquid is a chemical liquid. The etching liquid is, for example, hydrofluoric nitric acid (a mixture of hydrofluoric acid (HF) and nitric acid (HNO3)), hydrofluoric acid, buffered hydrofluoric acid (BHF), ammonium fluoride, HFEG (a mixture of hydrofluoric acid and ethylene glycol), or phosphoric acid (H3PO4).

[0040] The substrate processing unit WU includes a spin chuck SC, a spin motor SM, a nozzle 311, and a nozzle moving mechanism 301. The spin chuck SC holds the substrate W horizontally. The spin motor SM has a first rotation axis AX1. The first rotation axis AX1 extends in the vertical direction. The spin chuck SC is attached to the upper end of the first rotation axis AX1 of the spin motor SM. When the spin motor SM rotates, the spin chuck SC rotates about the first rotation axis AX1. The spin motor SM is a stepping motor. The substrate W held by the spin chuck SC rotates about the first rotation axis AX1. Therefore, the rotation speed of the substrate W is the same as the rotation speed of the stepping motor. Note that if an encoder is provided to generate a rotation speed signal indicating the rotation speed of the spin motor, the rotation speed of the substrate W may be obtained from the rotation speed signal generated by the encoder. In this case, a motor other than a stepping motor may be used as the spin motor.

[0041] The nozzle 311 supplies the etching liquid to the substrate W. The nozzle 311 receives the etching liquid from an etching liquid supply unit (not shown) and ejects the etching liquid toward the substrate W while it is rotating.

[0042] The nozzle movement mechanism 301 moves the nozzle 311 in a substantially horizontal direction. Specifically, the nozzle movement mechanism 301 has a nozzle motor 303 having a second rotation axis AX2, and a nozzle arm 305. The nozzle motor 303 is disposed so that the second rotation axis AX2 is aligned in a substantially vertical direction. The nozzle arm 305 has a longitudinal shape that extends linearly. One end of the nozzle arm 305 is attached to the upper end of the second rotation axis AX2 so that the longitudinal direction of the nozzle arm 305 is in a different direction from the second rotation axis AX2. The nozzle 311 is attached to the other end of the nozzle arm 305 so that its outlet faces downward.

[0043] When the nozzle motor 303 operates, the nozzle arm 305 rotates in a horizontal plane around the second rotation axis AX2. As a result, the nozzle 311 attached to the other end of the nozzle arm 305 moves (pivots) in the horizontal direction around the second rotation axis AX2. The nozzle 311 ejects the etching liquid toward the substrate W while moving in the horizontal direction. The nozzle motor 303 is, for example, a stepping motor.

[0044] The control device 10 includes a CPU (Central Processing Unit) and a memory, and the CPU executes a program stored in the memory to control the entire substrate processing apparatus 300. The control device 10 controls the spin motor SM and the nozzle motor 303.

[0045] The learning device 200 receives experimental data from the substrate processing apparatus 300, performs machine learning to create a learning model using the experimental data, and outputs the learned learning model to the information processing device 100.

[0046] The information processing apparatus 100 uses the trained learning model to determine processing conditions for processing a substrate that is to be processed by the substrate processing apparatus 300. The information processing apparatus 100 outputs the determined processing conditions to the substrate processing apparatus 300.

[0047] Fig. 2 is a diagram showing an example of the configuration of an information processing device. Referring to Fig. 2, information processing device 100 is configured with CPU 101, RAM (random access memory) 102, ROM (read only memory) 103, storage device 104, operation unit 105, display device 106, and input / output I / F (interface) 107. CPU 101, RAM 102, ROM 103, storage device 104, operation unit 105, display device 106, and input / output I / F 107 are connected to bus 108.

[0048] The RAM 102 is used as a working area for the CPU 101. The ROM 103 stores a system program. The storage device 104 includes a storage medium such as a hard disk or semiconductor memory, and stores the program. The program may be stored in the ROM 103 or another external storage device.

[0049] A CD-ROM 109 is detachably attached to the storage device 104. The recording medium for storing the program executed by the CPU 101 is not limited to the CD-ROM 109, and may be an optical disk (MO (Magnetic Optical Disc) / MD (Mini Disc) / DVD (Digital Versatile Disc)), an IC card, an optical card, a semiconductor memory such as a mask ROM, or an EPROM (Erasable Programmable ROM). Furthermore, the CPU 101 may download the program from a computer connected to the network and store it in the storage device 104, or the computer connected to the network may write the program to the storage device 104, and the program stored in the storage device 104 may be loaded into the RAM 102 and executed by the CPU 101. The program here includes not only a program that can be directly executed by the CPU 101, but also a source program, a compressed program, an encrypted program, etc.

[0050] The operation unit 105 is an input device such as a keyboard, a mouse, or a touch panel. A user can give predetermined instructions to the information processing device 100 by operating the operation unit 105. The display device 106 is a display device such as a liquid crystal display device, and displays a GUI (Graphical User Interface) or the like for receiving instructions from the user. The input / output I / F 107 is connected to a network.

[0051] Fig. 3 is a diagram showing an example of the configuration of a learning device. Referring to Fig. 3, learning device 200 is composed of CPU 201, RAM 202, ROM 203, storage device 204, operation unit 205, display device 206, and input / output I / F 207. CPU 201, RAM 202, ROM 203, storage device 204, operation unit 205, display device 206, and input / output I / F 207 are connected to bus 208.

[0052] The RAM 202 is used as a working area for the CPU 201. The ROM 203 stores a system program. The storage device 204 includes a storage medium such as a hard disk or semiconductor memory, and stores the program. The program may be stored in the ROM 203 or another external storage device. A CD-ROM 209 is detachably attached to the storage device 204.

[0053] The operation unit 205 is an input device such as a keyboard, a mouse, a touch panel, etc. The input / output I / F 207 is connected to a network.

[0054] 3. Functional configuration of the substrate processing system 4 is a diagram showing an example of a functional configuration of a substrate processing system according to one embodiment of the present invention. Referring to FIG. 4, a control device 10 included in a substrate processing apparatus 300 controls a substrate processing unit WU to process a substrate W according to processing conditions. The processing conditions are conditions for processing a substrate W for a predetermined processing time. The processing time is a time determined for processing a substrate. In this embodiment, the processing time is the time during which a nozzle 311 discharges an etching solution onto the substrate W.

[0055] In this embodiment, the processing conditions include the temperature of the etching liquid, the concentration of the etching liquid, the flow rate of the etching liquid, the rotation speed of the substrate W, and the relative position of the nozzle 311 and the substrate W. The processing conditions also include variable conditions that change over time. In this embodiment, the variable condition is the relative position of the nozzle 311 and the substrate W. The relative position is indicated by the rotation angle of the nozzle motor 303. The processing conditions also include fixed conditions that do not change over time. In this embodiment, the fixed conditions are the temperature of the etching liquid, the concentration of the etching liquid, the flow rate of the etching liquid, and the rotation speed of the substrate W.

[0056] The learning device 200 trains the learning model using the learning data to generate an inference model that infers an etching profile from processing conditions. Hereinafter, the inference model generated by the learning device 200 will be referred to as a predictor.

[0057] Learning device 200 includes a variable condition generation unit 251, a compressor generation unit 253, a compressor transmission unit 255, an experimental data acquisition unit 261, a first compression unit 263, a predictor generation unit 265, and a predictor transmission unit 267. The functions of learning device 200 are realized by CPU 201 included in learning device 200 as CPU 201 executes a learning program stored in RAM 202.

[0058] The variable condition generation unit 251 generates a variable condition. For example, the variable condition generation unit 251 generates multiple variable conditions through calculation. In this embodiment, the substrate W rotates about a first rotation axis AX1, and the nozzle 311 rotates about a second rotation axis AX2. Therefore, a change in the relative position between the nozzle 311 and the substrate W is indicated by a change in the position of the nozzle 311. The change in the position of the nozzle 311 is determined by the rotation angle of the nozzle motor 303. Furthermore, the range of the rotation angle of the nozzle motor 303 is limited to a predetermined range. Furthermore, the processing time is a predetermined period. Therefore, the change in the position of the nozzle 311 can be classified into multiple operation patterns. For example, multiple operation patterns are determined in which at least one of the movement speed of the nozzle 311, the speed change position indicating the position where the speed is changed, the movement direction of the nozzle 311, and the number of times the nozzle 311 reverses its speed is fixed. The variable condition generation unit 251 determines a function for each of the multiple operation patterns and generates a variable condition using the function.

[0059] 5 is a diagram showing an example of a nozzle operation pattern. Referring to FIG. 5, three variation conditions are shown in which the number of times the nozzle 311 reverses its speed is varied while other variables are kept the same. The upper row shows a variation condition in which there are three reversals from the positive side to the negative side, the middle row shows a variation condition in which there are four reversals from the positive side to the negative side, and the lower row shows a variation condition in which there are five reversals from the positive side to the negative side.

[0060] Fig. 6 is a diagram showing an example of a nozzle operation pattern. Referring to Fig. 6, three variation conditions are shown where the acceleration at which the nozzle 311 accelerates or decelerates is varied while other variables are kept the same. The upper row shows a variation condition for a constant speed, the middle row shows a variation condition for which the acceleration gradually decreases, and the lower row shows a variation condition for which the acceleration gradually increases.

[0061] 4 , the compressor generation unit 253 generates a compressor by machine learning the plurality of variable conditions generated by the variable condition generation unit 251. The compressor transmission unit 255 transmits the compressor generated by the compressor generation unit 253 to the information processing device 100.

[0062] The compressor generation unit 253 performs machine learning using a neural network. In this embodiment, the compressor generation unit 253 is configured with an autoencoder. The compressor generation unit 253 causes the autoencoder to machine-learn the variation conditions generated by the variation condition generation unit 251. Specifically, the compressor generation unit 253 inputs the variation conditions generated by the variation condition generation unit 251 to the autoencoder as generation input data and determines parameters so that the output of the autoencoder is equal to the generation input data. The autoencoder is configured with an input layer, a hidden layer, and an output layer. The portion from the input layer to the hidden layer corresponds to the compressor, and the portion from the hidden layer to the output layer corresponds to the decoder. The compressor generation unit 253 generates a neural network as a compressor, incorporating parameters set in the input layer to hidden layer portion of a trained autoencoder. The compressor is an inference program incorporating parameters set in the input layer to hidden layer portion of a trained autoencoder.

[0063] The compressor generation unit 253 may normalize the variation condition and have the autoencoder learn the normalized variation condition by machine learning. The variation condition indicates the relative position of the nozzle 311 with respect to the substrate W, and will be described as being in the range of, for example, -150 mm to 150 mm. Normalizing the variation condition scales it to the range of 0 to 1. Normalizing the variation condition unifies the range that the parameters can take, facilitating machine learning by the autoencoder.

[0064] The compressor generation unit 253 classifies multiple variable conditions prepared in advance for machine learning into first variable conditions used for machine learning and second variable conditions used for evaluating the compressor, has the autoencoder learn the first variable conditions by machine learning, and evaluates the performance of the autoencoder under the second variable conditions.

[0065] The experimental data acquisition unit 261 acquires experimental data from the substrate processing apparatus 300. The experimental data includes processing conditions used when the substrate processing apparatus 300 actually processes the substrate W, and film thickness characteristics of the coating formed on the substrate W before and after processing. The film thickness characteristics are represented by film thicknesses of the coating formed on the substrate W at a plurality of different positions in the radial direction of the substrate W.

[0066] Fig. 7 is a diagram showing an example of film thickness characteristics. Referring to Fig. 7, the horizontal axis represents the radial position of the substrate, and the vertical axis represents film thickness. The origin of the horizontal axis represents the center of the substrate. The solid line represents the film thickness of the film formed on the substrate W before processing by the substrate processing apparatus 300. The substrate processing apparatus 300 performs a process of applying an etching solution in accordance with processing conditions, thereby adjusting the film thickness of the film formed on the substrate W. The dotted line represents the film thickness of the film formed on the substrate W after processing by the substrate processing apparatus 300.

[0067] The difference between the thickness of the film formed on the substrate W before processing by the substrate processing apparatus 300 and the thickness of the film formed on the substrate W after processing by the substrate processing apparatus 300 is the processing amount (etching amount). The processing amount indicates the thickness of the film reduced by the process of applying an etching solution by the substrate processing apparatus 300. The radial distribution of the processing amount is called the etching profile. The etching profile includes the processing amount at each of multiple positions in the radial direction of the substrate W.

[0068] Furthermore, it is desirable that the film thickness formed by the substrate processing apparatus 300 be uniform over the entire surface of the substrate W. For this reason, a target film thickness is set for the processing performed by the substrate processing apparatus 300. The target film thickness is indicated by a dashed dotted line. The deviation characteristic is the difference between the film thickness of the film formed on the substrate W after processing by the substrate processing apparatus 300 and the target film thickness. The deviation characteristic includes the difference at each of multiple positions in the radial direction of the substrate W.

[0069] Returning to Fig. 4, the first compression unit 263 converts the variable conditions included in the processing conditions of the experimental data input from the experimental data acquisition unit 261 into a low-dimensional data set using a compressor generated by the compressor generation unit 253. Here, the compressor generated by the compressor generation unit 253 is a neural network. The first compression unit 263 inputs the variable conditions to the neural network, which is the compressor, and outputs the conversion result by the compressor to the predictor generation unit 265.

[0070] The predictor generation unit 265 receives the conversion result of the variable conditions converted from the first compression unit 263, and receives the experimental data from the experimental data acquisition unit 261. The predictor generation unit 265 generates a predictor by performing supervised learning on a neural network.

[0071] Specifically, the training data includes input data and correct answer data. The input data includes the conversion result of the variable conditions converted by the first compression unit 263 and fixed conditions other than the variable conditions of the processing conditions included in the experimental data. The correct answer data includes an etching profile. The etching profile is the difference between the film thickness characteristics of the coating before processing included in the experimental data and the film thickness characteristics of the coating after processing included in the experimental data. The predictor generation unit 265 inputs the input data to a neural network and determines the parameters of the neural network so that the output of the neural network is equal to the correct answer data. The predictor generation unit 265 generates a neural network as a predictor, incorporating the parameters set in the trained neural network. The predictor is an inference program incorporating the parameters set in the trained neural network. The predictor generation unit 265 transmits the predictor to the information processing device 100.

[0072] FIG. 8 is a diagram illustrating a predictor. Referring to FIG. 8, the predictor includes an input layer, a hidden layer, and an output layer, and each layer includes multiple nodes indicated by circles. Note that the figure shows one hidden layer, but the number of hidden layers may be greater than this. Also, five nodes are shown in the input layer, four in the hidden layer, and three in the output layer, but the number of nodes is not limited to this. The output of a higher node is connected to the input of a lower node. The parameters include a coefficient that weights the output of the higher node. Also, the number of hidden layers is one or more, and is not limited.

[0073] When the predictor receives input of the fixed conditions and the conversion results obtained by converting the variable conditions into a low-dimensional data set, it outputs an etching profile. The etching profile is represented by the difference E[n] in film thickness before and after processing at each of a plurality of positions P[n] (n is an integer equal to or greater than 1) in the radial direction of the substrate W. Note that although the figure shows three output nodes for the predictor, in reality the number of output nodes is n.

[0074] 4 , the information processing device 100 includes a processing condition determination unit 151, a compressor receiving unit 153, a predictor receiving unit 155, a second compression unit 157, a prediction unit 159, an evaluation unit 161, and a processing condition transmission unit 163. The functions of the information processing device 100 are realized by the CPU 101 provided in the information processing device 100 as the CPU 101 executes a processing condition determination program stored in the RAM 102.

[0075] The compressor receiving unit 153 receives the compressor transmitted from the learning device 200 and outputs the received compressor to the second compression unit 157. The predictor receiving unit 155 receives the predictor transmitted from the learning device 200 and outputs the received predictor to the prediction unit 159.

[0076] The processing condition determination unit 151 determines processing conditions for a substrate W to be processed by the substrate processing apparatus 300. The processing condition determination unit 151 outputs variable conditions included in the processing conditions to the second compression unit 157, and outputs fixed conditions included in the processing conditions to the prediction unit 159. The processing condition determination unit 151 selects one from a plurality of variable conditions prepared in advance using experimental design, pairwise analysis, or Bayesian estimation, and determines the processing condition including the selected variable condition and the fixed condition as the processing condition for the prediction unit 159 to estimate. It is preferable that the plurality of variable conditions prepared in advance are the plurality of variable conditions generated by the learning device 200 to generate a compressor.

[0077] The second compression unit 157 uses a compressor to compress the variable conditions input from the processing condition determination unit 151. Specifically, the second compression unit 157 inputs the variable conditions to the compressor and outputs the output of the compressor to the prediction unit 159 as the conversion result.

[0078] The prediction unit 159 uses a predictor to estimate an etching profile from the conversion result of the variable conditions by the compressor and the fixed conditions. Specifically, the prediction unit 159 inputs the conversion result input from the second compression unit 157 and the fixed conditions input from the processing condition determination unit 151 to the predictor, and outputs the etching profile output by the predictor to the evaluation unit 161.

[0079] The evaluation unit 161 evaluates the etching profile input from the prediction unit 159 and outputs the evaluation result to the processing condition determination unit 151. Specifically, the evaluation unit 161 acquires pre-processing film thickness characteristics of a substrate W to be processed by the substrate processing apparatus 300. The evaluation unit 161 calculates film thickness characteristics predicted after the etching process from the etching profile input from the prediction unit 159 and the pre-processing film thickness characteristics of the substrate W, and compares the film thickness characteristics with the target film thickness characteristics. If the comparison result satisfies the evaluation criterion, the evaluation unit 161 outputs the processing conditions determined by the processing condition determination unit 151 to the processing condition transmission unit 163. For example, the evaluation unit 161 calculates deviation characteristics and determines whether the deviation characteristics satisfy the evaluation criterion. The deviation characteristics are the difference between the film thickness characteristics of the substrate W after the etching process and the target film thickness characteristics. The evaluation criterion can be set arbitrarily. For example, the evaluation criterion may be that the maximum difference in the deviation characteristics is equal to or less than a threshold, or that the average difference is equal to or less than a threshold.

[0080] The processing condition transmission unit 163 transmits the processing conditions determined by the processing condition determination unit 151 to the substrate processing apparatus 300. The substrate processing apparatus 300 processes the substrate W in accordance with the processing conditions.

[0081] If the evaluation result does not satisfy the evaluation criteria, the evaluation unit 161 outputs the evaluation result to the processing condition determination unit 151. The evaluation result includes a film thickness characteristic predicted after the etching process or a difference between the film thickness characteristic predicted after the etching process and the target film thickness characteristic.

[0082] In response to the evaluation result input from the evaluation unit 161, the processing condition determination unit 151 determines new processing conditions to be estimated by the prediction unit 159. The processing condition determination unit 151 selects one from a plurality of variable conditions prepared in advance using an experimental design method, a pairwise method, or Bayesian estimation, and determines the processing conditions including the selected variable condition and fixed conditions as new processing conditions to be estimated by the prediction unit 159.

[0083] The processing condition determination unit 151 may search for processing conditions using Bayesian estimation. When multiple evaluation results are output by the evaluation unit 161, there will be multiple pairs of processing conditions and evaluation results. Based on the tendency of the etching profile in each of the multiple pairs, the processing condition determination unit 151 searches for processing conditions that will result in a uniform film thickness or processing conditions that will minimize the difference between the film thickness characteristics predicted after the etching process and the target film thickness characteristics.

[0084] Specifically, the processing condition determination unit 151 searches for processing conditions to minimize an objective function. The objective function is a function indicating the uniformity of the film thickness or a function indicating the agreement between the film thickness characteristics of the film and the target film thickness characteristics. For example, the objective function is a function indicating, using parameters, the difference between the film thickness characteristics predicted after the etching process and the target film thickness characteristics. Here, the parameters are the conversion results of the corresponding variable conditions converted by the second compression unit 157. The corresponding variable conditions are the variable conditions before the conversion results used by the predictor to estimate the etching profile were converted. The processing condition determination unit 151 selects, from among the multiple variable conditions, the variable conditions corresponding to the conversion results, which are the parameters determined by the search, and determines new processing conditions including the selected variable conditions and fixed conditions.

[0085] Instead of selecting one of the prepared variation conditions, the processing condition determination unit 151 may select a variation condition obtained by decoding the parameters obtained by the search using a decoder. The decoder is part of the autoencoder configured by the compressor generation unit 253 of the learning device 200.

[0086] 9 is a flowchart showing an example of the flow of autoencoder generation processing. The autoencoder generation processing is performed by CPU 201 included in learning device 200 as CPU 201 executes an autoencoder generation program stored in RAM 202. The autoencoder generation program is part of the learning program.

[0087] 9, CPU 201 included in learning device 200 generates a variation condition, and the process proceeds to step S02. A plurality of operation patterns indicating changes in the position of nozzle 311 are determined in advance, and one of the operation patterns is selected. Then, a function corresponding to the selected operation pattern is selected, and the parameters are varied in order to generate a variation condition.

[0088] In step S02, the CPU 201 trains the autoencoder and proceeds to step S03. The variation conditions generated in step S01 are set to the input and output, and the parameters of the neural network are adjusted. The variation conditions are sampled at a predetermined sampling interval. In this embodiment, the processing time is 20 seconds, and the sampling interval is 0.01 seconds. Therefore, the variation conditions are data in which 2000 pieces of position data indicating positions are arranged in time series.

[0089] In step S03, it is determined whether the adjustment is complete. Variable conditions used to evaluate the compressor are prepared in advance, and the performance of the autoencoder is evaluated under the variable conditions for evaluation. The adjustment is determined to be complete if the evaluation result satisfies the predetermined evaluation criteria. If the evaluation result does not satisfy the evaluation criteria (NO in step S03), the process returns to step S01, but if the evaluation result satisfies the evaluation criteria (YES in step S03), the process proceeds to step S04.

[0090] When the process returns to step S01, a new motion pattern is generated in step S01. In the loop of steps S01 to S03, CPU 201 performs machine learning on the autoencoder using a plurality of variable conditions. As a result, the parameters of the autoencoder are adjusted to appropriate values.

[0091] In step S04, the autoencoder is stored in the storage device 104, and the process proceeds to step S05. The autoencoder is a neural network in which parameters adjusted by machine learning are set.

[0092] In step S05, the compressor is transmitted, and the process ends. The CPU 201 controls the input / output I / F 107 to transmit the part from the input layer to the hidden layer of the autoencoder to the information processing device 100 as a compressor.

[0093] 10 is a flowchart showing an example of the flow of the predictor generation process. The predictor generation process is performed by CPU 201 included in learning device 200 as CPU 201 executes a predictor generation program stored in RAM 202. The predictor generation program is part of the learning program.

[0094] 10, the CPU 201 included in the learning device 200 acquires experimental data. The CPU 201 controls the input / output I / F 107 to acquire the experimental data from the substrate processing apparatus 300 (step S11). The experimental data may be acquired by reading experimental data recorded on a recording medium such as a CD-ROM 209 using the storage device 104. The acquired experimental data includes processing conditions and film thickness characteristics of a coating formed on the substrate W before and after processing. The film thickness characteristics are represented by the film thickness of the coating formed on the substrate W at each of multiple different positions in the radial direction of the substrate W.

[0095] In the next step S12, experimental data to be processed is selected, and the process proceeds to step S13. In step S13, the variable conditions included in the experimental data are compressed, and the process proceeds to step S14. The variable conditions are input to a compressor generated by learning device 200, and the output of the compressor is obtained as the conversion result.

[0096] In step S14, the conversion result, the fixed conditions included in the experimental data, and the etching profile are set as training data. The etching profile is the difference between the film thickness characteristics of the coating before processing included in the experimental data and the film thickness characteristics of the coating after processing included in the experimental data. The training data includes input data and correct data. The conversion result calculated by the compressor in step S13 and the fixed conditions included in the experimental data are set as input data. The etching profile is set as correct data.

[0097] In the next step S15, the CPU 201 performs machine learning on the predictor, and proceeds to step S16. Input data is input to the predictor, which is a neural network, and parameters are determined so that the output of the predictor is equal to the correct data. This adjusts the parameters of the predictor. The predictor is a neural network with parameters determined by machine learning using training data.

[0098] In step S16, it is determined whether the adjustment is complete. Learning data used to evaluate the predictor is prepared in advance, and the performance of the predictor is evaluated using the learning data for evaluation. The adjustment is determined to be complete when the evaluation result satisfies a predetermined evaluation criterion. If the evaluation result does not satisfy the evaluation criterion (NO in step S16), the process returns to step S12, but if the evaluation result satisfies the evaluation criterion (YES in step S16), the process proceeds to step S17.

[0099] When the process returns to step S12, in step S12, experimental data that has not been selected as a processing target is selected from the experimental data acquired in step S11. In the loop of steps S12 to S16, CPU 201 performs machine learning on the predictor using a plurality of learning data. As a result, the parameters of the predictor, which is a neural network, are adjusted to appropriate values. In step S18, the predictor is transmitted, and the process ends. CPU 201 controls input / output I / F 107 to transmit the predictor to information processing device 100.

[0100] 11 is a flowchart showing an example of the flow of processing condition determination processing. The processing condition determination processing is performed by CPU 101 provided in information processing device 100 as CPU 101 executes a processing condition determination program stored in RAM 102.

[0101] 11, CPU 101 included in information processing device 100 selects one of a plurality of variable conditions prepared in advance (step S21), and proceeds to step S22. The plurality of variable conditions are generated by learning device 200 to generate a compressor. One of the plurality of variable conditions prepared in advance is selected using experimental design, pairwise estimation, Bayesian estimation, or the like.

[0102] In step S22, the variable conditions are compressed, and the process proceeds to step S23. The variable conditions are compressed using a compressor generated by the learning device 200. Here, the variable conditions are input to the compressor, and the output is obtained as the conversion result.

[0103] In step S23, a predictor is used to estimate an etching profile from the conversion result and the fixed conditions, and the process proceeds to step S24. The conversion result and the fixed conditions are input to the predictor, and the etching profile output by the predictor is obtained. In step S24, the film thickness characteristic after processing is compared with the target film thickness characteristic. The film thickness characteristic after processing the substrate W is calculated from the film thickness characteristic before processing of the substrate W to be processed by the substrate processing apparatus 300 and the etching profile estimated in step S23. Then, the film thickness characteristic after processing is compared with the target film thickness characteristic. Here, the difference between the film thickness characteristic after processing the substrate W and the target film thickness characteristic is calculated.

[0104] In step S25, it is determined whether the comparison result satisfies the evaluation criterion. If the comparison result satisfies the evaluation criterion (YES in step S25), the process proceeds to step S26; otherwise, the process returns to step S21. For example, if the maximum value of the differences is equal to or less than a threshold, it is determined that the evaluation criterion is met. Also, if the average of the differences is equal to or less than a threshold, it is determined that the evaluation criterion is met.

[0105] In step S26, processing conditions including the variable condition selected immediately before in step S21 are set as candidates for processing conditions for driving the substrate processing apparatus 300, and the process proceeds to step S27. In step S27, it is determined whether an instruction to end the search has been accepted. If an instruction to end the search has been accepted by the user operating the information processing apparatus 100, the process proceeds to step S28; otherwise, the process returns to step S21. Note that instead of an instruction to end the search being accepted by the user, it may be determined whether a predetermined number of processing conditions have been set as candidates.

[0106] In step S28, one of the one or more candidate processing conditions is determined, and processing proceeds to step S29. The user operating the information processing apparatus 100 may select one of the one or more candidate processing conditions. This broadens the user's range of choices. Alternatively, a variable condition with the simplest nozzle operation may be automatically selected from among the variable conditions included in the plurality of processing conditions. The variable condition with the simplest nozzle operation may be, for example, a variable condition with the fewest number of speed change points. This makes it possible to present a plurality of variable conditions for processing results for complex nozzle operation for processing the substrate W. Selecting a variable condition with easy nozzle control from among the plurality of variable conditions makes it easier to control the substrate processing apparatus 300.

[0107] In step S29, the processing conditions including the variable conditions determined in step S28 are transmitted to the substrate processing apparatus 300, and the processing ends. The CPU 101 controls the input / output I / F 107 to transmit the processing conditions to the substrate processing apparatus 300. When the substrate processing apparatus 300 receives the processing conditions from the information processing apparatus 100, the substrate processing apparatus 300 processes the substrate W in accordance with the processing conditions.

[0108] 4. Specific Examples In this embodiment, the variable conditions are time-series data sampled at a sampling interval of 0.01 seconds over a nozzle operation processing time of 20 seconds. The variable conditions are made up of 2000 values. This makes it possible to express complex nozzle operations using the variable conditions. In particular, the variable conditions can accurately express nozzle operations with a relatively large number of speed change points at which the nozzle movement speed is changed.

[0109] The compressor generation unit 253 in this embodiment trains an autoencoder to convert variable conditions into a two-dimensional data set. Therefore, the compressor generated by the compressor generation unit 253 converts variable conditions consisting of 2000 values ​​into a two-dimensional data set. The inventors have found through experiments that even when variable conditions consisting of 2000 values ​​that indicate complex nozzle operation are converted into a two-dimensional data set, a desired result can be obtained as the etching profile predicted by the predictor.

[0110] This reduces the amount of data input to the predictor, simplifying the predictor configuration and making it easier to train the neural network. Furthermore, it also makes it possible to adjust the neural network parameters to appropriate values, improving the accuracy of the predictor.

[0111] Furthermore, since variable conditions with 2000 dimensions are converted into a two-dimensional data set, there may be multiple variable conditions among the multiple variable conditions that result in the same conversion. In this case, the etching profile predicted by the predictor from each of the multiple variable conditions that result in the same conversion will be the same. In this embodiment, when the processing condition determination unit 151 searches for processing conditions, processing conditions corresponding to different etching profiles are searched for, and therefore processing conditions corresponding to multiple different etching profiles are selected. Therefore, the processing condition determination unit 151 can efficiently search for processing conditions that predict a target etching profile from among the multiple processing conditions.

[0112] Although an example in which the sampling interval is 0.01 seconds has been described, the sampling interval is not limited to this. It may be a longer or shorter sampling interval. For example, the sampling interval may be 0.1 seconds or 0.005 seconds. In short, regardless of the number of dimensions of the variable conditions, it is sufficient that the number of dimensions of the conversion result obtained by converting the variable conditions by the compressor is two.

[0113] 5. Other Embodiments (1) In the above-described embodiment, the learning device 200 generates a predictor based on learning data. The learning device 200 may perform additional learning of the predictor. After the predictor is generated, the learning device 200 acquires the film thickness characteristics and processing conditions of the coating before and after processing of the substrate W processed by the substrate processing apparatus 300. The learning device 200 then generates learning data from the film thickness characteristics and processing conditions of the coating before and after processing, and performs machine learning of the predictor, thereby additionally learning the predictor. The additional learning does not change the configuration of the neural network that constitutes the predictor, but adjusts the parameters.

[0114] The accuracy of the predictor can be improved by machine learning the predictor using information obtained as a result of the substrate W actually being processed by the substrate processing apparatus 300. In addition, the number of pieces of learning data used to generate the predictor can be minimized.

[0115] 12 is a flowchart showing an example of the flow of the additional learning process. The additional learning process is performed by CPU 201 included in learning device 200 as CPU 201 executes an additional learning program stored in RAM 202. The additional learning program is part of the learning program.

[0116] 12, the CPU 201 included in the learning device 200 acquires production data (step S31) and proceeds to step S32. The production data includes processing conditions when the substrate processing apparatus 300 processes the substrate W after the predictor is generated, and film thickness characteristics of the coating before and after the processing. The CPU 201 controls the input / output I / F 107 to acquire the production data from the substrate processing apparatus 300. The production data may be acquired by reading experimental data recorded on a recording medium such as a CD-ROM 209 using the storage device 104.

[0117] In step S32, the variable conditions are compressed, and the process proceeds to step S33. The variable conditions are input to a compressor generated by the learning device 200, and the output of the compressor is obtained as the conversion result. In step S33, the conversion result, the fixed conditions included in the processing conditions of the production data, and the etching profile are set as the learning data. The etching profile is the difference between the film thickness characteristics of the coating before processing included in the production data and the film thickness characteristics of the coating after processing included in the production data. The conversion result by the compressor and the fixed conditions included in the processing conditions are set as the input data. The etching profile is set as the correct data.

[0118] In the next step S34, CPU 201 performs additional learning on the predictor and proceeds to step S35. Input data is input to the predictor, which is a neural network, and parameters are determined so that the output of the predictor is equal to the correct data. This further adjusts the parameters of the predictor.

[0119] In step S35, it is determined whether the adjustment is complete. The performance of the predictor is evaluated using the learning data for evaluation. The adjustment is determined to be complete if the evaluation result satisfies a predetermined evaluation criterion for additional learning. The evaluation criterion for additional learning is a higher standard than the evaluation criterion used when the predictor was generated. If the evaluation result does not satisfy the evaluation criterion for additional learning (NO in step S35), the process returns to step S31, but if the evaluation result satisfies the evaluation criterion for additional learning (YES in step S35), the process ends.

[0120] (2) The learning device 200 may generate a distillation model by machine learning a new learning model using distillation data including processing conditions determined by the information processing device 100 and an etching profile estimated by a predictor from the processing conditions. This makes it easier to prepare data for training a new learning model.

[0121] (3) Although the learning device 200 generates a compressor by machine learning an autoencoder, the present invention is not limited to this. The compressor may reduce the dimension of the temporal change in the relative position between the nozzle 311 and the substrate W using other methods.

[0122] (4) In this embodiment, the learning data used to generate a predictor includes the fixed conditions and the conversion results of the variable conditions converted from the input data. The present invention is not limited to this. The input data may include only the conversion results of the variable conditions, and may not include the fixed conditions.

[0123] (5) In the present embodiment, the relative position between the nozzle 311 and the substrate W is shown as an example of a variable condition, but the present invention is not limited to this. When at least one of the temperature of the etching liquid, the concentration of the etching liquid, the flow rate of the etching liquid, and the rotation speed of the substrate W varies over time, it may be set as a variable condition. Furthermore, the variable condition is not limited to one type, and may include a combination of multiple types.

[0124] (6) Although the information processing device 100 and the learning device 200 have been described as being separate from the substrate processing device 300, the present invention is not limited to this. The information processing device 100 may be incorporated into the substrate processing device 300. Furthermore, the information processing device 100 and the learning device 200 may be incorporated into the substrate processing device 300. Furthermore, although the information processing device 100 and the learning device 200 have been described as being separate devices, they may also be configured as an integrated device.

[0125] 6. Effects of the embodiment The learning device 200 in this embodiment acquires a processing amount indicating the difference in film thickness between before and after processing of a film formed on a substrate W after the substrate processing apparatus 300 is operated under processing conditions including variable conditions, and generates a predictor, which is a learning model that estimates an etching profile, by machine learning a neural network using training data that includes, as input data, the conversion results obtained by converting the variable conditions using a compressor and the etching profile corresponding to the processing conditions as ground truth data. Since the training data includes, as input data, the conversion results obtained by converting the variable conditions so as to reduce the number of dimensions of the variable conditions that vary over time, the number of dimensions of the training data can be reduced. This makes it possible to generate a learning device suitable for machine learning conditions that change over time for processing a substrate W.

[0126] Furthermore, the learning device 200 generates a plurality of variable conditions within a range in which the substrate processing apparatus 300 can be operated, and generates a compressor by having an autoencoder learn the generated plurality of variable conditions by machine learning. Therefore, since the compressor is generated by having the autoencoder learn the variable conditions, it becomes easy to generate the compressor.

[0127] Furthermore, the processing conditions include variable conditions and fixed conditions that do not change over time. This allows for processing with different fixed conditions, eliminating the need to generate multiple learning models with different fixed conditions.

[0128] Furthermore, after generating the predictor, the learning device 200 acquires a processing amount indicating the difference in film thickness before and after processing of a film formed on a substrate W processed by the substrate processing apparatus 300 according to the processing conditions, and trains the learning model using additional learning data including the conversion result obtained by converting the tentative variable conditions by the compressor and the acquired processing amount. Therefore, the learning model is additionally trained, and the performance of the learning model can be improved.

[0129] Furthermore, the learning device 200 provides the learning model with the conversion result obtained by converting the tentative variable conditions by the compressor, and if the processing amount estimated by the learning model satisfies the tolerance condition, generates a new learning model using distillation data including the conversion result and the processing amount estimated by the learning model. This makes it easy to prepare data for training a new learning model.

[0130] The substrate processing apparatus 300 also includes a nozzle 311 that supplies a processing liquid to the substrate W, and a nozzle movement mechanism 301 that changes the relative position between the nozzle 311 and the substrate W, and the variable condition is the relative position between the nozzle 311 and the substrate W that is changed by the nozzle movement mechanism 301. A learning model is generated that estimates the processing amount of a coating that is processed by supplying the processing liquid to the substrate W from the nozzle 311 while changing the relative position between the nozzle 311 and the substrate W. This makes it possible to generate a learning model that estimates the processing amount in an etching process.

[0131] Furthermore, the information processing device 100 provides the learning model generated by the learning device 200 with the conversion results of the provisional variable conditions generated by the compressor. If the etching profile estimated by the learning model satisfies the tolerance conditions, the information processing device 100 determines the processing conditions, including the provisional variable conditions, as the processing conditions for driving the substrate processing device 300. Since the etching profile is estimated from the processing conditions, there is no need to determine the effect of complex nozzle operation on the etching process results through experiments or the like. Furthermore, since multiple provisional variable conditions are determined for a processing volume that satisfies the tolerance conditions, multiple variable conditions corresponding to multiple etching profiles that satisfy the tolerance conditions can be determined. Therefore, multiple variable conditions can be presented for the processing results of a complex substrate processing process. Selecting processing conditions that are easy to control for nozzle operation from among the multiple variable conditions facilitates control of the substrate processing device 300.

[0132] Furthermore, the information processing apparatus 100 determines multiple processing conditions for a processing volume that satisfies the tolerance conditions, and thus can determine multiple processing conditions corresponding to multiple etching profiles that satisfy the tolerance conditions. The fixed conditions include the temperature of the etching liquid. Therefore, multiple etching liquid temperatures can be presented for the processing results of a complex process for processing a substrate. Furthermore, from among the multiple etching liquid temperatures, an etching liquid temperature that is easily applicable to the etching process can be selected. Furthermore, because the temperature of the etching liquid that is easily applicable can be selected, temperature control of the etching liquid used in the etching process becomes easier.

[0133] 7. Correspondence between each element of the claims and each part of the embodiment The substrate W is an example of a substrate, the etching liquid is an example of a processing liquid, the substrate processing apparatus 300 is an example of a substrate processing apparatus, the experimental data acquisition unit 261 is an example of an experimental data acquisition unit, the first compression unit 263 is an example of a first compressor, the predictor is an example of a learning model, and the predictor generation unit 265 is an example of a model generation unit. Also, the information processing apparatus 100 is an example of an information processing apparatus, the variable condition generation unit 251 is an example of a variable condition generation unit, the compressor generation unit 253 is an example of a compressor generation unit, the second compression unit 157 is an example of a second compressor, the nozzle 311 is an example of a nozzle that supplies processing liquid to the substrate, the nozzle movement mechanism 301 is an example of a movement unit, and the prediction unit 159, the evaluation unit 161, and the processing condition determination unit 151 are examples of a processing condition determination unit. [Explanation of symbols]

[0134] 1...substrate processing system, 100...information processing device, 200...learning device, 300...substrate processing device, 301...nozzle moving mechanism, 10...control device, 303...nozzle motor, 305...nozzle arm, 311...nozzle, 101, 201...CPU, 102, 202...RAM, 103, 203...ROM, 104, 204...storage device, 105, 205...operation unit, 106, 206...display device, 107, 207...input / output I / F, 108, 20 8...bus, 109,209...CD-ROM, 151...processing condition determination unit, 153...compressor receiving unit, 155...predictor receiving unit, 157...second compression unit, 159...prediction unit, 161...evaluation unit, 163...processing condition transmitting unit, 251...variable condition generating unit, 253...compressor generating unit, 255...compressor transmitting unit, 261...experimental data acquisition unit, 263...first compression unit, 265...predictor generating unit, 267...predictor transmitting unit, W...substrate, WU...substrate processing unit.

Claims

1. an experimental data acquisition unit that acquires a processing amount indicating a difference in film thickness between before and after processing of a coating formed on a substrate after the substrate processing apparatus that processes the substrate by supplying a processing liquid to the substrate on which a coating is formed is operated under processing conditions including variable conditions that vary over time; a compressor that converts the variable conditions to reduce the number of dimensions; a model generation unit that performs machine learning on learning data including the conversion result of the variable conditions converted by the compressor and the processing amount corresponding to the processing conditions to generate a learning model that estimates the processing amount indicating the difference in film thickness before and after processing of a coating formed on a substrate before processing by the substrate processing apparatus.

2. a variable condition generating unit that generates a plurality of variable conditions; The learning device according to claim 1 , further comprising: a compressor generation unit that generates the compressor by having an autoencoder learn the plurality of variation conditions generated by the variation condition generation unit.

3. The processing conditions further include fixed conditions that do not vary over time; The learning device according to claim 1 , wherein the learning data includes the fixed condition.

4. a performance acquisition unit that provides the learning model with a conversion result obtained by converting the tentative variable conditions by the compressor, and operates the substrate processing apparatus under processing conditions including the tentative variable conditions when the processing amount estimated by the learning model satisfies an allowable condition, thereby acquiring a processing amount indicating a difference in film thickness between before and after processing of a film formed on a substrate processed by the substrate processing apparatus; The learning device according to any one of claims 1 to 3, further comprising an additional learning unit that trains the learning model using additional learning data including the conversion result of the virtual variable conditions converted by the compressor and the processing amount acquired by the performance acquisition unit.

5. A learning device as described in any one of claims 1 to 4, further comprising a distillation unit that provides the learning model with the conversion result obtained by converting the tentative variable conditions by the compressor, and generates a new learning model using distillation data including the conversion result and the processing amount estimated by the learning model if the processing amount estimated by the learning model satisfies the tolerance condition.

6. The substrate processing apparatus includes: a nozzle for supplying the processing liquid to the substrate; a moving unit that changes the relative position between the nozzle and the substrate, 6. The learning device according to claim 1, wherein the variable condition is a relative position between the nozzle and the substrate that is changed by the moving unit.

7. An information processing device for managing a substrate processing device, the substrate processing apparatus processes the substrate by supplying a processing liquid to the substrate on which the coating is formed under processing conditions including variable conditions that vary over time; a second compressor that converts the variable conditions to reduce the number of dimensions; a processing condition determination unit that determines processing conditions for driving the substrate processing apparatus using a learning model that estimates a processing amount indicating a difference in film thickness between before and after processing of a film formed on a substrate before processing by the substrate processing apparatus, the learning model is an inference model that is machine-learned on learning data that includes a conversion result obtained by converting, by a first compressor that is the same as the second compressor, variable conditions included in the processing conditions under which the substrate was processed by the substrate processing apparatus, and a processing amount that indicates a difference in film thickness between before and after processing of a film formed on the substrate processed by the substrate processing apparatus, The processing condition determination unit is an information processing device that provides the conversion results obtained by converting the provisional variable conditions by the second compressor to the learning model, and determines the processing conditions including the provisional variable conditions as the processing conditions for driving the substrate processing device if the processing amount estimated by the learning model satisfies the tolerance conditions.

8. A substrate processing apparatus comprising the information processing apparatus according to claim 7.

9. A substrate processing system that manages a substrate processing apparatus that processes a substrate, A learning device and an information processing device are provided, the substrate processing apparatus processes the substrate by supplying a processing liquid to the substrate on which the coating is formed under processing conditions including variable conditions that vary over time; the learning device includes an experimental data acquisition unit that acquires a processing amount indicating a difference between a film thickness before and after processing of a film formed on a substrate after the substrate processing apparatus is operated under the processing conditions; a first compressor that converts the variable conditions to reduce the number of dimensions; a model generation unit that performs machine learning on learning data including a conversion result of the variable condition converted by the first compressor and the processing amount corresponding to the processing condition to generate a learning model that estimates a processing amount indicating a difference in film thickness between before and after processing of a coating formed on a substrate before processing by the substrate processing apparatus, The information processing device includes a second compressor identical to the first compressor; a processing condition determination unit that determines processing conditions for driving the substrate processing apparatus using the learning model generated by the learning device, The processing condition determination unit provides the learning model with the conversion results obtained by converting the provisional variable conditions by the second compressor, and determines the processing conditions including the provisional variable conditions as the processing conditions for driving the substrate processing apparatus if the processing volume estimated by the learning model satisfies the tolerance conditions.

10. a process of acquiring a processing amount indicating a difference in film thickness between before and after processing of a coating formed on a substrate, after the substrate processing apparatus, which processes the substrate by supplying a processing liquid to the substrate on which a coating has been formed, is operated under processing conditions including variable conditions that vary over time; a compression process for converting the variable conditions so as to reduce the number of dimensions; a process of machine learning learning data including the conversion result of the variable conditions converted in the compression process and the processing amount corresponding to the processing conditions to generate a learning model that estimates the processing amount indicating the difference in film thickness before and after processing of a coating formed on a substrate before processing by the substrate processing apparatus.

11. A processing condition determination method executed by a computer that manages a substrate processing apparatus, comprising: the substrate processing apparatus processes the substrate by supplying a processing liquid to the substrate on which the coating is formed under processing conditions including variable conditions that vary over time; A process of converting the variable conditions so as to reduce the number of dimensions; determining processing conditions for operating the substrate processing apparatus using a learning model that estimates a processing amount indicating a difference in film thickness between before and after processing of a film formed on a substrate before processing by the substrate processing apparatus; the learning model is an inference model obtained by machine learning learning data including a conversion result obtained by converting a variable condition included in a processing condition under which the substrate was processed by the substrate processing apparatus in the same processing as the processing to be converted, and a processing amount indicating a difference in film thickness between before and after the processing of a film formed on the substrate processed by the substrate processing apparatus, The processing condition determination method includes a process of providing the conversion results obtained by converting the provisional variable conditions by the conversion process to the learning model, and determining the processing conditions including the provisional variable conditions as the processing conditions for driving the substrate processing apparatus if the processing amount estimated by the learning model satisfies the tolerance condition.

Images