Data processing device, substrate processing system, data processing method, and computer-readable program

Abstract

A data reception unit (722) of a data processing device (72) receives sensor time-series data and control time-series data. The sensor time-series data is time-series data of the output value of a sensor that measures a device state in a substrate processing device (1). The control time-series data is time-series data of the control state of the substrate processing device (1). A time synchronization unit (724) performs time stamp processing on the sensor time-series data and the control time-series data. A cycle synchronization unit (725) resamples short-cycle time-series data, which is time-series data having a shorter cycle between the sensor time-series data and the control time-series data, at the cycle of long-cycle time-series data, which is time-series data having a longer cycle. This makes it possible to facilitate acquisition of the correlation between the control state and the device state in the substrate processing device (1).

Description

Data Processing Apparatus, Substrate Processing System, Data Processing Method, and Computer Readable Program

[0001] The present invention relates to a technique for processing data measured in a substrate processing apparatus. [Reference to Related Application] This application claims the benefit of priority from Japanese Patent Application JP2024-164284 filed on September 20, 2024, and all disclosures of the said application are incorporated herein.

[0002] Conventionally, in the manufacturing process of a semiconductor substrate (hereinafter simply referred to as "substrate"), various processes are performed on the substrate by a substrate processing apparatus. In the said substrate processing apparatus, time-series data of various measured values such as temperature, humidity, vibration of the apparatus, and flow rate of the processing liquid are acquired and used for various purposes such as control of the apparatus and failure prediction.

[0003] For example, in Japanese Unexamined Patent Application Publication No. 2024-2710 (Document 1), a technique for comparing time-series data representing the state of each of a plurality of substrate processing apparatuses in order to analyze the individual differences of the plurality of substrate processing apparatuses is proposed. Specifically, the time-series data of the apparatus state output from the same type of sensors (for example, temperature sensors) provided in the plurality of substrate processing apparatuses are analyzed by the dynamic time warping method to obtain the similarity. Thereby, even when the lengths of the plurality of time-series data are somewhat different, the similarity of the apparatus states of the plurality of substrate processing apparatuses can be obtained.

[0004] By the way, among the sensors for measurement in the substrate processing apparatus, some have a relatively long sampling period (for example, 100 milliseconds to 1 second) like a temperature sensor, and some have a very short sampling period (for example, 0.1 milliseconds to 0.25 milliseconds) like a vibration sensor. Also, the period of the control signal (for example, 1 millisecond to 10 milliseconds) for controlling and operating each component in the substrate processing apparatus may be different from the sampling period of the above-mentioned sensors.

[0005] Therefore, in order to obtain a correlation between time-series data of the control state of the substrate processing device and time-series data of measured values ​​(i.e., the device state of the substrate processing device) measured by sensors, the operator must visually compare these time-series data and manually match data values ​​at the same time. However, since these time-series data are transmitted using different communication protocols and are not time-synchronized, obtaining the aforementioned correlation is not easy.

[0006] The present invention aims to facilitate the acquisition of the correlation between the control state and the state of the substrate processing apparatus.

[0007] One aspect of the present invention is a data processing device for processing data measured in a substrate processing device, comprising: a data receiving unit that receives sensor time series data, which is time series data of the output value of a sensor that measures the state of the substrate processing device, and control time series data, which is time series data of the control state of the substrate processing device; a time synchronization unit that performs a timestamp processing on the sensor time series data and the control time series data; and a period synchronization unit that resamples the short-period time series data, which is the time series data with the shorter period among the sensor time series data and the control time series data, to the period of the long-period time series data, which is the time series data with the longer period.

[0008] According to the present invention, it is possible to easily obtain the correlation between the control state and the device state in a substrate processing apparatus.

[0009] Aspect 2 of the present invention is a data processing device of aspect 1, wherein in the resampling of the short-period time series data in the periodic synchronization unit, the effective value of the short-period time series data at the period of the long-period time series data is obtained.

[0010] A third aspect of the present invention is a data processing device according to aspect 1 (or aspect 1 or 2), wherein the short-period time series data is the sensor time series data. The data processing device further includes a data transmission unit that sends the resampled sensor time series data to a processing control unit that controls the substrate processing device. The resampled sensor time series data is used for the control of the substrate processing device by the processing control unit.

[0011] Aspect 4 of the present invention is a data processing device according to aspect 1 (which may be any one of aspects 1 to 3), wherein the sensor is a soft sensor.

[0012] Aspect 5 of the present invention is a data processing device according to aspect 1 (which may be any one of aspects 1 to 4), further comprising a data transmission unit that sends the long-period time series data and the resampled short-period time series data to a display device. The display device displays graphs of the long-period time series data and the resampled short-period time series data in parallel.

[0013] Aspect 6 of the present invention is a data processing device according to aspect 1 (which may be any one of aspects 1 to 5), further comprising a correlation information generation unit that generates correlation information showing the correlation between the long-period time series data and the resampled short-period time series data.

[0014] Embodiment 7 of the present invention is a substrate processing system comprising a substrate processing apparatus for processing a substrate and one data processing apparatus from any one of embodiments 1 to 6 for processing data measured in the substrate processing apparatus.

[0015] Aspect 8 of the present invention is a data processing method for processing data measured in a substrate processing apparatus, comprising: a) receiving sensor time series data, which is time series data of the output value of a sensor that measures the state of the apparatus in the substrate processing apparatus, and control time series data, which is time series data of the control state of the substrate processing apparatus; b) performing a timestamp processing on the sensor time series data and the control time series data; and c) resampling the short-period time series data, which is the time series data with the shorter period among the sensor time series data and the control time series data, by the period of the long-period time series data, which is the time series data with the longer period.

[0016] Aspect 9 of the present invention is a computer-readable program for processing data measured in a substrate processing apparatus, wherein when the program is executed on a computer, the following steps are performed: a) receiving sensor time series data, which is time series data of the output value of a sensor that measures the state of the substrate processing apparatus, and control time series data, which is time series data of the control state of the substrate processing apparatus; b) performing a timestamp processing on the sensor time series data and the control time series data; and c) resampling the short-period time series data, which is the time series data with the shorter period among the sensor time series data and the control time series data, by the period of the long-period time series data, which is the time series data with the longer period.

[0017] The aforementioned objectives, as well as other objectives, features, embodiments, and advantages, will be revealed by the detailed description of the present invention below, with reference to the attached drawings.

[0018] This is a plan view of a substrate processing system according to one embodiment. This is a side view of a substrate processing device. This is a diagram showing the configuration of the control unit. This is a block diagram showing the functions of the control unit. This is a diagram showing an example of the data processing flow. This is a diagram schematically showing an example of sensor time-series data and control time-series data. This is a diagram schematically showing an example of sensor time-series data and control time-series data.

[0019] Figure 1 is an illustrative plan view showing the layout of a substrate processing system 10 according to one embodiment of the present invention. The substrate processing system 10 is a system for processing semiconductor substrates 9 (hereinafter simply referred to as "substrate 9").

[0020] The substrate processing system 10 comprises an indexer block 101 and a processing block 102 coupled to the indexer block 101. The indexer block 101 and the processing block 102 are also called an indexer cell and a processing cell, respectively. The indexer block 101 is also called an Equipment Front End Module (EFEM) unit, etc.

[0021] The indexer block 101 comprises a carrier holding unit 104, an indexer robot 105, and an IR moving mechanism 106. The carrier holding unit 104 holds a plurality of carriers 107 capable of accommodating a plurality of substrates 9. The plurality of carriers 107 (for example, FOUP) are held in the carrier holding unit 104 in an arrangement in a predetermined carrier arrangement direction. The IR moving mechanism 106 moves the indexer robot 105 in the carrier arrangement direction. The indexer robot 105 performs an unloading operation to unload the substrates 9 from the carriers 107, and an loading operation to load the substrates 9 into the carriers 107 held in the carrier holding unit 104. The substrates 9 are transported in a horizontal position by the indexer robot 105.

[0022] The processing block 102 comprises a plurality of processing units 108 (for example, four or more) for processing the substrate 9, and a central robot 109. The plurality of processing units 108 are arranged to surround the central robot 109 in a plan view. Various processing is performed on the substrate 9 in the plurality of processing units 108. The substrate 9 is transported in a horizontal position by the central robot 109. The central robot 109 performs an loading operation to load the substrate 9 into the processing units 108, and an unloading operation to unload the substrate 9 from the processing units 108. Furthermore, the central robot 109 transports the substrate 9 between the plurality of processing units 108. The central robot 109 receives the substrate 9 from the indexer robot 105 and passes the substrate 9 to the indexer robot 105.

[0023] Figure 2 is a side view showing the configuration of the substrate processing apparatus 1. The substrate processing apparatus 1 is one of the multiple processing units 108 shown in Figure 1. In the substrate processing system 10, for example, multiple substrate processing apparatuses 1 having a structure similar to that shown in Figure 2 are provided as multiple processing units 108.

[0024] The substrate processing apparatus 1 is a single-wafer type apparatus that processes substrates 9 one at a time. The substrate processing apparatus 1, for example, supplies a processing liquid to the substrate 9 and performs liquid processing. In this embodiment, the processing liquid is a cleaning liquid, and the liquid processing is a cleaning process for the substrate 9. The substrate processing apparatus 1 comprises a substrate holding section 21, a substrate rotating mechanism 22, a cup section 23, a processing liquid supply section 24, a nozzle moving mechanism 25, and a chamber 11. The substrate holding section 21, the substrate rotating mechanism 22, the cup section 23, the processing liquid supply section 24, and the nozzle moving mechanism 25, etc., are housed in the internal space of the chamber 11. The top of the chamber 11 is provided with an airflow forming section 12 that supplies gas into the internal space to form a downward-flowing airflow (so-called downflow). For example, an FFU (Fan Filter Unit) is used as the airflow forming section 12.

[0025] The substrate holding section 21 holds the substrate 9 in a horizontal position from below. The substrate holding section 21 is a mechanical chuck equipped with, for example, a plurality of chuck pins 211 that directly contact the outer circumference of the substrate 9 to hold the substrate 9. The substrate rotation mechanism 22 rotates the substrate 9 held by the substrate holding section 21 by rotating the substrate holding section 21 around the central axis J1. The substrate rotation mechanism 22 is, for example, a motor. In the substrate processing apparatus 1, the substrate holding section 21 and the substrate rotation mechanism 22 constitute a spin chuck that holds and rotates the substrate 9.

[0026] The cup section 23 comprises an annular cup 231 centered on the central axis J1. The cup 231 is arranged around the entire circumference of the substrate 9 and the substrate holder 21, covering the sides of the substrate 9 and the substrate holder 21. The cup 231 is a liquid receiving container that receives liquids such as processing liquid that are scattered outwards from the rotating substrate 9. A drain port (not shown) is provided at the bottom of the cup 231 to discharge the processing liquid received in the cup 231 to the outside of the chamber 11. The cup 231 moves vertically by a lifting mechanism (not shown).

[0027] The cup portion 23 may include a plurality of cups 231 stacked radially (hereinafter also simply referred to as "radial direction") around the central axis J1. When the cup portion 23 includes a plurality of cups 231, each of the cups 231 can move independently in the vertical direction, and the plurality of cups 231 are switched and used to receive the processing liquid according to the type of processing liquid scattered from the substrate 9.

[0028] The processing liquid supply unit 24 includes a nozzle 241 positioned above the substrate 9. The nozzle 241 discharges the processing liquid toward the upper surface 91 of the substrate 9. In this embodiment, the processing liquid supplied to the substrate 9 from the nozzle 241 is a cleaning liquid used to clean the substrate 9, as described above. In the processing liquid supply unit 24, for example, multiple types of processing liquids may be switched and discharged from the nozzle 241. Alternatively, the processing liquid supply unit 24 may include multiple nozzles 241, each discharging multiple types of processing liquids.

[0029] The processing liquid supply unit 24 further includes a pump 242. The pump 242 is connected to the nozzle 241 and the processing liquid supply source 244 via piping 243. The pump 242 pumps the processing liquid stored in the processing liquid supply source 244 to the nozzle 241. The pump 242 is, for example, a centrifugal magnetic pump.

[0030] A vibration sensor 26 is attached to the pump 242. The vibration sensor 26 measures the vibration of the pump 242. The vibration of the pump 242 measured by the vibration sensor 26 is one of the apparatus states in the substrate processing apparatus 1. The vibration sensor 26 measures the vibration of the pump 242 in three axial directions perpendicular to each other (e.g., X, Y, and Z directions).

[0031] The nozzle moving mechanism 25 is a swinging mechanism that swings the nozzle 241 substantially horizontally in the space above the substrate 9. The nozzle moving mechanism 25 comprises an arm 251 and an arm rotation mechanism 252. The arm 251 is a rod-shaped member that extends substantially horizontally. The nozzle 241 is fixed to one end of the arm 251, and the other end is connected to the arm rotation mechanism 252, which is located radially outside the cup portion 23. The arm rotation mechanism 252 rotates the arm 251 substantially horizontally around a rotation axis that extends vertically. The arm rotation mechanism 252 is, for example, a motor.

[0032] The nozzle movement mechanism 25 reciprocates the nozzle 241, which discharges the processing liquid onto the rotating substrate 9, between a first position that is vertically opposite to the center of the upper surface 91 of the substrate 9, and a second position that is radially outward from the first position. Preferably, the second position is vertically opposite to the outer peripheral region of the upper surface 91 of the substrate 9. This ensures that the processing liquid is supplied substantially evenly over substantially the entire upper surface 91 of the substrate 9.

[0033] As shown in Figure 1, the substrate processing system 10 further comprises a control unit 8. Figure 3 is a diagram showing the configuration of the control unit 8. The control unit 8 has the configuration of a general computer system, including a CPU 81, a GPU 82, a ROM 83, a RAM 84, a fixed disk 85, a display 86, an input unit 87, a reader 88, a communication unit 89, and a bus 80. The CPU 81 performs various arithmetic processing. The GPU 82 performs various arithmetic processing related to image processing. The ROM 83 stores the basic program. The RAM 84 stores various information. The fixed disk 85 stores information. The display 86 is a display unit that displays various information such as images.

[0034] The input unit 87 includes a keyboard 87a and a mouse 87b that accept input from the operator. The reader 88 reads information from a computer-readable recording medium 881 such as an optical disk, magnetic disk, magneto-optical disk, or memory card. The display 86, keyboard 87a, mouse 87b, and reader 88 are connected to the bus 80 via an interface I / F. The communication unit 89 sends and receives signals to and from external devices of the control unit 8. The bus 80 is a signal circuit that connects the CPU 81, GPU 82, ROM 83, RAM 84, fixed disk 85, display 86, input unit 87, reader 88, and communication unit 89.

[0035] In the control unit 8, the program 882 is read in advance from the recording medium 881 via the reader 88 and stored in the fixed disk 85. The program 882 may also be stored in the fixed disk 85 via a network. The CPU 81 and GPU 82 perform arithmetic processing using the RAM 84 and fixed disk 85 according to the program 882. The CPU 81 and GPU 82 function as the arithmetic unit in the control unit 8. Other configurations besides the CPU 81 and GPU 82 that function as the arithmetic unit may also be employed.

[0036] Figure 4 is a block diagram showing some of the functions realized by the execution of program 882 in the control unit 8. The control unit 8 comprises a processing control unit 71 and a data processing device 72. The processing control unit 71 transmits control signals to the substrate processing device 1 and controls each component of the substrate processing device 1. The data processing device 72 is a device that processes data measured by various sensors, etc., in the substrate processing device 1. The data processing device 72 may be realized, for example, by a gateway board incorporated into the computer system that constitutes the control unit 8.

[0037] The data processing device 72 includes a storage unit 721, a data receiving unit 722, a data transmission unit 723, a time synchronization unit 724, a periodic synchronization unit 725, and a correlation information generation unit 726. The storage unit 721 is implemented by a RAM 84 and a fixed disk 85, etc. The data receiving unit 722, the data transmission unit 723, the time synchronization unit 724, the periodic synchronization unit 725, and the correlation information generation unit 726 are implemented by a CPU 81, a GPU 82, a ROM 83, a RAM 84, a fixed disk 85, and their peripheral configurations.

[0038] Next, the data processing flow by the data processing device 72 will be explained with reference to Figures 5, 6A, and 6B. Figure 5 is a diagram showing an example of the data processing flow. Figures 6A and 6B are schematic diagrams showing examples of sensor time-series data and control time-series data, which will be described later.

[0039] When data is processed by the data processing device 72, first, the device status of the substrate processing device 1 is measured by a sensor, and the measured value from the sensor is continuously output to the data processing device 72. In the data processing device 72, the time-series data of the output value from the sensor (hereinafter also referred to as "sensor time-series data") is received by the data receiving unit 722. In addition, the control signals sent from the processing control unit 71 to the substrate processing device 1 are also continuously output to the data processing device 72. In the data processing device 72, the time-series data of the control signals from the processing control unit 71 (i.e., time-series data of the control status of the substrate processing device 1, hereinafter also referred to as "control time-series data") is received by the data receiving unit 722 (step S11).

[0040] In this embodiment, the sensor described above is a vibration sensor 26 attached to the pump 242, and the sensor time-series data is time-series data of the vibration of the pump 242. The vibration sensor 26 measures the vibration of the pump 242 at a very short period and outputs it to the data processing device 72. The measurement period (i.e., sampling period) by the vibration sensor 26 is, for example, 0.1 milliseconds. The data receiving unit 722 may receive multiple sensor time-series data from multiple types of sensors provided in the substrate processing device 1.

[0041] Furthermore, in this embodiment, the control signal from the processing control unit 71 is a signal that controls the operation of the pump 242, and the control time-series data is time-series data of the control state of the pump 242. The processing control unit 71 outputs the control signal to the substrate processing device 1 and the data processing device 72 at a period longer than the sampling period of the vibration sensor 26. The output period of the control signal from the processing control unit 71 is, for example, 10 milliseconds. The data receiving unit 722 may receive multiple control time-series data corresponding to multiple configurations of the substrate processing device 1.

[0042] The sensor time-series data and control time-series data received by the data receiving unit 722 are sent to the time synchronization unit 724. The time synchronization unit 724 performs timestamp processing on the sensor time-series data and control time-series data to synchronize the time between them (step S12).

[0043] FIG. 6A is a diagram schematically showing sensor time-series data and control time-series data for which time synchronization has been performed. The upper graph in FIG. 6A shows the sensor time-series data, and the lower graph shows the control time-series data. The horizontal axis in FIG. 6A indicates time, and the vertical axis of the upper graph indicates the measured value of the acceleration output from the vibration sensor 26. In FIG. 6A, only one acceleration among the accelerations in three axial directions output from the vibration sensor 26 is shown as the sensor time-series data (the same applies to FIG. 6B described later). Also, the vertical axis of the lower graph in FIG. 6A indicates the rotational speed of the pump 242 instructed by the control signal transmitted from the process control unit 71 to the pump 242. In FIG. 6A, the timing at which the control signal is transmitted is indicated by a black circle (the same applies to FIG. 6B described later).

[0044] As shown in FIG. 6A, the period of the sensor time-series data (i.e., the sampling period) is shorter than the period of the control time-series data (i.e., the time between each two adjacent black circles in the horizontal axis direction in FIG. 6A). If the time-series data with the shorter period among the sensor time-series data and the control time-series data is called "short-period time-series data" and the time-series data with the longer period is called "long-period time-series data", in the present embodiment, the sensor time-series data is short-period time-series data and the control time-series data is long-period time-series data. In FIG. 6A, for the sake of illustration, the period of the sensor time-series data is drawn longer than the actual one.

[0045] The sensor time-series data and the control time-series data for which the timestamp processing has been completed are sent to the period synchronization unit 725. In the period synchronization unit 725, the short-period time-series data is resampled at the period of the long-period time-series data (step S13). That is, in step S13, the period synchronization between the sensor time-series data and the control time-series data is performed. In the present embodiment, the sensor time-series data with a sampling period of 0.1 milliseconds is resampled at 10 milliseconds, which is the period of the control time-series data.

[0046] The resampling in step S13 may be performed by various known methods. For example, the resampling is performed by thinning out, averaging, or obtaining the effective value of the short-period time-series data in accordance with the period of the long-period time-series data. In the present embodiment, the sensor time-series data is resampled as shown in FIG. 6B by obtaining the effective value at a period of 10 milliseconds, which is the period of the control time-series data.

[0047] Specifically, in the period synchronization unit 725, first, the sensor time-series data is divided into a plurality of sections every 10 milliseconds. The plurality of sections of the sensor time-series data respectively correspond to a plurality of transmission timings of the control signal in the control time-series data (that is, the black circles in FIG. 6A). Then, in each section of the sensor time-series data, after the square of the data value (that is, the instantaneous value) is integrated for 10 milliseconds, the square root of the value obtained by dividing the integrated value by 10 milliseconds (that is, the effective value) is obtained as the resampled data value.

[0048] When step S13 ends, the long-period time-series data and the resampled short-period time-series data are sent from the period synchronization unit 725 to the storage unit 721 and stored in the storage unit 721. Then, these time-series data are sent to the display 86 by the data transmission unit 723, and graphs of these time-series data are displayed in parallel on the screen of the display 86 (step S14).

[0049] In the present embodiment, a graph of the control time-series data with a period of 10 milliseconds and a graph of the sensor time-series data resampled at a period of 10 milliseconds are displayed in parallel on the screen of the display 86. Note that the parallel display of a plurality of graphs means a display method in which a plurality of graphs can be visually recognized without the operator switching the screen display, such as arranging the plurality of graphs vertically and / or horizontally, or collectively displaying a plurality of data groups corresponding to the plurality of graphs in one graph.

[0050] If, for example, an operator of the substrate processing system 10 experiences a malfunction in the processing result of the substrate 9 by the substrate processing apparatus 1 that is thought to be caused by an abnormality in the flow rate of the processing liquid, they will check the display on the display 86. On the display 86, as described above, the time-synchronized and period-synchronized control time-series data of the pump 242 and the sensor time-series data of the vibration sensor 26 are displayed in parallel. By comparing the control time-series data and the sensor time-series data on the display 86, the operator can easily grasp the correlation between the control state and the apparatus state of the substrate processing apparatus 1. As a result, if the cause of the abnormality in the flow rate of the processing liquid is an abnormality in the control of the pump 242, or an abnormal operation contrary to the control of the pump 242, the cause can be easily identified.

[0051] In the substrate processing system 10, as described above, if the short-period time-series data resampled in the data processing device 72 is sensor time-series data, the substrate processing device 1 may be controlled using the resampled sensor time-series data. Specifically, the resampled sensor time-series data is sent to the processing control unit 71 by the data transmission unit 723, and the resampled sensor time-series data is used by the processing control unit 71 to control the substrate processing device 1.

[0052] In this embodiment, the control signal for the pump 242, sent from the processing control unit 71 to the substrate processing device 1, is adjusted based on the resampled sensor time-series data of the vibration sensor 26, and feedback control of the pump 242 is performed. As described above, the period of the resampled sensor time-series data is the same as the period of the control time-series data (i.e., the period of the control signal). Therefore, feedback control that suitably reflects the operation of the pump 242 is performed.

[0053] In the data processing device 72, the correlation information generation unit 726 may generate correlation information showing the correlation between long-period time series data and resampled short-period time series data. In this embodiment, the correlation information generation unit 726 generates correlation information showing the correlation between the control time series data of the pump 242 and the resampled sensor time series data of the vibration sensor 26. The generation of this correlation information may be performed by various known methods (for example, methods using Kalman filters or neural networks). As described above, the period of the resampled sensor time series data is the same as the period of the control time series data. Therefore, the correlation information can be automatically generated by the correlation information generation unit 726 without any preprocessing such as period synchronization by the operator.

[0054] The correlation information generated by the correlation information generation unit 726 may be, for example, a function showing the correlation relationship described above, or a data table showing the correlation relationship described above. Furthermore, the feedback control of the pump 242 described above may be performed based on the correlation information and the resampled sensor time-series data of the vibration sensor 26.

[0055] In step S11, if the number of sensor time-series data and control time-series data received by the data receiving unit 722 is three or more, then in step S12, timestamp processing is performed on the three or more time-series data. In step S13, the remaining time-series data (i.e., two or more time-series data) from the three or more time-series data, excluding the one with the longest period, are resampled to the period of the one time-series data with the longest period. The time-series data with the longest period does not necessarily have to be sensor time-series data; it may be control time-series data. In step S14, the one time-series data with the longest period and the remaining resampled time-series data are displayed in parallel on the screen of the display 86.

[0056] In the data processing device 72, if two or more sensor time-series data are received in step S11, time synchronization and period synchronization of the two or more sensor time-series data are also performed in steps S12 and S13. Therefore, even if the communication protocols and sampling periods of the two or more sensors that acquire the two or more sensor time-series data are different, the two or more sensor time-series data can be handled in substantially the same manner.

[0057] Furthermore, in step S11 described above, if the data receiving unit 722 receives two or more sensor time-series data, the correlation information generation unit 726 may generate correlation information showing the correlation between the two sensor time-series data. In this case, one of the two sensors that acquire the two sensor time-series data may be provided in the substrate processing device 1 only during trial operation of the substrate processing device 1, and may be omitted when the substrate processing device 1 is actually used for continuous processing of the substrate 9. Then, based on the output from the other of the two sensors and the correlation information described above, sensor time-series data corresponding to the omitted sensor may be acquired (i.e., estimated). In other words, the omitted sensor may be replaced with a soft sensor.

[0058] In steps S11 to S14 described above, the sensor measurement target corresponding to the sensor time-series data does not necessarily have to be the vibration of the pump 242, and can be changed in various ways. For example, the measurement target may be the temperature and humidity inside the chamber 11 of the substrate processing apparatus 1, the flow rate and velocity of the gas supplied from the airflow forming unit 12, the concentration and flow rate of the processing liquid discharged from the nozzle 241, etc. Alternatively, the measurement target may be something that is difficult to measure in real time during the processing of the substrate 9 (for example, the amount of etching on the upper surface 91 of the substrate 9, or the in-plane uniformity of the processing on the upper surface 91 of the substrate 9, etc.), in which case the sensor time-series data is acquired using a soft sensor.

[0059] Furthermore, the controlled object corresponding to the control time-series data does not necessarily have to be the pump 242, but may be various components of the substrate processing apparatus 1. For example, the controlled object may be the substrate rotation mechanism 22, the processing liquid supply unit 24, the nozzle moving mechanism 25, the airflow forming unit 12, etc.

[0060] As described above, the data processing device 72 is a device that processes data measured in the substrate processing device 1. The data processing device 72 comprises a data receiving unit 722, a time synchronization unit 724, and a period synchronization unit 725. The data receiving unit 722 receives sensor time-series data and control time-series data. The sensor time-series data is time-series data of the output values ​​of a sensor (vibration sensor 26 in the above example) that measures the device state in the substrate processing device 1. The control time-series data is time-series data of the control state of the substrate processing device 1 (control state of pump 242 in the above example).

[0061] The time synchronization unit 724 performs timestamp processing on the sensor time series data and the control time series data. The period synchronization unit 725 resamples the short-period time series data, which is the time series data with the shorter period among the sensor time series data and the control time series data, to the period of the long-period time series data, which is the time series data with the longer period. As a result, the period of the long-period time series data and the period of the resampled short-period time series data become the same, making it easy to obtain the correlation between the control state and the device state in the substrate processing apparatus 1.

[0062] As described above, in the resampling of short-period time-series data in the periodic synchronization unit 725, it is preferable to obtain the RMS value of the short-period time-series data at the period of the long-period time-series data. This makes it possible to suppress the effects of noise, distortion, and other factors in the short-period time-series data. As a result, it is possible to suitably obtain the correlation between the control state and the device state in the substrate processing apparatus 1.

[0063] In the example described above, the short-period time-series data is sensor time-series data. The data processing device 72 further includes a data transmission unit 723. Preferably, the data transmission unit 723 sends the resampled sensor time-series data to a processing control unit 71 that controls the substrate processing device 1, and the resampled sensor time-series data is used by the processing control unit 71 to control the substrate processing device 1. This allows for suitable feedback control of the operation of the substrate processing device 1.

[0064] As described above, it is also preferable that the sensor is a soft sensor. This makes it easy to obtain a correlation between the control state of the substrate processing apparatus 1 and the state of the substrate processing apparatus 1, even when the apparatus 1 does not have an actual sensor installed.

[0065] In the above example, the data processing device 72 further includes a data transmission unit 723. The data transmission unit 723 sends long-period time-series data and resampled short-period time-series data to a display device (i.e., a display 86). Preferably, the display 86 displays graphs of the long-period time-series data and the resampled short-period time-series data in parallel. By comparing the time-synchronized and period-synchronized control time-series data and sensor time-series data, the operator can easily grasp the correlation between the control state and the device state in the substrate processing device 1. As a result, when an operational abnormality occurs in the substrate processing device 1, the cause can be easily identified.

[0066] As described above, it is preferable that the data processing device 72 includes a correlation information generation unit 726. The correlation information generation unit 726 generates correlation information that shows the correlation between long-period time series data and resampled short-period time series data. This allows the correlation information to be acquired automatically. As a result, it becomes easy to control the device and predict failures using the correlation information.

[0067] The above-described substrate processing system 10 comprises a substrate processing device 1 that performs processing on the substrate 9, and the above-described data processing device 72 that processes data measured in the substrate processing device 1. This allows for the acquisition of a correlation between the control state and the device state of the substrate processing device 1 in the substrate processing system 10.

[0068] The data processing method described above includes a step (S11) of receiving sensor time-series data, which is time-series data of the output value of a sensor that measures the device state in the substrate processing apparatus 1, and control time-series data, which is time-series data of the control state of the substrate processing apparatus 1; a step (S12) of performing a timestamp processing on the sensor time-series data and the control time-series data; and a step (S13) of resampling the short-period time-series data, which is the time-series data with the shorter period among the sensor time-series data and the control time-series data, to the period of the long-period time-series data, which is the time-series data with the longer period. This makes it easy to obtain the correlation between the control state and the device state in the substrate processing apparatus 1, as described above.

[0069] When the above-described program 882 is executed on the computer, the following steps are performed: (step S11) receiving sensor time-series data, which is time-series data of the output value of a sensor that measures the state of the substrate processing apparatus 1, and control time-series data, which is time-series data of the control state of the substrate processing apparatus 1; (step S12) performing a timestamp processing on the sensor time-series data and the control time-series data; and (step S13) resampling the short-period time-series data, which is the time-series data with the shorter period among the sensor time-series data and the control time-series data, to the period of the long-period time-series data, which is the time-series data with the longer period. As a result, as described above, it is possible to easily obtain the correlation between the control state and the state of the substrate processing apparatus 1.

[0070] Various modifications are possible to the data processing device 72, the substrate processing system 10, the data processing method, and the program 882 described above.

[0071] For example, displaying the graph on the display 86 in step S14 is not necessarily required.

[0072] Furthermore, the data processing device 72 does not necessarily need to generate the aforementioned correlation information; for example, the correlation information may be generated by another device. In this case, the correlation information generation unit 726 may be omitted from the data processing device 72.

[0073] In the substrate processing system 10, control of the substrate processing apparatus 1 using the resampled sensor time-series data is not necessarily required.

[0074] The substrate processing apparatus 1 may be used to process glass substrates used in flat panel displays such as liquid crystal displays or organic electroluminescence (EL) displays, or glass substrates used in other displays, in addition to semiconductor substrates. Furthermore, the substrate processing apparatus 1 may be used to process substrates for optical discs, magnetic discs, magneto-optical discs, photomasks, ceramic substrates, and solar cell substrates.

[0075] The data processing device 72 does not necessarily have to be implemented by the computer that implements the processing control unit 71, and may be provided in a separate device from the processing control unit 71. Furthermore, the data processing device 72 does not necessarily have to be provided in the substrate processing system 10 together with the substrate processing device 1, and may be used as a device independent of the substrate processing device 1.

[0076] The configurations in the above embodiments and each modified example may be combined as appropriate, as long as they do not contradict each other.

[0077] Although the invention has been described in detail, the above description is illustrative and not limiting. Therefore, it can be said that numerous modifications and embodiments are possible as long as they do not deviate from the scope of the present invention.

[0078] 1 Substrate processing device 9 Substrate 10 Substrate processing system 26 Vibration sensor 71 Processing control unit 72 Data processing device 86 Display 722 Data receiving unit 723 Data transmitting unit 724 Time synchronization unit 725 Period synchronization unit 726 Correlation information generation unit 882 Program S11-S14 Steps

Claims

1. A data processing device for processing data measured in a substrate processing device, comprising: a data receiving unit that receives sensor time-series data, which is time-series data of the output value of a sensor that measures the state of the substrate processing device, and control time-series data, which is time-series data of the control state of the substrate processing device; a time synchronization unit that performs timestamp processing on the sensor time-series data and the control time-series data; and a period synchronization unit that resamples the short-period time-series data, which is the time-series data with the shorter period among the sensor time-series data and the control time-series data, to the period of the long-period time-series data, which is the time-series data with the longer period.

2. A data processing device according to claim 1, wherein in the resampling of the short-period time series data in the periodic synchronization unit, the effective value of the short-period time series data at the period of the long-period time series data is obtained.

3. A data processing device according to claim 1, wherein the short-period time series data is the sensor time series data, the data processing device further comprises a data transmission unit that sends the resampled sensor time series data to a processing control unit that controls the substrate processing device, and the resampled sensor time series data is used for the control of the substrate processing device by the processing control unit.

4. A data processing device according to claim 1, wherein the sensor is a soft sensor.

5. A data processing device according to claim 1, further comprising a data transmission unit for sending the long-period time series data and the resampled short-period time series data to a display device, wherein the display device displays graphs of the long-period time series data and the resampled short-period time series data in parallel.

6. A data processing device according to claim 1, further comprising a correlation information generation unit that generates correlation information showing the correlation between the long-period time series data and the resampled short-period time series data.

7. A substrate processing system comprising: a substrate processing device for processing a substrate; and a data processing device according to any one of claims 1 to 6 for processing data measured in the substrate processing device.

8. A data processing method for processing data measured in a substrate processing apparatus, comprising: a) receiving sensor time series data, which is time series data of the output value of a sensor that measures the state of the apparatus in the substrate processing apparatus, and control time series data, which is time series data of the control state of the substrate processing apparatus; b) performing a timestamp processing on the sensor time series data and the control time series data; and c) resampling the short-period time series data, which is the time series data with the shorter period among the sensor time series data and the control time series data, with the period of the long-period time series data, which is the time series data with the longer period.

9. A computer-readable program for processing data measured in a substrate processing apparatus, wherein when the program is executed on a computer, the following steps are performed: a) receiving sensor time-series data, which is time-series data of the output value of a sensor that measures the state of the substrate processing apparatus, and control time-series data, which is time-series data of the control state of the substrate processing apparatus; b) performing a timestamp processing on the sensor time-series data and the control time-series data; and c) resampling the short-period time-series data, which is the time-series data with the shorter period among the sensor time-series data and the control time-series data, to the period of the long-period time-series data, which is the time-series data with the longer period.

Images