Information processing device, substrate polishing device, inference device, machine learning device, information processing method, inference method, and machine learning method

Abstract

To provide an information processing device that enables the processing quality of a substrate to be improved according to chemical mechanical polishing.SOLUTION: An information processing device 6 comprises: an information acquisition section 210a for acquiring substrate polishing information including measured temperature distribution information indicating measured values of a temperature distribution of a polishing surface when a polishing process for polishing a substrate with the polishing surface of a polishing pad is performed by a substrate polishing device 2 and device state information indicating a device state of the substrate polishing device 2; and an information generation section 210c for generating polishing surface temperature control information indicating a control amount of a polishing surface temperature adjustment section when a polishing process is performed by the substrate polishing device 2, by inputting the substrate polishing information acquired by the information acquisition section 210a into a learning model 10A. The learning model 10A is a trained model having learned a correlation between the substrate polishing information and the polishing surface temperature control information according to machine learning.SELECTED DRAWING: Figure 13

Description

[Technical Field]

[0001] The present invention relates to an information processing device, an inference device, a machine learning device, an information processing method, an inference method, and a machine learning method. [Background technology]

[0002] A substrate polishing apparatus that performs chemical mechanical polishing (CMP) processing is known as one type of substrate processing apparatus that performs various processes on substrates such as semiconductor wafers. In CMP processing, for example, a polishing table having a polishing pad is rotated, a polishing liquid (slurry) is supplied to the polishing pad from a polishing fluid supply nozzle, and the substrate is pressed against the polishing pad by a polishing head called a top ring, thereby chemically and mechanically polishing the substrate.

[0003] In this case, the processing quality when a substrate is polished by chemical mechanical polishing depends not only on the pressing load of the substrate against the polishing pad but also on the temperature distribution on the polishing surface of the polishing pad. This is because the chemical action of the polishing liquid on the substrate depends on temperature. Therefore, when performing chemical mechanical polishing, a temperature control device is used to adjust the temperature distribution of the polishing pad to maintain the temperature distribution on the polishing surface of the polishing pad at an appropriate target temperature (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2001-179613 Summary of the Invention [Problem to be solved by the invention]

[0005] The temperature adjustment device for a polishing pad disclosed in Patent Document 1 determines the control amount of a heating device or a cooling device so that the polishing pad has a predetermined temperature distribution. However, the temperature distribution of the polishing pad fluctuates in a complex manner due to the various device states that the substrate polishing apparatus can assume during chemical mechanical polishing processing, so it has been difficult to analyze all behaviors and factors and accurately determine the control amount of the temperature adjustment device.

[0006] In view of the above problems, an object of the present invention is to provide an information processing device, an inference device, a machine learning device, an information processing method, an inference method, and a machine learning method that enable improving the processing quality of substrates by chemical mechanical polishing processing. [Means for solving the problem]

[0007] In order to achieve the above object, an information processing device according to one aspect of the present invention comprises: an information acquiring unit that acquires substrate polishing information including measured temperature distribution information indicating a measured value of a temperature distribution on a polishing surface when a polishing process for polishing a substrate with a polishing surface of a polishing pad is performed by a substrate polishing apparatus, and apparatus status information indicating an apparatus status of the substrate polishing apparatus; an information generating unit that generates polishing surface temperature control information indicating a control amount of a polishing surface temperature adjusting unit that adjusts the temperature distribution of the polishing surface when the polishing process is performed by the substrate polishing apparatus in the apparatus state indicated by the apparatus state information included in the substrate polishing information, for the polishing surface having the temperature distribution indicated by the measured temperature distribution information included in the substrate polishing information, by inputting the substrate polishing information acquired by the information acquiring unit into a learning model, The learning model is This is a trained model that has been trained by machine learning to determine the correlation between the substrate polishing information and the polishing surface temperature control information. [Effects of the Invention]

[0008] According to an information processing device of one aspect of the present invention, substrate polishing information, including measured temperature distribution information indicating measured values ​​of the temperature distribution of the polishing surface and apparatus status information indicating the apparatus status of the substrate polishing apparatus, is input to a learning model, whereby polishing surface temperature control information for the substrate polishing information is generated. Therefore, the temperature distribution of the polishing surface is appropriately adjusted based on the polishing surface temperature control information, thereby improving the processing quality of the substrate by chemical mechanical polishing.

[0009] Problems, configurations, and effects other than those described above will become apparent from the detailed description of the invention that follows. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is an overall configuration diagram showing an example of a substrate polishing system 1. FIG. [Figure 2] 1 is a schematic diagram showing an example of a substrate polishing apparatus 2. FIG. [Figure 3] FIG. 2 is a schematic diagram showing an example of a polishing head unit 202. [Figure 4] FIG. 2 is a schematic diagram showing an example of a polishing surface heating unit 206. [Figure 5] FIG. 2 is a schematic diagram showing an example of a polishing surface cooling unit 207. [Figure 6] 1 is a block diagram showing an example of a substrate polishing apparatus 2 (information processing apparatus 6) according to a first embodiment. [Figure 7] FIG. 9 is a hardware configuration diagram showing an example of a computer 900. [Figure 8] FIG. 2 is a data structure diagram showing an example of a database 30. [Figure 9] FIG. 2 is a block diagram showing an example of a machine learning device 4 according to the first embodiment. [Figure 10] 1 is a schematic diagram showing an example of learning data 11A according to the first embodiment and a relationship with reinforcement learning. FIG. [Figure 11] FIG. 2 is a diagram showing an example of a learning model 10A according to the first embodiment. [Figure 12] 4 is a flowchart showing an example of a machine learning method performed by the machine learning device 4 according to the first embodiment. [Figure 13] 1 is a functional explanatory diagram showing an example of a substrate polishing apparatus 2 (information processing apparatus 6) according to a first embodiment. [Figure 14] 1 is a flowchart showing an example of an information processing method by the substrate polishing apparatus 2 (information processing apparatus 6) according to the first embodiment. [Figure 15] FIG. 10 is a block diagram showing an example of a machine learning device 4a according to a second embodiment. [Figure 16] 10 is a diagram showing an example of a learning model 10B and learning data 11B according to the second embodiment. FIG. [Figure 17] 10 is a flowchart showing an example of a machine learning method performed by a machine learning device 4a according to the second embodiment. [Figure 18] FIG. 10 is a block diagram showing an example of a substrate polishing apparatus 2a (information processing apparatus 6a) according to a second embodiment. [Figure 19] FIG. 10 is a functional explanatory diagram showing an example of a substrate polishing apparatus 2a (information processing apparatus 6a) according to a second embodiment. [Figure 20] 10 is a flowchart showing an example of an information processing method by a substrate polishing apparatus 2a (information processing apparatus 6a) according to a second embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, an embodiment for carrying out the present invention will be described with reference to the drawings. The scope necessary for the explanation to achieve the object of the present invention will be schematically shown, and the scope necessary for explaining the relevant parts of the present invention will be mainly explained, and the parts that are omitted from the explanation will be based on publicly known techniques.

[0012] (First embodiment) 1 is an overall configuration diagram showing an example of a substrate polishing system 1. The substrate polishing system 1 according to this embodiment functions as a system for managing a chemical mechanical polishing process (hereinafter referred to as a "polishing process") for polishing the surface of a substrate (hereinafter referred to as a "wafer") W such as a semiconductor wafer to a flat surface.

[0013] The substrate polishing system 1 mainly comprises a substrate polishing apparatus 2, a database device 3, a machine learning device 4, and a user terminal device 5. Each of the devices 2 to 5 is configured, for example, as a general-purpose or dedicated computer (see FIG. 7 described later), and is connected to a wired or wireless network 7 so as to be able to transmit and receive various data to and from each other. The number of the devices 2 to 5 and the connection configuration of the network 7 are not limited to the example shown in FIG. 1 and may be changed as appropriate.

[0014] The substrate polishing apparatus 2 is an apparatus that performs a polishing process to polish a wafer W. At that time, the substrate polishing apparatus 2 controls the operation of the substrate polishing apparatus 2 using a trained learning model 10A (trained model) generated by the machine learning device 4, while referring to apparatus setting information 215 consisting of a plurality of apparatus parameters and substrate recipe information 216 that defines the polishing conditions of the polishing process.

[0015] Furthermore, as the substrate polishing apparatus 2 performs the polishing process, it accumulates various types of information as operation history information 217 and transmits the operation history information 217 to the database device 3. The operation history information 217 includes, for example, apparatus status information indicating the apparatus status of the substrate polishing apparatus 2 when the polishing process is performed, event information detected by the substrate polishing apparatus 2, and operation information of the user (operator, production manager, maintenance manager, etc.) on the substrate polishing apparatus 2.

[0016] The database device 3 includes a database 30 that stores operation history information 217 transmitted from each substrate polishing apparatus 2. In addition to the operation history information 217 transmitted from each substrate polishing apparatus 2, the database 30 may also store information on polishing process tests performed using dummy wafers. The polishing process tests may be performed by the substrate polishing apparatus 2, by a polishing test apparatus capable of reproducing the polishing process, or by a polishing simulation apparatus capable of simulating the polishing process. The database 30 may also store apparatus setting information 215 and substrate recipe information 216.

[0017] The machine learning device 4 acquires, for example, a portion of the information stored in the database 30 as learning data 11A, and generates, by machine learning, a learning model 10A to be used in the substrate polishing apparatus 2. The trained learning model 10A (trained model) is provided to the substrate polishing apparatus 2 via the network 7, a recording medium, or the like.

[0018] The user terminal device 5 is a terminal device used by a user, and may be a stationary device or a portable device. The user terminal device 5 accepts various input operations via a display screen of, for example, an application program, a web browser, or the like, and displays various information (for example, event notifications, device setting information 215, substrate recipe information 216, operation history information 217, etc.) via the display screen. Note that some of the information may be editable via the user terminal device 5.

[0019] The learning model 10A according to this embodiment employs reinforcement learning as a machine learning technique. The machine learning device 4 is installed, for example, in an assembly factory or evaluation facility for the substrate polishing apparatus 2, and executes a learning phase of reinforcement learning. The substrate polishing apparatus 2 is installed, for example, in a wafer W manufacturing factory, and executes an inference phase of reinforcement learning using the learning model 10A trained by the machine learning device 4, and also executes a further learning phase of reinforcement learning to adapt to the individual environment of the wafer W manufacturing factory, etc.

[0020] (Substrate polishing device 2) 2 is a schematic diagram showing an example of the configuration of the substrate polishing apparatus 2. The substrate polishing apparatus 2 includes a polishing unit 20 that performs polishing processing on the wafer W, and a control unit 21 that controls the operation of the substrate polishing apparatus 2. The substrate polishing apparatus 2 is required to include at least one polishing unit 20. In addition to the polishing unit 20, the substrate polishing apparatus 2 may also include, for example, a substrate transport unit that transports the wafer W, a cleaning unit that cleans the wafer W, etc.

[0021] The polishing unit 20 includes a polishing table portion 201 that rotatably supports a polishing pad 200 having a polishing surface, a polishing head portion 202 (top ring) that presses a wafer W against the polishing surface of the polishing pad 200 to polish the wafer W, a polishing fluid supply portion 203 that supplies a polishing fluid to the polishing pad 200, a dressing portion 204 (dresser) that dresses the polishing surface of the polishing pad 200, a cleaning fluid injection portion 205 (atomizer) that injects a cleaning fluid onto the polishing pad 200, a polishing surface heating portion 206 and a polishing surface cooling portion 207 that function as polishing surface temperature adjustment portions that adjust the temperature distribution on the polishing surface of the polishing pad 200, a polishing surface temperature measurement portion 208 that measures the temperature distribution on the polishing surface of the polishing pad 200, and a polishing unit measurement portion 209 that measures the state of the wafer W and the state of the processing environment in which the polishing process is performed.

[0022] The polishing table section 201 includes a polishing table 201a to which a polishing pad 200 is replaceably attached, a polishing table shaft 201b that supports the polishing table 201a, and a rotational movement mechanism section 201c that rotates the polishing table 201a around the axis of the polishing table shaft 201b.

[0023] The polishing head unit 202 includes a polishing head 202a that holds the wafer W, a polishing head shaft 202b that supports the polishing head 202a, a rotational movement mechanism unit 202c that rotates the polishing head 202a around the axis of the polishing head shaft 202b, a vertical movement mechanism unit 202d that moves the polishing head 202a in the vertical direction, a support arm 202e that supports the polishing head shaft 202b, a support shaft 202f that supports the support arm 202e, and a swinging movement mechanism unit 202g that swings (swings) the polishing head 202a around the support shaft 202f as the center of rotation.

[0024] The polishing-fluid supply unit 203 includes a polishing-fluid supply nozzle 203a capable of supplying the polishing-fluid, a support shaft 203b supporting the polishing-fluid supply nozzle 203a, a swinging movement mechanism 203c for swinging (swinging) the polishing-fluid supply nozzle 203a around the support shaft 203b, a flow rate adjusting unit 203d for adjusting the flow rate of the polishing fluid, and a temperature adjusting mechanism 203e for adjusting the temperature of the polishing fluid. The polishing fluid is a polishing liquid (slurry) or pure water, and may further contain a chemical solution or may be a polishing liquid to which a dispersant has been added.

[0025] The dressing unit 204 includes a dressing head 204a to which a dressing pad (not shown) can be replaceably attached, a dressing shaft 204b that supports the dressing head 204a, a rotational movement mechanism 204c that rotates the dressing head 204a around the axis of the dressing shaft 204b, a vertical movement mechanism 204d that moves the dressing head 204a in the vertical direction, a support arm 204e that supports the dressing shaft 204b, a support shaft 204f that supports the support arm 204e, and a swinging movement mechanism 204g that swings (swings) the dressing head 204a around the support shaft 204f as the center of rotation.

[0026] The cleaning fluid ejection unit 205 includes a cleaning fluid ejection nozzle 205a capable of ejecting cleaning fluid, a support shaft 205b supporting the cleaning fluid ejection nozzle 205a, and a swinging movement mechanism 202 that swings (swings) the cleaning fluid ejection nozzle 205a around the support shaft 205b. 5c, a flow rate adjusting unit 205d that adjusts the flow rate of the cleaning fluid, and a temperature adjusting mechanism 205e that adjusts the temperature of the polishing fluid. The cleaning fluid is a mixed fluid of a liquid (e.g., pure water) and a gas (e.g., nitrogen gas) or a liquid (e.g., pure water).

[0027] The polishing surface heating unit 206 includes a heating fluid injection nozzle 206a capable of injecting a heating fluid, a rotational movement mechanism 206b that rotates the heating fluid injection nozzle 206a around its axis, a support shaft 206c that supports the heating fluid injection nozzle 206a, a vertical movement mechanism 206d that moves the heating fluid injection nozzle 206a up and down, a swinging movement mechanism 206e that swings the heating fluid injection nozzle 206a around the support shaft 206c, and a heating fluid supply mechanism 206f that supplies the heating fluid. The heating fluid may be a high-temperature gas (e.g., air or an inert gas such as nitrogen or argon) or high-temperature steam.

[0028] The polishing surface cooling unit 207 includes a cooling fluid injection nozzle 207a capable of injecting a cooling fluid, a rotational movement mechanism 207b that rotates the cooling fluid injection nozzle 207a around its axis, a support shaft 207c that supports the cooling fluid injection nozzle 207a, a vertical movement mechanism 207d that moves the cooling fluid injection nozzle 207a up and down, a swing movement mechanism 207e that swings (swings) the cooling fluid injection nozzle 207a around the support shaft 207c, and a cooling fluid supply mechanism 207f that supplies the cooling fluid. The cooling fluid may be a gas at room temperature (e.g., air or an inert gas such as nitrogen or argon) or a gas cooled to a set temperature lower than room temperature. In addition, the moving mechanism units provided for the polishing surface heating unit 206 and the polishing surface cooling unit 207 may be shared, in which case the polishing surface heating unit 206 and the polishing surface cooling unit 207 are moved by the shared moving mechanism unit.

[0029] The polishing surface temperature measuring unit 208 measures the temperature distribution on the polishing surface of the polishing pad 200 in a contactless or contact manner, and outputs measured temperature distribution information indicating the measured value. The measured temperature distribution information is, for example, a record of the temperature distribution according to the radial position of the polishing pad 200. Note that the measured temperature distribution information may also be a record of the temperature distribution according to the radial position of the polishing pad 200, for each circumferential direction of the polishing pad 200.

[0030] The polishing surface temperature measuring unit 208 is configured with temperature sensors such as a thermograph, a thermopile, and an infrared camera. The polishing surface temperature measuring unit 208 may be configured by arranging a plurality of temperature sensors such as an infrared radiation thermometer and a thermocouple thermometer. The polishing surface temperature measuring unit 208 may also be configured by combining a plurality of types of temperature sensors.

[0031] The polishing unit measuring section 209 includes a substrate measuring section 209a that measures the state of the wafer W, and an environment measuring section 209b that measures the state of the processing environment in which the polishing process is performed.

[0032] The substrate measuring unit 209a measures the state of the wafer W before the polishing process, the state of the wafer W during the polishing process, and the state of the wafer W after the polishing process. The substrate measuring unit 209a is configured with a sensor that measures the state of the wafer W, such as, but not limited to, the film thickness of the wafer W and the temperature distribution on the surface of the wafer W.

[0033] The environment measurement unit 209b is configured with sensors that measure the temperature, humidity, air pressure, etc. of the processing environment as the state of the processing environment, but is not limited to these.

[0034] 2, the specific configurations of the rotation movement mechanism units 201c, 202c, 204c, 206b, 207b, the vertical movement mechanism units 202d, 204d, 206d, 207d, and the swing movement mechanism units 202g, 203c, 204g, 205c, 206e, 207e are omitted, but for example, modules for generating driving force such as motors and fluid pressure cylinders, linear guides, etc. , ball screws, gears, belts, couplings, bearings, and other driving force transmission mechanisms, and sensors such as linear sensors, encoder sensors, limit sensors, and torque sensors. While the specific configuration of the flow rate adjustment units 203d and 205d is omitted in Fig. 2, they may be configured, for example, by appropriately combining fluid adjustment modules such as pumps, valves, and regulators, and sensors such as flow rate sensors, pressure sensors, and liquid level sensors. While the specific configuration of the temperature adjustment mechanisms 203e and 205e is omitted in Fig. 2, they may be configured, for example, by appropriately combining temperature adjustment modules (contact or non-contact) such as heaters and heat exchangers, and sensors such as temperature sensors.

[0035] (Polishing head part 202) 3 is a schematic diagram showing an example of the polishing head unit 202. The polishing head 202a includes a top ring body 2020 attached to a polishing head shaft 202b, a substantially disc-shaped carrier 2021 housed in the top ring body 2020, a membrane 2022 disposed below the carrier 2021 and pressing the wafer W against the polishing pad 200, a substantially annular retaining ring 2023 disposed on the outer periphery of the carrier 2021 and the membrane 2022 and directly pressing the polishing pad 200, and a retaining ring air bag 2024 disposed between the top ring body 2020 and the retaining ring 2023 and pressing the retaining ring 2023 against the polishing pad 200.

[0036] The membrane 2022 is formed of an elastic membrane and has a plurality of concentric partition walls 2022e therein, thereby providing first to fourth membrane pressure chambers 2022a to 2022d arranged concentrically from the center to the outer periphery of the top ring body 2020. The membrane 2022 also has a plurality of holes 2022f on its underside for suction of the wafer W, functioning as a substrate holding surface for holding the wafer W. The retainer ring airbag 2024 is formed of an elastic membrane and has a retainer ring pressure chamber 2024a therein. The configuration of the polishing head 202a may be modified as needed. It may include a pressure chamber for pressing the entire carrier 2021. The number and shape of the membrane pressure chambers in the membrane 2022 may be modified as needed. The number and arrangement of the suction holes 2022f may also be modified as needed. The membrane 2022 may not necessarily have the suction holes 2022f.

[0037] First to fourth flow paths 2026A to 2026D are connected to the first to fourth membrane pressure chambers 2022a to 2022d, respectively, and a fifth flow path 2026E is connected to the retaining ring pressure chamber 2024a. The first to fifth flow paths 2026A to 2026E communicate with the outside via a rotary joint 2025 provided on the polishing head shaft 202b and branch into first branch flow paths 2027A to 2027E and second branch flow paths 2028A to 2028E, respectively. Pressure sensors PA to PE are installed in the first to fifth flow paths 2026A to 2026E, respectively. The first branch flow paths 2027A to 2027E are connected to a gas supply source GS of pressurized fluid (air, nitrogen, etc.) via valves V1A to V1E, flow sensors FA to FE, and pressure regulators RA to RE. The second branch flow paths 2028A to 2028E are connected to a vacuum source VS via valves V2A to V2E, respectively, and are configured to be able to communicate with the atmosphere via valves V3A to V3E.

[0038] The wafer W is held by suction on the underside of the polishing head 202a and moved to a predetermined polishing position on the polishing table 201. After that, the polishing head 202a presses the wafer W against the polishing surface of the polishing pad 200 to which polishing fluid is supplied from the polishing fluid supply unit 203, thereby polishing the wafer W. At this time, the polishing head 202a independently controls the pressure regulators RA to RE to adjust the pressure of the pressure fluid supplied to the first to fourth membrane pressure chambers 2022a to 2022d to press the wafer W against the polishing pad 200 for each region of the wafer W, and also adjusts the pressure of the pressure fluid supplied to the retainer ring pressure chamber 2024a to press the retainer ring 2023 against the polishing pad 200. The pressures of the pressurized fluids supplied to the retaining ring pressure chambers 2022d and 2024a are measured by the pressure sensors PA to PE, respectively, and the flow rates of the pressurized gases are measured by the flow rate sensors FA to FE, respectively.

[0039] (Polished surface heating section 206) FIG. 4 is a schematic diagram showing an example of the polishing surface heating unit 206. The heating fluid ejection nozzle 206a includes a nozzle body 2060 attached to a support shaft 206c and a heating fluid ejection port 2061 for ejecting a heating fluid toward the polishing surface of the polishing pad 200. The nozzle body 2060 is formed in a cylindrical shape and has a flow path (not shown) for the heating fluid inside. As shown in FIG. 4, the heating fluid ejection port 2061 may be an elongated hole formed along the longitudinal direction of the nozzle body 2060, or may be a plurality of small holes formed and arranged in the longitudinal direction of the nozzle body 2060. The heating fluid ejection port 2061 may be provided with an aperture adjustment mechanism that adjusts the aperture of the heating fluid ejection port 2061 using a piezoelectric element, a shutter member, or the like.

[0040] The heated fluid supply mechanism 206f includes a flow rate adjustment valve 2063 connected to the heated fluid injection nozzle 206a via a first flow path 2062A, and a heated fluid generator 2064 connected to the flow rate adjustment valve 2063 via a second flow path 2062B. As shown in FIG. 4, a gas supply source GS and a water supply source WS are connected to the heated fluid generator 2064. When the heated fluid is high-temperature steam, the gas supply source GS may be omitted, and when the heated fluid is high-temperature gas, the water supply source WS may be omitted. A flow rate sensor FA, a pressure sensor PA, and a temperature sensor TA are installed in the first flow path 2062A.

[0041] (Polished surface cooling section 207) FIG. 5 is a schematic diagram showing an example of the polishing surface cooling unit 207. The cooling fluid injection nozzle 207a includes a nozzle body 2070 attached to a support shaft 207c and a cooling fluid injection port 2071 for injecting a cooling fluid toward the polishing surface of the polishing pad 200. The nozzle body 2070 is formed in a cylindrical shape and has a cooling fluid flow path (not shown) therein. The cooling fluid injection port 2071 may be an elongated hole formed along the longitudinal direction of the nozzle body 2070, or may be a plurality of small holes formed and arranged in the longitudinal direction of the nozzle body 2070 as shown in FIG. 5. The cooling fluid injection port 2071 may be provided with an aperture adjustment mechanism that adjusts the aperture of the cooling fluid injection port 2071 using a piezoelectric element, a shutter member, or the like.

[0042] The cooling fluid supply mechanism 207f includes a flow rate adjustment valve 2073 connected to the cooling fluid injection nozzle 207a via a first flow path 2072A, and a cooling fluid generator 2074 connected to the flow rate adjustment valve 2073 via a second flow path 2072B. A gas supply source GS is connected to the cooling fluid generator 2074. A flow rate sensor FB, a pressure sensor PB, and a temperature sensor TA are installed in the first flow path 2072A. Note that if the cooling fluid is a room temperature gas, the cooling fluid generator 2074 may be omitted.

[0043] The heating fluid supply mechanism 206f and the cooling fluid supply mechanism 207f are controlled by the control unit 21 so that the measured temperature distribution information indicating the measured value of the temperature distribution of the polishing surface by the polishing surface temperature measuring unit 208 satisfies the target temperature distribution information indicating the target value of the temperature distribution of the polishing surface.

[0044] The target temperature distribution information of the polishing pad 200 is included in, for example, the apparatus setting information 215 or the substrate recipe information 216, and is referenced by the control unit 21. The target temperature distribution information is defined in the same way as the measured temperature distribution information, and is, for example, a record of the temperature distribution according to the radial position of the polishing pad 200. The target temperature distribution information may be a constant value from the start to the end of the polishing process, or may be a variable value.

[0045] The heating fluid supply mechanism 206f and the cooling fluid supply mechanism 207f are controlled by the control unit 21. 4 and 5, the opening of the fluid ejection port 2061 and the opening of the cooling fluid ejection port 2071 are changed from pre-control states to post-control states so as to reduce the difference between the measured temperature distribution information and the target temperature distribution information. As control variables for the polishing surface heating unit 206 and the polishing surface cooling unit 207, in addition to the above-mentioned openings, the flow rate, pressure, temperature, and supply position of the heating fluid and the flow rate, pressure, temperature, and supply position of the cooling fluid are used, and these may be combined as appropriate.

[0046] The configurations of the polishing surface heating unit 206 and the polishing surface cooling unit 207 are not limited to the above examples. For example, the shapes and arrangements of the heating fluid ejection port 2061 and the cooling fluid ejection port 2071 may be changed as appropriate. The number of heating fluid ejection port 2061 and the cooling fluid ejection port 2071 may be one or more. When there are multiple heating fluid ejection ports 2061 and cooling fluid ejection ports 2071, the flow rate, pressure, temperature, and supply position of the heating fluid or cooling fluid may be controlled for each heating fluid ejection port 2061 or each cooling fluid ejection port 2071, or the opening degree of each heating fluid ejection port 2061 or each cooling fluid ejection port 2071 may be controlled for each heating fluid ejection port 2061 or each cooling fluid ejection port 2071.

[0047] 6 is a block diagram showing an example of the substrate polishing apparatus 2 (information processing device 6) according to the first embodiment. The control unit 21 is electrically connected to each of the components 201 to 209 included in the polishing unit 20, and functions as a control unit that comprehensively controls the polishing unit 20.

[0048] The polishing unit 20 comprises a plurality of modules 20a to be controlled, each of which is arranged in each of the sections 201 to 209 of the polishing unit 20; a plurality of sensors 20b arranged in each of the modules 20a to detect data (detected values ​​and measured values) necessary for controlling each module 20a; and a sequencer 20c that controls the operation of each module 20a based on the detected values ​​and measured values ​​of each sensor 20b.

[0049] The sensors 20b of the polishing unit 20 include, for example, a sensor for detecting the rotation speed and rotation torque of the polishing table 201, a sensor for detecting the rotation speed, rotation torque, swing torque, and height of the polishing head 202, a sensor for detecting the swing position of the polishing head 202 which can be converted into the polishing position of the polishing head 202, a sensor for detecting the lifting torque of the polishing head 202 which can be converted into the pressing load of the polishing head 202, and a sensor for detecting the pressure (positive pressure) in the first to fourth membrane pressure chambers 2022a to 202d and the retaining ring pressure chamber 2024a. and negative pressure), a sensor for detecting the flow rate of the pressure fluid supplied to the first to fourth membrane pressure chambers 2022a to 2022d and the retaining ring pressure chamber 2024a, a sensor for detecting the surface roughness of the polishing pad 200, a sensor for detecting the flow rate and temperature of the polishing fluid supplied from the polishing fluid supply unit 203, a sensor for detecting the oscillation position of the polishing fluid supply unit 203 which can be converted into a dropping position of the polishing fluid, sensors for detecting the rotation speed, rotation torque, oscillation torque and height of the dressing unit 204, a sensor for detecting the swing position of the dressing unit 204, which can be converted into a dressing position of the dressing unit 204; a sensor for detecting the lifting torque of the dressing unit 204, which can be converted into a pressing load of the dressing unit 204; a sensor for detecting the flow rate, pressure, and temperature of the cleaning fluid supplied from the cleaning fluid spray unit 205; a sensor for detecting the swing position of the cleaning fluid spray unit 205, which can be converted into a dripping position of the cleaning fluid; a sensor for detecting the flow rate, pressure, and temperature of the heating fluid supplied from the polishing surface heating unit 206; These include sensors that detect the rotation angle, height, oscillation position, and opening of the heating unit 206, sensors that detect the opening of the heating fluid nozzle 2061 of the polishing surface heating unit 206, sensors that detect the flow rate, pressure, and temperature of the cooling fluid supplied from the polishing surface cooling unit 207, sensors that detect the rotation angle, height, and oscillation position of the polishing surface cooling unit 207 which can be converted into the cooling fluid supply position, sensors that detect the opening of the cooling fluid nozzle 2071 of the polishing surface cooling unit 207, sensors possessed by the polishing surface temperature measuring unit 208, and sensors possessed by the polishing unit measuring unit 209.

[0050] The control unit 21 includes a control unit 210, a communication unit 211, an input unit 212, an output unit 213, and a storage unit 214. The control unit 21 is configured, for example, by a general-purpose or dedicated computer (see FIG. 7 described later).

[0051] The communication unit 211 is connected to the network 7 and functions as a communication interface for transmitting and receiving various types of data. The input unit 212 accepts various input operations, and the output unit 213 functions as a user interface by outputting various types of information via a display screen, signal tower lighting, and buzzer sound.

[0052] The memory unit 214 stores various programs (operating system (OS), application programs, web browsers, etc.) and data (apparatus setting information 215, substrate recipe information 216, operation history information 217, learning model 10A, etc.) used in the operation of the substrate polishing apparatus 2.

[0053] The control unit 210 functions as a polishing unit 210a, an information acquisition unit 210b, an information generation unit 210c, and a machine learning unit 210d.

[0054] The polishing processing unit 210a acquires detection values ​​and measurement values ​​of the plurality of sensors 20b (hereinafter referred to as a "sensor group") via the sequencer 20c, and operates the plurality of modules 20a (hereinafter referred to as a "module group") in cooperation with each other, thereby polishing the wafer W. The polishing processing unit 210a also stores the detection values ​​and measurement values ​​of each sensor 20b in the memory unit 214 as operation history information 217.

[0055] When the polishing process is performed, the information acquisition unit 210b and the information generation unit 210c execute the inference phase of reinforcement learning using the learning model 10A. When the polishing process is performed, the machine learning unit 210d executes the learning phase of reinforcement learning for the learning model 10A. The information acquisition unit 210b, the information generation unit 210c, and the machine learning unit 210d constitute an information processing device 6. Details of the information processing device 6 will be described later.

[0056] (Hardware configuration of each device) 7 is a hardware configuration diagram showing an example of a computer 900. Each of the substrate polishing apparatus 2 (particularly the control unit 21), the database device 3, the machine learning device 4, and the user terminal device 5 is configured by a general-purpose or dedicated computer 900.

[0057] 7, the computer 900 includes, as its main components, a bus 910, a processor 912, a memory 914, an input device 916, an output device 917, a display device 918, a storage device 920, a communication I / F (interface) unit 922, an external device I / F unit 924, an I / O (input / output) device I / F unit 926, and a media input / output unit 928. Note that the above components may be omitted as appropriate depending on the application of the computer 900.

[0058] The processor 912 is composed of one or more arithmetic processing devices (such as a CPU (Central Processing Unit), MPU (Micro-Processing Unit), DSP (Digital Signal Processor), GPU (Graphics Processing Unit), or NPU (Neural Processing Unit)), and operates as a control unit that controls the entire computer 900. The memory 914 stores various data and programs 930, and is composed of, for example, a volatile memory (such as a DRAM or SRAM) that functions as a main memory, a non-volatile memory (ROM), a flash memory, etc.

[0059] The input device 916 is composed of, for example, a keyboard, a mouse, a numeric keypad, an electronic pen, etc., and functions as an input unit. The output device 917 is composed of, for example, a sound (audio) output device, a vibration device, etc., and functions as an output unit. The display device 918 is composed of, for example, a liquid crystal display, an organic EL display, electronic paper, a projector, etc., and functions as an output unit. The input device 916 and the display device 918 may be integrated into one device, such as a touch panel display. The storage device 920 is composed of, for example, an HDD, an SSD, etc., and functions as a storage unit. The storage device 920 stores various data necessary for executing the operating system and the program 930.

[0060] The communication I / F unit 922 is connected to a network 940 such as the Internet or an intranet (which may be the same as network 7 in FIG. 1) via a wired or wireless connection and functions as a communication unit that transmits and receives data to and from other computers in accordance with a predetermined communication protocol. The external device I / F unit 924 is connected to an external device 950 such as a camera, printer, scanner, or reader / writer via a wired or wireless connection and functions as a communication unit that transmits and receives data to and from the external device 950 in accordance with a predetermined communication protocol. The I / O device I / F unit 926 is connected to an I / O device 960 such as various sensors and actuators and functions as a communication unit that transmits and receives various signals and data, such as detection signals from sensors and control signals to actuators, to and from the I / O device 960. The media input / output unit 928 is composed of a drive device such as a DVD drive or CD drive, a memory card slot, and a USB connector, and reads and writes data from and to media (non-transitory storage media) 970 such as a DVD, CD, memory card, or USB memory.

[0061] In the computer 900 having the above configuration, the processor 912 loads a program 930 stored in the storage device 920 into the memory 914, executes the program, and controls each unit of the computer 900 via the bus 910. The program 930 may be stored in the memory 914 instead of the storage device 920. The program 930 may be recorded on the medium 970 in an installable file format or an executable file format and provided to the computer 900 via the media input / output unit 928. The program 930 may be provided to the computer 900 by being downloaded via the communication I / F unit 922 over the network 940. Furthermore, the computer 900 may implement various functions realized by the processor 912 executing the program 930 using hardware such as an FPGA (Field-Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit).

[0062] The computer 900 is, for example, a desktop computer or a portable computer, and is an electronic device of any type. The computer 900 may be a client computer, a server computer, a cloud computer, or an embedded computer called, for example, a control panel or a controller (including a microcomputer, a programmable logic controller, or a sequencer). The computer 900 may also be applied to devices other than the devices 2 to 5.

[0063] (Database 30) 8 is a data configuration diagram showing an example of the database 30. In the database 30, operation history information 217 acquired when a polishing process of a wafer W is performed in the substrate polishing apparatus 2 is classified and registered. The database 30 includes, for example, a wafer history table 300 related to each wafer W, and a polishing history table 301 related to apparatus status information in the polishing process. In addition to the above, the database 30 also includes an event history table related to event information, an operation history table related to operation information, and the like, but detailed description thereof will be omitted.

[0064] Each record in the wafer history table 300 stores, for example, a wafer ID for identifying the wafer W, the start time and end time of the polishing process, and the like.

[0065] Each record in the polishing history table 301 includes, for example, a wafer ID, polishing table status information, polishing head status information, polishing fluid supply status information, dressing status information, cleaning fluid injection status information, polishing surface temperature adjustment status information, measured temperature distribution information, substrate status information, and environmental information.

[0066] The polishing table status information is information that indicates the status of the polishing table 201 during the polishing process. The polishing table status information is, for example, detection values ​​and measurement values ​​of each sensor sampled at predetermined time intervals by a group of sensors included in the polishing table 201.

[0067] The polishing head status information is information indicating the status of the polishing head unit 202 during the polishing process. The polishing head status information is, for example, detection values ​​and measurement values ​​of each sensor sampled at predetermined time intervals by a group of sensors included in the polishing head unit 202.

[0068] The polishing-fluid supply state information is information indicating the state of the polishing-fluid supply unit 203 during the polishing process. The polishing-fluid supply state information is, for example, detection values ​​and measurement values ​​of each sensor sampled at predetermined time intervals by a group of sensors included in the polishing-fluid supply unit 203.

[0069] The dressing state information is information indicating the state of the dressing unit 204 during the polishing process. The polishing fluid supply state information is, for example, detection values ​​and measurement values ​​of each sensor sampled at predetermined time intervals by a group of sensors included in the dressing unit 204.

[0070] The cleaning fluid ejection state information is information indicating the state of the cleaning fluid ejection unit 205 during the polishing process. The polishing fluid supply state information is, for example, detection values ​​and measurement values ​​of each sensor sampled at predetermined time intervals by a group of sensors included in the cleaning fluid ejection unit 205.

[0071] The polishing surface temperature adjustment status information is information indicating the status of the polishing surface heating unit 206 and the polishing surface cooling unit 207, which function as polishing surface temperature adjustment units in the polishing process. The polishing surface temperature adjustment status information is, for example, the detection values ​​and measurement values ​​of each sensor sampled at predetermined time intervals by the sensor groups included in the polishing surface heating unit 206 and the polishing surface cooling unit 207.

[0072] The measured temperature distribution information is information that indicates the temperature distribution during the polishing process on the polishing surface of the polishing pad 200. The measured temperature distribution information is, for example, measurement values ​​sampled by the polishing surface temperature measuring unit 208 at predetermined time intervals.

[0073] The substrate state information is information indicating the state of the wafer W. The substrate state information is, for example, detection values ​​and measurement values ​​of each sensor sampled by the substrate measuring unit 209a at predetermined time intervals.

[0074] The environmental information is information that indicates the state of the processing environment in which the polishing process is performed, such as the detected values ​​and measured values ​​of each sensor sampled by the environment measurement unit 209b at predetermined time intervals.

[0075] By referring to the polishing history table 301, it is possible to extract time series data of the detected values ​​and measured values ​​of each sensor as apparatus status information of the substrate polishing apparatus 2 when the polishing process is performed on the wafer W identified by the wafer ID. Note that the polishing history table 301 may also register command values ​​to each module instead of the detected values ​​and measured values ​​of each sensor. In this case, each module The command values ​​to the module are set based on the device setting information 215 and the substrate recipe information 216, but the setting values ​​set in the device setting information 215 and the substrate recipe information 216 may also be registered as device status information.

[0076] (Machine Learning Device 4) 9 is a block diagram showing an example of a machine learning device 4 according to the first embodiment. The machine learning device 4 includes a control unit 40, a communication unit 41, a learning data storage unit 42, and a trained model storage unit 43.

[0077] The control unit 40 functions as a learning data acquisition unit 400 and a machine learning unit 401. The communication unit 41 is connected to external devices (e.g., the substrate polishing apparatus 2, the database device 3, the user terminal device 5, the polishing test device (not shown), the polishing simulation device (not shown), etc.) via the network 7, and functions as a communication interface for transmitting and receiving various data.

[0078] The learning data acquisition unit 400 is connected to an external device via the communication unit 41 and the network 7, and acquires learning data 11A that is configured at least of substrate polishing information as input data. The learning data 11A is acquired, for example, by referring to the database 30. In this embodiment, the learning data 11A is data used for reinforcement learning.

[0079] The learning data storage unit 42 is a database that stores a plurality of sets of learning data 11A acquired by the learning data acquisition unit 400. The specific configuration of the database that constitutes the learning data storage unit 42 may be designed as appropriate.

[0080] The machine learning unit 401 performs machine learning using multiple sets of training data 11A. That is, the machine learning unit 401 inputs multiple sets of training data 11A to the training model 10A and generates a trained training model 10A by having the training model 10A learn the correlation between the substrate polishing information included as input data in the training data 11A and the polishing surface temperature control information through reinforcement learning. The machine learning unit 401 may perform predetermined pre-processing on the input data (substrate polishing information) input to the training model 10A, or may perform predetermined post-processing on the output data (polishing surface temperature control information) output from the training model 10A.

[0081] The trained model storage unit 43 is a database that stores the trained learning model 10A (specifically, a group of adjusted weight parameters) generated by the machine learning unit 401. The trained learning model 10A stored in the trained model storage unit 43 is provided to a real system (e.g., the substrate polishing apparatus 2) via the network 7, a recording medium, or the like. Note that although the training data storage unit 42 and the trained model storage unit 43 are shown as separate storage units in FIG. 9, they may also be configured as a single storage unit.

[0082] The number of learning models 10A stored in the trained model storage unit 43 is not limited to one, and multiple learning models with different conditions, such as machine learning techniques, types of wafers W (size, thickness, film type, etc.), differences in the configuration of the polishing units 20, types of membranes 2022, types of retaining rings 2023, types of polishing pads 200, types of polishing fluids, types of cleaning fluids, types of heating fluids, types of cooling fluids, types of data included in substrate polishing information, types of data included in polishing surface temperature control information, etc. may be stored and used selectively or in parallel. In this case, the training data storage unit 42 may store multiple types of training data having data configurations corresponding to multiple training models with different conditions.

[0083] FIG. 10 is a schematic diagram showing an example of learning data 11A according to the first embodiment and its relationship with reinforcement learning. The machine learning unit 401 functions as a reinforcement learning agent. In the basic mechanism of reinforcement learning, the agent observes the state of the environment under predetermined conditions and selects an action according to a predetermined policy for the observed state. When the selected action changes the state of the environment, the agent receives a reward according to the change in state and evaluates the value of the selected action. By repeating this series of processes, namely, observing the state, selecting an action, and evaluating the value, the learning model 10A learns a policy for selecting an action that will maximize reward acquisition.

[0084] When the reinforcement learning by the machine learning unit 401 is adapted to the basic mechanism of reinforcement learning described above, the environment corresponds to the substrate polishing apparatus 2 that performs the polishing process so that the measured temperature distribution information of the polishing pad 200 satisfies the target temperature distribution information of the polishing pad 200.

[0085] The state s is represented by substrate polishing information corresponding to the input data. The substrate polishing information includes measured temperature distribution information of the polishing pad 200 and apparatus status information of the substrate polishing apparatus 2. The substrate polishing information may be time-series data from the past to the present.

[0086] The measured temperature distribution information of the polishing pad 200 is the temperature distribution of the polishing surface of the polishing pad 200 measured by the polishing surface temperature measuring unit 208. In this case, the measured temperature distribution information of the polishing pad 200 may be processed values ​​obtained by normalizing the measured values ​​measured by the polishing surface temperature measuring unit 208. The normalization process is a process for normalizing the temperature to a predetermined range (for example, 0 to 1).

[0087] The apparatus status information of the substrate polishing apparatus 2 includes polishing surface temperature adjustment status information indicating the status of the polishing surface heating unit 206 and the polishing surface cooling unit 207 that function as polishing surface temperature adjustment units.

[0088] The polishing-surface temperature adjustment state information includes at least one of the flow rate, pressure, temperature, supply position of the heating fluid, and the aperture of the heating fluid nozzle 2061 of the polishing-surface heating unit 206 as the current polishing-surface heating state, and at least one of the flow rate, pressure, temperature, supply position of the cooling fluid, and the aperture of the cooling fluid nozzle 2071 of the polishing-surface cooling unit 207 as the current polishing-surface cooling state. In this embodiment, the substrate polishing apparatus 2 is equipped with the polishing-surface heating unit 206 and the polishing-surface cooling unit 207 as the polishing-surface temperature adjustment units, and the polishing-surface temperature adjustment state information is described as including both the polishing-surface heating state and the polishing-surface cooling state. However, if the substrate polishing apparatus 2 is equipped with either the polishing-surface heating unit 206 or the polishing-surface cooling unit 207, the polishing-surface temperature adjustment state information may include either the corresponding polishing-surface heating state or the polishing-surface cooling state.

[0089] The apparatus status information of the substrate polishing apparatus 2 may include other information as long as it has an effect on the temperature distribution of the polishing surface of the polishing pad 200.

[0090] For example, the apparatus status information of the substrate polishing apparatus 2 may further include at least one of polishing table status information indicating the status of the polishing table portion 201, polishing head status information indicating the status of the polishing head portion 202, polishing fluid supply status information indicating the status of the polishing fluid supply portion 203, dressing status information indicating the status of the dressing portion 204, cleaning fluid injection status information indicating the status of the cleaning fluid injection portion 205, and substrate status information indicating the status of the wafer W, as shown in FIG.

[0091] The polishing table status information includes at least one of the rotation speed and rotation torque of the polishing table portion 201, the type and surface roughness of the polishing pad 200, and the polishing time during which the polishing process is performed by the polishing pad 200.

[0092] The polishing head status information includes at least one of the rotation speed, rotation torque, polishing position, and pressing load of the polishing head unit 202.

[0093] The polishing-fluid supply status information includes at least one of the flow rate, temperature, dropping position, and type of the polishing fluid. If the polishing fluid is a plurality of types (e.g., polishing liquid, pure water, chemical liquid, dispersant, etc.), it is sufficient that the information includes at least one of the flow rate, dropping position, and temperature for each type of polishing fluid. For example, if the polishing fluid is a polishing liquid and pure water, it is sufficient that the information includes at least one of the flow rate, dropping position, temperature, flow rate, and temperature of the pure water.

[0094] The dressing state information includes at least one of the rotation speed of the dressing unit 204, the rotation torque, the dressing position, the pressing load, the type of dressing pad, and the dressing time during which dressing is performed by the dressing pad.

[0095] The cleaning fluid ejection status information includes at least one of the flow rate, temperature, drop position, and type of the cleaning fluid.

[0096] The substrate state information includes at least one of the film thickness of the wafer W, the type of film on the wafer W, and the temperature distribution on the surface of the wafer W.

[0097] The apparatus status information of the substrate polishing apparatus 2 may further include environmental information indicating the state of the processing environment. The environmental information includes at least one of the temperature, humidity, and atmospheric pressure of the processing environment.

[0098] Action a is a candidate for polishing surface temperature control information that indicates the amount of control when adjusting the temperature distribution of the polishing pad 200 by the polishing surface heating unit 206 and the polishing surface cooling unit 207 that function as polishing surface temperature adjustment units.

[0099] The polishing surface temperature control information is a control amount for the polishing surface heating unit 206 and the polishing surface cooling unit 207. Specifically, the polishing surface temperature control information includes at least one of the flow rate, pressure, temperature, supply position of the heating fluid when the polishing surface of the polishing pad 200 is heated by the polishing surface heating unit 206, and the opening of the heating fluid jetting orifice 2061 of the polishing surface heating unit 206, and at least one of the flow rate, pressure, temperature, supply position of the cooling fluid when the polishing surface of the polishing pad 200 is cooled by the polishing surface cooling unit 207, and the opening of the cooling fluid jetting orifice 2071 of the polishing surface cooling unit 207. In this embodiment, the substrate polishing apparatus 2 is equipped with a polishing surface heating unit 206 and a polishing surface cooling unit 207 as polishing surface temperature adjustment units, and the polishing surface temperature control information is described as including information for both the polishing surface heating unit 206 and the polishing surface cooling unit 207. However, if the substrate polishing apparatus 2 is equipped with either the polishing surface heating unit 206 or the polishing surface cooling unit 207, the polishing surface temperature control information may include any of the corresponding information.

[0100] The control amounts for the polishing surface heating unit 206 and the polishing surface cooling unit 207 may be expressed as absolute amounts, or as amounts of change relative to the control amounts before control, or as step values ​​or continuous values. The control amounts for the polishing surface heating unit 206 and the polishing surface cooling unit 207 may be expressed as control patterns that arbitrarily combine the above parameters (flow rate, pressure, temperature, supply position of the heating fluid, opening of the heating fluid jetting orifice 2061, flow rate, pressure, temperature, supply position of the cooling fluid, and opening of the cooling fluid jetting orifice 2071). In this embodiment, action a is expressed as a control pattern consisting of a plurality of heating patterns (+P1, +P2, ..., +Pj), a maintenance pattern (±P0), and a plurality of cooling patterns (-P1, -P2, ..., -Pj), as shown in FIG. 10 . The case where the candidate is a .

[0101] The reward r is calculated based on the difference between the target temperature distribution information of the polishing pad 200 and the measured temperature distribution information of the polishing pad 200 based on the state s2 after the action a is taken for the state s1. In this case, the reward r is calculated so that the smaller the difference between the target temperature distribution information of the polishing pad 200 and the measured temperature distribution information, the larger the reward r becomes.

[0102] The measured temperature distribution information of the polishing pad 200 based on state s2 is the temperature distribution of the polishing pad 200 after action a is taken for state s1, i.e., after the polishing surface heating unit 206 and the polishing surface cooling unit 207 adjust the temperature of the polishing pad 200 based on the control amounts of the polishing surface heating unit 206 and the polishing surface cooling unit 207 indicated by the polishing surface temperature control information corresponding to action a. The measured temperature distribution information of the polishing pad 200 may be a measurement value measured by the polishing surface temperature measuring unit 208, or may be obtained based on the past operation history information 217 registered in the database 30, or may be estimated by performing a simulation using a polishing simulation device.

[0103] When reinforcement learning is used as the machine learning method, the training data includes only input data corresponding to state s. In other words, the training data does not include output data. The input data constituting the training data in this embodiment is substrate polishing information including measured temperature distribution information of the polishing pad 200 and apparatus status information of the substrate polishing apparatus 2, as shown in FIG.

[0104] Fig. 11 is a diagram showing an example of a learning model 10A according to the first embodiment. In Fig. 11, the evaluation of a predetermined action a in a state s is performed using an action value function Q(s, a) of the Q-learning method.

[0105] The action-value function Q(s, a) can be approximately calculated using a neural network model, for example, using a technique called DQN (Deep Q-Network), in which the state s is an input variable and the action-value function Q(s, a) when each action aj (+PM, ..., +P2, +P1, ±P0, -P1, -P2, ..., -PN) is taken in the state s as an output variable. In this case, the machine learning unit 401 updates the action-value function Q(s, aj) by adjusting the weights wk of the neural network model so that an error function (e.g., TD error) including the reward r, learning rate α, and discount rate γ as variables is minimized, and the learning model 10A learns the correlation between the input data (state s) and the polishing surface temperature control information (action aj). Note that any technique may be used for reinforcement learning, and in addition to Q-learning, for example, SARSA, Monte Carlo, etc. may also be used.

[0106] In order to approximately calculate the action value function Q(s, aj), the learning model 10A is configured as a neural network model shown in Fig. 11. The neural network model shown in Fig. 11 is configured from i neurons (x1 to xi) in the input layer 100, p neurons (y11 to y1p) in the first hidden layer 101A, q neurons (y21 to y2q) in the second hidden layer 101B, and j neurons (=M+N+1) in the output layer (+PM, ..., +P2, +P1, ±P0, -P1, -P2, ..., -PN).

[0107] Each neuron in the input layer 100 is associated with substrate polishing information as input data (state s) included in the training data.

[0108] Each neuron in the output layer 102 is assigned an action value function Q(s, a) when each action aj (+PM, ..., +P2, +P1, ±P0, -P1, -P2, ..., -PN) is taken for the state s. Each neuron in the output layer 102 is associated with each of the actions aj, and outputs the value of the action-value function Q(s, aj) for each action aj.

[0109] The first hidden layer 101A and the second hidden layer 101B are also called hidden layers, and the neural network may have a plurality of hidden layers in addition to the first hidden layer 101A and the second hidden layer 101B, or may have only the first hidden layer 101A as a hidden layer. Furthermore, synapses connecting neurons in each layer are established between the input layer 100 and the first hidden layer 101A, between the first hidden layer 101A and the second hidden layer 101B, and between the second hidden layer 101B and the output layer 102, and each synapse is assigned a weight wk (k is a natural number).

[0110] (machine learning methods) FIG. 12 is a flowchart showing an example of a machine learning method performed by the machine learning device 4 according to the first embodiment.

[0111] First, in step S100, the learning data acquisition unit 400 prepares a desired number of pieces of learning data as a preliminary preparation for starting machine learning, and stores the prepared learning data in the learning data storage unit 42. Several methods can be used to prepare the learning data.

[0112] Next, in step S110, the machine learning unit 401 prepares a pre-learning learning model 10A to start machine learning. The pre-learning learning model 10A prepared here is configured with the neural network model exemplified in Fig. 11, and the weight wk of each synapse is set to an initial value.

[0113] Next, in step S120, the machine learning unit 401 acquires, for example, one piece of learning data at random from the multiple sets of learning data stored in the learning data storage unit .

[0114] Next, in step S121, the machine learning unit 401 acquires target temperature distribution information of the polishing pad 200 for the input data included in the one learning data acquired in step S120.

[0115] Next, in step S130, the machine learning unit 401 inputs the input data (state s1) included in the one learning data acquired in step S120 to the input layer 100 of the prepared learning model 10A before learning (or during learning). As a result, the value of each action aj (the value of the action-value function Q(s, aj)) is output as an inference result from the output layer 102 of the learning model 10A.

[0116] Next, in step S140, the machine learning unit 401 selects, for example, a specific action a corresponding to the maximum value based on the value of the action-value function Q(s, aj) of each action aj output as an inference result from the output layer 102 in step S130. As a method for selecting a specific action a, for example, the greedy method, the ε-greedy method, or the like may be adopted.

[0117] Next, in step S150, the machine learning unit 401 acquires measured temperature distribution information of the polishing pad 200 when the action a selected in step S140 is taken for the state s1. That is, for the polishing pad 200 having a temperature distribution indicated by the measured temperature distribution included in the substrate polishing information (state s1) as input data, the machine learning unit 401 acquires the measured temperature distribution information of the polishing pad 200 after adjusting the temperature of the polishing pad 200 based on the control amounts of the polishing surface heating unit 206 and the polishing surface cooling unit 207 indicated by the polishing surface temperature control information corresponding to the action a selected in step S140. Obtain constant temperature distribution information.

[0118] Next, in step S160, the machine learning unit 401 calculates a reward r based on the difference between the target temperature distribution information of the polishing pad 200 acquired in step S121 and the measured temperature distribution information of the polishing pad 200 acquired in step S150.

[0119] Next, in step S170, the machine learning unit 401 updates the action value function Q(s, aj) by adjusting the weights wk of the neural network model based on the reward r calculated in step S160 so that the error function is minimized. As a result, the machine learning unit 401 causes the learning model 10A to learn the correlation between the substrate polishing information (state s) as input data and the polishing surface temperature control information (action aj). Note that the action value function Q(s, aj) does not need to be updated every time, and may be updated only when a predetermined condition is met, for example.

[0120] Next, in step S180, the machine learning unit 401 determines whether or not it is necessary to continue the machine learning. As a result, if it is determined to continue (No in step S180), the process returns to step S120 and performs steps S120 to S170 on the learning model 10A being trained. If it is determined to end the machine learning (Yes in step S180), the process proceeds to step S190.

[0121] Then, in step S190, the machine learning unit 401 stores the trained learning model 10A generated by adjusting the weights wk associated with each synapse in the trained model storage unit 43, thereby completing the series of machine learning methods shown in Fig. 12. As the trained learning model 10A, for example, parameters representing the structure of a neural network and the adjusted values ​​of the weights wk are stored. In the machine learning method, step S100 corresponds to a training data acquisition step, steps S110 to S180 correspond to a machine learning step, and step S190 corresponds to a trained model storage step.

[0122] As described above, the machine learning device 4 and machine learning method according to this embodiment can provide the learning model 10A that generates polishing surface temperature control information for substrate polishing information. Therefore, the temperature distribution of the polishing pad 200 can be appropriately adjusted based on the polishing surface temperature control information generated using the learning model 10A, thereby improving the processing quality of the wafer W during the polishing process.

[0123] (Substrate polishing apparatus 2 and information processing apparatus 6) FIG. 13 is a functional explanatory diagram showing an example of the substrate polishing apparatus 2 (information processing apparatus 6) according to the first embodiment.

[0124] The information acquiring unit 210b acquires substrate polishing information including measured temperature distribution information of the polishing pad 200 and apparatus status information of the substrate polishing apparatus 2. The measured temperature distribution information of the polishing pad 200 is acquired as measured values ​​by the polishing surface temperature measuring unit 208. The apparatus status information of the substrate polishing apparatus 2 is acquired as detected values ​​and measured values ​​by a group of sensors included in the substrate polishing apparatus 2. Note that the apparatus status information may be acquired as command values ​​to a group of modules instead of detected values ​​and measured values ​​by the group of sensors.

[0125] The information generation unit 210c inputs the substrate polishing information acquired by the information acquisition unit 210b as input data into the learning model 10A, thereby generating polishing surface temperature control information when polishing processing is performed on a polishing pad 200 having a temperature distribution indicated by the measured temperature distribution included in the substrate polishing information by a substrate polishing apparatus 2 in an apparatus state indicated by the apparatus state information included in the substrate polishing information.

[0126] For example, the information generation unit 210c inputs substrate polishing information as input data into the learning model 10A, and based on the value of the action value function Q(s, aj) of each action aj output as an inference result from the output layer 102, selects, for example, a specific action a corresponding to the maximum value, and generates polishing surface temperature control information corresponding to the specific action a as polishing surface temperature control information for the substrate polishing information of the input data.

[0127] 13 illustrates a case where action a(+P1) is selected and acceleration pattern 1 corresponding to action a(+P1) is generated as the polishing surface temperature control information. The polishing surface temperature control information generated by the information generator 210c is output to the polishing processing unit 210a, which controls the polishing surface heating unit 206 and the polishing surface cooling unit 207. That is, the polishing surface heating unit 206 and the polishing surface cooling unit 207 are controlled to execute acceleration pattern 1 corresponding to action a(+P1).

[0128] While performing the polishing process on the wafer W, the machine learning unit 210d performs machine learning of the learning model 10A by reinforcement learning based on the operation history information 217 acquired at the time.

[0129] Specifically, the machine learning unit 210d calculates the reward r based on the difference between the target temperature distribution information of the polishing pad 200 and the measured temperature distribution information obtained when the polishing surface heating unit 206 and the polishing surface cooling unit 207, which function as the polishing surface temperature adjusting unit, are controlled based on the control amount indicated by the polishing surface temperature control information generated by the information generating unit 210c. Then, based on the reward r, the machine learning unit 210d causes the learning model 10A to learn the correlation between the input data (substrate polishing information) included in the learning data 11A and the polishing surface temperature control information through reinforcement learning. Note that if the difference between the target temperature distribution information of the polishing pad 200 and the measured temperature distribution information is smaller than a predetermined value, the machine learning of the learning model 10A may be omitted. The machine learning technique used by the machine learning unit 210d is similar to that used by the machine learning unit 401 of the machine learning device 4, and therefore a detailed description thereof will be omitted.

[0130] The number of learning models 10A stored in the memory unit 214 is not limited to one, and multiple learning models with different conditions, such as machine learning techniques, types of wafers W (size, thickness, film type, etc.), differences in the configuration of the polishing units 20, types of membranes 2022, types of retaining rings 2023, types of polishing pads 200, types of polishing fluids, types of cleaning fluids, types of heating fluids, types of cooling fluids, types of data included in substrate polishing information, types of data included in polishing surface temperature control information, etc., may be stored and used selectively or in parallel. Furthermore, the learning model 10A may be stored in a memory unit of an external computer (e.g., a server-type computer or a cloud-type computer), and in that case, the information generation unit 210c and the machine learning unit 210d may access the external computer.

[0131] (Information processing method) 14 is a flowchart showing an example of an information processing method by the substrate polishing apparatus 2 (information processing apparatus 6) according to the first embodiment. An example of the operation of the substrate polishing apparatus 2 when polishing a specific wafer W will be described below.

[0132] First, in step S200, the polishing processing unit 210a of the substrate polishing apparatus 2 acquires detection values ​​and measurement values ​​from the group of sensors and starts polishing processing on a specific wafer W by operating the group of modules in cooperation with each other.

[0133] Next, in step S210, the information acquiring unit 210b acquires substrate polishing information including measured temperature distribution information of the polishing pad 200 and apparatus status information of the substrate polishing apparatus 2 at a specific time point.

[0134] Next, in step S211, the information generation unit 210c performs inference by inputting the substrate polishing information acquired in step S210 as input data into the learning model 10A, and acquires the values ​​of the action value functions Q(s, +PM), ..., Q(s, +P2), Q(s, +P1), Q(s, ±P0), Q(s, -P1), Q(s, -P2), ..., Q(s, -PN) of each action a as output data output from the output layer 102 of the learning model 10A.

[0135] Next, in step S212, as an example of post-processing of reinforcement learning, the information generator 210c selects the action a that gives the maximum value among the values ​​of the action value function Q(s, aj) of each action amn output from each neuron in the output layer 102 as output data.

[0136] Next, in step S213, the output processing unit 53 generates polishing surface temperature control information corresponding to the action a selected in step S211 (in the example of Figure 13, acceleration pattern 1 corresponding to action a (+P1)) and outputs it to the polishing processing unit 210a.

[0137] Next, in step S220, the polishing processing unit 210a adjusts the temperature of the polishing pad 200 based on the control amounts of the polishing surface heating unit 206 and the polishing surface cooling unit 207 indicated by the polishing surface temperature control information generated in step S213.

[0138] Next, in step S230, the machine learning unit 210d calculates a reward r based on the difference between the target temperature distribution information of the polishing pad 200 and the measured temperature distribution information when the temperature of the polishing pad 200 was adjusted in step S220. The measured temperature distribution information here is measured when a predetermined sampling time interval has elapsed since the specific point in time in step S210.

[0139] Next, in step S240, the machine learning unit 210d updates the action value function Q(s, aj) based on the reward r calculated in step S230, and the machine learning unit 401 causes the learning model 10A to learn the correlation between the substrate polishing information (state s) as input data and the polishing surface temperature control information (action aj).

[0140] Next, in step S250, the polishing processing unit 210a determines whether the polishing processing has ended. As a result, if it is determined that the polishing processing has not ended (No in step S250), the process returns to step S210 after the sampling time interval has elapsed, and the polishing processing continues. On the other hand, if it is determined that the polishing processing has ended (Yes in step S250), the series of information processing methods shown in Fig. 14 ends. In the information processing method, steps S200 and S220 correspond to the polishing processing step, step S210 corresponds to the information acquisition step, steps S211 to S213 correspond to the information generation step, and steps S230 and S240 correspond to the machine learning step.

[0141] As described above, according to the substrate polishing apparatus 2 (information processing device 6) and information processing method of this embodiment, the substrate polishing information is input to the learning model 10A, and polishing surface temperature control information for the substrate polishing information is generated. Therefore, the temperature distribution of the polishing pad 200 is appropriately adjusted based on the polishing surface temperature control information, thereby improving the processing quality of the wafer W by the polishing process.

[0142] Furthermore, since machine learning of the learning model 10A is performed while the polishing process of the wafer W is being performed, the inference accuracy of the learning model 10A can be further improved depending on the environment in which the polishing process is actually performed.

[0143] (Second embodiment) In the first embodiment, reinforcement learning is used as the machine learning method for the learning model 10A. In contrast, the second embodiment differs from the first embodiment in that it employs supervised learning. The following describes the substrate polishing apparatus 2a (particularly the information processing device 6a) and the machine learning device 4a according to the second embodiment, focusing on the differences from the first embodiment.

[0144] Fig. 15 is a block diagram showing an example of a machine learning device 4a according to the second embodiment. Fig. 16 is a diagram showing an example of a learning model 10B and learning data 11B.

[0145] The learning data acquisition unit 400 is connected to an external device via the communication unit 41 and the network 7, and acquires learning data 11B, which is composed of substrate polishing information as input data and polishing surface temperature control information as output data. The learning data 11B is acquired, for example, by referring to the database 30. In this embodiment, the learning data 11B is data used as teacher data (training data), verification data, and test data in supervised learning. The polishing surface temperature control information is data used as a correct answer label in supervised learning.

[0146] As in the first embodiment, the substrate polishing information includes measured temperature distribution information of the polishing pad 200 and apparatus status information of the substrate polishing apparatus 2. The substrate polishing information may also include target temperature distribution information of the polishing pad 200, as shown in FIG.

[0147] Furthermore, the apparatus status information of the substrate polishing apparatus 2 may further include at least one of polishing table status information, polishing head status information, polishing fluid supply status information, and substrate status information, as in the first embodiment, or may include polishing surface temperature adjustment status information. Furthermore, the apparatus status information of the substrate polishing apparatus 2 may further include at least one of dressing status information, cleaning fluid injection status information, and environmental information. Note that the various data included in the substrate polishing information are the same as in the first embodiment, and therefore description thereof will be omitted.

[0148] As in the first embodiment, the polishing surface temperature control information includes at least one of the flow rate, pressure, temperature, supply position of the heating fluid when the polishing surface of the polishing pad 200 is heated by the polishing surface heating unit 206, and the opening of the heating fluid injection port 2061 of the polishing surface heating unit 206, and at least one of the flow rate, pressure, temperature, supply position of the cooling fluid when the polishing surface of the polishing pad 200 is cooled by the polishing surface cooling unit 207, and the opening of the cooling fluid injection port 2071 of the polishing surface cooling unit 207.

[0149] The machine learning unit 401 performs machine learning using multiple sets of training data 11B. That is, the machine learning unit 401 inputs multiple sets of training data 11B to the training model 10B, and generates a trained training model 10B by having the training model 10B learn the correlation between the substrate polishing information and the polishing surface temperature control information contained in the training data 11B through supervised learning.

[0150] The learning model 10B employs, for example, a neural network structure and includes an input layer 100, an intermediate layer 101, and an output layer 102. Synapses (not shown) that connect each neuron are laid between each layer, and each synapse is associated with a weight. A group of weight parameters consisting of the weights of each synapse is adjusted by machine learning.

[0151] The input layer 100 has neurons in a number corresponding to the substrate polishing information as input data, and each value of the substrate polishing information is input to each neuron. The output layer 102 has neurons in a number corresponding to the polishing surface temperature control information as output data, and the inference result (inference result) of the polishing surface temperature control information for the substrate polishing information is output as output data. When the learning model 10B is configured as a regression model, the polishing surface temperature control information is calculated based on a predetermined range (for example, If the learning model 10B is configured as a classification model, the polishing surface temperature control information is output as a score (accuracy) for each class, normalized to a predetermined range (for example, 0 to 1).

[0152] (machine learning methods) FIG. 17 is a flowchart showing an example of a machine learning method performed by the machine learning device 4a according to the second embodiment.

[0153] First, in step S300, the learning data acquisition unit 400 acquires a desired number of pieces of learning data 11B as a preliminary preparation for starting machine learning, and stores the acquired learning data 11B in the learning data storage unit 42.

[0154] Next, in step S310, the machine learning unit 401 prepares a pre-learning learning model 10B to start machine learning. The pre-learning learning model 10B prepared here is configured with the neural network model exemplified in Fig. 16, and the weights of each synapse are set to initial values.

[0155] Next, in step S320, the machine learning unit 401 acquires, for example, one set of training data 11B at random from the multiple sets of training data 11B stored in the training data storage unit .

[0156] Next, in step S330, the machine learning unit 401 inputs the substrate polishing information (input data) contained in one set of training data 11B to the input layer 100 of the prepared learning model 10B before (or during) training. As a result, polishing surface temperature control information (output data) is output as an inference result from the output layer 102 of the learning model 10B, and this output data was generated by the learning model 10B before (or during) training. Therefore, in the before (or during) training state, the output data output as an inference result indicates information different from the polishing surface temperature control information (correct label) contained in the training data 11B.

[0157] Next, in step S340, the machine learning unit 401 performs machine learning by comparing the polishing surface temperature control information (correct label) included in the set of learning data 11B acquired in step S320 with the polishing surface temperature control information (output data) output from the output layer as an inference result in step S330, and performing a process (backpropagation) to adjust the weight of each synapse. In this way, the machine learning unit 401 causes the learning model 10B to learn the correlation between the substrate polishing information and the polishing surface temperature control information.

[0158] Next, in step S350, the machine learning unit 401 determines whether a predetermined learning termination condition has been met, for example, based on the evaluation value of an error function based on the polishing surface temperature control information (correct label) included in the learning data 11B and the polishing surface temperature control information (output data) output as an inference result, or the remaining number of unlearned learning data 11B stored in the learning data memory unit 42.

[0159] If the machine learning unit 401 determines in step S350 that the learning termination condition is not satisfied and that machine learning should continue (No in step S350), the process returns to step S320, and performs steps S320 to S340 multiple times on the learning model 10B being trained using untrained training data 11B. On the other hand, if the machine learning unit 401 determines in step S350 that the learning termination condition is satisfied and that machine learning should end (Yes in step S350), the process proceeds to step S360.

[0160] Then, in step S360, the machine learning unit 401 calculates the The trained learning model 10B (adjusted weight parameter group) generated by adjusting the weights is stored in the trained model storage unit 43, and the series of machine learning methods shown in Fig. 17 is terminated. In the machine learning method, step S300 corresponds to the training data acquisition step, steps S310 to S350 correspond to the machine learning step, and step S360 corresponds to the trained model storage step.

[0161] As described above, the machine learning device 4a and the machine learning method according to this embodiment can provide the learning model 10B that generates polishing surface temperature control information for the substrate polishing information. Therefore, the temperature distribution of the polishing pad 200 can be appropriately adjusted based on the polishing surface temperature control information generated by using the learning model 10B, thereby improving the processing quality of the wafer W in the polishing process.

[0162] Fig. 18 is a block diagram showing an example of a substrate polishing apparatus 2a (information processing apparatus 6a) according to the second embodiment. Fig. 19 is a functional explanatory diagram showing an example of the substrate polishing apparatus 2a (information processing apparatus 6a) according to the second embodiment.

[0163] The control unit 210 functions as a polishing unit 210a, an information acquisition unit 210b, and an information generation unit 210c, and constitutes an information processing device 6a.

[0164] As in the first embodiment, the information acquiring unit 210b acquires substrate polishing information including measured temperature distribution information of the polishing pad 200 and apparatus status information of the substrate polishing apparatus 2. The substrate polishing information may also include target temperature distribution information of the polishing pad 200, as shown in Fig. 18. The various data included in the substrate polishing information are the same as those in the first embodiment, and therefore will not be described here.

[0165] The information generation unit 210c inputs the substrate polishing information acquired by the information acquisition unit 210b as input data into the learning model 10B, thereby generating polishing surface temperature control information when polishing processing is performed on a polishing pad 200 having a temperature distribution indicated by the measured temperature distribution included in the substrate polishing information by a substrate polishing apparatus 2 in an apparatus state indicated by the apparatus state information included in the substrate polishing information.

[0166] (Information processing method) 20 is a flowchart showing an example of an information processing method by the substrate polishing apparatus 2a (information processing apparatus 6a) according to the second embodiment. An example of the operation of the substrate polishing apparatus 2a when polishing a specific wafer W will be described below.

[0167] First, in step S400, the polishing processing unit 210a of the substrate polishing apparatus 2 acquires detection values ​​and measurement values ​​from the group of sensors and starts polishing processing on a specific wafer W by operating the group of modules in cooperation with each other.

[0168] Next, in step S410, the information acquisition unit 210b acquires substrate polishing information at a specific time point.

[0169] Next, in step S411, the information generating unit 210c inputs the substrate polishing information acquired in step S410 as input data into the learning model 10A, thereby generating polishing surface temperature control information.

[0170] Next, in step S412, the polishing processing unit 210a adjusts the temperature of the polishing pad 200 based on the control amounts of the polishing surface heating unit 206 and the polishing surface cooling unit 207 indicated by the polishing surface temperature control information generated in step S411.

[0171] Next, in step S420, the polishing processing unit 210a determines whether the polishing processing has ended. As a result, if it is determined that the polishing processing has not ended (No in step S420), the process returns to step S410 after the sampling time interval has elapsed, and the polishing processing continues. On the other hand, if it is determined that the polishing processing has ended (Yes in step S420), the series of information processing methods shown in Fig. 20 ends. In the information processing method, steps S400 and S412 correspond to the polishing processing step, step S410 corresponds to the information acquisition step, and step S411 corresponds to the information generation step.

[0172] As described above, according to the substrate polishing apparatus 2a (information processing device 6a) and information processing method of this embodiment, the substrate polishing information is input to the learning model 10B, and polishing surface temperature control information for the substrate polishing information is generated. Therefore, the temperature distribution of the polishing pad 200 is appropriately adjusted based on the polishing surface temperature control information, thereby improving the processing quality of the wafer W by the polishing process.

[0173] (Other embodiments) The present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the spirit and scope of the present invention, all of which are included in the technical concept of the present invention.

[0174] In the above embodiment, the information processing devices 6, 6a are described as being incorporated into the substrate polishing apparatuses 2, 2a. However, the information processing devices 6, 6a may be configured as devices separate from the substrate polishing apparatuses 2, 2a, and may be configured as, for example, a general-purpose or dedicated computer 900. Furthermore, the information processing devices 6, 6a may be incorporated into the user terminal device 5, and the machine learning devices 4, 4a may be incorporated into the user terminal device 5.

[0175] In the above embodiment, a case has been described in which a neural network is used as a learning model for realizing machine learning by the machine learning unit 401, but other machine learning models may also be used. Examples of other machine learning models include tree types such as decision trees and regression trees, ensemble learning such as bagging and boosting, and neural network types (including deep learning) such as recurrent neural networks, convolutional neural networks, and LSTM. (including hierarchical clustering, non-hierarchical clustering, k-nearest neighbors, k-means, etc.) Examples include multivariate analysis such as filtering, principal component analysis, factor analysis, and logistic regression, as well as support vector machines.

[0176] (Machine learning programs and information processing programs) The present invention can also be provided in the form of a program (machine learning program) that causes the computer 900 to function as each unit included in the machine learning device 4, 4a, or a program (machine learning program) that causes the computer 900 to execute each step included in the machine learning method. The present invention can also be provided in the form of a program (information processing program) that causes the computer 900 to function as each unit included in the substrate polishing apparatus 2, 2a, or a program (information processing program) that causes the computer 900 to execute each step included in the information processing method according to the above embodiment.

[0177] (Inference device, inference method and inference program) The present invention can be provided not only in the form of the substrate polishing apparatus 2, 2a (information processing method or information processing program) according to the above embodiment, but also in the form of an inference device (inference method or inference program) used to infer polishing surface temperature control information. In this case, the inference device (inference method or inference program) can include a memory and a processor, and the processor can execute a series of processes. The method includes an information acquisition process (information acquisition step) for acquiring substrate polishing information, and an inference process (inference step) for inferring polishing surface temperature control information when polishing is performed on a polishing surface having a temperature distribution indicated by measured temperature distribution information included in the substrate polishing information by a substrate polishing apparatus 2, 2a in an apparatus state indicated by apparatus state information included in the substrate polishing information, once the substrate polishing information is acquired by the information acquisition process.

[0178] By providing it in the form of an inference device (inference method or inference program), it can be more easily applied to various devices than when it is implemented in an information processing device. It will be naturally understood by those skilled in the art that when the inference device (inference method or inference program) infers polishing surface temperature control information, it may apply an inference method implemented by an information generating unit using a trained learning model generated by the machine learning device and machine learning method according to the above embodiments. [Explanation of symbols]

[0179] 1...substrate polishing system, 2, 2a...substrate polishing apparatus, 3...database device, 4, 4a...machine learning device, 5...user terminal device, 6, 6a...information processing device, 10A, 10B...Learning model, 11A, 11B...Learning data 20...polishing unit, 21...control unit, 30...database, 40...control unit, 41...communication unit, 42...learning data storage unit, 43...Trained model memory unit, 200... polishing pad, 201... polishing table portion, 202... polishing head portion, 203... polishing fluid supply unit, 204... dressing unit, 205... cleaning fluid spray unit, 206...polishing surface heating section, 207...polishing surface cooling section, 208...polishing surface temperature measurement section, 209... Polishing unit measuring section, 210...control unit, 210a...polishing processing unit, 210b...information acquisition unit, 210c...information generation unit, 210d...machine learning unit, 211...communication unit, 212...input unit, 213...output unit, 214...storage unit, 215...equipment setting information, 216...substrate recipe information, 217...operation history information, 400...Learning data acquisition unit, 401...Machine learning unit

Claims

1. an information acquiring unit that acquires substrate polishing information including measured temperature distribution information indicating a measured value of a temperature distribution on a polishing surface when a polishing process for polishing a substrate with a polishing surface of a polishing pad is performed by a substrate polishing apparatus, and apparatus status information indicating an apparatus status of the substrate polishing apparatus; an information generating unit that generates polishing surface temperature control information indicating a control amount of a polishing surface temperature adjusting unit that adjusts the temperature distribution of the polishing surface when the polishing process is performed by the substrate polishing apparatus in the apparatus state indicated by the apparatus state information included in the substrate polishing information, for the polishing surface having the temperature distribution indicated by the measured temperature distribution information included in the substrate polishing information, by inputting the substrate polishing information acquired by the information acquiring unit into a learning model, The learning model is a trained model that has been trained by machine learning to determine the correlation between the substrate polishing information and the polishing surface temperature control information; Information processing device.

2. The apparatus status information included in the substrate polishing information is polishing surface temperature adjustment status information indicating the status of the polishing surface temperature adjustment unit; The learning model is A trained model in which the correlation is trained by reinforcement learning. The information processing device according to claim 1 .

3. a machine learning unit that calculates a reward based on a difference between target temperature distribution information indicating a target value of the temperature distribution of the polishing surface and the measured temperature distribution information when the polishing surface temperature adjustment unit is controlled based on the control amount indicated by the polishing surface temperature control information generated by the information generation unit, and causes the learning model to learn the correlation by reinforcement learning based on the reward; The information processing device according to claim 2 .

4. The substrate polishing information includes: target temperature distribution information indicating a target value of the temperature distribution of the polishing surface; The learning model is The correlation is a trained model trained by supervised learning. The information processing device according to claim 1 .

5. The measured temperature distribution information included in the substrate polishing information is A processed value obtained by normalizing the measurement value. The information processing device according to claim 1 .

6. The apparatus status information included in the substrate polishing information is polishing table status information indicating the status of a polishing table portion that rotatably supports the polishing pad; polishing head status information indicating the status of a polishing head portion that presses the substrate against the polishing pad; polishing fluid supply status information indicating the status of a polishing fluid supply unit that supplies a polishing fluid to the polishing pad; dressing state information indicating the state of a dressing unit that dresses the polishing pad; cleaning fluid ejection status information indicating the status of a cleaning fluid ejection unit that ejects cleaning fluid onto the polishing pad; and at least one piece of substrate status information indicating a status of the substrate; The information processing device according to claim 1 .

7. The polishing table state information is the rotation speed of the polishing table; a rotation torque of the polishing table; The type of the polishing pad; the surface roughness of the polishing pad; and at least one of a polishing time during which the polishing process is performed by the polishing pad, The polishing head status information is the rotation speed of the polishing head unit; the rotational torque of the polishing head unit; a polishing position of the polishing head unit; and at least one of the pressing loads of the polishing head portion, The polishing fluid supply status information is the flow rate of the polishing fluid; the temperature of the polishing fluid; a drop position of the polishing fluid; and at least one of the polishing fluid types; The dressing status information is the rotation speed of the dressing unit; a rotational torque of the dressing portion; a dressing position of the dressing portion; a pressing load of the dressing portion; The type of dressing pad that the dressing unit has, and at least one of a dressing time during which the dressing is performed by the dressing pad; The cleaning fluid ejection status information is the flow rate of the cleaning fluid; the temperature of the cleaning fluid; a drop position of the cleaning fluid; and The type of cleaning fluid and The substrate state information is the film thickness of the substrate; The type of film on the substrate, and a temperature distribution on the surface of the substrate; The information processing device according to claim 6 .

8. The polishing surface temperature control information is at least one of a flow rate, a pressure, a temperature, a supply position of a heating fluid when heating the polishing surface, and an opening of a heating fluid injection port for injecting the heating fluid; the cooling fluid supply position when cooling the polishing surface, and at least one of the flow rate, pressure, temperature, and supply position of the cooling fluid, and the opening of a cooling fluid injection port for injecting the cooling fluid, The information processing device according to claim 1 .

9. a polishing table that rotatably supports a polishing pad; a polishing head unit that presses a substrate against a polishing surface of the polishing pad to polish the substrate; a polishing fluid supply unit that supplies a polishing fluid to the polishing pad; a polishing surface temperature adjusting unit that adjusts the temperature distribution of the polishing surface; a polishing surface temperature measuring unit for measuring a temperature distribution of the polishing surface; a control unit that controls a polishing process of polishing the substrate by the polishing surface, The control unit an information acquiring unit that acquires substrate polishing information including measured temperature distribution information indicating a measured value of the temperature distribution of the polishing surface measured by the polishing surface temperature measuring unit and apparatus status information indicating an apparatus status of the substrate polishing apparatus; an information generating unit that generates polishing surface temperature control information indicating a control amount of the polishing surface temperature adjusting unit when the polishing process is performed on the polishing surface having the temperature distribution indicated by the measured temperature distribution information included in the substrate polishing information by the substrate polishing apparatus in the apparatus state indicated by the apparatus state information included in the substrate polishing information by inputting the substrate polishing information acquired by the information acquiring unit into a learning model; a polishing processing unit that controls the polishing surface temperature adjusting unit based on the control amount indicated by the polishing surface temperature control information generated by the information generating unit, The learning model is a trained model that has been trained by machine learning to determine the correlation between the substrate polishing information and the polishing surface temperature control information; Substrate polishing equipment.

10. The polishing surface temperature adjusting unit is one or more heating fluid injection ports for injecting a heating fluid onto the polishing surface; and one or more cooling fluid injection ports for injecting a cooling fluid onto the polishing surface to cool the polishing surface, The polishing surface temperature control information is At least one of the flow rate, pressure, temperature, and supply position of the heating fluid, and the opening of the heating fluid injection port; and at least one of a flow rate, a pressure, a temperature, a supply position of the cooling fluid, and an opening degree of the cooling fluid injection port.

10. The substrate polishing apparatus according to claim 9.

11. The apparatus status information included in the substrate polishing information is polishing surface temperature adjustment status information indicating the status of the polishing surface temperature adjustment unit; The learning model is a trained model in which the correlation is trained by reinforcement learning, The substrate polishing apparatus includes: a machine learning unit that calculates a reward based on a difference between target temperature distribution information indicating a target value of the temperature distribution of the polishing surface and the measured temperature distribution information when the polishing surface temperature adjustment unit is controlled based on the control amount indicated by the polishing surface temperature control information generated by the information generation unit, and causes the learning model to learn the correlation by reinforcement learning based on the reward; 11. The substrate polishing apparatus according to claim 9 or 10.

12. An inference device comprising a memory and a processor, The processor: an information acquisition process for acquiring substrate polishing information including measured temperature distribution information indicating a measured value of a temperature distribution on a polishing surface when a polishing process for polishing a substrate with a polishing surface of a polishing pad is performed by a substrate polishing apparatus, and apparatus status information indicating an apparatus status of the substrate polishing apparatus; When the substrate polishing information is acquired by the information acquisition process, a polishing surface temperature controller is provided to adjust the temperature distribution of the polishing surface when the polishing process is performed by the substrate polishing apparatus in the apparatus state indicated by the apparatus state information included in the substrate polishing information, for the polishing surface having the temperature distribution indicated by the measured temperature distribution information included in the substrate polishing information. an inference process for inferring control information; Reasoning device.

13. a learning data acquisition unit that acquires a plurality of sets of learning data that are at least composed of input data; a machine learning unit that uses the plurality of sets of learning data acquired by the learning data acquisition unit to train a learning model by machine learning to learn a correlation between the input data and the output data; a learned model storage unit that stores the learned model in which the correlation is learned by the machine learning unit, The input data is substrate polishing information including measured temperature distribution information indicating a measured value of a temperature distribution on a polishing surface when a polishing process for polishing a substrate with a polishing surface of a polishing pad is performed by a substrate polishing apparatus, and apparatus status information indicating an apparatus status of the substrate polishing apparatus, The output data is polishing surface temperature control information indicating a control amount of a polishing surface temperature adjuster that adjusts the temperature distribution of the polishing surface when the polishing process is performed by the substrate polishing apparatus; Machine learning device.

14. The learning data is It consists of only the input data, The apparatus status information included in the substrate polishing information is polishing surface temperature adjustment status information indicating the status of the polishing surface temperature adjustment unit; The machine learning unit causing the learning model to learn the correlation by reinforcement learning; The machine learning device according to claim 13.

15. The learning data is The input data and the output data are configured as follows: The substrate polishing information includes: target temperature distribution information indicating a target value of the temperature distribution of the polishing surface; The machine learning unit The correlation is learned by the learning model through supervised learning. The machine learning device according to claim 13.

16. an information acquiring step of acquiring substrate polishing information including measured temperature distribution information indicating a measured value of a temperature distribution on a polishing surface when a polishing process for polishing a substrate with a polishing surface of a polishing pad is performed by a substrate polishing apparatus, and apparatus status information indicating an apparatus status of the substrate polishing apparatus; an information generating step of inputting the substrate polishing information acquired by the information acquiring step into a learning model to generate polishing surface temperature control information indicating a control amount of a polishing surface temperature adjusting unit that adjusts the temperature distribution of the polishing surface when the polishing process is performed by the substrate polishing apparatus in the apparatus state indicated by the apparatus state information included in the substrate polishing information, for the polishing surface having the temperature distribution indicated by the measured temperature distribution information included in the substrate polishing information; The learning model is a trained model that has been trained by machine learning to determine the correlation between the substrate polishing information and the polishing surface temperature control information; Information processing methods.

17. An inference method executed by an inference device having a memory and a processor, The processor: an information acquisition process for acquiring substrate polishing information including measured temperature distribution information indicating a measured value of a temperature distribution on a polishing surface when a polishing process for polishing a substrate with a polishing surface of a polishing pad is performed by a substrate polishing apparatus, and apparatus status information indicating an apparatus status of the substrate polishing apparatus; When the substrate polishing information is acquired by the information acquisition process, an inference process is executed to infer polishing surface temperature control information indicating a control amount of a polishing surface temperature adjuster that adjusts the temperature distribution of the polishing surface when the polishing process is performed by the substrate polishing apparatus in the apparatus state indicated by the apparatus state information included in the substrate polishing information, with respect to the polishing surface having the temperature distribution indicated by the measured temperature distribution information included in the substrate polishing information. Reasoning method.

18. 1. A computer-implemented machine learning method comprising: a learning data acquisition step of acquiring a plurality of sets of learning data composed at least of input data; a machine learning step of causing a learning model to learn a correlation between the input data and the output data by machine learning using the plurality of sets of learning data acquired by the learning data acquisition step; a learned model storage step of storing the learned model, which has learned the correlation through the machine learning step, in a learned model storage unit; The input data is substrate polishing information including measured temperature distribution information indicating a measured value of a temperature distribution on a polishing surface when a polishing process for polishing a substrate with a polishing surface of a polishing pad is performed by a substrate polishing apparatus, and apparatus status information indicating an apparatus status of the substrate polishing apparatus, The output data is polishing surface temperature control information indicating a control amount of a polishing surface temperature adjuster that adjusts the temperature distribution of the polishing surface when the polishing process is performed by the substrate polishing apparatus; Machine learning methods.

Images