Parameter optimization apparatus, parameter optimization method, and computer program product

Abstract

This invention provides a technique for efficiently optimizing control parameters even when there is an error between the learned model and the device being optimized. A sliding parameter generation unit (912) determines the sliding amount that changes the control parameters and generates sliding control parameters after changing the sliding amount of the control parameters used for testing. An optimal sliding amount determination unit (916) determines the optimal sliding amount for the coating apparatus (1) based on first film thickness information of the coating film when the control parameters for testing are applied, and second film thickness information of the coating film when the sliding control parameters are applied. An optimization unit (917) optimizes the control parameters applicable to the coating apparatus (1) using a basic estimation model (Y1) and the optimal sliding amount.

Description

Technical Field

[0001] The subject matter disclosed in this specification relates to parameter optimization devices, parameter optimization methods, and computer program products. Background Technology

[0002] In the manufacturing process of flat panel displays (FPDs), a device called a coating machine is used. A coating machine is a substrate processing device that uses a pump to spray a processing liquid from a slit nozzle, coating the entire substrate with the processing liquid. In recent years, with the increasing demand for higher product quality, it has become important for the processing liquid to form a uniform film thickness across the entire surface of the substrate. Previously, to achieve uniform film thickness, control parameters such as the spraying of the processing liquid were repeatedly measured and adjusted to control the coating liquid application onto the substrate. In general, industrial machinery also contains various control parameters in its control devices, which are often adjusted before shipment. This adjustment is typically performed by technicians. However, to reduce costs and minimize deviations by saving labor, there is a strong demand for automation of the adjustment process. Therefore, many efficient methods for optimizing control parameters have been proposed to date. Prior art related to this invention includes, for example, the techniques described in Patent Documents 1 and 2.

[0003] Existing technical documents

[0004] Patent documents

[0005] Patent Document 1: Japanese Patent Application Publication No. 2020-027370

[0006] Patent Document 2: Japanese Patent Application Publication No. 2022-074880

[0007] Patent Document 1 repeatedly performs simulations and experiments to optimize control parameters, but when the error between the learned model and the device being optimized is large, the optimization efficiency decreases, sometimes requiring a significant amount of time. The main causes of the error between the learned model and the device include individual differences in the device and differences caused by the type of processing fluid. While Patent Document 2 discloses an efficient method for exploring optimization, in this case, the optimization efficiency of control parameters may also decrease significantly when the error between the learned model and the device being optimized is large. Summary of the Invention

[0008] The purpose of this invention is to provide a technique that enables efficient optimization of control parameters even when there is an error between the learned model and the device being optimized.

[0009] To address the aforementioned issues, a first approach is a parameter optimization apparatus comprising: a test parameter determination unit that determines control parameters for testing; a sliding parameter generation unit that, based on the output of a basic estimation model that takes the control parameters as input and film thickness information as output, determines a sliding amount that changes at least one parameter element of the control parameters for testing, and generates sliding control parameters that change the control parameters for testing by the sliding amount; a film thickness information acquisition unit that, for a coating apparatus being optimized, acquires first film thickness information of the coating film when the control parameters for testing are applied, and second film thickness information of the coating film when the sliding control parameters are applied; an optimal sliding amount determination unit that, based on the first film thickness information and the second film thickness information, determines an optimal sliding amount that changes the at least one parameter element; and an optimization unit that optimizes the control parameters applicable to the coating apparatus being optimized using the basic estimation model and the optimal sliding amount.

[0010] The second approach is a parameter optimization apparatus, comprising: a sliding parameter generation unit that determines a sliding amount that changes at least one parameter element of a control parameter, and generates a sliding dataset including sliding control parameters after changing the control parameter by the sliding amount, based on a basic dataset consisting of a set of pairs of the control parameter and film thickness information associated with the control parameter; a learning unit that uses the sliding dataset to learn a sliding estimation model that takes the sliding control parameter as input and the film thickness information as output; and a verification parameter determination unit that uses the basic estimation model that takes the control parameter as input and the film thickness information as output to determine control parameters for verification, and makes... The sliding estimation model is used to determine the sliding control parameters for inspection; the film thickness information acquisition unit acquires, for the coating apparatus being optimized, first film thickness information of the coating film when the control parameters for inspection are applied, and second film thickness information of the coating film when the sliding control parameters for inspection are applied; the optimal estimation model determination unit determines the optimal estimation model from the basic estimation model and the sliding estimation model based on the first film thickness information and the second film thickness information; and the optimization unit optimizes the control parameters applicable to the coating apparatus being optimized using the optimal estimation model determined by the optimal estimation model determination unit.

[0011] The third approach is a parameter optimization device based on the first or second approach, wherein the sliding parameter generation unit determines the sliding amount based on the distribution of evaluation values ​​shown by a plurality of film thickness information output from the basic estimation model.

[0012] The fourth approach is a third-party parameter optimization device, wherein the sliding parameter generation unit determines the sliding amount based on the characteristics of the evaluation value distribution on a straight line passing through the minimum value of the evaluation value distribution.

[0013] The fifth method is a parameter optimization device of the fourth method, wherein the straight line is parallel to the axis of a parameter element.

[0014] The sixth method is a parameter optimization device of the fourth method, wherein the straight line intersects the regression line of the evaluation value distribution.

[0015] The seventh method is a parameter optimization device for any one of the first to sixth methods, wherein the control parameters used for verification are control parameters optimized using the basic estimation model.

[0016] The eighth method is a parameter optimization method, comprising: a test parameter determination step, determining control parameters for testing; a sliding parameter generation step, determining a sliding amount that changes at least one parameter element of the control parameters for testing based on the output of a basic estimation model that takes the control parameters as input and film thickness information as output, and generating sliding control parameters that change the control parameters for testing by the sliding amount; a film thickness information acquisition step, acquiring, for the coating apparatus being optimized, first film thickness information of the coating film when the control parameters for testing are applied, and second film thickness information of the coating film when the sliding control parameters are applied; an optimal sliding amount determination step, determining the optimal sliding amount that changes the at least one parameter element based on the first film thickness information and the second film thickness information; and an optimization step, using the basic estimation model and the optimal sliding amount to optimize the control parameters applicable to the coating apparatus being optimized.

[0017] The ninth method is a parameter optimization method, comprising: a sliding parameter generation step, determining a sliding amount that changes at least one parameter element of the control parameter, and generating a sliding dataset including sliding control parameters after changing the control parameter by the sliding amount, based on a set of pairs of the control parameter and film thickness information associated with the control parameter, i.e., a basic dataset; a learning step, using the sliding dataset, learning a sliding estimation model that takes the sliding control parameter as input and the film thickness information as output; and a verification parameter determination step, using the basic estimation model that takes the control parameter as input and the film thickness information as output to determine the control parameters for verification, and making... The sliding estimation model is used to determine the sliding control parameters for inspection; in the film thickness information acquisition step, for the coating apparatus as the optimization target, first film thickness information of the coating film when the control parameters for inspection are applied, and second film thickness information of the coating film when the sliding control parameters for inspection are applied; in the optimal estimation model determination step, based on the first film thickness information and the second film thickness information, the optimal estimation model is determined from the basic estimation model and the sliding estimation model; and in the optimization step, the optimal estimation model determined by the optimal estimation model determination step is used to optimize the control parameters applicable to the coating apparatus as the optimization target.

[0018] The tenth method is a computer-executable computer program product, including a computer program, wherein when the computer program is executed by a computer, it implements the steps of the parameter optimization method of the eighth or ninth method.

[0019] The effects of the invention

[0020] According to the first to tenth methods, by varying the control parameters applicable to the coating apparatus to the optimal slip amount, the film thickness information of the formed coating film can be made close to the output of the available learned basic estimation model. Therefore, even when using a learned basic estimation model that has errors with the coating apparatus, the control parameters can be optimized efficiently by applying the optimal slip amount.

[0021] According to the parameter optimization device of the fourth method, by acquiring the axial evaluation value distribution of a parameter element, the change in the evaluation value for that parameter element can be understood. Therefore, the slip amount of a parameter element that reduces the error between the basic estimation model and the coating device can be appropriately calculated.

[0022] According to the parameter optimization device of the fifth method, the direction intersecting the regression line indicates the direction of a sharp change in the evaluation value. That is, by setting a slip amount in this direction, the change in the evaluation value can be relatively increased. Therefore, by using the characteristics of the distribution of evaluation values ​​in this direction as an indicator to determine the slip amount, it is possible to explore a slip amount that effectively reduces the error between the basic estimation model and the coating device. Attached Figure Description

[0023] Figure 1 This is a schematic diagram showing the overall structure of the coating apparatus according to the first embodiment.

[0024] Figure 2 It means Figure 1 A diagram showing the structure of the treatment liquid supply mechanism in the coating apparatus.

[0025] Figure 3 It is a block diagram showing the structure of the control unit.

[0026] Figure 4 It is a diagram that shows the function blocks of the control unit and the flow of data.

[0027] Figure 5 This is a flowchart illustrating the parameter optimization method involved in the first embodiment.

[0028] Figure 6 It means Figure 5 The flowchart shown details the process for determining the optimal slip amount.

[0029] Figure 7A It is a graph representing the distribution of evaluation values ​​in the parameter space.

[0030] Figure 7B It means Figure 7A The graph shows the distribution of evaluation values ​​along the straight line shown.

[0031] Figure 7C It is a diagram showing the positions of the two control parameters generated by the sliding parameter generation unit in the parameter space.

[0032] Figure 8 yes Figure 5 The detailed flowchart of the optimization process is shown.

[0033] Figure 9 This is a block diagram illustrating the functions of the control unit involved in the second embodiment.

[0034] Figure 10 This is a flowchart illustrating the parameter optimization method of the second embodiment.

[0035] Figure 11 It means Figure 10The flowchart shown illustrates the detailed process for determining the optimal estimation model.

[0036] Figure 12 This is a graph showing the distribution of evaluation values ​​for the film thickness information output by the basic estimation model and the sliding estimation model.

[0037] Figure 13 It is shown Figure 11 The flowchart shows the detailed process of optimization.

[0038] Figure 14A It is a graph representing the distribution of evaluation values ​​in the parameter space.

[0039] Figure 14B It means Figure 14A The graph shows the distribution of evaluation values ​​along the straight line shown.

[0040] Figure 14C It is a diagram showing the positions of the two control parameters generated by the sliding parameter generation unit in the parameter space.

[0041] Explanation of reference numerals in the attached figures

[0042] 1: Coating device

[0043] 911: Inspection Parameter Determination Department

[0044] 912: Sliding Parameter Generation Unit

[0045] 913: Film Thickness Information Acquisition Department

[0046] 914: Study Department

[0047] 916: Optimal Slippage Determination Section

[0048] 917: Determination of the Optimal Estimated Model

[0049] 918: Optimization Department

[0050] 931: Computer Programs

[0051] DB1: Basic Dataset

[0052] DB2, DB3: Sliding Datasets

[0053] DB3: Sliding Dataset

[0054] Y1: Basic Presumed Model

[0055] Y2, Y3: Sliding estimation model

[0056] Y _opt : Best Estimated Model Detailed Implementation

[0057] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. Furthermore, in the drawings, for ease of understanding, the dimensions and quantities of various parts are sometimes exaggerated or simplified.

[0058] <1. First Implementation Method>

[0059] Figure 1 This is a schematic diagram showing the overall structure of the coating apparatus 1 according to the first embodiment. The coating apparatus 1 is a substrate processing apparatus that forms a coating film on a substrate S by spraying a processing liquid onto the upper surface Sf of the substrate S. As will be described later, the control unit 9 of the coating apparatus 1 functions as a parameter optimization device, which optimizes the control parameters used to control the spraying of the processing liquid.

[0060] The substrate S is, for example, a glass substrate for a liquid crystal display device. Alternatively, the substrate S can be various substrates processed for electronic devices, such as semiconductor wafers, glass substrates for photomasks, glass substrates for plasma displays, substrates (glass or ceramic substrates) for magneto-optical discs, glass substrates for organic EL devices, glass substrates or silicon substrates for solar cells, as well as flexible substrates and printed circuit boards. The coating apparatus 1 is, for example, a slot coater.

[0061] exist Figure 1 In this diagram, an XYZ coordinate system is defined to represent the configuration of the elements of the coating apparatus 1. The transport direction Dt of the substrate S is the "X direction". The direction in which the substrate S travels in the X direction (downstream of the transport direction Dt) is designated as the +X direction, and its opposite direction (upstream of the transport direction Dt) is designated as the -X direction. Furthermore, the direction orthogonal to the X direction is the Y direction, and the direction orthogonal to both the X and Y directions is designated as the Z direction. In the following description, the Z direction is designated as the vertical direction, and the X and Y directions are designated as the horizontal directions. Within the Z direction, the +Z direction is designated as the upward direction, and the -Z direction is designated as the downward direction. It should be noted that these directions are not intended to limit the configuration orientation of the coating apparatus.

[0062] The coating apparatus 1, arranged sequentially in the +X direction, includes an input conveyor 100, an input transfer unit 2, a floating stage unit 3, an output transfer unit 4, and an output conveyor 110. The input conveyor 100, input transfer unit 2, floating stage unit 3, output transfer unit 4, and output conveyor 110 form a transport path for the substrate S to pass through. Additionally, the coating apparatus 1 also includes a substrate transport unit 5, a coating mechanism 7, a processing liquid supply mechanism 8, and a control unit 9.

[0063] The substrate S is conveyed from a device located upstream of the coating apparatus 1 to the input conveyor 100. The input conveyor 100 includes a roller conveyor 101 and a rotary drive mechanism 102. The rotary drive mechanism 102 rotates each roller of the roller conveyor 101. By rotating each roller of the roller conveyor 101, the substrate S is conveyed downstream (in the +X direction) in a horizontal orientation. "Horizontal orientation" means that the main surface (the surface with the largest area) of the substrate S is parallel to the horizontal plane (XY plane).

[0064] The input transfer unit 2 includes a roller conveyor 21 and a rotation / lifting drive mechanism 22. The rotation / lifting drive mechanism 22 rotates each roller of the roller conveyor 21 and lifts the roller conveyor 21. By rotating each roller constituting the roller conveyor 21, the substrate S is conveyed downstream (in the +X direction) in a horizontal position. In addition, by lifting the roller conveyor 21, the position of the substrate S in the Z direction is changed. The substrate S is transferred from the input conveyor 100 to the floating platform unit 3 via the input transfer unit 2.

[0065] like Figure 1 As shown, the floating stage section 3 has a generally flat structure. The floating stage section 3 is divided into three regions along the X direction. The floating stage section 3 has, sequentially in the +X direction, an inlet floating stage 31, a coating stage 32, and an outlet floating stage 33. The upper surfaces of the inlet floating stage 31, the coating stage 32, and the outlet floating stage 33 are on the same plane. The floating stage section 3 also includes a lifting pin drive mechanism 34, a floating control mechanism 35, and a lifting drive mechanism 36. The lifting pin drive mechanism 34 raises and lowers a plurality of lifting pins disposed on the inlet floating stage 31. The floating control mechanism 35 supplies compressed air for levitizing the substrate S to the inlet floating stage 31, the coating stage 32, and the outlet floating stage 33, respectively. The lifting drive mechanism 36 raises and lowers the outlet floating stage 33.

[0066] On the upper surfaces of the inlet floating stage 31 and the outlet floating stage 33, a plurality of ejection holes are arranged in a matrix to eject compressed air supplied from the floating control mechanism 35. When compressed air is ejected from each ejection hole, the substrate S floats upward relative to the floating stage 3. The lower surface Sb of the substrate S then separates from the upper surface of the floating stage 3, and the substrate S is supported in a horizontal position. The distance (floatation amount) between the lower surface Sb of the substrate S and the upper surface of the floating stage 3 when the substrate S is floating is, for example, in the range of 10 μm to 500 μm.

[0067] The upper surface of the coating stage 32 is provided with an ejector hole for discharging compressed air supplied from the levitation control mechanism 35 and an suction hole for drawing in gas. The ejector hole and suction hole are alternately arranged in the X and Y directions. The levitation control mechanism 35 controls the amount of compressed air ejected from the ejector hole and the amount of air drawn from the suction hole. By controlling the amount of ejection and suction, the levitation amount of the substrate S relative to the coating stage 32 is precisely controlled so that the position of the upper surface Sf of the substrate S above the coating stage 32 in the Z direction reaches a target value. Furthermore, the levitation amount of the substrate S relative to the coating stage 32 is calculated by the control unit 9 based on the detection results of the sensor 61 or sensor 62 described later. In addition, the levitation amount of the substrate S relative to the coating stage 32 can be adjusted with high precision by airflow control.

[0068] The substrate S, which is fed into the floating platform section 3, is given a pushing force from the roller conveyor 21 in the +X direction and is conveyed onto the inlet floating platform 31. The inlet floating platform 31, the coating platform 32, and the outlet floating platform 33 support the substrate S in a floating state. The floating platform section 3 can be, for example, the structure described in Japanese Patent No. 5346643.

[0069] The substrate transport unit 5 is positioned below the floating stage unit 3. The substrate transport unit 5 includes a chuck mechanism 51 and an adsorption / travel control mechanism 52. The chuck mechanism 51 has an adsorption pad (not shown) formed on the adsorption member. The chuck mechanism 51 supports the substrate S from the lower surface side by bringing the adsorption pad into contact with the periphery of the lower surface Sb of the substrate S. The adsorption / travel control mechanism 52 adsorbs the substrate S onto the adsorption pad by applying a negative pressure less than atmospheric pressure. Furthermore, the adsorption / travel control mechanism 52 causes the substrate transport unit 5 to reciprocate in the X direction.

[0070] The chuck mechanism 51 holds the substrate S at a position where the lower surface Sb of the substrate S is higher than the upper surface of the floating stage 3. While the peripheral portion is held by the chuck mechanism 51, the substrate S maintains a horizontal posture by the buoyancy provided by the floating stage 3.

[0071] like Figure 1 As shown, the coating apparatus 1 includes a sensor 61 for measuring plate thickness. The sensor 61 is disposed adjacent to the roller conveyor 21. The sensor 61 detects the position of the upper surface Sf of the substrate S held in the chuck mechanism 51 in the Z direction. In addition, by disposing of a chuck (not shown) that does not hold the substrate S directly below the sensor 61, the sensor 61 can detect the position of the adsorption surface of the upper surface of the adsorption member in the vertical Z direction.

[0072] The chuck mechanism 51 moves in the +X direction while holding the substrate S that has been loaded into the floating stage 3. This transports the substrate S from above the inlet floating stage 31, over the coating stage 32, to above the outlet floating stage 33. Then, the substrate S is transported from the outlet floating stage 33 to the output transfer unit 4.

[0073] The output transfer unit 4 moves the substrate S from above the outlet floating platform 33 toward the output conveyor 110. The output transfer unit 4 includes a roller conveyor 41 and a rotation / lifting drive mechanism 42. The rotation / lifting drive mechanism 42 rotates the roller conveyor 41 and lifts it in the Z direction. The rotation of each roller of the roller conveyor 41 moves the substrate S in the +X direction. Furthermore, the Z-direction position of the substrate S is changed by lifting the roller conveyor 41.

[0074] The output conveyor 110 includes a roller conveyor 111 and a rotary drive mechanism 112. The output conveyor 110 transports the substrate S along the +X direction by rotating the rollers of the roller conveyor 111, and moves the substrate S out of the coating apparatus 1. In this embodiment, the input conveyor 100 and the output conveyor 110 are configured as part of the coating apparatus 1, but they may also be configured to be assembled in a device different from the coating apparatus 1.

[0075] The coating mechanism 7 applies a processing liquid to the upper surface Sf of the substrate S. The coating mechanism 7 is positioned above the transport path of the substrate S. The coating mechanism 7 has a nozzle 71. The nozzle 71 is a slit nozzle with a slit-shaped outlet formed on its lower surface. The nozzle 71 is connected to a positioning mechanism (not shown). The positioning mechanism positions the nozzle 71 at a coating position above the coating stage 32. Figure 1 The position shown by the solid line in the middle moves between the maintenance position described later. The treatment fluid supply mechanism 8 is connected to the nozzle 71. Treatment fluid is sprayed from the spray outlet disposed on the lower surface of the nozzle 71 by the supply of treatment fluid from the treatment fluid supply mechanism 8.

[0076] Figure 2 It means Figure 1 The diagram shows the structure of the processing liquid supply mechanism 8 in the coating apparatus 1. The processing liquid supply mechanism 8 includes a pump 81, piping 82, a processing liquid replenishment unit 83, piping 84, an on / off valve 85, a pressure sensor 86, and a drive unit 87. The pump 81 is the source of the processing liquid, supplying it through volume changes. For example, a bellows-type pump as described in Japanese Patent Application Publication No. 10-61558 can be used as the pump 81. Figure 2 As shown, the pump 81 has a flexible tube 811 that expands and contracts freely in the radial direction. One end of the flexible tube 811 is connected to the treatment fluid replenishment unit 83 via a piping 82. The other end of the flexible tube 811 is connected to the nozzle 71 via a piping 84.

[0077] Pump 81 has a bellows 812 capable of elastic deformation in the axial direction. The bellows 812 includes a small bellows section 813, a large bellows section 814, a pump chamber 815, and a working disc section 816. The pump chamber 815 is disposed between the flexible tube 811 and the bellows 812. An incompressible medium is sealed within the pump chamber 815. The working disc section 816 is connected to a drive unit 87.

[0078] The processing fluid replenishment unit 83 has a storage tank 831 for storing processing fluid. The storage tank 831 is connected to the pump 81 via a pipe 82. An on / off valve 833 is provided on the pipe 82. The on / off valve 833 opens and closes according to instructions from the control unit 9. When the on / off valve 833 is open, processing fluid can be replenished from the storage tank 831 to the flexible pipe 811 of the pump 81. Conversely, when the on / off valve 833 is closed, the replenishment of processing fluid from the storage tank 831 to the flexible pipe 811 of the pump 81 is cut off.

[0079] Pipe 84 is connected to the output side of pump 81. An on / off valve 85 is provided on pipe 84. The on / off valve 85 opens and closes pipe 84 according to commands from control unit 9. The opening and closing of the on / off valve 85 switches between supplying treatment fluid to nozzle 71 and stopping treatment fluid supply. A pressure sensor 86 is provided on pipe 84. The pressure sensor 86 detects the pressure (ejection pressure) applied to the treatment fluid supplied to nozzle 71 and outputs a signal indicating the detected pressure value to control unit 9.

[0080] like Figure 1 and Figure 2 As shown, a sensor 62 is disposed on the nozzle 71 from which the processing liquid is supplied from the processing liquid supply mechanism 8. The sensor 62 detects the height of the substrate S in the Z direction in a non-contact manner. The sensor 62 is electrically connected to the control unit 9. Based on the detection result of the sensor 62, the control unit 9 measures the distance (interval distance) between the floating substrate S and the upper surface of the coating stage 32. Then, based on the measured interval distance, the control unit 9 adjusts the coating position of the nozzle 71 by means of a positioning mechanism. Furthermore, as the sensor 62, for example, an optical sensor or an ultrasonic sensor can be used.

[0081] Furthermore, the substrate S removed from the output conveyor 110 is dried by a drying device, etc., to form a coating film. Then, as... Figure 1 As shown, the substrate S with the coated film is transported to the film thickness measuring device AP1 as needed to measure the film thickness. The film thickness measuring device AP1 can be, for example, a spectrophotometer with elliptic polarization or an X-ray reflectance measuring device.

[0082] The coating unit 7 includes a nozzle cleaning standby unit 72. The nozzle cleaning standby unit 72 performs prescribed maintenance on the nozzles 71, which are positioned in the maintenance location. The nozzle cleaning standby unit 72 includes a roller 721, a cleaning section 722, and a roller holder 723. The nozzle cleaning standby unit 72 adjusts the nozzle 71's outlet position to a state suitable for coating processing by cleaning the nozzles 71 and forming accumulated liquid.

[0083] Figure 3 This is a block diagram showing the structure of the control unit 9. The control unit 9 controls the operation of each component within the coating apparatus 1. A computer can be used as the control unit 9. The control unit 9 has a processor 91 and a memory 93. The processor 91 may be, for example, a CPU (Central Processing Unit). The memory 93 may be a temporary storage device such as RAM (Random Access Memory). Alternatively, the memory 93 may be a non-temporary storage device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive). The memory 93 is connected to the processor 91 via bus wiring.

[0084] The control unit 9 includes a display device 95 for displaying various information and an input device 97 for accepting user commands. The display device 95 and the input device 97 are connected to the processor 91 via bus wiring. The display device 95 is, for example, a liquid crystal display (LCD). The input device 97 includes, for example, a mouse or keyboard. Alternatively, the display device 95 can function as an input device by having a touch panel.

[0085] A computer program 931 is stored in memory 93. The computer program 931 is provided to the control unit 9 via recording medium M. That is, the computer program 931 is recorded in recording medium M in a manner that can be read by the control unit 9, which is a computer. Recording medium M is a removable medium such as a USB (Universal Serial Bus) memory, a DVD (Digital Versatile Disc), or a magnetic disk.

[0086] Figure 4 This is a diagram showing the function blocks and data flow of the control unit 9. Figure 4 The ejection control unit 910, inspection parameter determination unit 911, sliding parameter generation unit 912, film thickness information acquisition unit 913, learning unit 914, estimation unit 915, optimal sliding amount determination unit 916, and optimization unit 918 shown are functional blocks implemented by the processor 91 executing the computer program 931.

[0087] The ejection control unit 910 controls the operation (transfer operation) of the pump 81 that delivers the processing liquid to the nozzle 71 based on preset control parameters. Specifically, in order to coat the processing liquid ejected from the nozzle 71 onto the upper surface Sf of the substrate S with a uniform film thickness, the control parameters closely related to the ejection pressure waveform are optimized before the start of substrate S production (or mass production) to ensure that the ejection pressure waveform has an ideal shape. The control parameters are setpoints for pump control, such as various parameters that define the movement of the working disc 816 (e.g., acceleration time, stabilization speed, time to maintain stabilization speed, deceleration time, etc.).

[0088] The inspection parameter determination unit 911 determines the control parameters for inspection. The sliding parameter generation unit 912 determines the sliding amount ΔS that changes at least one parameter element of the control parameters for inspection. Furthermore, the sliding parameter generation unit 912 generates sliding control parameters after changing the control parameters for inspection by the aforementioned sliding amount ΔS.

[0089] The film thickness information acquisition unit 913 acquires first film thickness information when using control parameters applicable to inspection and second film thickness information when using sliding control parameters applicable to inspection for the coating apparatus 1. The optimal sliding amount determination unit 916 determines the optimal sliding amount ΔS based on the evaluation values ​​shown by the first and second film thickness information. opt .

[0090] The learning unit 914 uses the basic dataset DB1 to learn a basic estimation model Y1 that takes control parameters as input and outputs film thickness information. The estimation unit 915 uses the basic estimation model Y1 to estimate film thickness information relative to the control parameters. The optimization unit 918, based on a prescribed algorithm, uses the basic estimation model Y1 and the optimal slip ΔS. opt To optimize the control parameters applicable to coating apparatus 1.

[0091] <Parameter Optimization Methods>

[0092] Figure 5 This is a flowchart illustrating the parameter optimization method according to the first embodiment. The parameter optimization method includes an optimal slip amount determination process S1 and a subsequent optimization process S2. These steps will be described in turn below.

[0093] <Optimal Slip Determination Process S1>

[0094] Figure 6 It means Figure 5 The flowchart shown details the process of determining the optimal slip amount, S1. Figure 6As shown, during the optimization of control parameters, the control unit 9 uses a basic dataset DB1. The basic dataset DB1 is a set of pairs of control parameters and film thickness information. Film thickness information is information related to the film thickness of the coating film formed on the substrate S, such as the film thickness distribution in one direction (film thickness profile). Furthermore, film thickness information is not limited to the film thickness distribution; it can also be an evaluation value used to assess the film thickness distribution. For example, as an evaluation value representing uniformity, an index representing the magnitude of the deviation in the film thickness distribution can be used. Specifically, the average value and deviation of the error relative to a specified target film thickness can be used as the index.

[0095] The basic dataset DB1 contains data pairs obtained from coating apparatuses different from the coating apparatus 1 that is being optimized, or data obtained from the coating apparatus 1 that is being optimized, including data obtained under different process conditions (e.g., type of coating liquid, type of substrate S, etc.). The basic dataset DB1 can be stored in memory 93 or in a device different from the control unit 9, such as a server. Furthermore, the control unit 9 can access the basic dataset DB1 via a network.

[0096] In the optimal slip determination process S1, firstly, the learning unit 914 uses the basic dataset DB1 as teacher data to learn a basic estimation model Y1 that takes control parameters as input and membrane thickness information as output. Figure 6 (First learning step S11). By using, for example, an RNN (Recurrent Neural Network) as the estimation model for learning, the membrane thickness distribution, which is one-dimensional data, can be estimated as membrane thickness information. Alternatively, when the membrane thickness information is set as an evaluation value representing uniformity, a random forest regression model can be used, for example. Furthermore, the learning of the basic estimation model Y1 does not necessarily have to be performed by the control unit 9; a basic estimation model Y1 learned by other devices can also be used.

[0097] Errors sometimes arise between the output (film thickness information) of the basic estimation model Y1 learned from the basic dataset DB1 and the actual output of the coating apparatus 1. This is due to differences in individual apparatuses, processing solutions, substrate types, and other process conditions. Such errors can potentially be reduced by varying (slipping) the values ​​of several control parameters. From this perspective, in the optimal slip amount determination process S1, the optimal amount by which the parameter elements are varied (optimal slip amount) is determined.

[0098] The test parameter determination unit 911 generates N. ini N ini Random control parameters (hereinafter referred to as "random control parameters") are integers greater than or equal to 1. Figure 6(Random control parameter generation process S12). Then, the verification parameter determination unit 911 uses the basic estimation model Y1 to determine the parameters from N. ini Among the random control parameters, the control parameter P that determines the film thickness information as the minimum evaluation value is... min_A ( Figure 6 (Minimum evaluation value parameter determination process S13). Specifically, the estimation unit 915 uses the basic estimation model Y1 to estimate the film thickness information corresponding to each random control parameter. Then, the inspection parameter determination unit 911 uses the estimated N... ini The smallest random control parameter among the evaluation values ​​shown by the film thickness information is taken as the control parameter P. min_A Furthermore, the minimum evaluation value implies that the uniformity of the film thickness distribution becomes the optimal control parameter.

[0099] Furthermore, the control parameters used for the test do not necessarily have to be determined using the basic presumption model Y1. The control parameters used for the test can also be randomly selected from the basic dataset DB1, for example.

[0100] The sliding parameter generation unit 912 is based on N obtained during the minimum evaluation value parameter determination process S13. ini The distribution of evaluation values ​​E shown by the film thickness information _A Determine the slippage amount. Evaluation value distribution E _A This represents the distribution of evaluation values ​​shown by the complex film thickness information output from the basic estimation model Y1. Slippage is the amount that causes a change in at least one parameter element of the control parameters. (Refer to...) Figure 7A , Figure 7B as well as Figure 7C The method for determining the slip amount is explained.

[0101] Figure 7A E represents the distribution of evaluation values ​​in the parameter space. _A The diagram illustrates the case where the control parameters consist of two parameter elements, x1 and x2. However, the same procedure can be followed when there are three or more parameter elements. Figure 7A The distribution of evaluation values ​​E shown _A In the diagram, dark shading represents groups of smaller evaluation values, and light shading represents groups of larger evaluation values. The control parameter P represents the minimum evaluation value. min_A Located in the evaluation value distribution E _A Roughly in the center.

[0102] In the evaluation value distribution E _A In the middle, the sliding parameter generation unit 912 calculates the control parameter P that passes through the minimum evaluation value. min_A And a straight line L1 (parallel line) parallel to the axis of parameter element x1 ( Figure 6 Linear calculation procedure S14).

[0103] The sliding parameter generation unit 912 determines the sliding amount based on the distribution of evaluation values ​​on the straight line L1. Figure 7B It means Figure 7A The graph shows the distribution of evaluation values ​​along line L1. Figure 7B In the diagram, the horizontal axis represents x1, and the vertical axis represents the evaluation value. For example... Figure 7B As shown, the evaluation value distribution takes the point with the minimum evaluation value as the lower end and increases monotonically in both the positive and negative directions of the x1 axis. The sliding parameter generation unit 912 calculates the standard deviation σ, which is one of the characteristics of the evaluation value distribution. _eval Then, the sliding parameter generation unit 912 will use the calculated standard deviation σ as the standard deviation. _eval The values ​​ΔS1 and ΔS2 obtained by taking positive and negative integer multiples of each other are set as slip amounts. Figure 6 : Sliding amount determination process S15). For example, the standard deviation σ _eval The value of ΔS1 is 3 times that of 3σ _eval ) and standard deviation σ _eval -3 times the value of ΔS2 (= -3σ) _eval This is determined as the sliding amount.

[0104] In the slip amount determination step S15, the slip amount ΔS1 in the positive direction that increases the value of the parameter element and the slip amount ΔS2 in the negative direction that decreases the value of the parameter element are determined. This effectively expands the search range for appropriate slip amounts. Furthermore, the slip amount determined in the slip amount determination step S15 can be one or more slip amounts.

[0105] In addition, based on the standard deviation σ of the evaluation value distribution _eval To determine the slip amounts ΔS1 and ΔS2, the variation characteristics of the evaluation value distribution can be used as an indicator to search for effective slip amounts. Furthermore, it is not necessary to base it on the standard deviation σ. _eval The slippage amount is determined. For example, the slippage amount can also be determined based on a predetermined value. Specifically, the slippage amount can also be a control parameter P that minimizes the evaluation value. min_A The value obtained by multiplying x1 by a decimal less than 1 (±0.1, ±0.2, ±0.3, etc.).

[0106] When the sliding amounts ΔS1 and ΔS2 are determined, the sliding parameter generation unit 912 calculates the control parameter P respectively. min_A The control parameter P after the change of the parameter element x1 and the slip amount ΔS1, ΔS2 min_B (=P) min_A +ΔS1), P min_C (=P) min_A +ΔS2)(Sliding control parameter)( Figure 6 S16: Sliding parameter calculation procedure.

[0107] Figure 7C This represents the two control parameters P generated by the sliding parameter generation unit 912. min_B P min_C A graph showing the location in parameter space. For example... Figure 7C As shown, the two generated control parameters P min_B P min_C The position becomes the control parameter P along the straight line L1 min_A The positions after sliding in the positive and negative directions, respectively.

[0108] When the control parameter P is determined min_A P min_B P min_C At that time, the ejection control unit 910 applied the control parameter P. min_A P min_B P min_C The formation of the coating film ( Figure 6 (Coating film formation process S17). Then, the film thickness information acquisition unit 913 acquires the film thickness information relative to the control parameter P obtained by measuring the film thickness through the film thickness measuring instrument AP1. min_A Film thickness information T min_A (First film thickness information), and relative to the control parameter P min_B P min_C Film thickness information T min_B T min_C (Second film thickness information) Figure 6 (S18: Film thickness information acquisition process)

[0109] The optimal sliding amount determination unit 916 determines the amount of sliding based on the obtained film thickness information T. min_A T min_B T min_C The evaluation values ​​shown determine the optimal slip ΔS. opt ( Figure 6 : Optimal slip amount determination process S19). Specifically, the estimation unit 915 uses the basic estimation model Y1 to estimate the slip amount relative to the control parameter P. min_A P min_B P min_C Film thickness information P min_A_est P min_B_est P min_C_est Then, the optimal sliding amount determination unit 916 calculates the measured film thickness information T. min_A T min_B T min_C The evaluation value shown is consistent with the estimated film thickness information P. min_A_est P min_B_est P min_C_est The error between them. Then, the optimal slip amount determination unit 916 sets the slip amount of the control parameter that minimizes the error as the optimal slip amount ΔS. optFor example, in the control parameter P min_B The optimal slip ΔS is obtained when the error is minimized. opt It is set as the slip amount ΔS1. Additionally, in the control parameter P... min_A The optimal slip ΔS is obtained when the error is minimized. opt The value is zero. The determined optimal slip ΔS opt It is stored in memory 93. The optimal sliding amount determination process S1 is now complete.

[0110] <Optimization Process S2>

[0111] Next, for Figure 5 The optimization process S2 shown will be explained. Figure 8 yes Figure 5 The flowchart shows the detailed content of the optimization process S2. In optimization process S2, as described later, the optimization unit 918 uses the optimal slip amount ΔS determined in the optimal slip amount determination process S1. opt Based on the basic presupposition model Y1, the control parameters are optimized.

[0112] Furthermore, during the optimization process S2 performed by the optimization unit 918, a pre-set upper limit number of times N is set. est And the maximum number of searches N all Estimated upper limit of N est This indicates the number of times the estimated step S23, described later, is repeated. Furthermore, the optimization unit 918 counts the number of times the estimated step S23 is executed, therefore the estimated execution count C is used. est As a variable. Additionally, the maximum number of searches N. all This refers to the number of times the control parameter should be searched in a single optimization process (i.e., the number of times the optimization unit 918 determines the control parameter). To count the number of searches performed, the optimization unit 918 uses the search execution count C. ser As a variable.

[0113] When optimization process S2 begins, optimization unit 918 determines the initial control parameters to be evaluated. Figure 8 (Initial parameter determination step S21). The initial control parameters can be parameters already used in the coating apparatus 1, or randomly determined values. Alternatively, parameters from... Figure 6 The control parameter P, determined in step S13 of the minimum evaluation value parameter determination process, is shown. min_A .

[0114] Next, the optimization department determined the estimated execution count C. est Has the estimated maximum number of attempts N been reached? est ( Figure 8 : Judgment step S22). In judgment step S22, the estimated number of executions C is determined.est The estimated maximum number of times N was not reached est In this case, the optimization unit 918 causes the estimation unit 915 to perform estimation processing. That is, the estimation unit 915 uses the basic estimation model Y1 to estimate the film thickness information corresponding to the control parameters determined through the initial parameter determination process S21 (or the evaluation parameter determination process S29 described later). Figure 8 (Estimated process S23). Additionally, the optimization unit 918 estimates the number of executions C. est Add 1 ( Figure 8 (Incremental process S24). After that, the optimization unit 918 enters the evaluation parameter determination process S29.

[0115] In the determination process S22, the estimated number of executions C is determined. est The estimated maximum number of times N has been reached est In this case, the film thickness information acquisition unit 913 acquires film thickness information ( Figure 8 (Film thickness information acquisition step S25). Specifically, the control parameters determined in the initial parameter determination step S21 (or the evaluation parameter determination step S29 described later) are applied to the coating apparatus 1. However, here, the value of the parameter element x1 of the control parameters will be changed by the aforementioned optimal sliding amount ΔS. opt The subsequent sliding control parameters are applied to the coating apparatus 1. More specifically, the ejection control unit 910 controls the pump 81 based on the sliding control parameters, thereby forming a coating film on the substrate S, and the film thickness is measured by the film thickness measuring device AP1. Then, the film thickness information acquisition unit 913 acquires film thickness information as the measurement result. After the film thickness information acquisition step S25, the optimization unit 918 estimates the number of executions C. est Set to 0 ( Figure 8 : Reset process S26).

[0116] After the reset process S26, the optimization unit 918 determines whether to update the basic estimation model Y1. Figure 8 (Update Judgment Step S27). The criterion for whether to update the basic estimation model Y1 is, for example, whether the amount of unused film thickness information in the film thickness information obtained through the film thickness information acquisition step S25 is greater than a specified amount. In this case, if the amount of unused film thickness information is greater than the specified amount, the basic estimation model Y1 is updated.

[0117] In the update determination process S27, if it is determined that the basic presumption model Y1 needs to be updated, the learning unit 914 updates the basic presumption model Y1. Figure 8Model update step S28). That is, the dataset containing the control parameters and membrane thickness information added in step S25 (obtaining membrane thickness information) is used as teacher data for relearning the basic estimation model Y1. Furthermore, in model update step S28, the control parameters of the dataset used for relearning are the applicable optimal slip ΔS. opt The original control parameters were previously used. This prevents the basic estimated model Y1 from being affected by the optimal slip ΔS. opt The impact.

[0118] When the model update process S28 is completed, the optimization unit 918 executes the evaluation parameter determination process S29. Alternatively, if the update determination process S27 determines that no update is needed, the optimization unit 918 skips the model update process S28 and executes the evaluation parameter determination process S29.

[0119] In the evaluation parameter determination process S29, the optimization unit 918 determines the control parameters to be evaluated next based on the evaluation values ​​shown by the film thickness information estimated through the estimation process S23 or the film thickness information obtained through the film thickness information acquisition process S25. Figure 8 (Evaluation parameter determination step S29). Algorithms for determining the control parameters used in the evaluation include, for example, reinforcement learning (RL), Bayesian optimization, or particle swarm optimization. Based on these methods, the control parameters can be updated at each trial to optimize the evaluation value.

[0120] After the evaluation parameter determination process S29, the optimization unit 918 will perform the search execution count C. ser Add 1 ( Figure 8 (Incremental process S30). Then, the optimization department 918 determines the number of search executions C. ser Has the maximum number of searches N been reached? all ( Figure 8 : Judgment step S31). In judgment step S31, the number of search executions is determined to be C. ser The maximum number of searches (N) has not been reached. all In this case, the optimization unit 918 returns to the decision process S22 to continue processing. On the other hand, in the decision process S31, it is determined that the number of search executions C is... ser The maximum number of searches N has been reached. all In this case, the optimization process ends at step 918.

[0121] As described above, the optimal control parameters obtained through optimization are stored in memory 93 and used in subsequent coating processes in coating apparatus 1. However, in coating apparatus 1, instead of directly using the optimal control parameters, the optimal slip amount ΔS that causes the optimal control parameters to change is used. opt The subsequent sliding control parameters.

[0122] <Effect>

[0123] According to the parameter optimization method of this embodiment, the optimal sliding amount ΔS that makes the control parameter change is achieved is determined by... opt The subsequent sliding control parameters are applied to coating apparatus 1, enabling the film thickness information of the formed coating film to closely approximate the output of the basic estimation model Y1. Therefore, even when using the learned basic estimation model Y1, which has errors with coating apparatus 1, by applying the optimal sliding amount ΔS... opt It can also efficiently optimize control parameters.

[0124] Furthermore, the distribution of evaluation values ​​along the straight line L1 parallel to the axis of the control parameter element x1 represents the change in evaluation values ​​when only x1 is varied while other parameter elements are fixed at their optimal values. Therefore, when the error between the output of the basic estimation model Y1 and the output of the coating apparatus 1 is caused by the parameter element x1, the optimal effective sliding amount can be searched by calculating the sliding amount with respect to this parameter element.

[0125] <2. Second Implementation Method>

[0126] Next, the second embodiment will be described. It should be noted that in the following description, elements having the same function as those already described will be marked with the same reference numerals or additional letter reference numerals, and detailed descriptions may sometimes be omitted.

[0127] Figure 9 This is a block diagram illustrating the function of the control unit 9a according to the second embodiment. Figure 9 The functional blocks shown are functional blocks implemented by the processor 91 executing the computer program 931. The control unit 9a has an optimal estimation model determination unit 917 instead of the optimal slip amount determination unit 916. Figure 4 The optimal estimation model determination unit 917 performs processing to determine the estimation model most suitable for estimating film thickness information. Additionally, the optimization unit 918 uses the optimal estimation model Y determined by the optimal estimation model determination unit. _opt Optimize the execution control parameters.

[0128] Figure 10 This is a flowchart illustrating the parameter optimization method of the second embodiment. The parameter optimization method of this embodiment includes an optimal estimated model determination process S1a and an optimization process S2. In this embodiment, the optimal estimated model determined by the optimal estimated model determination process S1a is used in the optimization process S2a.

[0129] Figure 11 To indicate Figure 10The flowchart shows the detailed content of the optimal estimation model determination process S1a. In this optimal estimation model determination process S1a, there is also a connection with the optimal slip amount determination process S1a of the first embodiment. Figure 6 Similarly, by executing the first learning process S11 to the sliding amount determination process S15, the sliding amounts ΔS1 and ΔS2 are determined.

[0130] Sliding parameter generation unit 912 ( Figure 9 Create two sliding datasets, DB2 and DB3, based on the base dataset DB1, including the sliding control parameters after changing the slip amount ΔS1 and ΔS2 of the control parameters. Figure 11 (Sliding dataset creation process S16a). For example, in the basic dataset DB1, when the pair of control parameter P and film thickness information T is [P, T], the pair of sliding dataset DB2 is represented as [P+ΔS1, T], and the pair of sliding dataset DB3 is represented as [P+ΔS2, T].

[0131] Learning Department 914 used two sliding datasets, DB2 and DB3, to learn two sliding estimation models, Y2 and Y3, respectively, with sliding control parameters as input and membrane thickness information as output. Figure 11 (Learning process S16b).

[0132] Following the learning process S16b, the inspection parameter determination unit 911 generates N in the same manner as the random control parameter generation process S12. ini N ini Random control parameters (integers greater than or equal to 1) Figure 11 (Random control parameter generation process S16c). Furthermore, in random control parameter generation process S16c, the random control parameters generated in random control parameter generation process S12 can be reused. Then, the verification parameter determination unit 911 uses two sliding estimation models Y2 and Y3 to determine the parameters from N. ini Among the random control parameters, control parameter P is determined to minimize the evaluation value based on the film thickness information estimated by each model. min_B P min_C ( Figure 11 (Minimum evaluation value parameter determination process S16d).

[0133] Figure 12 E represents the distribution of evaluation values ​​for the film thickness information output by the basic estimation model Y1 and the sliding estimation models Y2 and Y3. _A E _B E _C The image. (As shown) Figure 12 As shown, the evaluation value distribution E represented by the outputs of the two sliding inference models Y2 and Y3 _B E _CThe distribution of evaluation values ​​E that makes the output of the basic presumption model Y1 represent is made possible by... _A The distributions of the sliding amounts ΔS1 and ΔS2 were shifted respectively.

[0134] The sliding estimation model Y2 is a model learned using the sliding dataset DB2, which is a set of pairs of sliding control parameters and membrane thickness information. Therefore, the control parameter P... min_B This becomes the control parameter corresponding to the sliding control parameter. Regarding the control parameter P... min_C The parameters are the same as those for sliding control.

[0135] After the minimum evaluation value parameter determination process S16d, and the optimal sliding amount determination process S1 ( Figure 6 Similarly, the coating film formation process S17 and the film thickness information acquisition process S18 are performed. Using the film thickness information acquisition process S18, the film thickness information is acquired relative to the control parameter P. min_A Film thickness information T min_A (First film thickness information), and relative to the control parameter P min_B P min_C Film thickness information T min_B T min_C (Second film thickness information).

[0136] The optimal estimation model determination unit 917 determines the film thickness based on the obtained film thickness information T. min_A T min_B T min_C The evaluation values ​​shown are used to determine the optimal estimation model (optimal estimation model determination step S19a). Specifically, the measured film thickness information T is compared with the measured film thickness information T. min_A T min_B T min_C The evaluation values ​​shown are relative to the control parameters P estimated using each model. min_A P min_B P min_C Film thickness information T min_A_est T min_B_est T min_C_est The error between them. Then, the estimation model with the smallest error (i.e., the estimation model whose output is close to the measured result) is determined as the optimal estimation model Y. _opt .

[0137] Figure 13 It means Figure 11 The flowchart shows the detailed content of the optimization process S2a. The basic flow of optimization process S2a is similar to the flow of optimization process S2 in the first embodiment. Figure 8 The same applies. However, in the optimization process S2a, in the estimation step S23, the estimation unit 915 replaces the basic estimation model Y1 and uses the optimal estimation model Y._opt This is used to estimate film thickness information. Furthermore, in the optimization process S2a, during the film thickness information acquisition step S25, the control parameters are directly applied to the coating apparatus 1 without any change.

[0138] In the second embodiment, the optimal estimation model Y for determining the film thickness information with a small error between the output and the film thickness information obtained based on the measured film thickness information formed by the coating apparatus 1 is determined. _opt Therefore, the optimal estimation model Y is used. _opt It can efficiently optimize the control parameters applicable to the coating device 1.

[0139] <3. Variations>

[0140] The embodiments have been described above, but the present invention is not limited to the embodiments described above and can be modified in various ways.

[0141] For example, the methods for determining the slip amounts ΔS1 and ΔS2 are not limited to... Figures 7A to 7C The method shown. Figure 14A E represents the distribution of evaluation values ​​in the parameter space. _A The diagram shows the variation. In this modified example, the sliding parameter generation unit 912 performs the evaluation value distribution E in the straight line calculation process S14. _A The regression line L2 is calculated. Furthermore, the sliding parameter generation unit 912 calculates the control parameter P that passes through the minimum evaluation value. min_A The line L1a intersects (or is orthogonal to) the regression line L2.

[0142] Next, in the sliding amount determination process S15, the sliding parameter generation unit 912 determines the sliding amount based on the distribution of evaluation values ​​on the straight line L1a. Figure 14B It means Figure 14A The graph shows the distribution of evaluation values ​​along the straight line L1a. Figure 14B In the diagram, the horizontal axis represents the parameter elements (x1 and x2), and the vertical axis represents the evaluation value. The sliding parameter generation unit 912 calculates the standard deviation σ indicated by the distribution of these evaluation values. _eval Furthermore, the sliding parameter generation unit 912 will generate the standard deviation σ. _eval The values ​​ΔS1 and ΔS2 obtained by positive and negative integer multiples are set as sliding amounts.

[0143] Figure 14C This represents the two control parameters P generated by the sliding parameter generation unit 912. min_B P min_C A graph showing the location in parameter space. For example... Figure 14C As shown, the two generated control parameters P min_B P min_C The position becomes the control parameter P min_AThe positions after sliding along the straight line L1a in the positive and negative directions, respectively. In addition, the sliding amount determined by this method is the sliding amount that causes the complex number of parameter elements (x1, x2) to change.

[0144] The direction in which the regression line L2 intersects indicates the direction of a sharp change in the evaluation value. That is, by setting a sliding amount in this direction, the change in the evaluation value can be relatively increased. Therefore, by considering the characteristics of the distribution of evaluation values ​​in this direction (e.g., standard deviation σ), _eval Using these as indicators to determine the sliding amounts ΔS1 and ΔS2, it is possible to search for the sliding amounts that effectively reduce the error between the basic presupposition model Y1 and the coating device 1.

[0145] The present invention has been described in detail, but the above description is illustrative in all respects and the invention is not limited thereto. It should be understood that numerous modifications not illustrated can be conceived without departing from the scope of the invention. The structures described in the above embodiments and modifications can be appropriately combined or omitted as long as they do not contradict each other.

Claims

1. A parameter optimization device, wherein, have: The inspection parameter determination unit determines the control parameters used for inspection. The sliding parameter generation unit determines, based on the output of a basic estimation model that takes control parameters as input and film thickness information as output, the sliding amount that causes at least one parameter element of the control parameters used for testing to change, and generates sliding control parameters that cause the control parameters used for testing to change by the sliding amount. The film thickness information acquisition unit acquires, for the coating apparatus which is to be optimized, first film thickness information of the coating film when the control parameters for inspection are applied, and second film thickness information of the coating film when the sliding control parameters are applied. The optimal sliding amount determination unit determines the optimal sliding amount that causes the at least one parameter element to change, based on the first film thickness information and the second film thickness information. as well as The optimization unit uses the basic estimation model and the optimal sliding amount to optimize the control parameters applicable to the coating apparatus that is the object of optimization.

2. A parameter optimization device, wherein, have: The sliding parameter generation unit determines the sliding amount that causes at least one parameter element of the control parameter to change, and generates a sliding dataset including the sliding control parameter after changing the control parameter by the sliding amount, based on a set of pairs of the control parameter and film thickness information associated with the control parameter, i.e., a basic dataset. The learning unit uses the sliding dataset to learn a sliding estimation model that takes the sliding control parameters as input and the film thickness information as output. The inspection parameter determination unit uses a basic estimation model that takes the control parameters as input and the film thickness information as output to determine the control parameters for inspection, and uses the sliding estimation model to determine the sliding control parameters for inspection. The film thickness information acquisition unit acquires, for the coating apparatus which is to be optimized, first film thickness information of the coating film when the control parameters for inspection are applied, and second film thickness information of the coating film when the sliding control parameters for inspection are applied. The optimal estimation model determination unit determines the optimal estimation model from the basic estimation model and the sliding estimation model based on the first film thickness information and the second film thickness information. as well as The optimization unit optimizes the control parameters applicable to the coating apparatus that is the object of optimization using the optimal estimation model determined by the optimal estimation model determination unit.

3. The parameter optimization device according to claim 1 or 2, wherein, The sliding parameter generation unit determines the sliding amount based on the distribution of evaluation values ​​represented by a plurality of film thickness information output from the basic estimation model.

4. The parameter optimization device according to claim 3, wherein, The sliding parameter generation unit determines the sliding amount based on the characteristics of the shape of the evaluation value distribution on the straight line passing through the minimum value of the evaluation value distribution.

5. The parameter optimization device according to claim 4, wherein, The straight line is parallel to the axis of a parameter feature.

6. The parameter optimization device according to claim 4, wherein, The straight line intersects the regression line of the evaluation value distribution.

7. The parameter optimization apparatus according to any one of claims 1 to 6, wherein, The control parameters used for the test are optimized using the basic estimation model.

8. A parameter optimization method, wherein, include: The inspection parameters are determined by the process steps, and the control parameters for inspection are determined accordingly. The sliding parameter generation process, based on the output of a basic estimation model that takes control parameters as input and film thickness information as output, determines the sliding amount that causes at least one parameter element of the control parameters used for testing to change, and generates sliding control parameters that cause the control parameters used for testing to change by the sliding amount. In the film thickness information acquisition process, for the coating apparatus which is to be optimized, the first film thickness information of the coating film when the control parameters for inspection are applied, and the second film thickness information of the coating film when the sliding control parameters are applied are acquired respectively. The optimal sliding amount determination process determines the optimal sliding amount that causes the at least one parameter element to change, based on the first film thickness information and the second film thickness information. as well as The optimization process involves using the basic estimation model and the optimal sliding amount to optimize the control parameters applicable to the coating apparatus, which is the object of optimization.

9. A parameter optimization method, wherein, have: The sliding parameter generation process involves determining the sliding amount that causes at least one parameter element of the control parameter to change, and generating a sliding dataset that includes the sliding control parameters after the control parameter has changed by the sliding amount, based on a set of pairs of the control parameter and film thickness information associated with the control parameter, i.e., a basic dataset. The learning process involves using the sliding dataset to learn a sliding estimation model that takes the sliding control parameters as input and the film thickness information as output. The inspection parameter determination process uses a basic estimation model that takes the control parameters as input and the film thickness information as output to determine the control parameters for inspection, and uses the sliding estimation model to determine the sliding control parameters for inspection. In the film thickness information acquisition process, for the coating apparatus which is to be optimized, the first film thickness information of the coating film when the control parameters for inspection are applied, and the second film thickness information of the coating film when the sliding control parameters for inspection are applied, are acquired respectively. The optimal estimation model determination process involves determining the optimal estimation model from the basic estimation model and the sliding estimation model based on the first film thickness information and the second film thickness information. as well as The optimization process involves using the optimal estimation model determined by the optimal estimation model determination process to optimize the control parameters applicable to the coating apparatus that is the object of optimization.

10. A computer program product comprising a computer program, wherein, When the computer program is executed by a computer, it implements the steps of the parameter optimization method as described in claim 8 or 9.

Images