Discharge pressure monitoring device, discharge pressure monitoring method, computer program, and coating device

Abstract

To provide a technology capable of appropriately detecting unknown abnormalities in discharge pressure that have not been anticipated in advance.SOLUTION: A discharge pressure measuring unit 911 acquires pressure data. A feature amount calculation unit 913 calculates a feature amount of pressure data. An abnormality level calculation unit 915 uses an unsupervised learning model M1, which takes the feature amount as input and outputs an abnormality level indicating the degree of deviation from the distribution of normal feature amounts, to calculate the abnormality level for the feature amount calculated by the feature calculation unit 913. An abnormality probability calculation unit 917 uses a supervised learning model M2, which takes the feature amount as input and outputs an abnormality probability indicating the probability corresponding to a specific abnormality A or B, to calculate an abnormality probability for the feature amount calculated by the feature calculation unit 913. An abnormality determination unit 919 determines whether the pressure data is abnormal by using the abnormality level calculated by the abnormality level calculation unit 915 and the abnormality probability calculated by the abnormality probability calculation unit 917 .SELECTED DRAWING: Figure 9

Description

[Technical field]

[0001] The subject matter disclosed herein relates to a discharge pressure monitoring device, a discharge pressure monitoring method, a computer program, and a coating apparatus. [Background technology]

[0002] In the manufacturing process of flat panel displays, a device called a coater is used. The coater discharges a processing liquid from a slit nozzle by driving a pump, and applies the processing liquid to the entire substrate being transported. With the recent trend toward higher product quality, such coaters are required to apply the processing liquid so that the film thickness of the processing liquid is uniform over the entire substrate. In order to keep the film thickness uniform, it is considered important to monitor the discharge pressure (discharge flow rate) of the processing liquid during production. For example, Patent Document 1 attempts to optimize parameters related to the discharge pressure.

[0003] However, it is difficult to detect minute changes in the discharge pressure obtained from the pressure sensor using a simple threshold value. In response to this, many methods and device systems that use machine learning to detect anomalies have been proposed as general techniques.

[0004] For example, Patent Document 2 proposes the construction of a system capable of determining the type (mode) of anomaly from given sensor information. Such a system construction not only makes it possible to plan specific countermeasures such as part replacement, but also allows the system to function as a failure prediction system capable of detecting minor changes (signs) that may develop into a breakdown. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2020-040046 A [Patent Document 2] JP 2022-125288 A Summary of the Invention [Problem to be solved by the invention]

[0006] Patent Document 2 proposes a method using unsupervised learning and a method using supervised learning as a system for predicting failures. However, since unsupervised learning is a method of learning data of a normal state and evaluating the degree of deviation of newly measured data from the normal state, it is possible to detect anomalies but is unable to distinguish the type of anomaly. Moreover, while supervised learning is able to distinguish the type of anomaly by learning teacher data that is previously labeled with the type of anomaly, it is unable to respond to unknown anomalies because any type of anomaly is classified into the type of anomaly that was previously assumed during learning.

[0007] An object of the present invention is to provide a technique capable of appropriately detecting an unknown abnormality in discharge pressure that has not been anticipated in advance. [Means for solving the problem]

[0008] In order to solve the above problem, a first aspect is a discharge pressure monitoring device comprising: a pressure data acquisition unit that acquires pressure data indicating a change over time in pressure inside a nozzle that discharges a treatment liquid; a feature calculation unit that calculates a feature of the pressure data; an anomaly calculation unit that calculates the degree of anomaly for the feature calculated by the feature calculation unit using an unsupervised learning model that takes the feature as an input and outputs an anomaly degree indicating the degree of deviation from a normal distribution of the feature; an anomaly probability calculation unit that calculates the anomaly probability for the feature calculated by the feature calculation unit using a supervised learning model that takes the feature as an input and outputs an anomaly probability indicating the probability that the feature corresponds to a specific anomaly; and an anomaly determination unit that determines an anomaly in the pressure data using the anomaly degree calculated by the anomaly degree calculation unit and the anomaly probability calculated by the anomaly probability calculation unit.

[0009] A second aspect is the discharge pressure monitoring device of the first aspect, wherein the unsupervised learning model is a model obtained by unsupervised learning using, as input data, feature quantities of only a plurality of normal pressure data.

[0010] A third aspect is the discharge pressure monitoring device of the first or second aspect, wherein the unsupervised learning model outputs a Mahalanobis distance as the degree of abnormality.

[0011] A fourth aspect is the discharge pressure monitoring device of the first or second aspect, wherein the supervised learning model outputs probabilities of falling under a plurality of types of specific abnormalities.

[0012] A fifth aspect is a discharge pressure monitoring device comprising: a pressure data acquisition unit that acquires pressure data indicating a time change in pressure inside a nozzle that discharges a treatment liquid; a feature calculation unit that calculates a feature of the pressure data; an anomaly calculation unit that calculates a degree of anomaly that indicates a degree of deviation from a normal feature for the feature calculated by the feature calculation unit; an anomaly probability calculation unit that calculates the anomaly probability for the feature calculated by the feature calculation unit using a supervised learning model that receives the feature as an input and outputs an anomaly probability that indicates a probability that the feature corresponds to a specific anomaly; and an anomaly determination unit that determines an anomaly in the pressure data using the degree of anomaly calculated by the anomaly degree calculation unit and the anomaly probability calculated by the anomaly probability calculation unit.

[0013] A sixth aspect is a method for monitoring an ejection pressure, comprising the steps of: a) acquiring pressure data indicating a change over time in pressure inside a nozzle that ejects a treatment liquid; b) calculating a feature of the pressure data; c) calculating the degree of anomaly for the feature calculated in step b) using an unsupervised learning model that takes the feature as an input and outputs an anomaly degree indicating the degree of deviation from a normal distribution of the feature; d) calculating the anomaly probability for the feature calculated in step b) using a supervised learning model that takes the feature as an input and outputs an anomaly probability indicating the probability that the feature corresponds to a specific anomaly; and e) determining an anomaly in the pressure data using the anomaly degree calculated in step c) and the anomaly probability calculated in step d).

[0014] A seventh aspect is a method for monitoring discharge pressure, comprising: A) acquiring pressure data indicating a change over time in pressure inside a nozzle that discharges a treatment liquid; B) calculating a feature of the pressure data; C) calculating an anomaly level for the feature calculated by step B, the anomaly level indicating the degree of deviation from a normal feature; D) calculating an anomaly probability for the feature calculated by step B using a supervised learning model that receives the feature as an input and outputs an anomaly probability indicating the probability that the feature corresponds to a specific anomaly; and E) determining an anomaly in the pressure data using the anomaly level calculated by step C) and the anomaly probability calculated by step D).

[0015] An eighth aspect is a computer program executable by a computer, causing the computer to execute the discharge pressure monitoring method of the sixth or seventh aspect.

[0016] A ninth aspect is a coating apparatus comprising a substrate holding part for holding a substrate, a nozzle for ejecting a processing liquid toward the substrate held by the substrate holding part, a pressure sensor for measuring the pressure inside the nozzle, and a ejection pressure monitoring device of the first, second or fifth aspect. Effect of the Invention

[0017] According to the first to ninth aspects, when a previously anticipated anomaly is determined, the anomaly can be determined by calculating the anomaly probability. Even when an unknown anomaly that has not been previously anticipated occurs, the unknown anomaly can be appropriately detected by calculating the anomaly degree.

[0018] According to the discharge pressure monitoring device of the second aspect, only the distribution of normal feature amounts can be learned, so that abnormality detection can be performed appropriately.

[0019] According to the third aspect of the discharge pressure monitoring device, the degree of abnormality can be calculated by taking into account the correlation between each dimension of the feature values ​​indicated by the pressure data, thereby making it possible to more accurately capture characteristic patterns and abnormal behavior of the pressure data.

[0020] According to the discharge pressure monitoring device of the fourth aspect, the probability of a plurality of types of specific abnormalities can be calculated, so that it is possible to determine whether the pressure data corresponds to any one of the plurality of types of abnormalities. [Brief description of the drawings]

[0021] [Figure 1] 1 is a diagram illustrating a schematic diagram of an overall configuration of a coating apparatus according to an embodiment. [Diagram 2] FIG. 2 is a diagram showing a configuration of a coating liquid supply mechanism. [Diagram 3] FIG. 4 is a diagram showing a pressure waveform of a discharge pressure. [Figure 4] FIG. 2 is a block diagram showing a configuration example of a control unit. [Diagram 5] FIG. 11 is a diagram for explaining an example of a feature amount. [Figure 6] FIG. 11 is a diagram for explaining another example of the feature amount. [Figure 7] FIG. 1 is a block diagram showing a control unit that performs machine learning. [Figure 8] FIG. 11 is a diagram showing the flow of machine learning processing executed by the control unit. [Figure 9]FIG. 4 is a block diagram showing a control unit that monitors abnormalities in pressure data. [Figure 10] FIG. 11 is a diagram showing a flow of a monitoring process for monitoring a discharge pressure executed by a control unit. [Figure 11] FIG. 13 is a diagram conceptually showing how an abnormality degree (Mahalanobis distance) is calculated. [Figure 12] FIG. 13 is a diagram conceptually showing an output of an anomaly probability calculation unit. [Figure 13] FIG. 13 is a diagram conceptually illustrating an example of comprehensive determination of anomalies using anomaly levels and anomaly probabilities. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0022] Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings. Note that the components described in the embodiment are merely examples, and are not intended to limit the scope of the present invention. In the drawings, the dimensions and numbers of each part may be exaggerated or simplified as necessary for ease of understanding.

[0023] FIG. 1 is a diagram showing a schematic overall configuration of a coating apparatus 1 according to an embodiment. The coating apparatus 1 is a substrate processing apparatus that coats a coating liquid on an upper surface Sf of a substrate S. The substrate S is, for example, a glass substrate for a liquid crystal display device. The substrate S may be a semiconductor wafer, a glass substrate for a photomask, a glass substrate for a plasma display, a glass or ceramic substrate for a magnetic or optical disk, a glass substrate for an organic electroluminescence (EL) display, a glass substrate or a silicon substrate for a solar cell, or any other substrate for electronic devices such as a flexible substrate and a printed circuit board. The coating apparatus 1 is, for example, a slit coater.

[0024] In FIG. 1, an XYZ coordinate system is defined to explain the positional relationship of each element of the coating apparatus 1. The transport direction of the substrate S is the "X direction". The direction in which the substrate S advances in the X direction (toward downstream in the transport direction) is the +X direction, and the opposite direction (toward upstream in the transport direction) is the -X direction. The direction perpendicular to the X direction is the Y direction, and the direction perpendicular to the X and Y directions is the Z direction. In the following description, the Z direction is the vertical direction, and the X and Y directions are the horizontal directions. In the Z direction, the +Z direction is the upward direction, and the -Z direction is the downward direction.

[0025] The coating apparatus 1 includes, in order in the +X direction, an input conveyor 100, an input transfer section 2, a floating stage section 3, an output transfer section 4, and an output conveyor 110. The input conveyor 100, the input transfer section 2, the floating stage section 3, the output transfer section 4, and the output conveyor 110 form a transport path through which the substrate S passes. The coating apparatus 1 further includes a substrate transport section 5, a coating mechanism 7, a coating liquid supply mechanism 8, and a control unit 9.

[0026] The substrate S is transported to the input conveyor 100 from an apparatus upstream of the coating apparatus 1. The input conveyor 100 includes a roller conveyor 101 and a rotation drive mechanism 102. The rotation drive mechanism 102 rotates each roller of the roller conveyor 101. By the rotation of each roller of the roller conveyor 101, the substrate S is transported downstream (+X direction) in a horizontal position. The "horizontal position" refers to a state in which the main surface of the substrate S (the surface with the largest area) is parallel to the horizontal plane (XY plane).

[0027] The input transfer section 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 raises and lowers the roller conveyor 21. The rotation of the roller conveyor 21 transports the substrate S downstream (+X direction) in a horizontal position. The elevation of the roller conveyor 21 also changes the position of the substrate S in the Z direction. The substrate S is transferred from the input conveyor 100 to the floating stage section 3 via the input transfer section 2.

[0028] As shown in FIG. 1, the floating stage part 3 is substantially flat. The floating stage part 3 is divided into three parts along the X direction. The floating stage part 3 includes, in order along the +X direction, an entrance floating stage 31, a coating stage 32, and an exit floating stage 33. The upper surface of the entrance floating stage 31, the upper surface of the coating stage 32, and the upper surface of the exit floating stage 33 are on the same plane. The floating stage part 3 further includes a lift pin driving mechanism 34, a floating control mechanism 35, and an elevation driving mechanism 36. The lift pin driving mechanism 34 raises and lowers a plurality of lift pins arranged on the entrance floating stage 31. The floating control mechanism 35 supplies compressed air for floating the substrate S to the entrance floating stage 31, the coating stage 32, and the exit floating stage 33. The elevation driving mechanism 36 raises and lowers the exit floating stage 33.

[0029] A large number of nozzle holes for spraying compressed air supplied from the levitation control mechanism 35 are arranged in a matrix on the upper surface of the entrance levitation stage 31 and the upper surface of the exit levitation stage 33. When compressed air is sprayed from each nozzle hole, the substrate S is levitated upward relative to the levitation stage part 3. Then, the lower surface Sb of the substrate S is supported in a horizontal position while being separated from the upper surface of the levitation stage part 3. When the substrate S is in a levitated state, the distance (levitation amount) between the lower surface Sb of the substrate S and the upper surface of the levitation stage part 3 is, for example, 10 μm or more and 500 μm or less.

[0030] On the upper surface of the coating stage 32, jet holes for jetting compressed air supplied from the floating control mechanism 35 and suction holes for sucking gas are alternately arranged in the X direction and the Y direction. The floating control mechanism 35 controls the amount of compressed air jetted from the jet holes and the amount of air sucked from the suction holes. This precisely controls the floating amount of the substrate S relative to the coating stage 32 so that the position in the Z direction of the upper surface Sf of the substrate S passing above the coating stage 32 becomes a specified value. The floating amount of the substrate S relative to the coating stage 32 is calculated by the control unit 9 based on the detection result of the sensor 61 or the sensor 62 described later. The floating amount of the substrate S relative to the coating stage 32 is preferably adjustable with high precision by airflow control.

[0031] The substrate S carried into the floating stage section 3 is given a propulsive force in the +X direction by the roller conveyor 21, and is transported onto the entrance floating stage 31. The entrance floating stage 31, the coating stage 32, and the exit floating stage 33 support the substrate S in a floating state. For example, the configuration described in Japanese Patent No. 5346643 may be adopted as the floating stage section 3.

[0032] The substrate transport unit 5 is disposed below the floating stage unit 3. The substrate transport unit 5 includes a chuck mechanism 51 and a suction / travel control mechanism 52. The chuck mechanism 51 includes a suction pad (not shown) provided on a suction member. The chuck mechanism 51 supports the substrate S from below by bringing the suction pad into contact with the peripheral portion of the lower surface Sb of the substrate S. The suction / travel control mechanism 52 applies negative pressure to the suction pad to suction the substrate S. The suction / travel control mechanism 52 also causes the substrate transport unit 5 to travel back and forth in the X direction.

[0033] The chuck mechanism 51 holds the substrate S in a state in which the lower surface Sb of the substrate S is located at a position higher than the upper surface of the floating stage part 3. With the peripheral edge of the substrate S held by the chuck mechanism 51, the substrate S maintains a horizontal position due to the buoyancy applied by the floating stage part 3.

[0034] 1, the coating device 1 is equipped with a sensor 61 for measuring plate thickness. The sensor 61 is disposed near the roller conveyor 21. The sensor 61 detects the position in the Z direction of the upper surface Sf of the substrate S held by the chuck mechanism 51. In addition, a chuck (not shown) not holding the substrate S is positioned directly below the sensor 61, so that the sensor 61 can detect the position in the vertical direction Z of the suction surface, which is the upper surface of the suction member.

[0035] The chuck mechanism 51 moves in the +X direction while holding the substrate S that has been carried into the floating stage section 3. As a result, the substrate S is transported from above the entrance floating stage 31, via above the coating stage 32, to above the exit floating stage 33. Then, the substrate S is moved from the exit floating stage 33 to the output transfer section 4.

[0036] The output transfer unit 4 moves the substrate S from a position above the exit floating stage 33 to 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 drives the roller conveyor 41 to rotate, and also raises and lowers the roller conveyor 41 in the Z direction. As each roller of the roller conveyor 41 rotates, the substrate S moves in the +X direction. Additionally, as the roller conveyor 41 rises and falls, the substrate S is displaced in the Z direction.

[0037] The output conveyor 110 includes a roller conveyor 111 and a rotation drive mechanism 112. The output conveyor 110 transports the substrate S in the +X direction by rotation of each roller of the roller conveyor 111, and discharges the substrate S to the outside of the coating apparatus 1. The input conveyor 100 and the output conveyor 110 are part of the coating apparatus 1. However, the input conveyor 100 and the output conveyor 110 may be incorporated in a device separate from the coating apparatus 1.

[0038] The coating mechanism 7 applies a coating liquid to the upper surface Sf of the substrate S. The coating mechanism 7 is disposed above the transport path of the substrate S. The coating mechanism 7 has a nozzle 71. The nozzle 71 is a slit nozzle having a slit-shaped outlet on the lower surface. The nozzle 71 is connected to a positioning mechanism (not shown). The positioning mechanism moves the nozzle 71 between a coating position (position indicated by a solid line in FIG. 1) above the coating stage 32 and a maintenance position described below. The coating liquid supply mechanism 8 is connected to the nozzle 71. The coating liquid supply mechanism 8 supplies the coating liquid to the nozzle 71, whereby the coating liquid is discharged from an outlet disposed on the lower surface of the nozzle 71.

[0039] FIG. 2 is a diagram showing the configuration of the coating liquid supply mechanism 8. The coating liquid supply mechanism 8 includes a pump 81, a pipe 82, a coating liquid refill unit 83, a pipe 84, an on-off valve 85, a pressure sensor 86, and a drive unit 87. The pump 81 is a supply source for supplying the coating liquid to the nozzle 71, and supplies the coating liquid by changing the volume. The pump 81 may be, for example, a bellows-type pump described in JP-A-10-61558. As shown in FIG. 2, the pump 81 has a flexible tube 811 that is elastically expandable and contractible in the radial direction. One end of the flexible tube 811 is connected to the coating liquid refill unit 83 via the pipe 82. The other end of the flexible tube 811 is connected to the nozzle 71 via the pipe 84.

[0040] The pump 81 has a bellows 812 that is elastically deformable in the axial direction. The bellows 812 has a small bellows portion 813, a large bellows portion 814, a pump chamber 815, and an operating disk portion 816. The pump chamber 815 is disposed between the flexible tube 811 and the bellows 812. A non-compressible medium is sealed in the pump chamber 815. The operating disk portion 816 is connected to the drive portion 87.

[0041] The coating liquid refill unit 83 has a storage tank 831 that stores the coating liquid. The storage tank 831 is connected to the pump 81 via a pipe 82. An on-off valve 833 is inserted in the pipe 82. The on-off valve 833 opens and closes in response to a command from the control unit 9. When the on-off valve 833 is opened, the coating liquid can be replenished from the storage tank 831 to the flexible tube 811 of the pump 81. When the on-off valve 833 is closed, the refilling of the coating liquid from the storage tank 831 to the flexible tube 811 of the pump 81 is restricted.

[0042] The pipe 84 is connected to the output side of the pump 81. The on-off valve 85 is interposed in the pipe 84. The on-off valve 85 opens and closes in response to a command from the control unit 9. The on-off valve 85 opens and closes to switch between sending and stopping the coating liquid to the nozzle 71. The pressure sensor 86 is disposed in the pipe 84. The pressure sensor 86 detects the pressure (discharge pressure) of the coating liquid sent to the nozzle 71, and outputs a signal indicating the detected pressure value to the control unit 9.

[0043] Fig. 3 is a diagram showing the pressure waveform of the discharge pressure. In Fig. 3, the horizontal axis indicates time, and the vertical axis indicates pressure value. In the coating device 1, an optimization process is appropriately performed to approximate the pressure waveform of the coating liquid discharged from the nozzle 71 to an ideal pressure waveform by adjusting various parameters (acceleration time, steady speed, steady speed time, deceleration time, etc.) that define the movement of the operating disk part 816.

[0044] As shown in Fig. 3, the discharge pressure is measured during a period from before the nozzle 71 starts discharging the coating liquid to after the nozzle 71 finishes discharging the coating liquid. In the example shown in Fig. 3, the discharge pressure at time ta when the nozzle 71 starts discharging the coating liquid and the discharge pressure at time te when the nozzle 71 finishes discharging the coating liquid are set to an initial pressure Pi. However, the pressures at the start and end of discharging do not always match the initial pressure Pi. As shown in Fig. 3, the discharge period is divided into a rise period T1, a transition period T2, a steady period T3, and a fall period T4.

[0045] The rise period T1 is a period from time ta when the coating liquid supply mechanism 8 starts discharging the coating liquid from the nozzle 71 (i.e., the time when the coating liquid supply mechanism 8 starts moving the operating disk portion 816) to time tb when the discharge pressure reaches the target pressure Pt. In other words, when the discharge of the coating liquid from the nozzle 71 starts at time ta, the discharge pressure increases from the initial pressure Pi to the target pressure Pt during the period from time ta to time tb.

[0046] The transition period T2 is a period from time tb to time tc, which is a predetermined vibration damping period. This vibration damping period is a period required for the time change of the discharge pressure to stabilize, and is set, for example, by a user through an input operation on the input device 97 and stored in the storage unit 93.

[0047] The steady period T3 is a period from time tc to time td when the coating liquid supply mechanism 8 starts to reduce the discharge pressure (i.e., the time when the coating liquid supply mechanism 8 starts to decelerate the operating disk portion 816 from the target speed). That is, the coating liquid supply mechanism 8 moves the operating disk portion 816 at a constant speed from time tc to time td, and starts to decelerate the operating disk portion 816 at time td. Note that in the steady period T3, the discharge pressure is basically stable at the target pressure Pt. However, even in the steady period T3, the time change in the discharge pressure includes minute vibrations, and the discharge pressure becomes larger or smaller than the target pressure Pt.

[0048] The falling period T4 is the period from time td to time te when the coating liquid supply mechanism 8 finishes discharging the coating liquid from the nozzle 71 (i.e., time te when the coating liquid supply mechanism 8 stops the operating disk portion 816). That is, the discharge pressure decreases to the initial pressure Pi during the period from time td to time te, and at time te, the discharge of the coating liquid from the nozzle 71 stops.

[0049] As shown in FIG. 1 and FIG. 2, a sensor 62 is disposed on the nozzle 71 to which the coating liquid is supplied from the coating 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 (separation distance) between the floating substrate S and the upper surface of the coating stage 32. Then, the control unit 9 adjusts the coating position of the nozzle 71 by the positioning mechanism based on the measured separation distance. Note that an optical sensor or an ultrasonic sensor can be used as the sensor 62.

[0050] The coating mechanism 7 includes a nozzle cleaning standby unit 72. The nozzle cleaning standby unit 72 performs a predetermined maintenance on the nozzle 71 arranged at the maintenance position. The nozzle cleaning standby unit 72 includes a roller 721, a cleaning section 722, and a roller pad 723. The nozzle cleaning standby unit 72 prepares the discharge port of the nozzle 71 in a state suitable for a coating process by cleaning the nozzle 71 and forming a liquid pool. In addition, in the coating device 1, in order to evaluate the discharge pressure applied to the coating liquid, a simulated discharge is performed in which the coating liquid is discharged from the nozzle 71 while the nozzle 71 is arranged at the maintenance position.

[0051] FIG. 4 is a block diagram showing a configuration example of the control unit 9. The control unit 9 controls the operation of each element of the coating device 1. The control unit 9 includes a calculation unit 91, a storage unit 93, a display 95, and an input device 97. For example, a desktop type, a laptop type, or a tablet type computer can be used as the control unit 9. As described later, the control unit 9 functions as a discharge pressure monitoring device that monitors abnormalities in the discharge pressure.

[0052] The calculation unit 91 is a processor including a CPU (Central Processing Unit), etc. The storage unit 93 includes a temporary storage device such as a RAM (Random Access Memory), and a non-transient auxiliary storage device such as a HDD (Hard Disk Drive) and a SDD (Solid State Drive).

[0053] The display 95 is a device that displays information to the user, and specifically, is a liquid crystal display, etc. The input device 97 is a device that accepts input operations by the user, and is, for example, a mouse and a keyboard.

[0054] The storage unit 93 stores a computer program 931. The computer program 931 is provided by a recording medium M. That is, the recording medium M records the computer program 931 so as to be readable by the control unit 9, which is a computer. The recording medium M is, for example, a Universal Serial Bus (USB) memory, an optical disk such as a Digital Versatile Disc (DVD), a magnetic disk, or the like.

[0055] The calculation unit 91 executes the computer program 931 to function as a discharge control unit 910 , a discharge pressure measurement unit 911 , a feature calculation unit 913 , an abnormality degree calculation unit 915 , an abnormality probability calculation unit 917 , and an abnormality determination unit 919 .

[0056] The discharge control unit 910 controls the operation (feeding operation) of the pump 81 that feeds the coating liquid to the nozzle 71 based on preset parameters.

[0057] The discharge pressure measuring unit 911 measures the discharge pressure. Specifically, the discharge pressure measuring unit 911 periodically acquires the discharge pressure measured by the pressure sensor 86 at a predetermined sampling period. The discharge pressure measuring unit 911 acquires the time change (time series data) of the discharge pressure applied to the coating liquid during the period in which the coating liquid is discharged from the nozzle 71, and stores the acquired data as pressure data in the storage unit 93. The pressure data is data indicating each time and the pressure measured at each time. The discharge pressure measuring unit 911 is an example of a pressure data acquiring unit.

[0058] The feature amount calculation unit 913 derives a feature amount from the discharge pressure measured by the discharge pressure measurement unit 911. As the feature amount, for example, some feature amounts described in JP 2022-138109 A may be adopted.

[0059] FIG. 5 is a diagram for explaining an example of a feature amount. In this example, the degree of overshoot that occurs at the rise of the discharge pressure is calculated as the feature amount F1. Specifically, the feature amount calculation unit 913 calculates the sign (positive or negative) of the second derivative Dif2 of the discharge pressure at time t11 when the discharge pressure reaches the maximum value Pmax. Then, the feature amount calculation unit 913 calculates the time t12 when the sign of the second derivative switches twice from the sign at time t11. Then, the feature amount calculation unit 913 calculates the feature amount F1 of the time change of the discharge pressure in the initial vibration period T2_s from time t11 to t12.

[0060] For example, the feature calculation unit 913 selects the smaller pressure between the minimum value Pmin of the discharge pressure in the initial vibration period T2_s and the steady pressure Pm (average value of the discharge pressure in the steady period T3) as the target pressure Pg. Then, the feature calculation unit 913 may calculate the difference between the maximum pressure Pmax and the target pressure Pg (=Pmax-Pg) as the feature F1.

[0061] Such a feature amount F1 becomes relatively larger as the overshoot of the time change in the discharge pressure increases due to the momentum of the rise in the discharge pressure.

[0062] 6 is a diagram for explaining another example of the feature amount. In this example, the stability of the time change of the discharge pressure in the transition period T2 is calculated as the feature amount F2. Specifically, the feature amount calculation unit 913 calculates the root mean square error RMSE(P_measure, Pm) between the discharge pressure in the transition period T2 and the steady pressure Pm which is the average value of the discharge pressure in the steady period T3 as the feature amount F2. Such a feature amount F2 becomes relatively larger as the ringing (waveform vibration) of the time change of the discharge pressure in the transition period T2 becomes larger.

[0063] The abnormality degree calculation unit 915 inputs the feature amount calculated by the feature amount calculation unit 913 into the unsupervised learning model M1 to calculate the abnormality degree of the pressure data.

[0064] The anomaly probability calculation unit 917 inputs the feature calculated by the feature calculation unit 913 into the supervised learning model M2 to calculate the probability of an anomaly. In detail, the pressure data can be classified into a normal state and one or more types of abnormal states assumed in advance. The feature calculation unit 913 calculates the probability that the pressure data corresponds to each of the one or more types of abnormalities assumed in advance (hereinafter also referred to as "anomaly probability"). For example, when two types of abnormalities ("abnormality A" and "abnormality B") are assumed in advance during learning of the supervised learning model M2, the pressure data is classified into a normal state, an abnormality A state, and an abnormality B state. Then, the anomaly probability calculation unit 917 calculates an anomaly probability indicating the probability that the pressure data corresponds to an abnormality A or B.

[0065] The abnormality determination unit 919 determines whether the pressure data is abnormal, using the degree of abnormality calculated by the abnormality degree calculation unit 915 and the abnormality probability calculated by the abnormality probability calculation unit 917 .

[0066] <Learning stage> Fig. 7 is a block diagram showing the control unit 9 that performs machine learning. Fig. 8 is a diagram showing the flow of the machine learning process executed by the control unit 9. Note that the machine learning shown in Figs. 8 and 9 may be performed by a computer device different from the control unit 9.

[0067] First, as shown in Fig. 7 and Fig. 8, in order to perform machine learning, a considerable number of normal pressure data D1, a considerable number of pressure data D2 corresponding to abnormality A, and a considerable number of pressure data D3 corresponding to abnormality B are acquired in advance by the discharge pressure measurement unit 911 (Fig. 8: step S11). The acquired pressure data D1 to D3 are stored in the storage unit 93. Then, the feature calculation unit 913 acquires the feature amounts of each of the pressure data D1 to D3 (Fig. 8: step S12).

[0068] The anomaly degree calculation unit 915 executes machine learning (unsupervised learning) to obtain the unsupervised learning model M1 (FIG. 8: step S13). The anomaly degree calculation unit 915 stores the unsupervised learning model M1 (more specifically, learned parameters) obtained by the machine learning in the storage unit 93 (step S14).

[0069] In unsupervised learning, specifically, a set of features of multiple normal pressure data D1 is prepared as learning data. Then, machine learning based on the k-nearest neighbor method is performed using the learning data. The unsupervised learning model M1 takes the features as input and outputs an anomaly level indicating the degree of deviation of the distribution of the features of normal pressure data. The anomaly level is preferably the Mahalanobis distance from the k-nearest neighbors, but may also be the Euclidean distance or Manhattan distance.

[0070] The anomaly probability calculation unit 917 executes machine learning (supervised learning) to obtain the supervised learning model M2 (FIG. 8: step S15). In addition, the anomaly probability calculation unit 917 stores the supervised learning model M2 (more specifically, the learned parameters) obtained by the machine learning in the storage unit 93 (FIG. 8: step S16).

[0071] In supervised learning, first, a plurality of normal pressure data feature values, a plurality of abnormal pressure data feature values, and a plurality of abnormal pressure data feature values ​​are prepared as input data. The supervised learning model M2 is based on, for example, a linear regression model. In supervised learning, optimal parameters for predicting target values ​​(labels indicating "normal", "abnormal A", and "abnormal B") from the input data are obtained. Specifically, in order to distinguish between the three states of normal, abnormal A, and abnormal B, the target value for normal is set to "0", the target value for abnormal A is set to "1", and the target value for abnormal B is set to "-1", and learning is advanced so that the model outputs the target values ​​of the input data.

[0072] It is also possible to use the rising period T1 as a feature quantity. However, when there are several patterns of the ejection waveform, such as when the target pressure Pt is different, the rising period T1 may vary for each pattern even in a normal state. For this reason, when there is variation in the rising period T1, it is preferable not to use it as a feature quantity.

[0073] <Prediction stage> Fig. 9 is a block diagram showing the control unit 9 for monitoring abnormalities in the pressure data, and Fig. 10 is a diagram showing the flow of a monitoring process executed by the control unit 9 for monitoring the discharge pressure.

[0074] In this monitoring process, first, the discharge pressure measurement unit 911 of the control unit 9 acquires pressure data (FIG. 10: step S21). Then, the feature calculation unit 913 calculates the feature of the acquired pressure data (FIG. 10: step S22).

[0075] The anomaly degree calculation unit 915 inputs the feature amount calculated by the feature amount calculation unit 913 to the unsupervised learning model M1 to calculate the anomaly degree (FIG. 10: step S23). Also, the anomaly probability calculation unit 917 inputs the feature amount calculated by the feature amount calculation unit 913 to the supervised learning model M2 to calculate the probability of each anomaly (FIG. 10: step S24).

[0076] FIG. 11 is a diagram conceptually illustrating how the degree of anomaly (Mahalanobis distance) is calculated. In the example shown in FIG. 11, a plurality of open circles ("○") indicate the positions (distribution) in the feature space of normal pressure data used in machine learning. Open triangles ("△") indicate the positions in the feature space of newly acquired pressure data. As shown in FIG. 11, the degree of anomaly calculation unit 915 calculates the Mahalanobis distance between the newly acquired pressure data (△) and the k-nearest (here, nearest) normal pressure data (○) as the degree of anomaly. The degree of anomaly is an index indicating the degree to which the newly acquired pressure data (△) deviates from the normal pressure data (○).

[0077] Fig. 12 is a diagram conceptually showing the output of the abnormality probability calculation unit 917. It is a diagram illustrating pressure waveforms of normal, abnormal A, and abnormal B. In the example shown in Fig. 12, a state in which the discharge pressure rises excessively higher than in the normal state and an overshoot occurs is determined as abnormality A. On the other hand, a state in which the discharge pressure does not rise sufficiently higher than in the normal state and the period until the discharge pressure reaches the steady pressure Pm becomes longer is determined as abnormality B.

[0078] The abnormality probability calculation unit 917 is configured to output a numerical value between -1 and +1 for newly acquired pressure data. As described above, in the machine learning, machine learning is performed so that normal is 0, the output value of abnormality A is +1, and the output of abnormality B is -1. Therefore, the closer the output value is to +1, the higher the probability of abnormality A, and the closer the output value is to -1, the higher the probability of abnormality B.

[0079] 9, the degree of abnormality calculated by the degree of abnormality calculation unit 915 and the abnormality probability (a value between -1 and 1) calculated by the abnormality probability calculation unit 917 are sent to an abnormality determination unit 919. The abnormality determination unit 919 compares the degree of abnormality and the abnormality probability with threshold values ​​to determine whether the newly obtained pressure data is normal or abnormal (FIG. 10: step S25).

[0080] Specifically, the abnormality determination unit 919 determines whether or not the degree of abnormality exceeds a predetermined threshold value Th1 (see FIG. 11). If the degree of abnormality exceeds the threshold value Th1, the abnormality determination unit 919 determines that the pressure data is abnormal.

[0081] Furthermore, the abnormality determination unit 919 determines whether the abnormality probability exceeds a predetermined threshold value. For example, as shown in FIG. 12, when determining whether the abnormality corresponds to abnormality A or abnormality B, threshold values ​​Th2 and Th3 are determined in advance. The threshold value Th2 is a numerical value greater than 0 and less than 1, and the threshold value Th3 is a numerical value greater than -1 and less than 0. When the abnormality probability exceeds the threshold value Th2 on the positive side (i.e., when the abnormality probability value is greater than the threshold value Th2), the abnormality determination unit 919 determines that the new pressure data corresponds to abnormality A. When the abnormality probability exceeds the threshold value Th3 on the negative side (i.e., when the abnormality probability falls below the threshold value Th3), the abnormality determination unit 919 determines that the new pressure data corresponds to abnormality B.

[0082] For example, in the example shown in Fig. 12, the abnormality probability of the first pressure data (data 1) does not exceed the thresholds Th2 and Th3, so the pressure data is determined to be normal. The abnormality probability of the second pressure data (data 2) exceeds the threshold Th2 on the positive side, so the pressure data is determined to be abnormal A. The abnormality probability of the third pressure data (data 3) exceeds the threshold Th3 on the negative side, so the pressure data is determined to be abnormal B.

[0083] Furthermore, the anomaly determination unit 919 may perform a final anomaly determination of the pressure data by combining the results of the anomaly determination using the degree of anomaly and the probability of anomaly. FIG. 13 is a diagram conceptually illustrating an example of a comprehensive anomaly determination using the degree of anomaly and the probability of anomaly. In FIG. 13, the horizontal axis indicates the probability of anomaly, and the vertical axis indicates the degree of anomaly. When both the degree of anomaly and the probability of anomaly do not exceed the threshold (i.e., when both are normal), the comprehensive determination of the pressure data is determined to be "normal" (corresponding to the third quadrant in FIG. 13). When both the degree of anomaly and the probability of anomaly exceed the threshold (i.e., when both are abnormal), the comprehensive determination of the new pressure data is determined to be "anomaly". Furthermore, when the probability of anomaly is high and the degree of anomaly is low, the comprehensive determination of the new pressure data is determined to be "anomaly close to normal distribution". Conversely, when the probability of anomaly is low and the degree of anomaly is high, the comprehensive determination of the new pressure data is determined to be "unknown anomaly", that is, an anomaly that is not assumed during learning.

[0084] The abnormality determination unit 919 outputs the determination result obtained in step S25 to the outside. As an example, the abnormality determination unit 919 causes the determination result to be displayed on the display 95 (step S26). The abnormality determination unit 919 may print out the determination result using a printer. The abnormality determination unit 919 may also output the determination result by turning on a lamp, outputting an alarm sound from a speaker, or the like.

[0085] As described above, the control unit 9 (discharge pressure monitoring device) includes the discharge pressure measurement unit 911 as a pressure data acquisition unit, the feature amount calculation unit 913, the abnormality degree calculation unit 915, the abnormality probability calculation unit 917, and the abnormality determination unit 919. The discharge pressure measurement unit 911 acquires pressure data indicating a time change in pressure inside the nozzle 71 that discharges the treatment liquid. The feature amount calculation unit 913 calculates a feature amount of the pressure data. The abnormality degree calculation unit 915 receives the feature amount as an input, and calculates an abnormality degree for the feature amount calculated by the feature amount calculation unit 913 using an unsupervised learning model M1 that outputs an abnormality degree indicating a degree of deviation from a distribution of normal feature amounts. The abnormality probability calculation unit 917 receives the feature amount as an input, and calculates an abnormality probability for the feature amount calculated by the feature amount calculation unit 913 using a supervised learning model M2 that outputs an abnormality probability indicating a probability that the feature amount corresponds to a predetermined abnormality. The abnormality determination unit 919 determines whether the pressure data is abnormal by using the degree of abnormality calculated by the abnormality degree calculation unit 915 and the abnormality probability calculated by the abnormality probability calculation unit 917 .

[0086] The control unit 9 also executes a pressure monitoring method, which includes the steps of: a) acquiring pressure data indicating a time change in pressure inside the nozzle 71 that ejects the processing liquid; b) calculating a feature of the pressure data; c) calculating an anomaly degree of the feature calculated in step b) using an unsupervised learning model M1 that receives the feature and outputs an anomaly degree indicating the degree of deviation from a distribution of normal feature values; d) calculating an anomaly probability for the feature calculated in step b) using a supervised learning model M2 that receives the feature and outputs an anomaly probability indicating the probability that the feature corresponds to a specific anomaly; and e) determining an anomaly of the pressure data using the anomaly degree calculated in step c) and the anomaly probability calculated in step d).

[0087] According to this configuration, when a previously anticipated anomaly is determined, the anomaly can be determined by calculating the anomaly probability. Even when an unknown anomaly that has not been previously anticipated occurs, the unknown anomaly can be appropriately detected by calculating the anomaly degree.

[0088] Moreover, the unsupervised learning model M1 is a model obtained by unsupervised learning using the features of only a plurality of normal pressure data as input data.

[0089] According to this configuration, only the distribution of normal feature amounts can be learned, so that abnormality detection can be performed appropriately.

[0090] The unsupervised learning model M1 outputs the Mahalanobis distance as the degree of anomaly.

[0091] According to this configuration, the degree of abnormality is calculated taking into consideration the correlation between each dimension of the feature amount indicated by the pressure data, so that characteristic patterns and abnormal behavior of the pressure data can be captured more accurately.

[0092] The supervised learning model M2 outputs the probability of multiple types of specific anomalies (e.g., anomalies A and B).

[0093] According to this configuration, since it is possible to calculate a plurality of types of probabilities, it is possible to determine whether the pressure data corresponds to any one of the plurality of types of abnormality.

[0094] <Other embodiments> In the above embodiment, the anomaly degree calculation unit 915 calculates the anomaly degree using the unsupervised learning model M1. However, it is not essential to calculate the anomaly degree using the unsupervised learning model M1. For example, the anomaly degree calculation unit 915 may calculate the difference between the normal feature amount and the feature amount of the newly obtained pressure data using a predetermined calculation formula. The difference corresponds to the anomaly degree indicating the degree of deviation of the feature amount of the newly obtained pressure data from the normal feature amount. In addition, the normal feature amount to be compared may be defined from the feature amounts of multiple normal pressure data, or may be the feature amount of a single normal pressure data.

[0095] The abnormality determination unit 919 compares the degree of abnormality (difference) calculated by the abnormality degree calculation unit 915 with a preset threshold value. Then, when the degree of abnormality exceeds the threshold value, the abnormality determination unit 919 determines that the newly acquired pressure data is abnormal.

[0096] For example, Japanese Patent Application Publication No. 2012-098133 describes that when detecting an abnormality in an inverter device based on the impulse component of a current waveform, the similarity between the feature amount of the impulse component in a normal state and the feature amount of the impulse component to be judged is calculated. Also, Japanese Patent Application Publication No. 2006-026584 describes that when detecting an ink ejection abnormality based on the voltage applied to an actuator that ejects ink, the difference between the voltage in a normal state and the voltage to be judged is calculated. These similarities and differences may be used as the abnormality degree.

[0097] Although the present invention has been described in detail, the above description is merely illustrative in all respects, and the present invention is not limited thereto. It is understood that countless variations not illustrated can be envisioned without departing from the scope of the present invention. The configurations described in the above embodiments and variations can be appropriately combined or omitted as long as they are not mutually contradictory. [Explanation of symbols]

[0098] 1 Coating device 51 Chuck mechanism (substrate holding part) 52 Driving control mechanism 7 Coating mechanism 71 Nozzle 86 Pressure Sensor 9 Control unit (discharge pressure monitoring device) 91 Arithmetic section 911 Discharge pressure measurement unit (pressure data acquisition unit) 913 Feature Calculation Unit 915 Abnormality calculation unit 917 Anomaly Probability Calculation Unit 919 Abnormality determination unit 931 Computer Programs M1 Unsupervised Learning Model M2 Supervised Learning Model

Claims

1. A discharge pressure monitoring device, comprising: a pressure data acquisition unit that acquires pressure data indicating a time change in pressure within a nozzle that ejects the treatment liquid; A feature amount calculation unit that calculates a feature amount of the pressure data; an anomaly degree calculation unit that calculates the anomaly degree for the feature calculated by the feature calculation unit using an unsupervised learning model that receives the feature and outputs an anomaly degree indicating a degree of deviation from a normal distribution of the feature; an anomaly probability calculation unit that uses the feature amount as an input and calculates the anomaly probability for the feature amount calculated by the feature amount calculation unit using a supervised learning model that outputs an anomaly probability indicating a probability that the feature amount corresponds to a specific anomaly; an abnormality determination unit that determines an abnormality in the pressure data by using the degree of abnormality calculated by the degree of abnormality calculation unit and the abnormality probability calculated by the abnormality probability calculation unit; A discharge pressure monitoring device comprising:

2. 2. The discharge pressure monitoring device according to claim 1, The unsupervised learning model is a model obtained by unsupervised learning using, as input data, feature quantities of only a plurality of normal pressure data, the discharge pressure monitoring device.

3. The discharge pressure monitoring device according to claim 1 or 2, The unsupervised learning model outputs a Mahalanobis distance as the degree of abnormality.

4. The discharge pressure monitoring device according to claim 1 or 2, The supervised learning model outputs probabilities of occurrence of a plurality of specific anomalies.

5. A discharge pressure monitoring device, comprising: a pressure data acquisition unit that acquires pressure data indicating a time change in pressure within a nozzle that ejects the treatment liquid; A feature amount calculation unit that calculates a feature amount of the pressure data; an anomaly degree calculation unit that calculates an anomaly degree indicating a degree of deviation from a normal feature amount for the feature amount calculated by the feature amount calculation unit; an anomaly probability calculation unit that uses the feature amount as an input and calculates the anomaly probability for the feature amount calculated by the feature amount calculation unit using a supervised learning model that outputs an anomaly probability indicating a probability that the feature amount corresponds to a specific anomaly; an abnormality determination unit that determines an abnormality in the pressure data by using the degree of abnormality calculated by the degree of abnormality calculation unit and the abnormality probability calculated by the abnormality probability calculation unit; A discharge pressure monitoring device comprising:

6. 1. A method for monitoring a discharge pressure, comprising: a) acquiring pressure data indicating a change over time in pressure within a nozzle that ejects a treatment liquid; b) calculating a feature amount of the pressure data; c) calculating the degree of anomaly for the feature calculated in step b) using an unsupervised learning model that receives the feature and outputs an anomaly degree indicating the degree of deviation from a normal distribution of the feature; d) calculating the anomaly probability for the feature calculated in step b) using a supervised learning model that receives the feature and outputs an anomaly probability indicating the probability that the feature corresponds to a specific anomaly; e) determining whether the pressure data is abnormal by using the degree of abnormality calculated in step c) and the probability of abnormality calculated in step d); A method for monitoring discharge pressure, comprising:

7. 1. A method for monitoring a discharge pressure, comprising: A) acquiring pressure data indicating a change over time in pressure within a nozzle that ejects a treatment liquid; B) calculating a feature amount of the pressure data; C) calculating an anomaly level indicating a degree of deviation from a normal feature amount for the feature amount calculated in the step B); D) calculating the anomaly probability for the feature calculated in step B) using a supervised learning model that receives the feature and outputs an anomaly probability indicating a probability that the feature corresponds to a specific anomaly; E) determining whether the pressure data is abnormal by using the degree of abnormality calculated in the step C) and the probability of abnormality calculated in the step D); A method for monitoring discharge pressure, comprising:

8. A computer program executable by a computer, A computer program causing the computer to execute the discharge pressure monitoring method according to claim 6 or 7.

9. A coating device comprising: A substrate holder for holding a substrate; a nozzle that ejects a processing liquid toward the substrate held by the substrate holder; a pressure sensor for measuring a pressure in the nozzle; A discharge pressure monitoring device according to claim 1, claim 2 or claim 5; The coating device comprises:

Images