State determination method and monitoring device

Abstract

To correctly determines a sign of an abnormality of a vacuum pump.SOLUTION: A state determination method for determining a state of a vacuum pump in a substrate processing device for performing a substrate processing on the basis of a recipe containing a plurality of steps contains steps of: (a) acquiring a vibration data from a sensor for detecting a vibration of the vacuum pump in a specification step of the plurality of steps; (b) calculating a vibration change feature amount of a predetermined frequency band from a vibration power spectrum of the predetermined frequency calculated on the basis of the vibration data; (c) calculating a first gap level indicating a gap level between a normal vibration feature amount previously calculated by executing the step (b) on the basis of the vibration data acquired from the sensor for detecting the vibration of a new vacuum pump or the vacuum pump at the time of replacement and a vibration change feature amount of the predetermined frequency band calculated in the step (b) after that; and (d) outputting a warning on a state of the vacuum pump in the case where the first gap level is a first threshold value that is previously set or larger.SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] The present disclosure relates to a status determination method and a monitoring device. [Background technology]

[0002] For example, Patent Document 1 describes a method for improving the accuracy of failure prediction in semiconductor manufacturing equipment by acquiring physical quantities from various sensors, calculating waveform features of each sensor value, comparing the values ​​with normal data to calculate the degree of deviation and the rate of increase or decrease, and outputting an alarm when the degree of deviation from the normal data exceeds an upper limit.

[0003] Patent Document 2 provides a technology for performing a predetermined process on a substrate by executing a process recipe including multiple steps, and for determining signs of abnormality based on vibration data detected by a vibration sensor at a specified step among the steps constituting the process recipe. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2021-102817 [Patent Document 2] International Publication No. 2020 / 188747 Summary of the Invention [Problem to be solved by the invention]

[0005] The present disclosure provides a technique for correctly determining a sign of an abnormality in a vacuum pump. [Means for solving the problem]

[0006] According to one aspect of the present disclosure, there is provided a status determination method for determining a status of a vacuum pump in a substrate processing apparatus having a vacuum pump and processing substrates based on a recipe including a plurality of steps, the status determination method including: (a) acquiring vibration data from a sensor that detects vibration of the vacuum pump in a specific step among the plurality of steps; (b) calculating a vibration power spectrum of a predetermined frequency based on the vibration data, and calculating a vibration change feature amount in a predetermined frequency band from the vibration power spectrum of the predetermined frequency; (c) calculating a first deviation degree indicating a degree of deviation between a normal vibration feature amount calculated in advance by executing (b) and a vibration change feature amount in a predetermined frequency band calculated in (b) thereafter, based on vibration data acquired from a sensor that detects vibration of the vacuum pump when the vacuum pump is new or has been replaced; and (d) outputting a warning regarding the status of the vacuum pump if the first deviation degree is equal to or greater than a predetermined first threshold value. Effect of the Invention

[0007] According to one aspect, a sign of an abnormality in a vacuum pump can be correctly determined. [Brief description of the drawings]

[0008] [Figure 1] 1 is a schematic cross-sectional view showing an example of a substrate processing apparatus according to an embodiment; [Diagram 2] 1 is a configuration diagram illustrating an example of a substrate processing system according to an embodiment. [Diagram 3] FIG. 2 is a diagram showing an example of a hardware configuration of a monitoring device according to an embodiment. [Figure 4] FIG. 2 is a diagram showing an example of a functional configuration of a monitoring device according to an embodiment. [Diagram 5] 4 is a flowchart showing an example of a state determination method according to an embodiment. [Figure 6] 6 is a diagram showing an example of an abnormality determination process in S6 of FIG. 5 according to an embodiment. [Figure 7] FIG. 4 is a diagram illustrating an example of a vibration change rate over an entire frequency band according to an embodiment. [Figure 8] FIG. 4 is a diagram showing an example of a vibration change rate in each frequency band according to an embodiment. [Figure 9] 1 is a two-dimensional map illustrating vibration power spectrum and rate of vibration change during substrate processing according to one embodiment. [Figure 10] FIG. 11 is a diagram showing an example of a result of a state determination method according to an embodiment. [Figure 11] FIG. 13 is a diagram showing an example of a multivariate anomaly detection model. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Hereinafter, an embodiment of the present disclosure will be described with reference to the drawings. In the drawings, the same components are denoted by the same reference numerals, and duplicated descriptions may be omitted.

[0010] [Substrate processing equipment] A batch-type substrate processing apparatus that performs a desired film formation process on a plurality of substrates will be described with reference to Fig. 1. Fig. 1 is a schematic cross-sectional view showing an example of a substrate processing apparatus 10 according to an embodiment. The substrate processing apparatus 10 accommodates a plurality of wafers 2 in a processing vessel 11, and simultaneously forms films on the plurality of wafers. The wafers 2 are an example of substrates.

[0011] The substrate processing apparatus 10 is a batch-type vertical heat processing apparatus that processes a plurality of wafers 2. The substrate processing apparatus 10 has a processing vessel 11 that accommodates the wafers 2 and forms an internal space in which the wafers 2 are processed, a lid 20 that airtightly closes an opening at the bottom end of the processing vessel 11, and a boat 30 that holds the wafers 2. The wafers 2 are, for example, semiconductor substrates (also simply referred to as "substrates"), such as silicon wafers.

[0012] The processing vessel 11 has a cylindrical processing vessel body 12 with a ceiling and an open lower end. The processing vessel body 12 is made of, for example, quartz. A flange portion 13 is formed at the lower end of the processing vessel body 12. The processing vessel 11 also has, for example, a cylindrical manifold 14. The manifold 14 is made of, for example, stainless steel. A flange portion 15 is formed at the upper end of the manifold 14, and the flange portion 13 of the processing vessel body 12 is installed on the flange portion 15. A seal member 16, such as an O-ring, is arranged between the flange portion 15 and the flange portion 13.

[0013] The lid 20 is airtightly attached to the opening at the lower end of the manifold 14 via a seal member 21 such as an O-ring. The lid 20 is made of, for example, stainless steel. A through hole is formed in the center of the lid 20, passing vertically through the lid 20. A rotating shaft 24 is disposed in the through hole. The gap between the lid 20 and the rotating shaft 24 is sealed by a magnetic fluid seal unit 23. The lower end of the rotating shaft 24 is rotatably supported by an arm 26 of a lifting unit 25. A rotating plate 27 is provided at the upper end of the rotating shaft 24. A boat 30 is placed on the rotating plate 27 via a heat retention stand 28.

[0014] The boat 30 holds a plurality of wafers 2 in a vertical direction. For example, in the boat 30 capable of holding 200 wafers 2, slots numbered 1 to 200 are arranged vertically. By arranging the wafers 2 in each slot, the plurality of wafers 2 are held horizontally at intervals. The boat 30 is formed of, for example, quartz (SiO2) or silicon carbide (SiC). When the lifting / lowering unit 25 is raised, the lid body 20 and the boat 30 are raised, the boat 30 is carried into the inside of the processing vessel 11, and the opening at the lower end of the processing vessel 11 is sealed by the lid body 20. When the lifting / lowering unit 25 is lowered, the lid body 20 and the boat 30 are lowered, and the boat 30 is carried out to the outside of the processing vessel 11. When the rotating shaft 24 is rotated, the boat 30 rotates together with the rotating plate 27.

[0015] The substrate processing apparatus 10 has three gas supply pipes 40A, 40B, and 40C. The gas supply pipes 40A, 40B, and 40C are made of, for example, quartz (SiO2). The gas supply pipes 40A, 40B, and 40C supply gas into the processing chamber 11. The types of gas will be described later. Note that one gas supply pipe may discharge one type or multiple types of gas in sequence. Also, multiple gas supply pipes may discharge the same type of gas.

[0016] The gas supply pipes 40A, 40B, and 40C include horizontal pipes 43A, 43B, and 43C that horizontally penetrate the manifold 14, and vertical pipes 41A, 41B, and 41C that are vertically disposed inside the processing vessel 11. The vertical pipes 41A, 41B, and 41C include a plurality of gas inlets 42A, 42B, and 42C that are spaced apart in the vertical direction. The various gases supplied to the horizontal pipes 43A, 43B, and 43C are sent to the vertical pipes 41A, 41B, and 41C, and are horizontally discharged from the plurality of gas inlets 42A, 42B, and 42C. The vertical pipe 41C is disposed inside the plasma box 19. The vertical pipes 41A and 41B are disposed inside the processing vessel 11.

[0017] The substrate processing apparatus 10 has an exhaust pipe 45. The exhaust pipe 45 is connected to an exhaust device 48. The exhaust device 48 includes a vacuum pump and exhausts the inside of the processing chamber 11. In FIG. 1, the exhaust device 48 has a dry pump (DRY PUMP) 46 and a turbo molecular pump (TMP) 47 as vacuum pumps. The dry pump 46 roughly evacuates the inside of the processing chamber 11, and the turbo molecular pump (TMP) 47 further evacuates the inside of the processing chamber 11 that has been roughly evacuated.

[0018] In order to exhaust the inside of the processing vessel 11, an exhaust port 18 is formed in the main body 12 of the processing vessel. The exhaust port 18 is disposed so as to face the air inlets 42A, 42B, and 42C. Gas discharged horizontally from the air inlets 42A, 42B, and 42C passes through the exhaust port 18 and is exhausted from an exhaust pipe 45. The exhaust device 48 sucks in the gas inside the processing vessel 11 and sends it to the detoxification device 49. The detoxification device 49 removes harmful components from the exhaust gas and then releases the exhaust gas into the atmosphere.

[0019] The substrate processing apparatus 10 further includes a heating unit 60. The heating unit 60 is disposed outside the processing vessel 11 and heats the inside of the processing vessel 11 from the outside of the processing vessel 11. For example, the heating unit 60 is formed in a cylindrical shape so as to surround the main body 12 of the processing vessel. The heating unit 60 is formed of, for example, a heater. The heating unit 60 heats the inside of the processing vessel 11, thereby improving the processing capacity of the gas supplied into the processing vessel 11.

[0020] An opening 17 is formed in a portion of the circumferential direction of the processing vessel body 12. A plasma box 19 is formed on the side surface of the processing vessel 11 so as to surround the opening 17. The plasma box 19 is formed so as to protrude radially outward from the processing vessel body 12, and is formed, for example, in a U-shape when viewed in the vertical direction.

[0021] A pair of electrodes are arranged on either side of the plasma box 19. The electrode pair is a pair of parallel electrodes installed facing each other on the outside of the plasma box 19. The electrode pair is formed to be elongated in the vertical direction facing each other, similar to the vertical tube 41C. The electrode pair is connected to an RF power supply via a matching box, and a high-frequency voltage is applied from the RF power supply.

[0022] The substrate processing apparatus 10 includes a control device 100. The control device 100 processes computer-executable instructions that cause the substrate processing apparatus 10 to perform various substrate processing steps. The control device 100 may be configured to control each element of the substrate processing apparatus 10 to perform various substrate processing steps. In an embodiment, a part or all of the control device 100 may be included in the substrate processing apparatus 10. The control device 100 may include a processing unit, a data storage unit, and a communication interface. The control device 100 is realized, for example, by a computer. The processing unit may be configured to read a recipe and a program from the data storage unit and execute the read recipe and program to perform various control operations. The recipe and the program may be stored in the data storage unit in advance, or may be acquired via a medium when necessary. The medium may be various storage media readable by a computer, or may be a communication line connected to the communication interface. The processing unit may be a CPU. The data storage unit may be a RAM, a ROM, a HDD, or the like. The communication interface may communicate between the substrate processing apparatus 10, the sensor group 200, and the monitoring device 300 (see FIG. 2) via a communication network such as a LAN.

[0023] [Substrate processing system] Next, the substrate processing system 1 will be described with reference to Fig. 2. Fig. 2 is a configuration diagram showing an example of the substrate processing system 1 according to an embodiment. The substrate processing system 1 has a substrate processing apparatus 10, a sensor group 200, a monitoring device 300, and a data storage unit 400, and these devices are connected to each other so as to be able to communicate data with each other via a communication network N such as the Internet or a LAN. The control device 100 is incorporated in the substrate processing apparatus 10. Although one substrate processing apparatus is shown, two or more substrate processing apparatuses may be provided.

[0024] A plurality of sensors are attached to the substrate processing apparatus 10. The sensor group 200 attached to the substrate processing apparatus 10 includes a film thickness sensor, a temperature sensor, a humidity sensor, a pressure sensor, a vibration sensor, a distance sensor, and the like. In this embodiment, as shown in FIG. 1, a vibration sensor 201 and a temperature sensor 203 are attached near the turbo molecular pump 47 and near the motor that drives the turbo molecular pump 47. In addition, a vibration sensor 202 and a temperature sensor 204 are attached near the dry pump 46 and near the motor that drives the dry pump 46.

[0025] 1, the substrate processing apparatus 10 carries a boat 30 on which a plurality of wafers 2 are placed into a processing vessel 11, and performs film formation on the plurality of wafers 2. After the boat 30 is carried into the processing vessel 11 and a film formation process is simultaneously performed on the plurality of wafers 2, the boat 30 is carried out from the processing vessel 11. This series of processes in which a predetermined process such as film formation is performed on the wafers 2 is called "substrate processing", and a substrate processing number is assigned to identify the substrate processing performed each time the boat 30 is carried into the processing vessel 11. In other words, the substrate processing number is an identification number for the substrate processing.

[0026] Log data of the sensor values ​​measured by the sensor group 200 is time-series data and is stored in the data storage unit 400. The monitoring device 300 acquires necessary log data from the data storage unit 400 and determines the state of the vacuum pumps (the dry pump 46 and the turbo molecular pump 47) of the exhaust device 48. In this embodiment, the monitoring device 300 acquires vibration data detected by the vibration sensors 201, 202 and temperature data detected by the temperature sensors 203, 204 from the data storage unit 400.

[0027] Repeated film formation causes by-products to accumulate inside the vacuum pump. If the amount of accumulated by-products becomes large, the vacuum pump will break down and stop in an emergency, which will lead to a decrease in productivity of the substrate processing apparatus 10. Furthermore, if the amount of accumulated by-products becomes large, the vibration of the vacuum pump will increase. Therefore, it is desirable to detect the vibration state of the vacuum pump, predict an abnormality in the vacuum pump before the vacuum pump stops due to a breakdown or the like, and take measures such as maintenance of the vacuum pump in advance.

[0028] In the conventional anomaly detection method, vibration data is acquired at a specified step in a process recipe, and a normal model is created using features such as the amplitude value of the vibration frequency spectrum in a specified frequency band under normal conditions.Then, the degree of deviation between the feature value of the specified frequency band of the vibration data and the normal model is calculated, and if the degree of deviation exceeds a predetermined threshold, it is determined to be a sign of an anomaly.

[0029] However, the vibration of the vacuum pump differs for each step of a recipe that includes multiple steps. The state of the vibration of the vacuum pump also changes when the substrate processing apparatus 10 is stopped or when the substrate processing apparatus 10 is maintained. For example, after the substrate processing apparatus 10 is restarted and left idling for a long period of time, the vibration of the vacuum pump may be abnormal for a while. Therefore, in the conventional method, the vibration of the vacuum pump varies depending on the selection of a specified step in the recipe, making it difficult to perform highly accurate monitoring.

[0030] In addition, since the normal region of the normal model is determined based on statistics such as the amplitude of a specified frequency in the normal state, the characteristics of the amplitude in the normal state have a large impact on the normal model, and the monitoring accuracy of the vacuum pump is easily affected by the learning results of the normal model.

[0031] In addition, in conventional methods, vibration is observed using only some feature quantities (fixed frequency, Nth harmonic, root mean square (RMS, etc.)), making it difficult to grasp the change in vibration over time in the entire frequency band. If an anomaly judgment model is created by the sum of the amplitude of a specific frequency and the amplitude of an arbitrary frequency range, when calculating the sum of the frequency range, the sum of the frequencies where the amplitude increases and decreases approaches 0 in the calculation, and the effect of the fluctuation does not appear in the sum result, so that the vibration change in the frequency range may not be accurately expressed.

[0032] Furthermore, the conventional method does not monitor the variation between substrate processing, and cannot determine that a large change between substrate processing is a sign of a vacuum pump abnormality. Also, when the temperature of the vacuum pump drops due to the stop of the substrate processing apparatus 10 or maintenance of the substrate processing apparatus 10, the vibration of the vacuum pump also changes. If it is determined that a sign of a vacuum pump abnormality is present when both the temperature sensor and the vibration sensor indicate an abnormality, there may be cases where the vacuum pump is determined to be abnormal due to the effects of maintenance of the substrate processing apparatus 10, even when the vacuum pump is normal.

[0033] Therefore, in this embodiment, a vibration power spectrum (vibration power intensity) of a predetermined frequency is calculated based on the vibration data, and a vibration change feature amount in a predetermined frequency band is calculated from the vibration power spectrum of the predetermined frequency. Then, a first deviation degree is calculated, which indicates the degree of deviation between the normal vibration feature amount calculated based on vibration data acquired from a sensor that detects the vibration of a new (new or replaced) vacuum pump, and the vibration change feature amount in the predetermined frequency band calculated from vibration data detected during the subsequent substrate processing. Then, if the first deviation degree is equal to or greater than a preset first threshold value, it is determined to be a sign of an abnormality in the vacuum pump, and a warning is output.

[0034] In addition, the vacuum pump is judged to have an abnormality or a sign of an abnormality based on the vibration data remaining after excluding vibration data corresponding to temperature data outside a preset temperature range from among the temperature data.

[0035] [Hardware configuration] The monitoring device 300 is realized by, for example, a computer having a hardware configuration shown in Fig. 3. Fig. 3 is a diagram showing an example of the hardware configuration of the monitoring device 300 according to the embodiment. The monitoring device 300 is configured by, for example, a computer, and has a CPU (Central Processing Unit) 311, a ROM (Read Only Memory) 312, and a RAM (Random Access Memory) 313. The monitoring device 300 also has an I / O port 314, an operation panel 315, and an HDD (Hard Disk Drive) 316. Each part is connected by a bus B.

[0036] The CPU 311 is a calculation device that realizes the overall control and functions of the monitoring device 300 by reading out programs and data from storage devices such as the ROM 312 and the HDD 316 onto the RAM 313 and executing substrate processing.

[0037] The ROM 312 is a storage medium that includes an EEPROM (Electrically Erasable Programmable ROM), a flash memory, a hard disk, etc., and stores programs used by the CPU 311. The RAM 313 functions as a work area, etc., for the CPU 311. The programs used by the CPU 311 include a program for executing a display method, which will be described later.

[0038] The I / O port 314 acquires measurement data such as vibration data and temperature data measured by the sensor group 200 and transmits the data to the CPU 311. In addition, an operation panel 315 through which a user operates the monitoring device 300 is connected to the I / O port 314.

[0039] The HDD 316 is an auxiliary storage device and may store programs, etc. The HDD 316 may also store log data measured by the sensor group 200.

[0040] [Function configuration] Next, the functional configuration of the monitoring device 300 will be described with reference to Fig. 4. Fig. 4 shows an example of the functional configuration of the monitoring device 300 according to an embodiment. The monitoring device 300 determines the states of the vacuum pumps, that is, the dry pump 46 and the turbo molecular pump 47, of the substrate processing apparatus 10.

[0041] The monitoring device 300 includes an acquisition unit 301, a pre-processing unit 302, a calculation unit 303, an abnormality determination unit 304, and a display control unit 305. The acquisition unit 301 acquires vibration data detected by a vibration sensor provided near a vacuum pump of the substrate processing apparatus 10 and temperature data detected by a temperature sensor in a specific step among a plurality of steps of a recipe used by the substrate processing apparatus 10 for substrate processing. The vibration data and temperature data may be time-series data. The acquisition unit 301 stores the vibration data and temperature data for each vacuum pump and each sensor in the data storage unit 400 (see FIG. 4).

[0042] In this embodiment, the specific steps are a step when the vacuum pump is idle and a step in which the pressure in the processing vessel 11 is reduced from atmospheric pressure to a predetermined pressure. However, the specific steps are not limited to these, and steps in which the vacuum pump operates under approximately the same conditions are identified. In order not to overlook a signal of an abnormality in the vacuum pump, it is preferable to obtain vibration data from both the step in which the vacuum pump is under the highest load (e.g., a step in which the processing vessel is depressurized) and the step in which the vacuum pump is under the lowest load (a step in which the vacuum pump is idle).

[0043] In a specific step, the pre-processing unit 302 excludes vibration data corresponding to temperature data outside a preset temperature range from among temperature data acquired together with the vibration data, and calculates a vibration power spectrum of a predetermined frequency based on the vibration data after the exclusion. The pre-processing unit 302 calculates a vibration change feature amount in a predetermined frequency band from the calculated vibration power spectrum of the predetermined frequency. The pre-processing unit 302 stores the calculated vibration power spectrum of the predetermined frequency and the vibration change feature amount in the predetermined frequency band in the data storage unit 400 for each vacuum pump and each sensor (see FIG. 4).

[0044] In a specific step, the calculation unit 303 learns a normal vibration feature amount based on vibration data acquired from a sensor that detects vibration of a new or replaced vacuum pump. After that, in a specific step, the calculation unit 303 calculates a vibration change feature amount in a predetermined frequency band based on the vibration data acquired from the vibration sensor, and calculates a first deviation degree indicating the degree of deviation from the normal vibration feature amount.

[0045] Note that the vacuum pump at the time of replacement is not limited to when the entire vacuum pump arranged in the substrate processing apparatus 10 is replaced, but also includes, for example, replacement of a trap attached to the vacuum pump, replacement of vacuum pump oil, etc. However, for example, when replacing a trap, this can be realized by attaching a vibration sensor to the trap itself and monitoring the vibration with the monitoring device 300 in the same manner as the vacuum pump.

[0046] The abnormality determination unit 304 determines whether the first deviation is equal to or greater than a preset first threshold. If the abnormality determination unit 304 determines that the first deviation is equal to or greater than the first threshold, it determines that there is a sign of an abnormality in the vacuum pump, and in response, the display control unit 305 outputs a warning about the vacuum pump. Even after the warning is output, the substrate processing continues, and the vacuum pump does not stop. This allows the user to take countermeasures against the output of the warning, such as replacing the vacuum pump, in a timely manner.

[0047] The acquisition unit 301 is realized, for example, by an operation panel 315 and an I / O port 314. The preprocessing unit 302, the calculation unit 303, the abnormality determination unit 304, and the display control unit 305 are realized, for example, by a CPU 311. The data storage unit 400 may be disposed inside the monitoring device 300, and in that case, is realized, for example, by a ROM 312, a RAM 313, and a HDD 316.

[0048] [Status determination method] Next, an example of a state determination method according to an embodiment will be described with reference to Fig. 5. Fig. 5 is a flowchart showing an example of a state determination method according to an embodiment.

[0049] The substrate processing apparatus 10 performs substrate processing based on a recipe including a plurality of steps. A vibration sensor near the vacuum pump detects vibrations due to the rotation of a motor that drives the vacuum pump during substrate processing. The state determination method according to the present embodiment is performed by a monitoring device 300, which determines the state of the vacuum pump based on vibration data detected by the vibration sensor.

[0050] First, in step S1, the acquisition unit 301 collects vibration data detected by a vibration sensor provided near a vacuum pump and temperature data detected by a temperature sensor during substrate processing.

[0051] Next, in step S2, the acquiring unit 301 acquires vibration data and temperature data of the vacuum pump in a specific step among a plurality of steps included in the recipe from the collected vibration data and temperature data. For example, the specific step may be a step when the vacuum pump is idling, and a step in which the pressure in the processing vessel 11 for processing the substrate is reduced from atmospheric pressure to a predetermined pressure.

[0052] Next, in step S3, the pre-processing unit 302 determines whether any of the acquired temperature data falls outside a preset temperature range. If the pre-processing unit 302 determines that any of the acquired temperature data falls outside a preset temperature range, the pre-processing unit 302 excludes the vibration data corresponding to the temperature data from the vibration data to be used for determining a sign of abnormality in the vacuum pump. This allows the vibration data containing a lot of noise when the temperature of the vacuum pump changes due to reasons such as the substrate processing apparatus 10 being stopped to be excluded from the vibration data to be analyzed, thereby improving the accuracy of determining a sign of abnormality in the vacuum pump.

[0053] Next, in step S4, the preprocessing unit 302 calculates the vibration power spectrum (vibration power intensity) of unit time and unit frequency based on the vibration data after the exclusion. The unit time may be set to any time by the user or may be set automatically. The unit frequency may be set to any frequency by the user or may be set automatically.

[0054] Next, in step S5, the calculation unit 303 calculates a vibration change feature amount for the entire frequency band and a vibration change feature amount for a predetermined frequency band, based on the vibration power spectrum for unit time and unit frequency.

[0055] Steps S4 and S5 are an example of a process of calculating a vibration change rate in a predetermined frequency band based on the vibration data, and calculating a vibration change feature amount in the predetermined frequency band from the vibration change rate in the predetermined frequency band.

[0056] In this step, the vibration change feature amount of the predetermined frequency band includes the vibration change feature amount of the entire frequency band and the vibration change feature amount of each frequency band when the entire frequency band is divided into a plurality of frequency bands.

[0057] Next, in step S56, the abnormality determination unit 304 executes an abnormality determination process, which will be described in detail with reference to FIG.

[0058] Fig. 7 is a diagram showing an example of a vibration change rate in the entire frequency band according to one embodiment. The horizontal axis of Fig. 7 is the substrate processing number, which is an identification number of a substrate processing given for each processing of a plurality of wafers 2 held in a boat 30. The vertical axis is the vibration change rate in the entire frequency band for each substrate processing. The vibration change rate is an example of a vibration change feature. The entire frequency band is 0 to 10000 Hz. However, the entire frequency band is not limited to this, and may be a frequency range that the vibration sensor can measure at a predetermined sampling rate.

[0059] Fig. 8 is a diagram showing an example of the vibration change rate of each frequency band according to one embodiment. The horizontal axis of Fig. 8 is the substrate processing number, and the vertical axis is the vibration change rate of each frequency band when the vibration change rate of the entire frequency band for each substrate processing is divided into a plurality of frequency bands. Each frequency band is divided every 1000 Hz.

[0060] FIG. 9(a) is a two-dimensional map showing a vibration power spectrum according to one embodiment, and FIG. 9(b) is a two-dimensional map showing a vibration rate of change according to one embodiment.

[0061] The vibration power spectrum of a predetermined frequency band is divided into vibration power spectra of unit frequencies for each substrate processing. The divided vibration power spectra are visualized and displayed on a two-dimensional map shown in FIG. 9(a) with the unit frequency and the substrate processing number (identification number of the substrate processing) as the two axes.

[0062] For example, in step S4, the calculated vibration power spectrum (vibration power intensity) of unit time and unit frequency may be visualized and displayed on the two-dimensional map by changing the color or shading. The waveform of the vibration power spectrum of unit time and unit frequency can be visualized by color or shading and shown to the user.

[0063] In addition, the vibration change rate calculated from the divided vibration power spectrum is visualized and displayed on a two-dimensional map with the value corresponding to the unit frequency and the substrate processing number (identification number of the substrate processing) as the two axes, as shown in Fig. 9(b). This makes it possible to visualize the vibration change rate between substrate processings, indicated by the substrate processing number on the horizontal axis, using colors or shading to show it to the user.

[0064] In Fig. 9(b), the vertical axis shows the vibration change rate between substrate treatments normalized to the range of 0 to 1 as an example of a value corresponding to a unit frequency. This is a visualization of the vibration power spectrum of all substrate treatments at each frequency normalized to the range of 0 to 1 as the vibration change rate between substrate treatments. This makes it easier for the user to see the vibration change rate between substrate treatments.

[0065] The two-dimensional maps of Figures 9(a) and (b) visualize the vibration power spectrum and vibration change rate for the entire frequency band, allowing the user to easily understand the vibration state of the vacuum pump.

[0066] (Abnormality determination process) Next, details of the abnormality determination process in Fig. 6 that is called when step S6 in Fig. 5 is executed will be described. In the abnormality determination process, in step S10, the calculation unit 303 determines whether a new vacuum pump has been installed. A brand new vacuum pump or a vacuum pump that has just been replaced is a new vacuum pump. If the calculation unit 303 determines that the vacuum pump is new, the process proceeds to step S11. If the calculation unit 303 determines that the vacuum pump is not new, the process proceeds to step S13.

[0067] In step S11, the calculation unit 303 learns the vibration data of the first substrate processing after starting to use the new vacuum pump as normal data. The "vibration data of the first substrate processing" referred to here may be not only the vibration data of the first substrate processing using the new vacuum pump, but also the vibration data of n substrate processing times (1st to nth (n is a positive integer)) using the new vacuum pump.

[0068] Next, in step S12, the calculation unit 303 calculates the vibration feature amount μ in the normal state by the formula (1). When the vibration feature amount μ in the normal state is calculated using the vibration data of n substrate processing operations using the new vacuum pump, the vibration feature amount μ in the normal state is the average value of the vibration feature amounts in the normal state.

[0069]

number

[0070] Next, in step S13, the calculation unit 303 starts vibration monitoring for each substrate processing from the (n+1)th processing, and calculates the vibration power spectrum P i,j When the vibration data during the first to fifth substrate processing are learned as normal data, the substrate processing number i in step S13 starts from 6.

[0071] Next, in step S14, the calculation unit 303 calculates the vibration change rate. The calculation unit 303 calculates the vibration feature amount μ j and the vibration power spectrum P of the i-th substrate processing (i>n) and frequency j i,j Using the above, the vibration change rate θ at frequency j during the i-th substrate processing is calculated. i,j is calculated using formula (2).

[0072]

number

[0073] Next, the calculation unit 303 calculates the vibration change rate θ i,j Using the above, the vibration change feature amount Q in the i-th predetermined frequency band (a to b Hz) for the i-th substrate processing is calculated. The vibration change feature amount Q in the i-th predetermined frequency band (a to b Hz) is shown below. In addition, the vibration change feature amount Q' in the i-th predetermined frequency band (a to b Hz) may be calculated.

[0074]

number

[0075] The calculation unit 303 calculates the difference between the vibration feature amount μ in a normal state and the vibration change feature amount Q' as the first deviation degree. Alternatively, the calculation unit 303 may calculate the difference between the vibration feature amount μ in a normal state and the vibration change feature amount Q' as the first deviation degree.

[0076] It is preferable to also calculate the vibration change feature quantity Q', taking into consideration the possibility that when the vibration change feature quantity Q is calculated as a sum including positive and negative values, the final value becomes 0, resulting in no vibration change.

[0077] In addition, the calculation unit 303 calculates a vibration power spectrum P c and the (i-1)th substrate processing (previous substrate processing), the vibration power spectrum PS of a predetermined frequency band at frequency j b The difference value between the i-th predetermined frequency band (a to b Hz) and the vibration power spectrum PS c is shown below.

[0078]

number

[0079]

number

[0080] Next, in step S15, the abnormality determination unit 304 determines whether the first deviation degree is equal to or greater than a preset first threshold value. If the abnormality determination unit 304 determines that the first deviation degree is equal to or greater than the first threshold value, it determines that the state of the vacuum pump indicates a sign of an abnormality.

[0081] In response to this, in step S16, display control unit 305 outputs a warning about the state of the vacuum pump, and in step S18, causes control device 100 to continue substrate processing, terminating this process. In the example of Fig. 10, the horizontal axis represents substrate processing number (identification number of substrate processing), and the vertical axis represents the vibration change rate. Vibration data during substrate processing of substrate processing numbers 1 to 5 was learned as normal data. For substrate processing numbers 61 to 67 and 76 to 79, the first deviation is equal to or greater than the first threshold, and a warning is output.

[0082] If the abnormality determination unit 304 determines in step S15 that the first deviation is less than the first threshold, the process proceeds to step S17, where it determines whether the second deviation is equal to or greater than a preset second threshold. If the abnormality determination unit 304 determines that the second deviation is equal to or greater than the second threshold, it determines that the state of the vacuum pump indicates a sign of an abnormality.

[0083] In response to this, in step S16, the display control unit 305 outputs a warning about the state of the vacuum pump, and in step S18, causes the control device 100 to continue the substrate processing, thereby terminating this process. In the example of Fig. 10, the second deviation degree becomes equal to or exceeds the second threshold value for substrate processing number 70, and a warning is output.

[0084] When the abnormality determination unit 304 determines in step S17 that the second deviation is less than the second threshold value, the abnormality determination unit 304 causes the control device 100 to continue the substrate processing, and ends this process.

[0085] It is preferable to calculate a plurality of first deviations in step S14. For example, the calculation unit 303 may calculate a first deviation for the entire frequency band indicating the degree of deviation between a vibration feature amount for the entire frequency band in a normal state calculated in advance and a vibration change feature amount for the entire frequency band calculated in each substrate processing after learning in a normal state. In addition, the calculation unit 303 may calculate a first deviation for each frequency band indicating the degree of deviation between a vibration feature amount for each frequency band in a normal state calculated in advance and a vibration change feature amount for each frequency band calculated in each substrate processing after learning in a normal state.

[0086] In this case, the abnormality determination unit 304 determines that the state of the vacuum pump indicates a sign of abnormality when the first deviation degree for the entire frequency band is equal to or greater than a first threshold value for the entire frequency band set in advance, or when the first deviation degree for each frequency band is equal to or greater than a first threshold value for each frequency band set in advance. In response to this, the display control unit 305 outputs a warning regarding the state of the vacuum pump.

[0087] It is preferable to calculate a plurality of second deviations in step S14. For example, the calculation unit 303 may calculate the second deviation of the entire frequency band, which indicates a difference value (degree of deviation) between the vibration power spectrum of the entire frequency band calculated based on the vibration data acquired when the substrate processing apparatus 10 processed the previous substrate and the vibration power spectrum of the entire frequency band calculated based on the vibration data acquired when the substrate processing apparatus 10 processed the current substrate. In addition, the calculation unit 303 may calculate the second deviation of each of the divided predetermined frequency bands, which indicates a difference value (degree of deviation) between the vibration power spectrum of each of the divided predetermined frequency bands when the entire frequency band is divided into a plurality of frequency bands based on the vibration data acquired when the substrate processing apparatus 10 processed the previous substrate, and the vibration power spectrum of each of the divided predetermined frequency bands when the entire frequency band is divided into a plurality of frequency bands based on the vibration data acquired when the substrate processing apparatus 10 processed the current substrate.

[0088] In this case, the abnormality determination unit 304 determines that the state of the vacuum pump indicates a sign of abnormality when the second deviation degree of the entire frequency band is equal to or greater than a second threshold value of the entire frequency band set in advance, or when the second deviation degree of each of the divided predetermined frequency bands is equal to or greater than a second threshold value of each of the divided frequency bands set in advance. In response to this, the display control unit 305 outputs a warning regarding the state of the vacuum pump.

[0089] As the amount of adhesion of by-products increases, the vibration of the vacuum pump increases, and the vibration increases significantly especially before the vacuum pump breaks down. Therefore, in the state determination method according to the present embodiment, the vibration change rate (vibration change feature amount) in each frequency range is observed, and early signs of vacuum pump failure can be found, and maintenance of the vacuum pump can be performed in advance before the failure occurs.

[0090] In addition, vibration data is monitored for both the step of reducing the pressure in the processing vessel 11 from atmospheric pressure to a predetermined pressure and the step of idling the vacuum pump. This allows the number of samplings of vibration data to be reduced while not overlooking any signs of abnormality and correctly determining the signs of abnormality in the vacuum pump.

[0091] In conventional methods, a normal model is created based on physical features such as the amplitude of vibration data. In contrast, in the state determination method according to the present embodiment, the state of the vacuum pump is determined based on the change in the physical feature of vibration intensity based on the vibration data, specifically, the vibration change feature. This makes it possible to reduce the influence of the amplitude-specific magnitude and characteristics on the model. This makes it possible to correctly determine signs of abnormality in the vacuum pump and output a warning.

[0092] In addition, when temperature data acquired at the same time as vibration data is outside the normal temperature range, the corresponding vibration data is excluded. This allows vibration data containing noise due to abnormal vibration caused by temperature changes to be excluded from the vibration data to be analyzed, thereby reducing false alarms of abnormalities.

[0093] The output of the warning enables the user to take some measure before the vacuum pump breaks down, such as replacing the vacuum pump, removing deposits on the vacuum pump by supplying a cleaning gas, or changing the oil in the vacuum pump.

[0094] Furthermore, the user can confirm whether the warning is for the turbo molecular pump 47 or the dry pump 46 from the position of the vibration sensor that detects the vibration data.

[0095] In addition, the vibration power spectrum of a predetermined frequency band and the vibration change feature amount of the predetermined frequency band can be visualized.

[0096] [others] In the state determination method according to the present embodiment, when monitoring vibrations in a predetermined frequency band, the first deviation degree based on the vibration change feature amount for all frequencies and the first deviation degree based on the vibration change feature amount for the predetermined frequency are calculated.

[0097] In this case, the method of calculating the vibration change rate (vibration change feature amount) may be a method of dividing the frequency band into several bands using a one-axis acceleration sensor and calculating the vibration change rate for each band. A method of automatically extracting the feature amount using an AE sensor (autoencoder) in the ultra-high frequency band may also be used.

[0098] In addition to calculating the first deviation and the second deviation, the maximum value max(θ i,j ), minimum value min(θ i,j ), standard deviation std(θ i,j ) may be calculated. By combining these statistical values ​​including the first deviation degree and the second deviation degree, the sign of an abnormality in the vacuum pump can be determined with even greater accuracy.

[0099] The predetermined frequency band can be defined by the user. An example of the predetermined frequency band is the 1000 Hz unit in FIG.

[0100] As described above, according to the state determination method and monitoring device of the present embodiment, it is possible to correctly determine the presence or absence of an abnormality in the vacuum pump.

[0101] (Variation 1) If there is data on the power (pump power) of the vacuum pump in the analysis target step, or if the time it takes for the power of the vacuum pump to stabilize in the analysis target step is known, the vibration intensity when the power of the vacuum pump is stable may be used to monitor and predict abnormalities in the vacuum pump.

[0102] (Variation 2: Building a multivariate anomaly detection model) As shown in FIG. 11, a prediction model may be created using the power of the vacuum pump, the temperature of the vacuum pump, and the cumulative flow rate (cumulative flow rate of film formation gas) through the vacuum pump as explanatory variables, and a model may be constructed that outputs the vibration change rate (predicted value) as a result.

[0103] Specifically, in the analysis target step, a prediction model such as a multiple regression or autoregression rule is created with the vibration change rate (predicted value) as the target variable and the power, temperature, and cumulative flow rate of the vacuum pump as explanatory variables, and the difference between the actual vibration change rate y and the predicted value y' of the predicted change rate is calculated. This difference (y-y') is an example of an index of deviation, and can be used to evaluate the degree of abnormality of the vacuum pump. For example, if the difference (y-y') is greater than a predetermined threshold, it may be determined that the vacuum pump is abnormal.

[0104] When sensor information from the vibration sensor and temperature sensor cannot be obtained, it is possible to partially eliminate noise such as power fluctuations that are determined to be an abnormality in the vacuum pump from stable vibration data, thereby improving the accuracy of determining an abnormality in the vacuum pump compared to conventional methods.

[0105] The state determination method and monitoring device according to the disclosed embodiments should be considered as illustrative and not restrictive in all respects. The embodiments can be modified and improved in various forms without departing from the scope and spirit of the appended claims. The matters described in the above embodiments can be configured in other ways as long as they are not inconsistent, and can be combined as long as they are not inconsistent.

[0106] The substrate processing apparatus disclosed in this specification is not limited to a batch-type heat treatment apparatus. For example, the substrate processing apparatus 10 may be a single-wafer type apparatus that processes wafers one by one. The substrate processing apparatus 10 may also be a semi-batch type apparatus that processes several substrates at once. The semi-batch type apparatus may be an apparatus that rotates a plurality of wafers arranged around the rotation center line of a rotary table together with the rotary table and passes the wafers sequentially through a plurality of regions to which different gases are supplied.

[0107] The substrate processing apparatus is not limited to a film forming apparatus, but may be an apparatus capable of processing a substrate, such as an etching apparatus, a sputtering apparatus, etc. Furthermore, the substrate processing apparatus 10 is not limited to an apparatus that processes a substrate using plasma, but may be an apparatus that processes a substrate without using plasma. [Explanation of symbols]

[0108] 1. Substrate Processing System 2 Wafer 10 Substrate processing equipment 11 Processing vessel 30 Boat 100 Control device 200 Sensors 300 Monitoring equipment 301 Acquisition Department 302 Pretreatment section 303 Arithmetic section 304 Abnormality determination section 305 Display control section 400 Data Storage Unit

Claims

1. 1. A state determination method for determining a state of a vacuum pump in a substrate processing apparatus that processes substrates based on a recipe including a plurality of steps, comprising: (a) acquiring vibration data from a sensor that detects vibration of the vacuum pump in a specific step among the plurality of steps; (b) calculating a vibration power spectrum of a predetermined frequency based on the vibration data, and calculating a vibration change feature amount of a predetermined frequency band from the vibration power spectrum of the predetermined frequency; (c) in the specific step, based on vibration data acquired from a sensor that detects vibration of the vacuum pump when it is new or when it has been replaced, calculating a first deviation degree indicating a degree of deviation between a normal vibration feature amount calculated in advance by executing (b) and a vibration change feature amount in a predetermined frequency band subsequently calculated in (b); (d) outputting a warning regarding a state of the vacuum pump when the first deviation is equal to or greater than a first threshold value set in advance; and A state determination method comprising:

2. the vibration change feature amount of the predetermined frequency band includes a vibration change feature amount of an entire frequency band and a vibration change feature amount of each frequency band when the entire frequency band is divided into a plurality of frequency bands, (c) calculates a first deviation degree for the entire frequency band indicating a degree of deviation between the vibration feature amount for the entire frequency band in a normal state calculated in advance by executing (b) and the vibration change feature amount for the entire frequency band calculated in (b), and calculates a first deviation degree for each frequency band indicating a degree of deviation between the vibration feature amount for each frequency band in a normal state calculated in advance by executing (b) and the vibration change feature amount for each frequency band calculated in (b); (d) outputs a warning regarding the state of the vacuum pump when the first deviation degree of the entire frequency band is equal to or greater than a first threshold value of the entire frequency band that is set in advance, or when the first deviation degree of each of the frequency bands is equal to or greater than a first threshold value of each of the frequency bands that is set in advance. The state determination method according to claim 1 .

3. (e) calculating a second deviation degree indicating a degree of deviation between a vibration power spectrum of a predetermined frequency band calculated based on the vibration data acquired when the substrate processing apparatus processed a previous substrate and a vibration power spectrum of a predetermined frequency band calculated based on the vibration data acquired when the substrate processing apparatus processed a current substrate; and (f) outputting a warning regarding a state of the vacuum pump when the second deviation is equal to or greater than a second threshold value set in advance; and The state determination method according to claim 1 , further comprising:

4. the vibration power spectrum of the predetermined frequency band includes a vibration power spectrum of an entire frequency band and a vibration power spectrum of each of the predetermined frequency bands obtained by dividing the entire frequency band into a plurality of frequency bands; (e) calculates a second degree of deviation for the entire frequency band indicating the degree of deviation between the vibration power spectrum for the entire frequency band calculated based on the vibration data acquired in (b) when the substrate processing apparatus processed a previous substrate and the vibration power spectrum for the entire frequency band calculated based on the vibration data acquired in (b) when the substrate processing apparatus processed a current substrate, and calculates a second degree of deviation for a predetermined frequency band indicating the degree of deviation between the vibration power spectrum for the predetermined frequency band calculated based on the vibration data acquired in (b) when the substrate processing apparatus processed a previous substrate and the vibration power spectrum for the predetermined frequency band calculated based on the vibration data acquired in (b) when the substrate processing apparatus processed a current substrate; (f) outputs a warning regarding the state of the vacuum pump when the second deviation degree of the entire frequency band is equal to or greater than a second threshold value of the entire frequency band that is set in advance, or when the second deviation degree of each of the divided specific frequency bands is equal to or greater than a second threshold value of each of the divided frequency bands that is set in advance. The state determination method according to claim 3 .

5. The specific step includes a step of idling the vacuum pump and a step of reducing the pressure of a processing vessel for processing the substrate from atmospheric pressure to a predetermined pressure. The state determination method according to any one of claims 1 to 4.

6. (a) includes, in the specific step, acquiring temperature data from a sensor that detects a temperature of the vacuum pump together with the vibration data; (b) calculating a vibration power spectrum of a predetermined frequency based on the vibration data obtained after excluding vibration data corresponding to temperature data outside a preset temperature range from the acquired temperature data; The state determination method according to any one of claims 1 to 4.

7. dividing the calculated vibration power spectrum of the predetermined frequency into vibration power spectra of unit frequencies for each of the substrate treatments, and visualizing and displaying the divided vibration power spectra on a two-dimensional map with the unit frequencies and the substrate treatment identification numbers as two axes; The state determination method according to any one of claims 1 to 4.

8. a vibration change feature amount between the substrate treatments calculated from the divided vibration power spectrum is visualized and displayed on a two-dimensional map having a value corresponding to a unit frequency and an identification number of the substrate treatment as two axes; The state determination method according to claim 7.

9. 1. A monitoring device for determining a state of a vacuum pump in a substrate processing apparatus for processing substrates based on a recipe including a plurality of steps, comprising: (a) an acquisition unit that acquires vibration data from a sensor that detects vibration of the vacuum pump in a specific step among the plurality of steps; (b) a preprocessing unit that calculates a vibration power spectrum of a predetermined frequency based on the vibration data and calculates a vibration change feature amount of a predetermined frequency band from the vibration power spectrum of the predetermined frequency; (c) a calculation unit that calculates a first deviation degree indicating a degree of deviation between a normal vibration feature amount calculated in advance by the pre-processing unit executing the process (b) and a vibration change feature amount in a predetermined frequency band subsequently calculated by the pre-processing unit in the process (b) based on vibration data acquired from a sensor that detects vibration of the vacuum pump when it is new or when it has been replaced, in the specific step; (d) a display control unit that outputs a warning regarding a state of the vacuum pump when the first deviation degree is equal to or greater than a preset first threshold value; and A monitoring device that performs control including:

Images