Substrate processing apparatus, abnormality detection method, and computer-readable medium storing an abnormality detection program

Abstract

A substrate processing apparatus, an abnormality detection method, and a computer-readable medium storing an abnormality detection program are provided. In the substrate processing apparatus, processing of a substrate using a processing liquid is performed. A first operation member and a second operation member are used for the processing of the substrate. A first operation value of the first operation member and a second operation value of the second operation member are acquired by an operation value acquisition section. Whether an abnormality has occurred is judged by an abnormality judgment section based on a correlation between the first operation value and the second operation value acquired by the operation value acquisition section.

Description

Technical Field

[0001] This invention relates to a substrate processing apparatus for processing substrates, an anomaly detection method, and a non-transitory computer readable medium storing an anomaly detection program. Background Technology

[0002] Substrate processing apparatuses are used to perform various processes such as film deposition, development, and cleaning on substrates including semiconductor substrates, glass substrates for liquid crystal display devices, glass substrates for photomasks, and glass substrates for optical discs. If a substrate processing apparatus is used for an extended period, abnormalities such as damage or deterioration of the apparatus's operating components may occur.

[0003] Japanese Patent Application Publication No. 2018-77764 discloses an anomaly detection device capable of predicting anomalies in a semiconductor manufacturing apparatus. This anomaly detection device collects status information indicating the state of various parts of the semiconductor manufacturing apparatus at predetermined intervals. The status information includes, for example, temperature, pressure, gas flow rate, or power consumption of each part of the semiconductor manufacturing apparatus. The collected status information is stored as a log in predetermined units.

[0004] Based on the saved logs, monitoring strips are generated to monitor the status of various parts of the semiconductor manufacturing apparatus. These monitoring strips are waveforms used to determine whether the collected status information is normal; for example, they are generated by interpolation based on upper and lower limits set for each specified period. Based on the status information and the monitoring strips, it is determined whether the status of each part of the semiconductor manufacturing apparatus is abnormal. Summary of the Invention

[0005] As described in Japanese Patent Application Publication No. 2018-77764, the anomaly detection device aims to detect anomalies before a malfunction occurs in the substrate processing device.

[0006] The purpose of this invention is to provide a substrate processing apparatus, an anomaly detection method, and a computer-readable medium storing an anomaly detection program that can detect anomalies at an early stage.

[0007] (1) A substrate processing apparatus according to one aspect of the present invention is a substrate processing apparatus comprising a first action member and a second action member used in processing a substrate using a processing liquid, comprising: an action value acquisition unit that acquires a first action value of the first action member and a second action value of the second action member; and an anomaly determination unit that determines whether an anomaly has occurred based on the correlation between the first action value and the second action value acquired by the action value acquisition unit.

[0008] In this substrate processing apparatus, substrates are processed using a processing solution. Furthermore, an anomaly is determined based on the correlation between the first operating value of the first operating member and the second operating value of the second operating member. Therefore, anomalies in the substrate processing apparatus can be detected at an early stage, before a malfunction occurs.

[0009] (2) Alternatively, the anomaly determination unit may determine whether an anomaly has occurred based on the proportion of data points that exceed an allowable range determined based on the correlation between the first and second action values ​​among the data points determined by the action value acquisition unit based on the set of the first and second action values. In this case, anomalies in the board processing device can be detected with simple processing.

[0010] (3) Alternatively, the allowable range may be defined as a predetermined proportion of data points that are determined based on a set of first and second action values ​​pre-acquired by the action value acquisition unit before processing the substrate. In this case, the allowable range used in the anomaly judgment can be easily determined.

[0011] (4) Alternatively, the anomaly detection unit may determine whether an anomaly has occurred after processing a specified number of substrates or after a specified time has elapsed. With this structure, it is easier to detect anomalies in the substrate processing apparatus at an earlier stage.

[0012] (5) Alternatively, the first actuating member includes a first regulating valve that adjusts the flow rate of the processing liquid supplied to the substrate based on a first pulse value, and the second actuating member includes a second regulating valve that adjusts the flow rate of the processing liquid supplied to the substrate based on a second pulse value, wherein the first actuating value is a first pulse value provided to the first regulating valve, and the second actuating value is a second pulse value provided to the second regulating valve.

[0013] In this case, anomalies in the substrate processing device can be detected early based on the correlation between the first pulse value provided to the first adjustment valve and the second pulse value provided to the second adjustment valve.

[0014] (6) Alternatively, the first actuating component includes an adjustment valve that adjusts the flow rate of the processing fluid supplied to the substrate based on a pulse value, and the second actuating component includes a pressure gauge that detects the pressure of the processing fluid directed to the adjustment valve. The first actuating value is the pulse value provided to the adjustment valve, and the second actuating value is the pressure detected by the pressure gauge.

[0015] In this case, abnormalities in the substrate processing device can be detected early based on the correlation between the pulse value provided by the regulating valve and the pressure detected by the pressure gauge.

[0016] (7) Alternatively, the first actuating component may include a first flow meter and a second flow meter that respectively detect the flow rates of the first treatment liquid and the second treatment liquid, and the second actuating component may include a concentration meter that detects the concentration of the mixed treatment liquid generated by mixing the first treatment liquid and the second treatment liquid. The first actuating value is a theoretical value of the concentration of the mixed treatment liquid calculated based on the ratio of the flow rate of the first treatment liquid to the flow rate of the second treatment liquid detected by the first flow meter and the second flow meter respectively, and the second actuating value is a measured value of the concentration of the mixed treatment liquid detected by the concentration meter.

[0017] In this case, anomalies in the substrate processing apparatus can be detected early based on the correlation between the theoretical value of the concentration of the mixed processing liquid calculated according to the ratio of the flow rate of the first processing liquid to the flow rate of the second processing liquid and the measured value of the concentration of the mixed processing liquid detected by the concentration meter.

[0018] (8) Alternatively, the substrate processing apparatus may further include a reservoir into which a mixed processing liquid formed by mixing the first processing liquid and the second processing liquid flows. The first operating member includes a first flow meter and a second flow meter for detecting the flow rates of the first processing liquid and the second processing liquid, respectively. The second operating member includes a concentration meter for detecting the concentration of the mixed processing liquid flowing out of the reservoir. The mixed processing liquid flowing into the reservoir reaches the concentration meter at a first time. The first operating value is a theoretical value of the concentration of the mixed processing liquid calculated based on the cumulative value of the flow rate of the first processing liquid detected by the first flow meter and the cumulative value of the flow rate of the second processing liquid detected by the first flow meter and the second flow meter, respectively, between the first time point and the second time point after the second time point. The second operating value is a statistical value of the concentration of the mixed processing liquid detected by the concentration meter between the third time point after the first time point and the fourth time point after the first time point.

[0019] In this case, the correlation between the theoretical value of the concentration of the mixed processing liquid, calculated from the cumulative flow rates of the first and second processing liquids, and the statistical value of the concentration of the mixed processing liquid is used to determine whether an abnormality has occurred in the substrate processing apparatus. Here, since there is a strong correlation between the theoretical value and the statistical value of the concentration of the mixed processing liquid, abnormalities in the substrate processing apparatus can be detected more reliably.

[0020] (9) Alternatively, the first actuating component may include a first flow meter and a second flow meter for detecting the flow rates of the first and second processed liquids, respectively, and the second actuating component may include a concentration meter for detecting the concentration of the mixed processed liquid generated by mixing the first and second processed liquids, thereby preparing a concentration prediction model. This concentration prediction model learns the flow rates of the first and second processed liquids and the concentration of the mixed processed liquid between the first time point and the second time point after the first time point, and the concentration of the mixed processed liquid between the first time point and the second time point after the second time point. The relationship between the concentrations of the mixed treatment liquids is as follows: the first action value is the predicted concentration of the mixed treatment liquid at the sixth time point (from the fifth time point through the second time point) obtained by applying the flow rate of the first treatment liquid detected by the first flow meter, the flow rate of the second treatment liquid detected by the second flow meter, and the concentration of the mixed treatment liquid detected by the concentration meter to the concentration prediction model between the fourth time point and the fifth time point (from the fourth time point through the first time point). The second action value is the measured concentration of the mixed treatment liquid detected by the concentration meter at the sixth time point.

[0021] In this case, based on the correlation between the predicted concentration of the mixed processing liquid obtained from the concentration prediction model and the measured concentration of the mixed processing liquid detected by the concentration meter, a single anomaly occurring in the substrate processing device can be detected within a relatively short period of time.

[0022] (10) Alternatively, the substrate processing apparatus may further include: a retention section for retaining a mixed processing liquid formed by mixing the first processing liquid and the second processing liquid; a substrate processing section for processing the substrate using the mixed processing liquid; a first flow path section for mixing the first processing liquid and the second processing liquid to generate a mixed processing liquid and guiding it to the retention section; a second flow path section for guiding the mixed processing liquid retained in the retention section to the substrate processing section; and a third flow path section for guiding unused mixed processing liquid in the substrate processing section to the retention section. The first actuating member includes a first flow meter and a second flow meter for detecting the flow rates of the first processing liquid and the second processing liquid, respectively. The second actuating member includes a concentration meter for detecting the concentration of the mixed processing liquid flowing out of the retention section. The mixed treatment liquid flowing into the storage section through the first flow path reaches the concentration meter at the first time. The first action value is the theoretical value of the concentration of the mixed treatment liquid stored in the storage section at each time point, calculated based on the volume of the mixed treatment liquid flowing into the storage section through the first flow path, the volume of the mixed treatment liquid flowing out of the storage section through the second flow path, the volume of the mixed treatment liquid flowing into the storage section through the third flow path, and the flow rates of the first and second treatment liquids detected by the first and second flow meters, respectively. The second action value is the measured value of the concentration of the mixed treatment liquid detected by the concentration meter at the time point after the first time point from the time point when the mixed treatment liquid flows into the storage section through the first flow path.

[0023] When the concentration of the mixed processing liquid supplied to the substrate processing unit changes, the concentration of the mixed processing liquid stored in the storage unit changes because unused mixed processing liquid in the substrate processing unit returns to the storage unit. Even in this case, according to the above structure, whether an abnormality has occurred in the substrate processing apparatus is determined based on the correlation between the theoretical value of the concentration of the mixed processing liquid stored in the storage unit calculated relative to the volume of the mixed processing liquid flowing into or out of the storage unit and the flow rates of the first and second processing liquids, and the measured value of the concentration of the mixed processing liquid detected by the concentration meter. In this case, the correlation between the theoretical value of the concentration of the mixed processing liquid and the statistical value of the concentration of the mixed processing liquid can be enhanced. As a result, abnormalities in the substrate processing apparatus can be detected more reliably.

[0024] (11) Alternatively, the first actuating member may include a first chuck pin that transitions between a first closed state holding the substrate and a first open state not holding the substrate, and the second actuating member may include a second chuck pin that transitions between a second closed state holding the substrate and a second open state not holding the substrate, the first actuating value being the transition time between the first closed state and the first open state in the first chuck pin, and the second actuating value being the transition time between the second closed state and the second open state in the second chuck pin.

[0025] In this case, anomalies in the substrate processing apparatus can be detected early based on the correlation between the transition time of the first chuck pin and the transition time of the second chuck pin.

[0026] (12) Alternatively, the first actuating member includes a chuck pin that changes between a first closed state holding the substrate and a first open state not holding the substrate, and the second actuating member includes a chuck drive unit that changes between a second closed state that changes the chuck pin to a first closed state and a second open state that changes the chuck pin to a first open state. The first actuating value is the transition time between the first closed state and the first open state in the chuck pin, and the second actuating value is the transition time between the second closed state and the second open state in the chuck drive unit.

[0027] In this case, abnormalities in the substrate processing apparatus can be detected early based on the correlation between the chuck pin transition time and the chuck drive unit transition time.

[0028] (13) Alternatively, the first action member may be a first processing tool that transitions between a first standby position of a non-processing substrate and a first processing position of a processing substrate, and the second action member may be a second processing tool that transitions between a second standby position of a non-processing substrate and a second processing position of a processing substrate, wherein the first action value is the transition time between the first standby position and the first processing position in the first processing tool, and the second action value is the transition time between the second standby position and the second processing position in the second processing tool.

[0029] In this case, anomalies in the substrate processing apparatus can be detected early based on the correlation between the transition time of the first processing tool and the transition time of the second processing tool.

[0030] (14) Alternatively, the substrate processing apparatus may also include an action value selection unit, which selects a second action value that meets a predetermined reference from the second action values ​​obtained by the action value acquisition unit, and selects a first action value that corresponds to the selected second action value from the first action values ​​obtained by the action value acquisition unit, and the anomaly judgment unit determines whether an anomaly has occurred based on the change in the time sequence of the first action value selected by the action value selection unit.

[0031] In this case, an anomaly is determined based on the change in the time series of the first action values ​​that meet the specified criteria among the acquired first action values. This allows for accurate detection of anomalies in the substrate processing apparatus.

[0032] (15) Alternatively, the first actuating member may include a torque sensor for detecting the torque of the rotary drive unit, and the second actuating member may include a temperature sensor for detecting the temperature of the rotary drive unit. The first actuating value is the torque of the rotary drive unit detected by the torque sensor, and the second actuating value is the temperature of the rotary drive unit detected by the temperature sensor.

[0033] In this case, based on the temperature of the rotary drive unit detected by the temperature sensor, the torque of the rotary drive unit that meets the specified reference can be easily selected from the torques of the rotary drive unit detected by the torque sensor. Therefore, malfunctions in the substrate processing apparatus can be easily detected.

[0034] (16) Alternatively, the first operating value is the torque of the rotary drive unit when it rotates at a predetermined constant rotational speed, which is part of the torque detected by the torque sensor, and the second operating value is the temperature of the rotary drive unit when it rotates at a constant rotational speed, which is part of the temperature detected by the temperature sensor. In this case, abnormalities in the substrate processing apparatus can be detected more accurately.

[0035] (17) An anomaly detection method according to another aspect of the present invention is an anomaly detection method for detecting anomalies in a substrate processing apparatus including a first action member and a second action member used in the processing of a substrate using a processing liquid, comprising: acquiring a first action value of the first action member and a second action value of the second action member that are related to each other; and determining whether an anomaly has occurred based on the acquired first action value and second action value.

[0036] According to this anomaly detection method, during the processing of a substrate using a processing liquid, an anomaly is determined based on the correlation between the first action value of the first actuating member and the second action value of the second actuating member. Therefore, anomalies in the substrate processing apparatus can be detected at an early stage, before a malfunction occurs.

[0037] (18) According to another aspect of the present invention, a non-transitory computer readable medium storing an anomaly detection program is a computer readable medium storing an anomaly detection program executed by a processing device and detecting anomalies of a first actuating member and a second actuating member used in a process involving a substrate using a processing liquid. The anomaly detection program causes the processing device to perform the following processes: acquiring a first actuation value of the first actuating member and a second actuation value of the second actuating member that are related to each other; and determining whether an anomaly has occurred based on the acquired first actuation value and second actuation value.

[0038] According to this anomaly detection procedure, during the processing of a substrate using a processing solution, an anomaly is determined based on the correlation between the first action value of the first actuating member and the second action value of the second actuating member. Therefore, anomalies in the substrate processing apparatus can be detected at an early stage, before a malfunction occurs. Attached Figure Description

[0039] Figure 1 This is a block diagram illustrating the schematic structure of a substrate processing apparatus according to one embodiment of the present invention.

[0040] Figure 2 This is a diagram showing the structure of the processing liquid supply unit in the first embodiment.

[0041] Figure 3 This is a functional block diagram representing the structure of the control unit.

[0042] Figure 4 It is a graph representing a correlation diagram.

[0043] Figure 5 It means based on Figure 3 The flowchart of the anomaly detection and processing performed by the control department.

[0044] Figure 6 It is a graph representing a correlation diagram.

[0045] Figure 7 This is a diagram showing the structure of the liquid generation unit in the second embodiment.

[0046] Figure 8 It is a graph representing a correlation diagram.

[0047] Figure 9This is a diagram used to illustrate the action values ​​in the third embodiment.

[0048] Figure 10 This is a diagram illustrating an example of the construction sequence of a concentration prediction model.

[0049] Figure 11 yes Figure 7 A magnified view of a portion of the drug solution generation area.

[0050] Figure 12 This is a side view showing the structure of the rotation holding part in the sixth embodiment.

[0051] Figure 13 This is a functional block diagram representing the structure of the rotating retainer.

[0052] Figure 14 This is a diagram used to illustrate the timing of the chuck pin transition when the indicator is closed.

[0053] Figure 15 This is a diagram used to illustrate the timing of the chuck pin transition when the indicator is open.

[0054] Figure 16 This is a diagram used to illustrate the transition time of the chuck pin and the chuck drive unit.

[0055] Figure 17 This is a diagram showing the structure of the processing fluid supply section.

[0056] Figure 18 This is a graph used to illustrate the transition time of cleaning tools.

[0057] Figure 19 This is a side view showing the structure of the rotation holding part in the 8th embodiment.

[0058] Figure 20 This is a functional block diagram representing the structure of the control unit.

[0059] Figure 21 This is a diagram showing the changes in the rotational speed and torque of the rotary drive unit during substrate processing.

[0060] Figure 22 It is a graph showing the obtained torque and temperature.

[0061] Figure 23 This is a diagram illustrating an example of an anomaly detection method.

[0062] Figure 24 It means based on Figure 20 The flowchart of the anomaly detection and processing performed by the control department. Detailed Implementation

[0063] <1> Structure of the substrate processing device

[0064] Hereinafter, the substrate processing apparatus, anomaly detection method, and a computer-readable medium storing anomaly detection program according to embodiments of the present invention will be described with reference to the accompanying drawings. In the following description, substrate refers to semiconductor substrate, substrate for FPD (Flat Panel Display) of liquid crystal display device or organic EL (Electro Luminescence) display device, substrate for optical disc, substrate for magnetic disk, substrate for optical disk, substrate for photomask, ceramic substrate, or substrate for solar cell, etc.

[0065] Figure 1 This is a block diagram illustrating a schematic structure of a substrate processing apparatus according to one embodiment of the present invention. Figure 1 As shown, the substrate processing apparatus 1 includes a substrate processing unit 100 and a control unit 200. The substrate processing unit 100 includes one or more processing units 110 that perform various processing on the substrate using a processing liquid.

[0066] Specifically, each processing unit 110 may be a cleaning unit that supplies cleaning solution to the substrate, a coating unit that supplies coating solution to the substrate, or a developing unit that supplies developing solution to the substrate. Each processing unit 110 is provided with a rotation holding part that holds and rotates the substrate, or a processing solution supply part that supplies processing solution to the substrate, etc.

[0067] In addition, when the processing unit 110 is an etching unit that supplies etching solution to the substrate, a solution generation unit 300 is sometimes provided in the substrate processing unit 100. The solution generation unit 300 generates a diluted solution obtained by diluting the original solution of the etching solution as an etching solution and supplies it to each substrate processing unit 100.

[0068] The control unit 200 consists of a CPU (Central Processing Unit) and a memory or microcomputer. The CPU of the control unit 200 controls the operation of various actuators in the board processing unit 100 or the liquid generation unit 300. The memory of the control unit 200 stores an anomaly detection program for detecting anomalies in the actuators of the board processing unit 1.

[0069] Furthermore, during the execution of substrate processing, the CPU of the control unit 200 collects prescribed information from various operating components in the substrate processing unit 100 or the liquid generation unit 300, and detects abnormalities in the substrate processing apparatus 1 based on the processing results of the collected information. Hereinafter, an example of detecting abnormalities in the substrate processing apparatus 1 associated with the various operating components of the substrate processing unit 100 or the liquid generation unit 300 will be described.

[0070] <2> First Embodiment

[0071] (1) Processing fluid supply unit

[0072] In the first embodiment, as an operating member of the substrate processing unit 100, an abnormality in the processing liquid supply unit included in each processing unit 110 is detected. Figure 2 This is a diagram showing the structure of the processing fluid supply unit in the first embodiment. (See diagram below.) Figure 2 As shown, multiple processing fluid supply units 400 are connected to a circulation piping 401 for circulating processing fluid. Additionally, a pressure gauge 402 for detecting the pressure of the processing fluid is inserted into the circulation piping 401.

[0073] The treatment fluid supply unit 400 includes an upper surface supply unit 410 and a lower surface supply unit 420. The upper surface supply unit 410 includes a piping 411, a flow meter 412, an on / off valve 413, an adjusting valve 414, and a nozzle 415. The piping 411 is connected to a circulation piping 401. Thus, the treatment fluid from the circulation piping 401 flows in the piping 411. In the piping 411, the flow meter 412, the on / off valve 413, the adjusting valve 414, and the nozzle 415 are sequentially inserted from upstream to downstream.

[0074] Flow meter 412 detects the flow rate of the processing fluid flowing in pipe 411 and provides the detection result to control unit 200. On / off valve 413 opens and closes the flow path of pipe 411. Adjustment valve 414 includes, for example, a motor needle valve, which adjusts the flow rate of the processing fluid flowing in pipe 411 based on pulse control performed by control unit 200. Nozzle 415 is disposed above substrate W and supplies processing fluid to the upper surface of substrate W.

[0075] The lower surface supply section 420 includes a piping 421, a flow meter 422, an on / off valve 423, an adjusting valve 424, and a nozzle 425. The upstream end of the piping 421 is connected to the piping 411 upstream of the flow meter 412. Thus, the treatment fluid from the circulation piping 401 flows in the piping 421. The flow meter 422, the on / off valve 423, the adjusting valve 424, and the nozzle 425 are sequentially inserted into the piping 421 from upstream to downstream.

[0076] Flow meter 422 detects the flow rate of the processing fluid flowing in pipe 421 and provides the detection result to control unit 200. On / off valve 423 opens and closes the flow path of pipe 421. Adjustment valve 424 includes, for example, a motorized needle valve, which adjusts the flow rate of the processing fluid flowing in pipe 421 based on pulse control performed by control unit 200. Nozzle 425 is disposed below substrate W and supplies processing fluid to the lower surface of substrate W.

[0077] In this example, nozzles 415 and 425 supply the same flow rate of treatment liquid to the upper and lower surfaces of the substrate W at approximately the same frequency. This processes the upper and lower surfaces of the substrate W. The substrate W can also be processed, for example, by means of the process described later. Figure 19 The rotating holding part 700 is processed in a state of holding and rotating.

[0078] (2) Control Department

[0079] Figure 3 This is a functional block diagram showing the structure of the control unit 200. For example... Figure 3 As shown, the control unit 200, as a functional unit, includes an action value acquisition unit 210, an exception detection unit 220, and a notification unit 230. The CPU of the control unit 200 implements the functional units of the control unit 200 by executing an exception detection program stored in memory. Some or all of the functional units of the control unit 200 may also be implemented by hardware such as circuits.

[0080] Based on the detection results provided by the flow meter 412, the action value acquisition unit 210 adjusts the flow rate of the processing liquid flowing in the flow path of the upper surface supply unit 410 to a predetermined value by providing a predetermined number of pulses to the adjustment valve 414 to perform pulse control. Thus, the action value acquisition unit 210 acquires the pulse value provided to the adjustment valve 414 as the first action value.

[0081] Similarly, based on the detection results provided by the flow meter 422, the action value acquisition unit 210 adjusts the flow rate of the processing liquid flowing in the flow path of the lower surface supply unit 420 to a predetermined value by providing a predetermined number of pulses to the adjustment valve 424 to perform pulse control of the adjustment valve 424. Thus, the action value acquisition unit 210 acquires the pulse value provided to the adjustment valve 424 as the second action value.

[0082] Furthermore, the flow rate of the processing fluid is sometimes unstable immediately after the start and end of the supply. Therefore, pulse values ​​may not be acquired immediately after the start and end of the supply. In this case, pulse values ​​are acquired during the period from a predetermined point after the start of the supply to a predetermined point before the end of the supply, when the flow rate of the processing fluid is stable, to serve as the first and second action values.

[0083] Each time the anomaly detection unit 220 performs pulse control on the regulating valves 414 and 424, it plots the data points determined based on the combination of the pulse values ​​for the regulating valve 414 and the pulse values ​​for the regulating valve 424 onto the correlation graph. Figure 4 This is a graph representing a correlation diagram. For example... Figure 4 As shown, the correlation diagram is a two-dimensional graph with the first and second action values ​​set as axes, respectively. Specifically, Figure 4The horizontal axis of the related graph represents the pulse value for regulating valve 414, and the vertical axis represents the pulse value for regulating valve 424.

[0084] In the correlation diagram, an allowable range R is predetermined based on the correlation between the pulse values ​​for adjustment valve 414 and adjustment valve 424. In this example, the allowable range R has an elliptical shape and is defined as including a predetermined proportion (e.g., 95%) of data points determined based on the combination of pulse values ​​for adjustment valve 414 and adjustment valve 424 obtained before substrate processing. The allowable range R can also be determined based on a regression curve obtained from the distribution of data points. Preferably, the allowable range R is determined based on data points obtained during the installation of the processing fluid supply unit 400, before the processing fluid supply unit 400 deteriorates over the years.

[0085] The anomaly detection unit 220 determines whether an anomaly has occurred based on the proportion of data points exceeding the allowable range R among the drawn data points at a predetermined time interval. Anomaly detection can be performed after processing a predetermined number of substrates W, or after each predetermined time interval. In this case, anomalies in the substrate processing apparatus 1 can be easily detected at an early stage.

[0086] In this example, an anomaly is determined to have occurred when the proportion of data points exceeding the allowable range R in the plotted data points exceeds a specified threshold. In this case, it is presumed that an actuating component associated with regulating valve 414 or regulating valve 424 has malfunctioned. Actuating components associated with regulating valve 414 or regulating valve 424 include valve needles, diaphragms, bearings, or motor coils, etc.

[0087] As another example of anomaly detection, the allowable range can be updated at the designated time of anomaly detection, in a manner that includes a specified proportion of the plotted data points. An anomaly is then determined based on the rate of change of the updated allowable range relative to a predetermined allowable range R. The rate of change of the allowable range can include the rate of change of the major axis of the allowable range, the rate of change of the minor axis of the allowable range, or the rate of change of the area of ​​the allowable range.

[0088] When the anomaly determination unit 220 determines that an anomaly has occurred, the notification unit 230 notifies the user of the situation. As an example of notification by the notification unit 230, if the board processing apparatus 1 includes a display device, it can display text indicating that an anomaly has occurred. If the board processing apparatus 1 includes a sound output device, it can output a sound indicating the same information, or it can output a warning sound such as a buzzer. If the board processing apparatus 1 includes an indicator light such as a lamp, it can turn the indicator light on, off, or flash in a manner corresponding to the warning content.

[0089] (3) Anomaly Detection and Handling

[0090] Figure 5 It means based on Figure 3 The flowchart of the anomaly detection and processing performed by the control unit 200. Figure 5 The anomaly detection processing is performed by executing an anomaly detection program stored in memory using the CPU of the control unit 200. The anomaly detection processing is performed in parallel with the board processing. Hereinafter, using... Figure 3 Control unit 200 and Figure 5 The flowchart illustrates the anomaly detection and handling process.

[0091] First, the action value acquisition unit 210 acquires the pulse value for the adjustment valve 414 (step S1). Then, the action value acquisition unit 210 acquires the pulse value for the adjustment valve 424 (step S2). Step S1 is performed during the processing of the upper surface of the substrate W, and step S2 is performed during the processing of the lower surface of the substrate W. The execution order of steps S1 and S2 is determined according to the substrate processing recipe. Therefore, steps S1 and S2 can be performed either first or simultaneously.

[0092] Next, the anomaly detection unit 220 plots the data points determined based on the pulse values ​​obtained in steps S1 and S2 onto a correlation diagram (step S3). Then, the anomaly detection unit 220 determines whether the current time point is an anomaly detection timing point (step S4). This can be determined after processing a predetermined number of substrates W, or after a predetermined time interval. If the current time point is not an anomaly detection timing point, the anomaly detection unit 220 returns to step S1. Steps S1 to S4 are repeated until the current time point becomes an anomaly detection timing point.

[0093] When the current time point is the anomaly detection timing, the anomaly detection unit 220 evaluates the proportion of data points in the data points depicted in the correlation graph that exceed the allowable range R (step S5). Then, the anomaly detection unit 220 determines whether the proportion evaluated in step S5 exceeds a predetermined threshold (step S6). If the proportion is below the threshold, the anomaly detection unit 220 does not determine that an anomaly has occurred and returns to step S1. If the proportion exceeds the threshold, the anomaly detection unit 220 determines that an anomaly has occurred. In this case, the notification unit 230 notifies the subject (step S7) and returns to step S1.

[0094] Furthermore, in step S4, the current time point can be determined as the anomaly detection timing when the processing of all substrates W to be processed by the substrate processing apparatus 1 has ended. In this structure, if the proportion is below the threshold in step S6, the anomaly detection process ends. Alternatively, the anomaly detection process ends after a notification is performed in step S7.

[0095] (4) Effect

[0096] In the substrate processing apparatus 1 of this embodiment, the pulse values ​​for the adjustment valves 414 and 424 are acquired by the operation value acquisition unit 210 and used as the first and second operation values, respectively. The anomaly determination unit 220 determines whether an anomaly has occurred based on the correlation between the pulse value for the adjustment valve 414 and the pulse value for the adjustment valve 424 acquired by the operation value acquisition unit 210. In this case, anomalies in the substrate processing apparatus 1 can be detected early based on the correlation between the pulse values ​​for the adjustment valve 414 and the pulse values ​​for the adjustment valve 424.

[0097] Whether an anomaly has occurred is determined based on the proportion of data points exceeding the allowable range R among the data points determined by the combination of pulse values ​​for adjustment valve 414 and pulse values ​​for adjustment valve 424 obtained by the action value acquisition unit 210. In this case, anomalies in the substrate processing apparatus 1 can be detected with simple processing.

[0098] Furthermore, the allowable range R is defined as a set of data points that include a predetermined proportion of the data points determined based on the combination of pulse values ​​for the adjustment valve 414 and pulse values ​​for the adjustment valve 424, which are pre-acquired by the action value acquisition unit 210 before substrate processing. Therefore, the allowable range R used in anomaly detection can be easily determined.

[0099] (5) Variations

[0100] In this embodiment, the pulse value for the regulating valve 414 is acquired as the first operating value, and the pulse value for the regulating valve 424 is acquired as the second operating value, but the embodiment is not limited to this. In the operating value acquisition unit 210, the pressure of the processing fluid flowing in the circulation pipe 401, detected by the pressure gauge 402, is provided. Therefore, it is also possible to acquire the pulse value for the regulating valve 414 as the first operating value and the pressure of the processing fluid flowing in the circulation pipe 401 as the second operating value.

[0101] Figure 6 It is a graph representing a correlation diagram. Figure 6 The horizontal axis of the related graph represents the pulse value for adjusting valve 414, and the vertical axis represents the pressure of the processed fluid. For example... Figure 6 As shown, each time the pulse control of the regulating valve 414 is performed, data points determined based on the combination of the pulse value of the regulating valve 414 and the pressure of the processed fluid are plotted on a correlation graph. The proportion of data points exceeding the allowable range R among the plotted data points is used to determine whether an anomaly has occurred.

[0102] In this structure, if an abnormality is determined to have occurred, it is presumed that an abnormality has occurred in the operating components associated with the regulating valve 414 or the circulation piping 401. Furthermore, in a modified example, the pulse value for the regulating valve 424 can also be obtained as the first operating value. In this structure, if an abnormality is determined to have occurred, it is presumed that an abnormality has occurred in the operating components associated with the regulating valve 424 and the circulation piping 401.

[0103] <3> Second Implementation Method

[0104] (1) Drug solution generation section

[0105] In the second embodiment, an abnormality is detected in the drug solution generation unit 300. Figure 7 This is a diagram showing the structure of the liquid drug generation unit 300 in the second embodiment. (See diagram for details.) Figure 7 As shown, the liquid generation unit 300 includes a stock solution supply unit 310, a dilution solution supply unit 320, a mixing pipe 330, a gas supply unit 340, a mixing tank 350, and a supply pipe 360.

[0106] The raw material supply unit 310 includes a piping 311, an on / off valve 312, a flow meter 313, and an adjusting valve 314. The upstream end of the piping 311 is connected to a raw material supply source 301 that supplies the raw material for the drug solution. Thus, the raw material from the raw material supply source 301 flows in the piping 311. In this example, the raw material is hydrofluoric acid.

[0107] In piping 311, an on / off valve 312, a flow meter 313, and an adjusting valve 314 are sequentially inserted from upstream to downstream. The on / off valve 312 opens and closes the flow path of piping 311. The flow meter 313 detects the flow rate of the raw liquid flowing in piping 311 and provides the detection result to the control unit 200. The adjusting valve 314 includes, for example, a motorized needle valve, which adjusts the flow rate of the raw liquid flowing in piping 311 based on pulse control performed by the control unit 200.

[0108] The dilution solution supply unit 320 includes a piping 321, an on / off valve 322, a flow meter 323, and an adjusting valve 324. The upstream end of the piping 321 is connected to a dilution solution supply source 302 that supplies the dilution solution. Thus, the dilution solution from the dilution solution supply source 302 flows in the piping 321. In this example, the dilution solution is DIW (De-ionized water).

[0109] In piping 321, an on / off valve 322, a flow meter 323, and an adjusting valve 324 are sequentially inserted from upstream to downstream. The on / off valve 322 opens and closes the flow path of piping 321. The flow meter 323 detects the flow rate of the diluent flowing in piping 321 and provides the detection result to the control unit 200. The adjusting valve 324 includes, for example, an electrically operated pressure regulator, which adjusts the flow rate of the diluent flowing in piping 321 based on pulse control performed by the control unit 200.

[0110] The mixing piping 330 has a main pipe 331 and two branch pipes 332 and 333. The upstream end of the main pipe 331 is connected to the downstream end of the piping 311 of the raw material supply unit 310 and the downstream end of the piping 321 of the dilution liquid supply unit 320. Branch pipe 332 is connected between the downstream end of the main pipe 331 and the mixing tank 350. Branch pipe 333 is connected between the downstream end of the main pipe 331 and the waste liquid tank 303. On / off valves 334 and 335 are respectively inserted into branch pipes 332 and 333.

[0111] In the main pipe 331, a diluted solution is generated by mixing the stock solution supplied from the stock solution supply unit 310 and the diluent supplied from the diluent supply unit 320. In this example, the diluent is dilute hydrofluoric acid. The diluted solution generated in the main pipe 331 is supplied to the mixing tank 350 through the branch pipe 332.

[0112] The gas supply unit 340 includes a piping 341 and an on / off valve 342. The upstream end of the piping 341 is connected to a gas supply source 304. Thus, gas from the gas supply source 304 flows through the piping 341. In this example, the gas is an inert gas such as nitrogen. The downstream end of the piping 341 is connected to a mixing chamber 350. The on / off valve 342 is inserted into the piping 341 to open and close the flow path of the piping 341.

[0113] The mixing tank 350 stores a mixture of the stock solution and the diluent as a diluent for the medicine. Four level sensors 351, 352, 353, and 354 are installed in the mixing tank 350. Level sensors 351 to 354 detect the first to fourth levels of the diluent stored in the mixing tank 350, respectively, and provide the detection results to the control unit 200.

[0114] Furthermore, the first, second, third, and fourth liquid levels are positioned from bottom to top in this order. Specifically, the first liquid level is slightly above the bottom surface of the mixing tank 350. The second liquid level is above the first liquid level at a predetermined height. The third liquid level is below the fourth liquid level at a predetermined height. The fourth liquid level is slightly below the upper surface of the mixing tank 350.

[0115] The supply piping 360 has a main pipe 361 and two branch pipes 362 and 363. The upstream end of the main pipe 361 is connected to the mixing tank 350. Branch pipe 362 is a circulation piping for circulating the diluted solution, connected between the downstream end of the main pipe 361 and the mixing tank 350. Branch pipe 363 is a processing piping for processing the substrate W, connected between the downstream end of the main pipe 361 and the substrate processing unit 100.

[0116] A concentration meter 364 and a heater 365 are clamped in the main pipe 361. A pump 366, a filter 367, and an on / off valve 368 are clamped in the branch pipe 362. An on / off valve 369 is clamped in the branch pipe 363. The concentration meter 364 measures the concentration of the diluted drug solution flowing in the main pipe 361 and provides the measurement result to the control unit 200.

[0117] While pump 366 is activated, valve 368 opens, thereby allowing the diluted solution from mixing tank 350, after being heated by heater 365, to circulate back to mixing tank 350 via filter 367. In this example, the diluted solution remaining in mixing tank 350 is always circulated through branch pipe 362.

[0118] Additionally, by opening the on / off valves 342 and 369, the diluted solution stored in the mixing tank 350 is pressurized by gas. As a result, the diluted solution stored in the mixing tank 350 is guided downstream through the main pipe 361, heated by the heater 365, and then supplied to the substrate processing unit 100 through the branch pipe 363.

[0119] like Figure 7 As shown by the dashed line in the middle, the liquid generation unit 300 may further include... Figure 2 The same circulation piping 370 as the circulation piping 401. In this case, the substrate processing unit 100 and the mixing tank 350 are connected via the circulation piping 370. In the substrate processing unit 100, a portion of the diluent supplied from the mixing tank 350 is used for substrate processing such as cleaning. The volume of the diluent used is determined according to the substrate processing formula. Another portion of the diluent supplied to the substrate processing unit 100 is not used for substrate processing, but is returned to the mixing tank 350 via the circulation piping 370.

[0120] (2) Control Department

[0121] The structure of the control unit 200 in this embodiment is basically the same as that in the first embodiment. Figure 3 The control unit 200 has the same structure, therefore it uses Figure 3 The operation of the control unit 200 will be briefly explained. The same applies to the control units 200 in the 3rd to 7th embodiments described later.

[0122] In the substrate processing unit 100, the level of the diluted drug solution stored in the mixing tank 350 drops by using the diluted drug solution supplied from the drug solution generation unit 300. Therefore, when a second liquid level is detected by the liquid level sensor 352, the diluted drug solution is replenished.

[0123] Specifically, Figure 3 The action value acquisition unit 210 opens the on / off valves 312 and 322. Furthermore, the action value acquisition unit 210 performs pulse control on the regulating valve 314 to supply a constant flow rate of stock solution from the stock solution supply unit 310. Similarly, the action value acquisition unit 210 performs pulse control on the regulating valve 324 to supply a constant flow rate of diluent solution from the diluent supply unit 320. Thus, constant flow rates of stock solution and diluent solution are supplied from the stock solution supply unit 310 and the diluent supply unit 320, respectively.

[0124] The supplied stock solution and diluent are mixed in the main pipe 331 of the mixing pipe 330 to generate a diluted drug solution. The operation value acquisition unit 210 calculates the theoretical value of the concentration of the generated diluted drug solution based on the ratio of the flow rate of the stock solution to the flow rate of the diluent detected by the flow meters 313 and 323, respectively, and obtains the calculated theoretical value of the concentration as the first operation value.

[0125] The generated diluted drug solution is stored in the mixing tank 350 through branch pipe 332, thereby causing the liquid level of the diluted drug solution stored in the mixing tank 350 to rise. When the liquid level sensor 353 detects a third liquid level, the action value acquisition unit 210 closes the on / off valves 312 and 322. This stops the replenishment of the diluted drug solution. The action value acquisition unit 210 acquires the measured value of the concentration of the replenished diluted drug solution detected by the concentration meter 364 as the second action value.

[0126] Furthermore, the flow rates of the diluent or stock solution are unstable at the beginning and immediately after the generation of the diluted solution, resulting in inconsistent concentrations of the generated diluted solution. Therefore, at the beginning of diluted solution generation, the on / off valve 335 is opened and the on / off valve 334 is closed after a predetermined time, thereby discarding the unstable diluted solution into the waste tank 303. Similarly, immediately after diluted solution generation, the on / off valve 335 is opened and the on / off valve 334 is closed after a predetermined time, thereby discarding the unstable diluted solution into the waste tank 303. Thus, the stable diluted solution is stored in the mixing tank 350.

[0127] Figure 8 It is a graph representing a correlation diagram. Figure 8 The horizontal axis of the correlation graph represents the theoretical concentration, and the vertical axis represents the measured concentration. For example... Figure 8As shown, each time the diluted drug solution is replenished, the anomaly detection unit 220 plots data points determined based on the combination of the theoretical and measured values ​​of the concentration of the replenished diluted drug solution onto a correlation graph. Furthermore, the anomaly detection unit 220 determines whether an anomaly has occurred based on the proportion of data points exceeding the allowable range R among the plotted data points. If an anomaly is detected, it is presumed that an anomalous operating component associated with the adjusting valves 314, 324, or the concentration meter 364 has malfunctioned.

[0128] If the anomaly detection unit 220 determines that an anomaly has occurred, the notification unit 230 will notify the user of this. Furthermore, in this embodiment, if the liquid level sensor 351 detects a first liquid level, or if the liquid level sensor 354 detects a fourth liquid level, the control of the medicine generation unit 300 will be stopped. In this case, the notification unit 230 can also notify the user that the control of the medicine generation unit 300 has been stopped.

[0129] In this example, the anomaly detection processing, besides obtaining the theoretical and measured values ​​of the diluted drug concentration in steps S1 and S2 respectively, also includes... Figure 5 The abnormality detection process is the same. Step S1 is performed, for example, during the replenishment of the diluted drug solution. Step S2 is performed, for example, after the replenishment of the diluted drug solution has stopped.

[0130] (3) Effect

[0131] In the substrate processing apparatus 1 of this embodiment, the operation value acquisition unit 210 acquires the theoretical value of the concentration of the diluted solution, calculated based on the ratio of the flow rate of the stock solution to the flow rate of the diluent solution detected by the flow meters 313 and 323, respectively, as the first operation value. Additionally, the operation value acquisition unit 210 acquires the measured value of the concentration of the diluted solution detected by the concentration meter 364 as the second operation value.

[0132] The anomaly detection unit 220 determines whether an anomaly has occurred based on the correlation between the theoretical value of the concentration of the diluted solution obtained by the operation value acquisition unit 210 and the measured value of the concentration of the diluted solution. In this case, anomalies in the substrate processing apparatus 1 can be detected early based on the correlation between the theoretical value of the concentration of the diluted solution and the measured value of the concentration of the diluted solution.

[0133] In this embodiment, the anomaly determination unit 220 may also determine whether an anomaly has occurred based on the difference between the first action value and the second action value, using the Hotelling method or cumulative summation, etc. This is also true in the following 3rd to 5th embodiments.

[0134] <4> Third Implementation Method

[0135] (1) Action value

[0136] In the second embodiment, the theoretical value of the concentration of the diluted drug solution is calculated as the first operating value based on the ratio of the instantaneous value of the flow rate of the stock solution detected by the flow meter 313 to the instantaneous value of the flow rate of the diluent detected by the flow meter 323. Additionally, the instantaneous value of the concentration of the diluted drug solution detected by the concentration meter 364 is used as the second operating value. However, the embodiment is not limited to this. Hereinafter, the operating values ​​in the third embodiment will be referred to as appropriate. Figure 7 The differences between the action values ​​of the liquid generation unit 300 and those in the second embodiment will be explained.

[0137] Figure 9 This is a diagram used to illustrate the action values ​​in the third embodiment. Figure 9 The diagram shows the time series of the flow rates of the stock solution, the dilution solution, and the concentration of the diluted drug solution, as detected by flow meters 313, 323, and 364, respectively. Figure 9 As shown, at point T1, after a specified time has elapsed since the opening and closing valves 312 and 322 were opened, the stock solution and the diluent solution are mixed in the mixing pipe 330, thereby starting to generate the diluted drug solution.

[0138] As described above, the flow rates of both the stock solution and the diluent are unstable after the initial generation of the diluted solution. Therefore, during the period from time T1 to time T2 after a predetermined time, valve 335 is open and valve 334 is closed. Consequently, the generated diluted solution is discarded into waste tank 303. At time T2, valve 335 closes and valve 334 opens, thereby beginning the supply of diluted solution to mixing tank 350.

[0139] At time T4, after a predetermined time ΔT1 elapsed from time T2, valve 335 opens and valve 334 closes. This ends the supply of diluted solution to the mixing tank 350. Figure 3 During the period from time T2 to time T4, the operation value acquisition unit 210 of the control unit 200 calculates the theoretical value of the concentration of the generated diluted drug solution based on the cumulative value of the flow rate of the stock solution detected by the flow meter 313 and the cumulative value of the flow rate of the diluent detected by the flow meter 323. Furthermore, the operation value acquisition unit 210 acquires the calculated theoretical value of the concentration of the diluted drug solution as the first operation value.

[0140] At time T3, which is a time ΔT2 shorter than time ΔT1 from time T2, the diluted drug solution supplied to the mixing tank 350 at time T2 reaches the concentration meter 364. At time T5, which is a time ΔT1 elapsed from time T3, i.e., a time ΔT2 elapsed from time T4, the diluted drug solution supplied to the mixing tank 350 at time T4, which is the end of the supply, reaches the concentration meter 364. During the period from time T3 to time T5, the operation value acquisition unit 210 calculates a statistical value of the concentration of the diluted drug solution detected by the concentration meter 364. In addition, the operation value acquisition unit 210 acquires the calculated statistical value of the concentration of the diluted drug solution as a second operation value.

[0141] In this embodiment, the statistical value is an average value, specifically, the sum of the concentrations detected at each time point from time point T3 to time point T5 divided by the number of detections. The statistical value can also be a weighted average or other calculated value. Furthermore, the time ΔT2 can be determined based on measurement or by calculations based on the length of the flow path, the cross-sectional area of ​​the flow path, and the flow rate of the diluted drug solution.

[0142] In this embodiment, the anomaly detection processing, besides obtaining the theoretical and statistical values ​​of the concentration of the diluted drug solution in steps S1 and S2 respectively, also includes... Figure 5 The abnormality detection process is the same. Steps S1 and S2 are performed, for example, during the replenishment and dilution of the drug solution. Step S1 can be performed at a time ΔT2 earlier than step S2, but step S2 can also begin before step S1 ends.

[0143] (2) Effect

[0144] In this embodiment, the anomaly judgment unit 220 of the control unit 200 determines whether an anomaly has occurred based on the correlation between the theoretical value of the concentration of the diluted solution and the statistical value of the concentration of the diluted solution. Here, the correlation between the theoretical value of the concentration of the diluted solution and the statistical value of the concentration of the diluted solution obtained in this embodiment is stronger than the correlation between the theoretical value of the concentration of the diluted solution and the measured value of the concentration of the diluted solution obtained in the second embodiment. Therefore, anomalies in the substrate processing apparatus 1 can be detected more reliably.

[0145] <5> Fourth Implementation Method

[0146] (1) Action value

[0147] Regarding the action values ​​in the fourth embodiment, please refer to the following: Figure 7 The differences between the operating values ​​of the drug solution generation unit 300 and those in the second embodiment will be explained. In this embodiment, a concentration prediction model for predicting the concentration of the diluted drug solution is constructed in advance. The constructed concentration prediction model is stored in... Figure 1 The control unit 200 includes a memory, etc.

[0148] When constructing the concentration prediction model, each part of the drug generation section 300 is controlled in the same way as during substrate processing. The concentration prediction model is preferably constructed during the installation of the drug generation section 300, before the drug generation section 300 deteriorates over the years. Figure 10 This is a diagram illustrating an example of the construction sequence of a concentration prediction model. Figure 10 The diagram shows a time series of the flow rate of the stock solution, the flow rate of the diluent, the amount of diluent discharged from the substrate processing unit 100 to the substrate W, and the concentration of the diluent.

[0149] like Figure 10 As shown, when constructing the concentration prediction model, the flow rate of the stock solution, the flow rate of the diluent, the discharge volume of the diluted drug solution, and the concentration of the diluted drug solution are measured sequentially at multiple time points. Furthermore, the flow rates of the stock solution, the diluent, and the diluted drug solution are measured by flow meters 313, 323, and 364, respectively. The discharge volume of the diluted drug solution can be measured by a flow meter not shown, or it can be measured based on calculations performed on the flow path capacity and discharge time.

[0150] The data set consisting of the flow rate of the stock solution, the flow rate of the diluent, the discharge volume of the diluent, and the concentration of the diluent, measured at each time point and between each time point and a specified time ΔT11 prior to that time point, is used as the explanatory variable corresponding to each time point. Additionally, the concentration of the diluent after a specified time ΔT12 prior to each time point is used as the objective variable corresponding to each time point. Based on the obtained explanatory and objective variables corresponding to each time point, training data representing the relationship between the explanatory and objective variables corresponding to that time point is generated.

[0151] exist Figure 10 In the example, the data detected between time point T11 and time point T12, which is ΔT11 after time point T11, constitutes the explanatory variable corresponding to time point T12. Similarly, the concentration of the diluted drug solution detected at time point T13, which is ΔT12 after time point T12, becomes the objective variable corresponding to time point T12. The same applies to time points later than time point T12; multiple training data points are generated by sequentially acquiring the explanatory and objective variables.

[0152] As the machine learning model, a LightGBM (Gradient Boosting Machine) is prepared in advance. A concentration prediction model is constructed using multiple training data points generated by the LightGBM. In this embodiment, the machine learning model is LightGBM, but the implementation is not limited to this. The machine learning model can also be linear regression, Lasso regression, or LSTM (Long Short Term Memory), etc.

[0153] During substrate processing, flow meters 313, 323 or concentration meters 364 are used to detect the flow rate of the stock solution, the flow rate of the diluent, the discharge volume of the diluent, and the concentration of the diluent at multiple time points. Figure 3 The control unit 200's action value acquisition unit 210 acquires the flow rate of the stock solution, the flow rate of the diluent, the discharge volume of the diluent, and the concentration of the diluent detected between each time point and a time point ΔT11 prior to that time point, and uses these as data sets corresponding to each time point.

[0154] Here, the action value acquisition unit 210 predicts the concentration of the diluted drug solution after time ΔT12 from each time point based on the acquired data sets corresponding to each time point and a pre-constructed concentration prediction model. The action value acquisition unit 210 acquires the predicted concentration of the diluted drug solution as the first action value. In addition, the action value acquisition unit 210 acquires the measured concentration of the diluted drug solution detected after time ΔT12 from each time point as the second action value. In this case, a correlation graph is created where the horizontal axis represents the predicted concentration and the vertical axis represents the measured concentration.

[0155] In this example, the anomaly detection processing, besides obtaining the predicted and measured values ​​of the diluted drug concentration in steps S1 and S2 respectively, also includes... Figure 5 The abnormal detection and handling are the same. Steps S1 and S2 are performed, for example, during the replenishment and dilution of the drug solution.

[0156] (2) Effect

[0157] In this embodiment, the anomaly detection unit 220 determines whether an anomaly has occurred based on the correlation between the predicted concentration of the diluted solution and the measured concentration of the diluted solution. Here, the correlation between the predicted and measured concentrations of the diluted solution obtained in this embodiment is stronger than the correlation between the theoretical and measured concentrations of the diluted solution obtained in the second embodiment. Therefore, single-occurrence anomalies occurring in the substrate processing apparatus 1 can be detected more reliably in a shorter period.

[0158] In this embodiment, the explanatory variables of the training data include the flow rate of the stock solution, the flow rate of the diluent, the discharge volume of the diluent, and the concentration of the diluent; however, the embodiment is not limited to this. If the discharge volume of the diluent has a relatively small effect on the target variable, the explanatory variables of the training data may not include the discharge volume of the diluent. On the other hand, the explanatory variables of the training data may also include characteristic quantities other than the flow rate of the stock solution, the flow rate of the diluent, the discharge volume of the diluent, and the concentration of the diluent. In this case, the concentration of the diluent can be predicted more accurately.

[0159] (3) Reference Example

[0160] In this embodiment, the flow rates of the stock solution and the diluent are controlled to be constant using regulating valves 314 and 324, respectively. Furthermore, during periods of unstable flow rates, the diluent and stock solution are not supplied to the mixing tank 350 but are discarded into the waste tank 303. In this configuration, since the flow rates of the stock solution and the diluent have a relatively small impact on the target variable, the explanatory variables of the training data may not include the flow rates of the stock solution or the diluent. Even in this case, it is possible to obtain a predicted value of the concentration of the diluent that is strongly correlated with the measured value of the concentration of the diluted drug solution.

[0161] <6> Fifth Implementation Method

[0162] (1) Action value

[0163] Regarding the action values ​​in the fifth embodiment, please refer to the following: Figure 7 And then Figure 11 The difference between the action values ​​in the liquid generation unit 300 and those in the third embodiment will be explained.

[0164] Figure 11 yes Figure 7 A magnified view of a portion of the drug solution generation section 300. (See image below.) Figure 11 As shown, at time point t when the diluted drug solution is replenished, the volume W1 (t) The diluted drug solution flows into the mixing tank 350 through the mixing pipe 330 per unit time. In this example, the unit time is the period τ1 of flow detection based on flow meters 313 and 323. Therefore, the capacity W1 (t) This refers to the volume of diluted solution flowing into mixing tank 350 between time point t and the time point before period τ1, calculated based on the flow rates of the original solution and the diluent detected by flow meters 313 and 323, respectively. The concentration C0 of the flowing diluent is... (t) Calculated based on the flow rates of the stock solution and the diluent, and the specific gravity of the stock solution.

[0165] Additionally, a volume W2 of diluted solution flows out of the mixing tank 350 per unit time via the supply piping 360. The volume W2 is a known fixed value independent of time, determined based on the operating parameters of the pump 366.

[0166] Moreover, at time t, the capacity is W3 (t) The diluted solution flows into the mixing tank 350 through the circulation pipe 370 per unit time. Here, the diluted solution passing through the substrate processing unit 100 arrives at the mixing tank 350 after a predetermined time τ0. Therefore, the capacity W3 (t) The calculation is performed by subtracting the volume of the diluent used by the substrate processing unit 100 at a time τ0 prior to time t from the volume W2. Time τ0 can be determined by measurement or by calculation based on the length of the flow path, the cross-sectional area of ​​the flow path, and the flow rate of the diluent.

[0167] Through the inflow and outflow of these diluted solutions, a volume W is stored in the mixing tank 350 at time point t. (t) The diluted drug solution. Volume W at time point t. (t) The concentration C of the diluted drug solution stored in the updated mixing tank 350 (t) The updated formula is calculated according to the following formulas (1) and (2).

[0168]

[0169]

[0170] In equation (2), τ2 is the period from when the diluted solution flowing from the mixing tank 350 through the supply pipe 360 ​​flows back into the mixing tank 350 through the circulation pipe 370. The period τ2 can be determined by measurement or by calculation based on the length of the flow path, the cross-sectional area of ​​the flow path, and the flow rate of the diluted solution. The initial value is the capacity W. (t-τ1) and concentration C (t -τ1) C (t-τ2) It can be appropriately determined in a way that matches the measured value.

[0171] Figure 3 The control unit 200's action value acquisition unit 210 calculates the concentration C of the diluted drug solution stored in the mixing tank 350 at each time point t based on equation (2). (t) The action value acquisition unit 210 acquires the calculated concentration C of the diluted drug solution. (t) The theoretical value is used as the first action value. Additionally, the action value acquisition unit 210 acquires the value obtained after each time point t has elapsed. Figure 9 The concentration (i.e., concentration C) detected by the concentration meter 364 at time point ΔT2. (t+ΔT2)The measured value is used as the second action value. Time ΔT2 is the time from when the diluted drug solution flowing from mixing tank 350 reaches concentration meter 364. In this case, a correlation graph is created with the theoretical concentration on the horizontal axis and the measured concentration on the vertical axis.

[0172] In this example, the anomaly detection processing, besides obtaining the predicted and measured values ​​of the diluted drug concentration in steps S1 and S2 respectively, also includes... Figure 5 The abnormal detection process is the same. Steps S1 and S2 can be performed during the replenishment of the diluent. On the other hand, steps S1 and S2 can also be performed during periods when the diluent is not replenished. In this case, W1 will be detected at times other than when the diluent is replenished. (t) and concentration C0 (t) Simply set it to 0.

[0173] (2) Effect

[0174] In this embodiment, the anomaly judgment unit 220 of the control unit 200 determines whether an anomaly has occurred based on the correlation between the theoretical value of the concentration of the diluent solution stored in the mixing tank 350 and the measured value of the concentration of the diluent solution. Here, if the concentration of the diluent solution supplied to the substrate processing unit 100 changes, the concentration of the diluent solution stored in the mixing tank 350 will change because the unused diluent solution in the substrate processing unit 100 returns to the mixing tank 350.

[0175] Even in this case, according to this embodiment, since the theoretical value of the concentration of the diluted solution stored in the mixing tank 350 is calculated based on the volume of the diluted solution flowing into or out of the mixing tank 350, the flow rate of the original solution, and the flow rate of the diluent, the correlation between the theoretical value of the concentration of the diluted solution and the statistical value of the concentration of the diluted solution can be enhanced. Therefore, abnormalities in the substrate processing apparatus 1 can be detected more reliably.

[0176] Sometimes, there is a certain deviation between the theoretical value of the concentration of the diluted drug solution calculated based on equation (2) and the measured value of the concentration of the diluted drug solution detected by the concentration meter 364. Therefore, the concentration C0 of the diluted solution flowing into the mixing tank 350 can also be adjusted. (t) By adding an appropriately determined offset value, the deviation can be eliminated. In this case, even after the initial calculation of the diluted drug concentration has just begun, the correlation between the theoretical and measured values ​​of the diluted drug concentration can be enhanced.

[0177] <7> 6th Implementation Method

[0178] (1) Rotation holding part

[0179] In the sixth embodiment, as an operating member of the substrate processing unit 100, an abnormality of the rotation holding unit included in each processing unit 110 is detected. Figure 12 This is a side view showing the structure of the rotation retainer in the sixth embodiment. (See attached image.) Figure 12 As shown, the rotation holding part 500 includes a rotary chuck 510, a plurality of chuck pins 520, and a chuck drive part 530. The rotary chuck 510 is configured to hold the substrate W horizontally and rotate it, and includes a rotation drive part 511, a rotation shaft 512, a plate support member 513, and a rotation plate 514.

[0180] Rotary drive unit 511 is provided in Figure 1 The upper part of the processing unit 110 is supported by a support member (not shown). A rotating shaft 512 extends downward from the rotating drive unit 511. A plate support member 513 is mounted on the lower end of the rotating shaft 512. The rotating plate 514 is circular and is horizontally supported by the plate support member 513. By rotating the rotating shaft 512 using the rotating drive unit 511, the rotating plate 514 rotates about a vertical axis.

[0181] Multiple chuck pins 520 are provided at equal angular intervals around the rotation axis 512 on the periphery of the rotating plate 514. In this example, eight chuck pins 520 are provided at 45-degree intervals around the rotation axis 512 on the periphery of the rotating plate 514. Each chuck pin 520 includes a shaft portion 521, a pin support portion 522, and a retaining portion 523. The shaft portion 521 is configured to pass through the rotating plate 514 in a vertical direction. The pin support portion 522 is configured to extend horizontally from the lower end of the shaft portion 521. The retaining portion 523 is configured to protrude downward from the front end of the pin support portion 522.

[0182] Each chuck pin 520 switches between a closed and an open state by rotating about a vertical axis with the shaft portion 521 as the center. In the closed state, each retaining portion 523 abuts against the outer peripheral end (beveled edge portion) of the substrate W. In the open state, each retaining portion 523 separates from the outer peripheral end of the substrate W.

[0183] The chuck drive unit 530 includes, for example, a rotary actuator, a magnet, a linear slider, and a cam, which switches between a closed state and an open state. Figure 13 This is a functional block diagram showing the structure of the rotary retaining part 500. For example... Figure 13 As shown, the control unit 200 provides the chuck drive unit 530 with an instruction to set the chuck drive unit 530 to a closed state (hereinafter referred to as a closed instruction). Additionally, the control unit 200 provides the chuck drive unit 530 with an instruction to set the chuck drive unit 530 to an open state (hereinafter referred to as an open instruction).

[0184] If a closed instruction is given to the chuck drive unit 530, the chuck drive unit 530 is closed by supplying air to the rotary actuator. In this case, the linear slider moves linearly forward by rotating the magnet. The linear forward movement of the linear slider is converted into rotational movement of each chuck pin 520 by the cam. Thus, each chuck pin 520 is closed. On the other hand, if an open instruction is given to the chuck drive unit 530, the air supply to the rotary actuator is stopped, thereby opening the chuck drive unit 530. In this case, each chuck pin 520 is open.

[0185] Additionally, the rotary holding unit 500 includes multiple Hall sensors 524 and Hall sensors 531. The multiple Hall sensors 524 correspond to multiple chuck pins 520 respectively. Each Hall sensor 524 detects the state of the corresponding chuck pin 520 based on a magnet (not shown) disposed on the corresponding chuck pin 520, and provides the detection result to the control unit 200. The Hall sensor 531 detects the state of the chuck drive unit 530 based on the magnet of the chuck drive unit 530, and provides the detection result to the control unit 200.

[0186] (2) Control Department

[0187] In the following description, the eight chuck pins 520 will be referred to as chuck pins 1 to 8. Chuck pins 1 and 2 are paired, chuck pins 3 and 4 are paired, chuck pins 520 are paired, chuck pins 520 are paired, and chuck pins 7 and 8 are paired. The operation of the control unit 200 with respect to chuck pins 1 and 2 will be described below; the operation of the control unit 200 with respect to chuck pins 3 to 8 is the same.

[0188] Figure 3 The action value acquisition unit 210 provides a closed or open instruction to the chuck drive unit 530. Furthermore, the action value acquisition unit 210 acquires the transition time of the first chuck pin 520 as a first action value based on the detection result provided by the Hall sensor 524 corresponding to the first chuck pin 520. Similarly, the action value acquisition unit 210 acquires the transition time of the second chuck pin 520 as a second action value based on the detection result provided by the Hall sensor 524 corresponding to the second chuck pin 520.

[0189] The switching time of the chuck pin 520 is the time from when the chuck drive unit 530 provides a closed instruction until the chuck pin 520 becomes closed, or the time from when the chuck drive unit 530 provides an open instruction until the chuck pin 520 becomes open.

[0190] Figure 14 This is a diagram illustrating the transition time of the chuck pin 520 when the indicator is closed. (See diagram for example.) Figure 14As shown, a closing instruction is provided to the chuck drive unit 530 at time t1. In this case, the chuck drive unit 530 switches from the open state to the closed state. As a result, each chuck pin 520 switches from the open state to the closed state, and at time t2, the Hall sensor 524 detects that each chuck pin 520 is in the closed state. The operation value acquisition unit 210 acquires the transition time Δt1 for each chuck pin 520 from the time t1 when the closing instruction was provided to the time t2 when it was detected to be in the closed state.

[0191] Figure 15 This is a diagram illustrating the transition time of the chuck pin 520 when the indicator is open. (See diagram for example.) Figure 15 As shown, an open instruction is provided to the chuck drive unit 530 at time t11. In this case, the chuck drive unit 530 switches from a closed state to an open state. As a result, each chuck pin 520 switches from a closed state to an open state, and at time t12, the Hall sensor 524 detects that each chuck pin 520 is in the open state. The operation value acquisition unit 210 acquires the transition time Δt11 from the time t11 when the open instruction is provided to the time t12 when the open state is detected for each chuck pin 520.

[0192] Each time, a close or open instruction is provided to the chuck drive unit 530. Figure 3 The anomaly detection unit 220 will plot the data points determined based on the combination of the transition time of the first chuck pin 520 and the transition time of the second chuck pin 520 onto the correlation diagram. Data points can also be plotted separately for the closing and opening indications. In this case, the allowable range R is determined separately for the closing and opening indications.

[0193] The anomaly detection unit 220 determines whether an anomaly has occurred based on the proportion of data points exceeding the allowable range R among the drawn data points. If an anomaly is detected, it presumes that an actuating component associated with the first or second chuck pin 520 has malfunctioned. Actuating components associated with the first or second chuck pin 520 include rotary actuators, magnets, linear slides, or cams of the chuck drive unit 530. The notification unit 230 notifies the user of this fact if the anomaly detection unit 220 determines that an anomaly has occurred.

[0194] In this example, the anomaly detection processing, besides obtaining the transition times of the first and second chuck pins 520 in steps S1 and S2 respectively, also includes... Figure 5 The abnormality detection and handling are the same. Steps S1 and S2 are executed approximately simultaneously in response to a closed or open instruction to the chuck drive unit 530. Alternatively, steps S1 and S2 may be executed only in response to either the closed or open instruction.

[0195] (3) Effect

[0196] In the substrate processing apparatus 1 of this embodiment, the operation value acquisition unit 210 acquires the transition times of the first and second chuck pins 520 as the first and second operation values, respectively. The anomaly determination unit 220 determines whether an anomaly has occurred based on the correlation between the transition times of the first and second chuck pins 520 acquired by the operation value acquisition unit 210. In this case, anomalies in the substrate processing apparatus 1 can be detected early based on the correlation between the transition times of the first and second chuck pins 520.

[0197] (4) Variations

[0198] In this embodiment, the transition time of the first chuck pin 520 is obtained as the first operation value, and the transition time of the second chuck pin 520 is obtained as the second operation value, but the embodiment is not limited to this. Sometimes a time difference occurs between the transition time of each chuck pin 520 and the transition time of the chuck drive unit 530.

[0199] Figure 16 This is a diagram used to illustrate the transition time of the chuck pin 520 and the chuck drive unit 530. (See diagram below.) Figure 16 As shown, a closing indication is provided to the chuck drive unit 530 at time point t1. In this case, with Figure 14 Similarly, at time t2, Hall sensor 524 detects that each chuck pin 520 is in the closed state. On the other hand, Hall sensor 531 detects that the chuck drive unit 530 is in the closed state at time t3, which is after time t2. The same time difference sometimes occurs in the open indication.

[0200] Therefore, the transition time Δt1 of a specific chuck pin 520 can be used as the first action value, and the transition time Δt2 of the chuck drive unit 530 can be used as the second action value. Alternatively, instead of the transition time Δt2, the time from time point t2 to time point t3 can be used as the second action value. In this structure, if an anomaly is determined to have occurred, it is presumed that an anomaly has occurred in the actuating component associated with a specific chuck pin 520 or chuck drive unit 530.

[0201] (5) Reference Example

[0202] The anomaly detection process in the reference example can monitor the change in transition time of each chuck pin 520 or chuck drive unit 530, and determine that an anomaly has occurred if the transition time exceeds the specified allowable range. However, in the reference example, sometimes even when no anomaly has actually occurred, it is determined to be an anomaly based on the fact that the transition time exceeds the specified allowable range. Therefore, in the reference example, it is difficult to accurately detect anomalies.

[0203] <8> 7th Implementation Method

[0204] (1) Processing fluid supply unit

[0205] In the seventh embodiment, as an operating member of the substrate processing unit 100, an abnormality in the processing liquid supply unit included in each processing unit 110 is detected. Figure 17 This is a diagram showing the structure of the processing fluid supply section. For example... Figure 17 As shown, the processing fluid supply unit 600 includes scanning drive units 610 and 620 and cleaning tools 630 and 640.

[0206] Scan drive units 610 and 620 each include, for example, a stepper motor and an encoder. Scan drive unit 610, based on pulse control from control unit 200, moves cleaning tool 630 between a standby position located outside the substrate W and a processing position located below the center of the substrate W. Furthermore, scan drive unit 610 uses an encoder to detect when cleaning tool 630 reaches the processing position and provides a positioning completion signal indicating that positioning is complete to control unit 200.

[0207] Based on pulse control from the control unit 200, the scan drive unit 620 moves the cleaning tool 640 between a standby position located outside the substrate W and a processing position located below the center of the substrate W. Furthermore, the scan drive unit 620 detects when the cleaning tool 640 reaches the processing position using an encoder and provides a positioning completion signal indicating that positioning is complete to the control unit 200.

[0208] Cleaning tools 630 and 640 are each, for example, nozzles, that supply processing liquid to the vicinity of the center of the lower surface of the substrate W at the processing position. This performs processing on the substrate W. The substrate W can be manufactured from, for example, a... Figure 12 The cleaning tool 630 is processed while held in a rotating holding state by the rotating holding part 500. Furthermore, the cleaning tool 630 may be a brush (including an abrasive brush) for cleaning the substrate W, instead of a nozzle. Similarly, the cleaning tool 640 may be a brush (including an abrasive brush) for cleaning the substrate W, instead of a nozzle.

[0209] In this example, the scan drive units 610 and 620 and the cleaning tools 630 and 640 are located in the same processing unit 110, but the implementation is not limited to this. As long as the substrate processing using the cleaning tool 630 and the substrate processing using the cleaning tool 640 are performed at approximately the same frequency, the scan drive units 610 and the cleaning tool 630 and the scan drive units 620 and the cleaning tool 640 can also be located in different processing units 110.

[0210] (2) Control Department

[0211] Figure 3During substrate processing, the motion value acquisition unit 210 performs pulse control on the scan drive unit 610, thereby moving the cleaning tool 630 from the standby position toward the processing position. Furthermore, the motion value acquisition unit 210 acquires the transition time of the cleaning tool 630 as a first motion value based on the positioning completion signal provided from the scan drive unit 610. The transition time of the cleaning tool 630 is the time from the start of pulse control by the scan drive unit 610 until the positioning completion signal is provided.

[0212] Figure 18 This is a diagram used to illustrate the transition time of cleaning tool 630. Figure 18 The horizontal axis represents time, and the vertical axis represents the pulse value for the scan drive unit 610. For example... Figure 18 As shown, at time t21, pulse control of the scanning drive unit 610 begins. In this case, the cleaning tool 630 moves from the standby position toward the processing position. At time t22, a positioning completion signal is provided. Consequently, the motion value acquisition unit 210 ends pulse control. Furthermore, the motion value acquisition unit 210 acquires the transition time Δt21 from the start of pulse control at time t21 to the time t22 when the positioning completion signal is provided.

[0213] Similarly, during substrate processing, the motion value acquisition unit 210 moves the cleaning tool 640 from the standby position to the processing position by pulse control of the scan drive unit 620. Furthermore, the motion value acquisition unit 210 acquires the transition time of the cleaning tool 640 as a second motion value based on the positioning completion signal provided from the scan drive unit 620. The transition time of the cleaning tool 640 is the time from the start of pulse control by the scan drive unit 620 until the positioning completion signal is provided.

[0214] Furthermore, after the substrate processing is completed, the cleaning tools 630 and 640 each return from the processing position to the standby position. In the above example, the transition times of the cleaning tools 630 and 640 from the standby position to the processing position are the first and second operation values, respectively, but the implementation is not limited to this. Alternatively, the transition times of the cleaning tools 630 and 640 from the processing position to the standby position may be the first and second operation values, respectively.

[0215] Each time, pulse control is performed on the scan drive units 610 and 620. Figure 3The anomaly detection unit 220 plots data points determined based on the combination of the changeover times of the cleaning tool 630 and the cleaning tool 640 onto a correlation diagram. Furthermore, the anomaly detection unit 220 determines whether an anomaly has occurred based on the proportion of data points exceeding the allowable range R among the plotted data points. If an anomaly is detected, it is presumed that an anomaly has occurred in the motion components associated with the scan drive units 610 and 620. The notification unit 230 notifies the user of this fact if the anomaly detection unit 220 determines that an anomaly has occurred.

[0216] In this example, the anomaly detection processing, besides obtaining the transition times of cleaning tools 630 and 640 in steps S1 and S2 respectively, also includes... Figure 5 The abnormality detection and processing are the same. Either step S1 or S2 can be executed first. When the scan drive unit 610 and cleaning tool 630 and the scan drive unit 620 and cleaning tool 640 are located in different processing units 110, steps S1 and S2 can be executed approximately simultaneously.

[0217] (3) Effect

[0218] In the substrate processing apparatus 1 of this embodiment, the operation value acquisition unit 210 acquires the transition times of cleaning tools 630 and 640 as first and second operation values, respectively. The anomaly determination unit 220 determines whether an anomaly has occurred based on the correlation between the transition times of cleaning tool 630 and cleaning tool 640 acquired by the operation value acquisition unit 210. In this case, anomalies in the substrate processing apparatus 1 can be detected early based on the correlation between the transition times of cleaning tool 630 and cleaning tool 640.

[0219] (4) Reference Example

[0220] As an example of anomaly detection processing, it is possible to monitor the changeover time of cleaning tool 630 or cleaning tool 640, and determine that an anomaly has occurred if the changeover time exceeds the specified allowable range. However, in the example, sometimes even when no anomaly has actually occurred, it is determined to be an anomaly based on the changeover time exceeding the specified allowable range. Therefore, in the example, it is difficult to accurately detect anomalies.

[0221] <9> Implementation Method 8

[0222] (1) Rotation holding part

[0223] In the eighth embodiment, as an operating member of the substrate processing unit 100, an abnormality of the rotation holding unit included in each processing unit 110 is detected. Figure 19 This is a side view showing the structure of the rotation retainer in the eighth embodiment. (See attached image.) Figure 19 As shown, the rotating holding part 700 includes a rotating drive part 710, a rotating shaft 720, and an adsorption part 730.

[0224] Rotary drive unit 710 is provided in Figure 1 The bottom of the processing unit 110. The rotation shaft 720 is provided to extend upward from the rotation drive unit 710. The adsorption unit 730 is mounted on the upper end of the rotation shaft 720 and holds the substrate W horizontally by adsorbing the substrate W. By rotating the rotation shaft 720 using the rotation drive unit 710, the adsorption unit 730 rotates about the vertical axis.

[0225] The rotary drive unit 710 is equipped with a torque sensor 711 and a temperature sensor 712. The torque sensor 711 detects the torque of the rotary drive unit 710 and provides the detection result to the control unit 200. The temperature sensor 712 detects the temperature of the rotary drive unit 710 and provides the detection result to the control unit 200. Furthermore, in this example, the detected torque is expressed as a percentage [%] relative to the rated torque.

[0226] (2) Control Department

[0227] Figure 20 This is a functional block diagram showing the structure of the control unit 200. For example... Figure 20 As shown, the control unit 200 in this embodiment also includes an action value selection unit 240. Hereinafter, the operation of the control unit 200 in this embodiment will be mainly explained in conjunction with... Figure 3 The differences between the control unit 200 and the control unit 200.

[0228] Figure 21 This is a diagram showing the changes in the rotational speed and torque of the rotational drive unit 710 during substrate processing. Figure 21 The upper part shows the time variation of the rotational speed of the rotational drive unit 710 when processing a substrate W. The time variation of the rotational speed of the rotational drive unit 710 is determined according to the substrate processing formula. Figure 21 The lower part shows the time variation of the torque of the rotary drive unit 710 when processing a substrate W in a manner corresponding to the rotational speed of the rotary drive unit 710.

[0229] like Figure 21 As shown in section A, when the rotational speed of the rotary drive unit 710 changes drastically, the torque of the rotary drive unit 710 becomes extremely large. Therefore, the motion value acquisition unit 210 acquires the torque value from the torque sensor 711 based on the substrate processing formula. Figure 21 The torque of the rotary drive unit 710 in part B when it rotates at a predetermined constant rotational speed is used as the first operating value. In addition, the operating value acquisition unit 210 acquires the temperature of the rotary drive unit 710 when it rotates at the aforementioned constant rotational speed from the temperature sensor 712 and uses it as the second operating value.

[0230] Furthermore, when the rotational speed of the rotary drive unit 710 during the substrate processing is approximately constant, the total torque detected by the torque sensor 711 can be used as the first operating value. Similarly, when the rotational speed of the rotary drive unit 710 during the substrate processing is approximately constant, the total temperature detected by the temperature sensor 712 can be used as the second operating value.

[0231] Figure 22 This is a graph showing the obtained torque and temperature. In Figure 22 The upper part shows the time variation of the acquired torque. Figure 22 The lower part shows the time-varying temperature, corresponding to the torque. Figure 22 The display indicates whether the obtained temperature is above the specified reference temperature. "High" indicates a temperature above the reference temperature, and "Low" indicates a temperature below the reference temperature.

[0232] like Figure 22 As shown in the upper part, since the torque is obtained when the rotary drive unit 710 rotates at a constant rotational speed, it does not fluctuate drastically, but rather varies within a relatively limited range. On the other hand, the inventors have conducted various experiments and studies, and have obtained the following insights regarding the torque and temperature of the rotary drive unit 710.

[0233] like Figure 22 As shown in section C, the torque is relatively high at the very beginning of substrate processing. Similarly, as... Figure 22 As shown in section D, the torque is relatively large even after a relatively long period has elapsed since the previous torque acquisition point. Furthermore, the temperature of the rotary drive unit 710 is relatively low at the very beginning of substrate processing, or after a relatively long period has elapsed since the previous torque acquisition point. In other words, torque is related to temperature; even if the rotational speed of the rotary drive unit 710 is constant, the torque is relatively large when the temperature of the rotary drive unit 710 is relatively low.

[0234] Therefore, the operation value selection unit 240 selects a temperature above the reference temperature from the temperatures acquired by the operation value acquisition unit 210. Additionally, the operation value selection unit 240 selects the torque corresponding to the selected temperature from the torques acquired by the operation value acquisition unit 210. The anomaly judgment unit 220 determines whether an anomaly has occurred based on the change in torque selected by the operation value selection unit 240.

[0235] Figure 23 This is a diagram illustrating an example of an anomaly detection method. Figure 23 The horizontal axis represents the date the torque was obtained, and the vertical axis represents the torque obtained. For example... Figure 23 As shown, in this example, a box-and-whisker diagram is created based on the torque obtained daily, and the created box-and-whisker diagram is arranged in chronological order. Additionally, it determines the specified torque reference value (in... Figure 23 In the example, 3% of the straight lines pass through the box of the box-and-whisker diagram. If the straight line representing the torque reference value fails to pass through the box portion of the box-and-whisker diagram more than a specified number of times (a specified number of days in this example), it is determined that an anomaly has occurred.

[0236] The method for determining an anomaly is not limited to the examples described above. Alternatively, it can involve monitoring the time variation of the torque selected by the action value selection unit 240, and determining that an anomaly has occurred if the torque exceeds a predetermined allowable range more than a predetermined number of times. The notification unit 230 then notifies the user of this fact if the anomaly determination unit 220 determines that an anomaly has occurred.

[0237] (3) Anomaly Detection and Handling

[0238] Figure 24 It means based on Figure 20 The flowchart of the anomaly detection and processing performed by the control unit 200. Figure 24 The anomaly detection process is performed by executing an anomaly detection program stored in memory via the CPU of the control unit 200. The anomaly detection process is executed daily in parallel with the board processing. The following is related to... Figure 20 Together with the control unit 200, use Figure 24 The flowchart illustrates the anomaly detection and handling process.

[0239] First, the operation value acquisition unit 210 acquires the torque of the rotation drive unit 710 when it rotates at a predetermined constant rotation speed from the torque sensor 711 (step S11). Additionally, the operation value acquisition unit 210 acquires the temperature corresponding to the torque acquired in step S11 from the temperature sensor 712 (step S12). Steps S11 and S12 are performed approximately simultaneously during the processing of the rotating substrate W, based on the substrate processing formula.

[0240] Next, the operation value selection unit 240 selects a temperature above the reference temperature from the temperatures obtained in step S12 (step S13). Next, the operation value selection unit 240 obtains the torque from the torques obtained in step S11 that corresponds to the temperature selected in step S13 (step S14).

[0241] Then, the fault determination unit 220 determines whether the processing of all substrates of the processing target has been completed (step S15). If the processing of all substrates of the processing target has not been completed, the fault determination unit 220 returns to step S11. Steps S11 to S15 are repeated until the processing of all substrates of the processing target is completed.

[0242] After processing of all substrates of the target object has been completed, the anomaly detection unit 220 uses the torque selected in step S14 to create a box-whisker diagram (step S16). Next, the anomaly detection unit 220 compares the box-whisker diagram created in step S16 with the box-whisker diagrams created up to the previous day. Figure 1 Configure according to the time series order (step S17).

[0243] Next, the anomaly detection unit 220 determines whether the straight line representing the torque reference value passes through the box portion of the box-and-whisker diagram arranged in step S17 for more than a predetermined number of days (step S18). If the straight line passes through the box portion of the box-and-whisker diagram for more than a predetermined number of days, the anomaly detection unit 220 does not determine that an anomaly has occurred and ends the anomaly detection process. If the straight line does not pass through the box portion of the box-and-whisker diagram for more than a predetermined number of days, the anomaly detection unit 220 determines that an anomaly has occurred. In this case, the notification unit 230 notifies the relevant party (step S19) and ends the anomaly detection process.

[0244] (4) Effect

[0245] In the substrate processing apparatus 1 of this embodiment, the torque of the rotation drive unit 710 is acquired by the operation value acquisition unit 210 as a first operation value. Additionally, the temperature of the rotation drive unit 710 is acquired by the operation value acquisition unit 210 as a second operation value. The operation value selection unit 240 selects a temperature above a reference temperature from the temperatures acquired by the operation value acquisition unit 210. Furthermore, the operation value selection unit 240 selects a torque from the torques acquired by the operation value acquisition unit 210 that corresponds to the selected temperature.

[0246] The anomaly determination unit 220 determines whether an anomaly has occurred based on the change in the time series of torque selected by the operation value selection unit 240. In this case, since the anomaly is determined based on the change in the time series of torques that meet the specified reference among the acquired torques, anomalies in the board processing apparatus 1 can be detected easily and accurately.

[0247] Furthermore, the torque obtained as the first operating value is the torque when the rotary drive unit 710 rotates at a constant rotational speed. Similarly, the temperature obtained as the second operating value is the temperature when the rotary drive unit 710 rotates at a constant rotational speed. Therefore, it is possible to detect abnormalities in the substrate processing apparatus 1 more accurately.

[0248] (5) Reference Example

[0249] As an example of anomaly detection processing, the aforementioned box-and-whisker diagram can be created using all the torques of the rotary drive unit 710 acquired by the action value acquisition unit 210, regardless of the temperature of the rotary drive unit 710. In this structure, if the straight line representing the torque reference value does not pass through the box portion of the box-and-whisker diagram for more than a specified number of days, an anomaly is determined to have occurred.

[0250] However, in the reference example, sometimes the temperature of the rotary drive unit 710 is low, causing the straight line representing the torque reference value to not pass through the box portion of the box-and-whisker diagram. That is, sometimes even when no abnormality has actually occurred, it is judged as an abnormality based on the fact that the straight line representing the torque reference value does not pass through the box portion of the box-and-whisker diagram for more than a specified number of days. Therefore, in the reference example, it is difficult to accurately detect the abnormality.

[0251] <10> Correspondence between the constituent elements of the claims and the parts of the embodiments

[0252] The following examples illustrate the correspondence between the constituent elements of the claims and the portions of the embodiments, but the present invention is not limited to these examples. Various other elements having the configuration or function described in the claims can also be used as constituent elements of the claims.

[0253] In the above embodiments, the substrate W is an example of a substrate, the substrate processing apparatus 1 is an example of a substrate processing apparatus, the action value acquisition unit 210 is an example of an action value acquisition unit, and the anomaly determination unit 220 is an example of an anomaly determination unit. The allowable range R is an example of an allowable range, the action value selection unit 240 is an example of an action value selection unit, and the control unit 200 is an example of a processing apparatus.

[0254] In the first embodiment, the regulating valve 414 is an example of a first actuating member or a first regulating valve, and the regulating valve 424 is an example of a second actuating member or a second regulating valve. Alternatively, in the first embodiment, the regulating valve 414 or the regulating valve 424 is an example of a first actuating member or a regulating valve, and the pressure gauge 402 is an example of a second actuating member or a pressure gauge.

[0255] In embodiments 2 to 5, flow meter 313 is an example of a first actuating member or a first flow meter, flow meter 323 is an example of a first actuating member or a second flow meter, and concentration meter 364 is an example of a second actuating member or a concentration meter. Mixing tank 350 is an example of a storage section, substrate processing section 100 is an example of a substrate processing section, mixing piping 330 is an example of a first flow path section, supply piping 360 is an example of a second flow path section, and circulation piping 370 is an example of a third flow path section.

[0256] In the sixth embodiment, the first chuck pin 520 is an example of a first actuating member or a first chuck pin, and the second chuck pin 520 is an example of a second actuating member or a second chuck pin. Alternatively, in the sixth embodiment, one of the chuck pins 520 is an example of a first actuating member or a chuck pin, and the chuck drive unit 530 is an example of a second actuating member or a chuck drive unit.

[0257] In the seventh embodiment, the cleaning tool 630 is an example of a first actuating member or a first processing tool, and the cleaning tool 640 is an example of a second actuating member or a second processing tool. In the eighth embodiment, the rotary drive unit 710 is an example of a rotary drive unit, the torque sensor 711 is an example of a first actuating member or a torque sensor, and the temperature sensor 712 is an example of a second actuating member or a temperature sensor.

Claims

1. A substrate processing apparatus comprising a first actuating member and a second actuating member used in processing a substrate using a processing liquid, characterized in that, have: An action value acquisition unit acquires a first action value of the first action member and a second action value of the second action member; and The anomaly detection unit determines whether an anomaly has occurred based on the correlation between the first action value and the second action value obtained by the action value acquisition unit. The anomaly determination unit determines whether an anomaly has occurred based on the proportion of data points that exceed an allowable range determined based on the correlation between the first action value and the second action value, among the data points determined according to the set of the first action value and the second action value obtained by the action value acquisition unit. The substrate processing apparatus further includes an operation value selection unit, which selects a second operation value from the second operation values ​​obtained by the operation value acquisition unit that satisfies a predetermined benchmark, and selects a first operation value from the first operation values ​​obtained by the operation value acquisition unit that corresponds to the selected second operation value. The anomaly detection unit determines whether an anomaly has occurred based on the change in the time series of the first action value selected by the action value selection unit. The first actuating component includes a torque sensor that detects the torque of the rotary drive unit. The second actuating component includes a temperature sensor that detects the temperature of the rotary drive unit. The first action value is the torque of the rotary drive unit detected by the torque sensor. The second action value is the temperature of the rotary drive unit detected by the temperature sensor.

2. The substrate processing apparatus according to claim 1, characterized in that, The anomaly detection unit updates the allowable range in a predetermined proportion that includes the data points when the detection is performed.

3. The substrate processing apparatus according to claim 2, characterized in that, The anomaly detection unit makes the judgment based on the rate of change of the updated tolerance range relative to the predetermined tolerance range.

4. The substrate processing apparatus according to any one of claims 1 to 3, characterized in that, The allowable range is defined as a specified proportion of data points that are determined based on the combination of the first action value and the second action value, which are obtained in advance by the action value acquisition unit before the substrate is processed.

5. The substrate processing apparatus according to any one of claims 1 to 3, characterized in that, The anomaly detection unit determines whether an anomaly has occurred after processing a specified number of substrates or after a specified time has elapsed.

6. The substrate processing apparatus according to claim 1, characterized in that, The first action value is the torque of the rotary drive unit detected by the torque sensor when the rotary drive unit rotates at a predetermined constant rotational speed. The second action value is the temperature of the rotary drive unit when it rotates at the constant rotational speed, which is detected by the temperature sensor.

7. An anomaly detection method for detecting anomalies in a substrate processing apparatus comprising a first actuating member and a second actuating member used in the processing of a substrate using a processing liquid, characterized in that, include: Obtain the first action value of the first action component and the second action value of the second action component that are related to each other; as well as Based on the obtained first action value and second action value, it is determined whether an anomaly has occurred. Determining whether the anomaly has occurred includes judging whether an anomaly has occurred based on the proportion of data points among the data points determined according to the group of the first action value and the second action value, which exceed the allowable range determined based on the correlation between the first action value and the second action value. The anomaly detection method further includes selecting a second action value from the acquired second action values ​​that meets a predetermined benchmark, and selecting a first action value from the acquired first action values ​​that corresponds to the selected second action value. Determining whether the anomaly has occurred includes judging whether an anomaly has occurred based on the change in the time series of the selected first action value. The first actuating component includes a torque sensor that detects the torque of the rotary drive unit. The second actuating component includes a temperature sensor that detects the temperature of the rotary drive unit. The first action value is the torque of the rotary drive unit detected by the torque sensor. The second action value is the temperature of the rotary drive unit detected by the temperature sensor.

8. A computer-readable medium storing an anomaly detection program executed by a processing device and detecting anomalies in a substrate processing apparatus comprising a first actuating member and a second actuating member used in a process involving a substrate using a processing liquid, characterized in that, The anomaly detection procedure causes the processing device to perform the following processing: The process of obtaining the first action value of the first action component and the second action value of the second action component that are related to each other; as well as Based on the obtained first action value and second action value, it is determined whether an abnormal processing has occurred. The process for determining whether the anomaly has occurred includes determining whether an anomaly has occurred based on the proportion of data points in the data points determined according to the group of the first action value and the second action value that exceed an allowable range determined based on the correlation between the first action value and the second action value. The anomaly detection procedure further causes the processing device to perform the following processing: selecting a second action value from the acquired second action values ​​that meets a predetermined benchmark, and selecting a first action value from the acquired first action values ​​that corresponds to the selected second action value. Determining whether the anomaly has occurred includes judging whether an anomaly has occurred based on the change in the time series of the selected first action value. The first actuating component includes a torque sensor that detects the torque of the rotary drive unit. The second actuating component includes a temperature sensor that detects the temperature of the rotary drive unit. The first action value is the torque of the rotary drive unit detected by the torque sensor. The second action value is the temperature of the rotary drive unit detected by the temperature sensor.

Images