JP7565393B2 - State determination method, state determination device, and storage medium - Google Patents

Description

This disclosure relates to a state determination method, a state determination device, and a storage medium.

Patent Document 1 describes a failure prediction method for predicting failures occurring in each of multiple substrate processing equipment. This failure prediction method includes a process of collecting the operation history of the equipment as historical data, and a process of predicting equipment failures based on the collected historical data and a failure pattern that is updated based on the historical data of the failed equipment.

JP 2010-225632 A

The substrate processing apparatus is equipped with a number of actuators for moving members as equipment. The present disclosure provides a state determination method, a state determination device, and a storage medium suitable for determining the state of the actuators.

A state determination method according to one aspect of the present disclosure is a method for determining the state of a target actuator included in a plurality of actuators mounted on a substrate processing apparatus. This state determination method includes acquiring a feature amount related to the operation of each of the plurality of actuators, calculating a probability distribution model for each of the plurality of actuators based on the feature amount, and determining the state of the target actuator based on the probability distribution model for each of the plurality of actuators.

The present disclosure provides a state determination method, a state determination device, and a storage medium suitable for determining the state of an actuator.

FIG. 1 is a schematic diagram showing an example of a schematic configuration of a substrate processing system. FIG. 2 is a schematic diagram showing an example of a schematic configuration of a coating and developing apparatus. FIG. 3 is a schematic diagram showing an example of the internal configuration of the heat treatment unit. FIG. 4 is a block diagram showing an example of a functional configuration of the state determination unit. FIG. 5 is a diagram showing an example of a detection result regarding the operation time of the actuator. 6A is a diagram showing an example of a transition regarding the operation time of the actuator for each period, and FIG 6B is a diagram showing an example of a transition regarding the index value for each period. FIG. 7 is a block diagram illustrating an example of a hardware configuration of the control device. FIG. 8 is a flowchart illustrating an example of a state determination procedure. FIG. 9 is a diagram for explaining an example of a method for determining whether an abnormal state is approaching.

Various exemplary embodiments are described below.

A state determination method according to one aspect of the present disclosure is a method for determining the state of a target actuator included in a plurality of actuators mounted on a substrate processing apparatus. This state determination method includes acquiring a feature amount related to the operation of each of the plurality of actuators, calculating a probability distribution model for each of the plurality of actuators based on the feature amount, and determining the state of the target actuator based on the probability distribution model for each of the plurality of actuators.

In this state determination method, the state of a target actuator is determined based on a probability distribution model for each of the multiple actuators. When the state of the target actuator changes from a normal state, fluctuations occur in the feature quantities related to the operation. As a result, the probability distribution model for the target actuator among the probability distribution models for each of the multiple actuators changes. By capturing the change in the probability distribution model of the target actuator based on the probability distribution models of the multiple actuators, it is possible to determine the state of the target actuator. In other words, a state determination method suitable for determining the state of an actuator is provided.

The above-mentioned determination may include determining a reference probability distribution model that serves as a basis for determination from the probability distribution models of each of the multiple actuators, and determining the state of the target actuator by comparing the reference probability distribution model with a target probability distribution model that is a probability distribution model of the target actuator. In this case, by creating a reference probability distribution model that serves as a basis for state determination from the probability distribution models of each of the multiple actuators and using this for state determination, it becomes possible to easily determine the state of the target actuator.

The reference probability distribution model may be obtained by calculating the average of the probability distribution models of each of the multiple actuators. In this case, by using the average of the probability distribution models of each of the multiple actuators, it is possible to obtain a reference that is more suitable for state judgment.

The above-mentioned determination may include determining the state of the target actuator based also on the skewness or kurtosis of the probability distribution model of the target actuator. By determining the state based also on the skewness or kurtosis, a determination is made that more accurately reflects the characteristics of the probability distribution model of the target actuator, making it possible to more reliably determine the state of the target actuator.

The above-mentioned determination may be a determination of whether the target actuator is abnormal. The above-mentioned state determination method may further include outputting a signal indicating that the target actuator is abnormal when it is determined that the target actuator is abnormal. In this case, when it is determined that the target actuator is in an abnormal state, it is possible to execute a process that is different from the process when the target actuator is in a normal state.

The above-mentioned determination may be a determination of whether the target actuator is approaching an abnormal state. The above-mentioned state determination method may further include outputting a signal indicating that the target actuator is approaching an abnormal state when it is determined that the target actuator is approaching an abnormal state. In this case, when it is determined that the target actuator is approaching an abnormal state, it is possible to execute a process that is different from the process when the target actuator is in a normal state.

The actuators may have a common operating target. The actuators may be included in a substrate processing apparatus and provided in different processing units that perform a common process. In this case, when the operation of the target actuator deviates from the normal state, a deviation occurs between the probability distribution model of the target actuator and other probability distribution models, making it possible to determine the state of the target actuator.

The actuators may operate on different targets. The actuators may be included in a substrate processing apparatus and provided in a single processing unit that performs a specified process. In this case, the state is determined based on a probability distribution model of features related to the actuator operation. Therefore, even if the actuators operate on different targets, it is possible to determine the state of the target actuator based on the deviation from other probability distribution models.

Each of the multiple actuators may be an air cylinder driven by air pressure. As the air cylinder moves away from its normal state, the feature quantities related to the operation of the air cylinder move away from the normal state values. As a result, the probability distribution model based on the feature quantities of the target air cylinder as the target actuator deviates from the probability distribution models of the other air cylinders, making it possible to determine the state of the target actuator.

A state determination device according to another aspect of the present disclosure is a device for determining the state of a target actuator included in a plurality of actuators mounted on a substrate processing apparatus. This state determination device includes an acquisition unit that acquires feature quantities related to the operation of each of the plurality of actuators, a probability distribution calculation unit that calculates a probability distribution model based on the feature quantities for each of the plurality of actuators, and a determination unit that determines the state of the target actuator based on the probability distribution model for each of the plurality of actuators.

A storage medium according to another aspect of the present disclosure is a computer-readable storage medium that stores a program for causing an apparatus to execute the above-mentioned state determination method.

Various exemplary embodiments will be described in detail below with reference to the drawings. In the description, identical elements or elements having the same functions are given the same reference numerals, and duplicate descriptions will be omitted.

[Substrate Processing System]
First, the schematic configuration of the substrate processing system will be described with reference to Fig. 1 and Fig. 2. The substrate processing system 1 is a system for forming a photosensitive film on a substrate, exposing the photosensitive film, and developing the photosensitive film. The substrate to be processed is, for example, a semiconductor wafer W. The photosensitive film is, for example, a resist film. The substrate processing system 1 includes a coating/developing apparatus 2 and an exposure apparatus 3. The exposure apparatus 3 performs an exposure process on a resist film (photosensitive film) formed on a wafer W (substrate). Specifically, the exposure apparatus 3 irradiates an energy beam onto an exposure target portion of the resist film by a method such as immersion exposure. The coating/developing apparatus 2 performs a process of forming a resist film on the surface of the wafer W (substrate) before the exposure process by the exposure apparatus 3, and performs a development process on the resist film after the exposure process.

[Substrate Processing Apparatus]
1 and 2 , the coating and developing apparatus 2 includes a carrier block 4, a processing block 5, an interface block 6, and a control device 100.

The carrier block 4 introduces wafers W into the coating/developing apparatus 2 and removes wafers W from the coating/developing apparatus 2. For example, the carrier block 4 can support multiple carriers C for wafers W, and has a built-in transfer device A1 including a transfer arm. The carrier C holds, for example, multiple circular wafers W. The transfer device A1 removes wafers W from the carrier C and transfers them to the processing block 5, and receives wafers W from the processing block 5 and returns them to the carrier C.

The processing block 5 has multiple processing modules 11, 12, 13, and 14. Each of the processing modules 11, 12, 13, and 14 incorporates a coating unit U1, a heat processing unit U2, and a transfer device A3 including a transfer arm that transfers the wafer W to these units. The coating unit U1 applies a processing liquid to the surface of the wafer W. The heat processing unit U2 incorporates, for example, a hot plate and a cooling plate, and heats the wafer W with the hot plate and cools the heated wafer W with the cooling plate to perform heat processing.

The processing module 11 forms an underlayer film on the surface of the wafer W using a coating unit U1 and a heat treatment unit U2. The coating unit U1 of the processing module 11 applies a treatment liquid for forming the underlayer film onto the wafer W. The heat treatment unit U2 of the processing module 11 performs various heat treatments associated with the formation of the underlayer film.

The processing module 12 forms a resist film on the underlayer film using a coating unit U1 and a thermal processing unit U2. The coating unit U1 of the processing module 12 applies a processing liquid for forming a resist film onto the underlayer film. The thermal processing unit U2 of the processing module 12 performs various thermal processes associated with the formation of the resist film. A specific example of the thermal process is a heating process (PAB: Pre Applied Bake) for hardening the coating film to form a resist film.

The processing module 13 forms an upper layer film on the resist film using a coating unit U1 and a thermal processing unit U2. The coating unit U1 of the processing module 13 applies a liquid for forming the upper layer film onto the resist film. The thermal processing unit U2 of the processing module 13 performs various thermal processes associated with the formation of the upper layer film.

The processing module 14 uses a coating unit U1 and a thermal processing unit U2 to perform development processing of the resist film after exposure. The coating unit U1 of the processing module 14 applies a developer onto the surface of the exposed wafer W, and then rinses it away with a rinsing liquid to perform development processing of the resist film. The thermal processing unit U2 performs various types of thermal processing associated with the development processing. Specific examples of thermal processing include a heating process before the development processing (PEB: Post Exposure Bake), a heating process after the development processing (PB: Post Bake), etc.

A shelf unit U10 is provided on the carrier block 4 side within the processing block 5. The shelf unit U10 is divided into multiple cells arranged in the vertical direction. A transfer device A7 including a lifting arm is provided near the shelf unit U10. The transfer device A7 raises and lowers the wafer W between the cells of the shelf unit U10.

A shelf unit U11 is provided on the interface block 6 side of the processing block 5. The shelf unit U11 is divided into multiple cells arranged vertically.

The interface block 6 transfers the wafer W to and from the exposure device 3. For example, the interface block 6 has a built-in transfer device A8 that includes a transfer arm, and is connected to the exposure device 3. The transfer device A8 transfers the wafer W placed on the shelf unit U11 to the exposure device 3, receives the wafer W from the exposure device 3, and returns it to the shelf unit U11.

(Heat treatment unit)
Next, an example of the thermal processing unit U2 will be described in detail with reference to Fig. 3. As shown in Fig. 3, the thermal processing unit U2 includes a housing 20, a temperature adjustment mechanism 30, and a heating mechanism 40.

The housing 20 forms a processing space (processing chamber 21) in which heat treatment is performed. The housing 20 houses a temperature adjustment mechanism 30 and a heating mechanism 40. An entrance 22 for loading the wafer W is opened in the side wall of the housing 20. The housing 20 has a floor plate 23 that divides the inside of the processing chamber 21 into an upper region and a lower region. The upper region is the region in which the wafer W moves, and the lower region is the rest of the region.

The temperature adjustment mechanism 30 is a mechanism that adjusts (cools) the temperature of the wafer W to a predetermined temperature in the processing chamber 21. As one of the heat treatments in the heat treatment unit U2, the temperature of the wafer W is adjusted in the temperature adjustment mechanism 30. The temperature adjustment mechanism 30 also has the function of transferring the wafer W to and from the external transfer device A3. The temperature adjustment mechanism 30 has a temperature adjustment plate 31, a connecting bracket 32, and an actuator 33.

The temperature adjustment plate 31 is a plate that adjusts the temperature of the wafer W placed on it. Specifically, the temperature adjustment plate 31 is a cool plate on which the wafer W heated by the heating mechanism 40 is placed and which cools the wafer W to a predetermined temperature. For example, the temperature adjustment plate 31 may be formed in a substantially circular plate shape. The temperature adjustment plate 31 may be made of a metal such as aluminum, silver, or copper that has high thermal conductivity. The temperature adjustment plate 31 may be made of the same material from the viewpoint of preventing deformation due to heat. A cooling flow path (not shown) for circulating cooling water or cooling gas is formed inside the temperature adjustment plate 31.

The connecting bracket 32 is connected to the temperature adjustment plate 31. The connecting bracket 32 is configured to be movable within the housing 20. Specifically, the connecting bracket 32 is movable along a guide rail (not shown) that extends between the loading entrance 22 of the housing 20 and the vicinity of the heating mechanism 40. As the connecting bracket 32 moves along the guide rail, the temperature adjustment plate 31 moves between a loading position that is near the loading entrance 22 and a heating position that is a position where it partially overlaps with the heating mechanism 40.

The actuator 33 operates based on instructions from the control device 100, and moves the connecting bracket 32 back and forth in the horizontal direction. The actuator 33 is a device that converts energy such as hydraulic pressure, pneumatic pressure, or magnetic force into mechanical motion. For example, the actuator 33 may be an air cylinder that drives a piston (piston rod) by air pressure, and may convert the energy of air pressure into a translational (straight) motion. The mechanical motion generated by the actuator 33 is transmitted to the connecting bracket 32, so that the temperature adjustment plate 31 moves between the loading position and the heating position within the housing 20. In other words, the actuator 33 operates on the temperature adjustment plate 31, and the function (role) of the actuator 33 is to move the temperature adjustment plate 31 between the loading position and the heating position. The actuator 33 may be configured so that when it receives an operation instruction from the control device 100 and starts its operation, the mechanical motion continues from either the loading position or the heating position to the other.

The actuator 33 includes a motion detection unit 34. The motion detection unit 34 acquires information related to the motion of the actuator 33. The motion detection unit 34 acquires, for example, information related to the start and end of the motion of the actuator as information related to the motion of the actuator 33. As an example, the motion detection unit 34 may be composed of a plurality of sensors that detect the position of a piston included in the actuator 33. The motion detection unit 34 outputs the acquired information related to the motion of the actuator 33 to the control device 100. The information related to the motion of the actuator will be described in detail later.

The heating mechanism 40 is a mechanism that heat-treats the wafer W in the processing chamber 21. As one of the heat treatments in the heat treatment unit U2, the heating mechanism 40 heats the wafer W. The heating mechanism 40 has a support table 41, a hot plate 42, a heater 43, a chamber 44, an actuator 45, multiple (e.g., three) support pins 46, and an actuator 47.

The support table 41 is disk-shaped with a recess in the center. The support table 41 supports the hot plate 42. The hot plate 42 is formed, for example, in a substantially disk shape and is accommodated in the recess of the support table 41. The hot plate 42 has a mounting surface 42a. The wafer W to be processed is placed on the mounting surface 42a, and the hot plate 42 supports the wafer W. The hot plate 42 heats the placed wafer W. A heater 43 for heating the hot plate 42 is provided on the lower surface of the hot plate 42 opposite the mounting surface 42a. For example, the heater 43 is composed of a resistance heating element. When a current flows through the heater 43, the heater 43 generates heat. Then, the heat from the heater 43 is transferred, and the temperature of the hot plate 42 increases. For example, a current of a value according to an instruction from the control device 100 may flow through the heater 43.

The chamber 44 is configured to surround the wafer W mounting surface 42a of the heat plate 42. The chamber 44 has a top plate portion 44a and a side wall portion 44b. The top plate portion 44a is configured to be a disk shape having a diameter similar to that of the support table 41. The top plate portion 44a is arranged to face the mounting surface 42a of the heat plate 42 in the vertical direction. The side wall portion 44b is configured to extend downward from the outer edge of the top plate portion 44a. The chamber 44 is configured to be movable between an elevated position where the space for performing the heat treatment of the wafer W is opened and a lowered position where the chamber 44 is lowered to close the space for performing the heat treatment of the wafer W (a processing space is formed).

The actuator 45 operates based on instructions from the control device 100 to raise and lower the chamber 44. The actuator 45 is a device that converts energy such as hydraulic pressure, pneumatic pressure, or magnetic force into mechanical motion. For example, the actuator 45 may be an air cylinder that drives a piston (piston rod) by air pressure, and may convert the energy of air pressure into a translational (straight) motion. The mechanical motion generated by the actuator 45 is transmitted to the chamber 44, so that the chamber 44 moves between the raised position and the lowered position. In other words, the actuator 45 operates on the chamber 44, and the function (role) of the actuator 45 is to move the chamber 44 between the raised position and the lowered position. The actuator 45 may be configured so that when it receives an operation instruction from the control device 100 and starts operating, the mechanical motion continues from either the raised position or the lowered position to the other.

The actuator 45 includes a motion detection unit 48. The motion detection unit 48 acquires information related to the motion of the actuator 45. The motion detection unit 48 acquires, for example, information related to the start and end of the motion of the actuator 45. As an example, the motion detection unit 48 may be composed of a plurality of sensors that detect the position of a piston included in the actuator 45. The motion detection unit 48 outputs the acquired information related to the motion of the actuator 45 to the control device 100.

The support pins 46 extend vertically through the support base 41 and the heat plate 42, and support the wafer W from below. The multiple support pins 46 may be arranged, for example, at equal intervals in the circumferential direction. The support pins 46 are configured to be able to move up and down between a placement position where the wafer W is placed on the heat plate 42 and a transfer position where the wafer W is transferred between the temperature adjustment plate 31.

The actuator 47 raises and lowers the support pin 46 based on an instruction from the control device 100. The actuator 47 is a device that converts energy such as hydraulic pressure, pneumatic pressure, or magnetic force into mechanical motion. For example, the actuator 47 may be an air cylinder that drives a piston (piston rod) by air pressure, and may convert the energy of air pressure into a translational (straight) motion. The mechanical motion generated by the actuator 47 is transmitted to the support pin 46, so that the support pin 46 moves between the transfer position and the placement position. In other words, the actuator 47 operates on the support pin 46, and the function (role) of the actuator 47 is to move the support pin 46 between the transfer position and the placement position. The actuator 47 may be configured so that when it receives an operation instruction from the control device 100 and starts its operation, the mechanical motion continues from either the transfer position or the placement position to the other.

The actuator 47 includes a motion detection unit 49. The motion detection unit 49 acquires information related to the motion of the actuator 47. The motion detection unit 49 acquires, for example, information related to the start and end of the motion of the actuator 47. As an example, the motion detection unit 49 may be composed of a plurality of sensors that detect the position of a piston included in the actuator 47. The motion detection unit 49 outputs the acquired information related to the motion of the actuator 47 to the control device 100.

In this way, in the coating/developing apparatus 2, a single heat treatment unit U2 that performs heat treatment is provided with multiple actuators (actuators 33, 45, 47) that have different operating targets (functions) and installation positions. In addition, multiple different heat treatment units U2 that perform a common process are each provided with actuators 33, 45, 47.

(Coating and developing process procedure)
The control device 100 controls the coating/developing device 2 to perform the coating/developing process, for example, in the following procedure: First, the control device 100 controls the transfer device A1 to transfer the wafer W in the carrier C to the shelf unit U10, and then controls the transfer device A7 to place the wafer W in a cell for the processing module 11.

The control device 100 then controls the transfer device A3 to transfer the wafer W from the shelf unit U10 to the coating unit U1 and heat treatment unit U2 in the processing module 11. The control device 100 also controls the coating unit U1 and heat treatment unit U2 to form an underlayer film on the surface of the wafer W. The control device 100 then controls the transfer device A3 to return the wafer W with the underlayer film formed thereon to the shelf unit U10, and controls the transfer device A7 to place the wafer W in the cell for the processing module 12. In these controls, the control device 100 drives the actuators 33, 45, and 47 included in the heat treatment unit U2 in the processing module 11 to move the corresponding members to be operated.

The controller 100 then controls the transfer device A3 to transfer the wafer W from the shelf unit U10 to the coating unit U1 and heat treatment unit U2 in the processing module 12. The controller 100 also controls the coating unit U1 and heat treatment unit U2 to form a resist film on the surface of the wafer W. The controller 100 then controls the transfer device A3 to return the wafer W to the shelf unit U10, and controls the transfer device A7 to place the wafer W in the cell for the processing module 13. In these controls, the controller 100 drives the actuators 33, 45, and 47 included in the heat treatment unit U2 in the processing module 12 to move the corresponding members to be operated.

The controller 100 then controls the transfer device A3 to transfer the wafer W from the shelf unit U10 to each unit in the processing module 13. The controller 100 also controls the coating unit U1 and the heat treatment unit U2 to form an upper layer film on the resist film of the wafer W. The controller 100 then controls the transfer device A3 to transfer the wafer W to the shelf unit U11. In these controls, the controller 100 drives the actuators 33, 45, and 47 included in the heat treatment unit U2 in the processing module 13 to move the corresponding members to be operated.

Next, the control device 100 controls the transfer device A8 to send the wafer W in the shelf unit U11 to the exposure device 3. The control device 100 then controls the transfer device A8 to receive the wafer W that has been subjected to the exposure process from the exposure device 3 and place it in a cell for the processing module 14 in the shelf unit U11.

The controller 100 then controls the transfer device A3 to transfer the wafer W from the shelf unit U11 to each unit in the processing module 14, and controls the coating unit U1 and heat treatment unit U2 to perform a development process on the resist film on the wafer W. The controller 100 then controls the transfer device A3 to return the wafer W to the shelf unit U10, and controls the transfer device A7 and transfer device A1 to return the wafer W into the carrier C. In these controls, the controller 100 drives the actuators 33, 45, and 47 included in the heat treatment unit U2 in the processing module 14 to move the corresponding members to be operated. This completes the coating and development process.

The specific configuration of the substrate processing apparatus is not limited to the configuration of the coating/developing apparatus 2 exemplified above. The substrate processing apparatus may be any type as long as it includes a processing unit that performs substrate processing, a plurality of actuators that move the members to be operated on, and a control device 100 that can control these.

[Status Determination Device]
Next, the specific configuration and functions of the control device 100 will be illustrated with reference to Figures 4 to 7. The control device 100 includes a state determination unit 200 (state determination device) as a functional configuration (hereinafter referred to as a "functional module") (see also Figure 3). The state determination unit 200 is configured to determine the state of a target actuator (any one or more actuators among the multiple actuators) included in the multiple actuators mounted on the coating and developing apparatus 2. Specifically, the state determination unit 200 is configured to execute the following operations: acquiring feature amounts related to the operation of each of the multiple actuators, calculating a probability distribution model based on the feature amounts for each of the multiple actuators, and determining the state of the target actuator based on the probability distribution model for each of the multiple actuators.

As shown in FIG. 4, the state determination unit 200 includes, as functional modules, an acquisition unit 201, a probability distribution calculation unit 202, a reference distribution calculation unit 203, an index value calculation unit 204, a determination unit 205, and an output unit 206. In the following description, the actuators for which the feature quantities are acquired and the probability distribution model is calculated are, for example, the actuators 33, 45, and 47 included in a single heat treatment unit U2. The target actuator for which the state is determined is, for example, at least one of the actuators 33, 45, and 47 in the heat treatment unit U2. The actuators for which the feature quantities are acquired and the probability distribution model is calculated are not limited to these actuators. When multiple heat treatment units U2 are provided in the coating/developing apparatus 2, the actuators included in the multiple heat treatment units U2 can be the actuators for which the feature quantities are acquired and the probability distribution model is calculated. The actuators included in a unit other than the heat treatment unit U2 can also be the actuators for which the feature quantities are acquired and the target actuators.

As mentioned above, the type of actuator is not particularly limited, but in the following, an example will be described in which each of the multiple actuators is a pneumatically driven actuator (air cylinder), and a common feature value is acquired from each of the multiple actuators to calculate a probability distribution model. In addition, in the following, an example will be described in which the operating time T of the air cylinder is used as the feature value.

The acquisition unit 201 is configured to acquire, for each of the multiple actuators, a feature amount related to the operation of each of the actuators. For example, the acquisition unit 201 acquires a feature amount related to the operation (e.g., operation time T) for each of the actuators 33, 45, and 47 via the operation detection units 34, 48, and 49. Specifically, the acquisition unit 201 may acquire information related to the start and end of the operation of the actuator 33 from the operation detection unit 34 included in the actuator 33.

As an example, the acquisition unit 201 can acquire information on the position of the piston of the actuator 33 from the operation detection unit 34 (for example, position information of the piston detected by a sensor). The acquisition unit 201 can use these pieces of information as information for specifying the start of an operation by the piston and information for specifying the end of an operation by the piston. The acquisition unit 201 may also associate the information for specifying the start of an operation by the piston and the information for specifying the end of an operation by the piston with information for specifying the time at which each operation occurred. In such a configuration, the acquisition unit 201 can calculate the operation time T of the actuator 33 according to the time at which the operation by the piston started and the time at which the operation by the piston ended. The acquisition unit 201 may also be configured to acquire (calculate) the operation time T of the actuators 45 and 47 in the same manner as the actuator 33. The acquisition unit 201 may acquire the operation time T of the actuator each time a series of operations of the actuator ends, that is, when the acquisition unit 201 acquires information for specifying the end of the operation.

Figure 5 shows an example of the acquisition of the operation time T for one actuator (one example of a detection result). In Figure 5, the horizontal axis indicates time, and the vertical axis indicates the operation time T. As shown in Figure 5, the acquisition unit 201 continuously acquires the operation time T for each of the multiple actuators. In the example shown in Figure 5, multiple (15 here) acquired values for the operation time T are divided into each predetermined period that is arbitrarily set. This predetermined period is a basic unit for calculating a probability distribution model, which will be described later. The predetermined period for calculating the probability distribution model may be set in units of several hours, half a day, or one day. In Figure 5, the period "0" indicates the immediately preceding (most recent) predetermined period among the periods that have already passed, and the period "(-k)" indicates a period prior to the immediately preceding predetermined period. Note that "k" is any positive integer. For example, if the width of the predetermined period is set to 1 hour and "-k" is set to "-3", the period "-k" indicates a period that starts 3 hours before the start time of the immediately preceding predetermined period and ends 2 hours before it.

The acquisition unit 201 may calculate (acquire) the maximum value Tmax and minimum value Tmin of the acquired values of the operation time T included in the period for each predetermined period. The acquisition unit 201 may calculate (acquire) the average value Tmean of the acquired values of the operation time T included in the period for each predetermined period. Note that hereinafter, the information on the operation time T will be described as including information on the maximum value Tmax, minimum value Tmin, and average value Tmean. The acquisition unit 201 outputs information indicating the acquired feature amount related to the operation (information on the operation time T) to the probability distribution calculation unit 202 for each predetermined period, for example.

The probability distribution calculation unit 202 calculates a probability distribution model for each of the multiple actuators based on the feature amount acquired by the acquisition unit 201. That is, the probability distribution calculation unit 202 calculates a probability distribution model for each of the multiple actuators based on the feature amount for each actuator. The probability distribution calculation unit 202 calculates a probability distribution model for each actuator for each predetermined period. In the example shown in FIG. 5, after acquiring information on the operation time T in period "0" from the acquisition unit 201, the probability distribution calculation unit 202 calculates a probability distribution model based on the information on the operation time T in this period "0".

Here, the probability distribution model will be described. A probability distribution model based on a feature quantity is a model of the operation of an actuator, assuming that an input variable related to a continuously obtained feature quantity follows a certain probability distribution. For example, the probability distribution model is calculated based on an operation time T. The probability distribution calculation unit 202 may combine a plurality of operation times T obtained within a predetermined period into one input variable X. For example, the probability distribution calculation unit 202 may combine a plurality of operation times T into an input variable X as shown in the following formulas (1) and (2).

As shown in formula (2), "Twidth" in formula (1) is the difference between the maximum value Tmax and the minimum value Tmin (hereinafter referred to as "operation time width Twidth"). Tmean(0) and Twidth(0) respectively indicate the average value Tmean of the operation time T and the operation time width Twidth in a specified period (period "0") immediately preceding the period for which the probability distribution model is to be calculated. Furthermore, Tmean(-k) and Twidth(-k) respectively indicate the average value Tmean of the operation time T and the operation time width Twidth in period "-k".

Note that the input variable X is not limited to the above variables. In other words, the probability distribution calculation unit 202 can arbitrarily select a variable according to the actuator operation time T as the input variable X. For example, the probability distribution calculation unit 202 may use the minimum value Tmin or maximum value Tmax of the operation time T in place of the average value Tmean in formula (1).

A common symptom of an abnormality common to pneumatic actuators (air cylinders) is a fluctuation (change) in the operating time due to a decrease in thrust or an increase in sliding resistance. Once an actuator transitions to an abnormal state, it is difficult to imagine that the actuator will naturally return to a normal state. In addition, it is believed that the fluctuation (change) in the operating time T will accelerate when the actuator transitions to an abnormal state. This tendency indicates that the instantaneous change amount dT of the operating time T depends on the operating time T at that time. Here, when the division value dT/T of the two is defined as the new change amount dY, the relationship of the following formula (3) is established. As a result, if it is assumed that the division value dT/T (change amount dY) follows a normal distribution, the operating time T itself follows a log-normal distribution. The fact that a random variable (here, the operating time T) follows a log-normal distribution indicates that the logarithm of the random variable (here, the operating time T) follows a normal distribution. Even if the distribution of the operation time T does not strictly conform to a log-normal distribution, the above assumption is empirically applicable as long as the distribution of the operation time T is similar to the log-normal distribution.

ｉ：アクチュエータごとに割り振られた装置番号（正の整数） It is known that random variables that follow a log-normal distribution also have the property that the remainder between random variables also follows a log-normal distribution. Therefore, the input variable X shown in formula (1) also follows a log-normal distribution. For this reason, the probability distribution calculation unit 202 calculates a probability distribution model for each of the multiple actuators by calculating parameters that define the normal distribution that the logarithm logX of the input variable X follows. Specifically, the probability distribution calculation unit 202 calculates two types of parameters as the above parameters: an average u that is the average of the normal distribution (log-normal distribution) and a variance σ ² that corresponds to the standard deviation σ. In other words, the probability distribution calculation unit 202 calculates a probability distribution model for one actuator by obtaining the average u and the variance σ ^2. The probability distribution calculation unit 202 is not particularly limited as a calculation method (estimation method) of the average u and the variance σ ² , but may use a known method such as a maximum likelihood method or Bayesian estimation, for example. The probability distribution calculation unit 202 estimates the mean u and variance ^σ2 of the normal distribution to which the logarithm logX follows by the above-mentioned estimation method, thereby calculating the probability distribution model p _i associated with each of the multiple actuators shown in formula (4). The probability distribution calculation unit 202 outputs information related to the calculated probability distribution model p _i to the reference distribution calculation unit 203.

i: device number assigned to each actuator (positive integer)

The reference distribution calculation unit 203 is configured to calculate a reference probability distribution model p _{s serving as a reference for determining the state of the target actuator from the probability distribution model p i of each of the multiple actuators acquired from the probability distribution calculation unit 202. In other words, the reference distribution calculation unit 203 calculates a single reference probability distribution model p s from the} probability distribution model p _i of each of the multiple actuators. For example, the reference distribution calculation unit 203 may calculate the reference probability distribution model p _s of a log-normal distribution by finding the average value or median value of parameters (mean u _i and variance σ _i ² ) included in the probability distribution model p _i .

The reason why the calculation of the reference probability distribution model p _s is based on the parameters of the probability distribution model p _i of multiple actuators is as follows. That is, if multiple actuators are included in the coating/developing apparatus 2, it is expected that many of the actuators are normal under normal operating conditions, and the number of actuators in an abnormal state is small or one. Furthermore, by including the probability distribution models of multiple actuators in the calculation, the standard for state judgment becomes more reliable.

ｎ：複数のアクチュエータの数（正の整数）

For example, when the average value is adopted, the parameters of the reference probability distribution model p _s are expressed as the following formulas (5) and (6). Note that, when the operation time T is substantially constant and stable within a predetermined period, the input variable X is substantially equal to 1, and the logarithm logX is substantially equal to 0, so that the reference distribution calculation unit 203 may calculate the average value u _s of the averages u _i as 0. In this case, the calculation of the reference probability distribution model p _s is simplified. In this way, the reference probability distribution model p _s may be obtained by calculating the average of the probability distribution models p _i of each of the multiple actuators. The reference distribution calculation unit 203 outputs information on the calculated reference probability distribution model p _s to the index value calculation unit 204.

n: the number of actuators (positive integer)

The index value calculation unit 204 calculates an index value indicating the state of the target actuator by comparing the reference probability distribution model p _s with a probability distribution model p _i of the target actuator (hereinafter referred to as a “target probability distribution model p _o ”). Specifically, the index value calculation unit 204 may calculate a distance D (dissimilarity) between the probability distributions of the reference probability distribution model p _s and the target probability distribution model p _o . As the distance D between such probability distributions, the well-known Kullback-Leibler distance, Pearson distance, or L ² -distance may be used, and the state of the target actuator may be quantified as an index value. The index ^value calculation unit 204 outputs information related to the calculated index value to the determination unit 205.

The determination unit 205 determines the state of the target actuator based on the index value calculated by the index value calculation unit 204. For example, the determination unit 205 determines whether the target actuator is in an abnormal state or a normal state based on the index value. FIG. 6(a) illustrates the transition of the operation time T (maximum, minimum, and average values of the operation time T) for each period over time, and FIG. 6(b) illustrates the transition of the index value for each period according to the operation time T. Note that FIGS. 6(a) and 6(b) show the results of obtaining the maximum value Tmax, minimum value Tmin, and average value Tmean of the operation time T for each predetermined period, and the calculation results of the index value.

As shown in FIG. 6(a), as the target actuator approaches an abnormal state (becomes an abnormal state) over time, the operation time width Twidth widens or the average operation time Tmean rises. For this reason, when the index value is shown as the distance D between the probability distributions, as shown in FIG. 6(b), the distance D (index value) tends to increase as the target actuator approaches an abnormal state (becomes an abnormal state). The target actuator is included in the multiple actuators that are the calculation targets of the probability distribution model p _i , and the target probability distribution model p _o of the target actuator is also used to calculate the reference probability distribution model p _s . Even in this case, if the value of "n" indicating the number of the above-mentioned multiple actuators is large, the fluctuation of the target probability distribution model p o included in the reference probability distribution model p _s calculated using the average value of the parameters is negligible. When the reference probability distribution model _{p s is} calculated by selecting the median value of the parameters, the influence of the fluctuation of the target probability distribution model p _o is small even if the value of "n" is not sufficiently large.

The determination unit 205 may use the index value to determine whether the target actuator is abnormal. For example, the determination unit 205 may determine that the target actuator is in an abnormal state when the index value exceeds a predetermined value (threshold Th1). The threshold Th1 may be set in advance by estimating or evaluating the index value when the actuator is in an abnormal state. The determination unit 205 may also determine that the target actuator is in a normal state when the index value is equal to or less than the threshold Th1. In this way, the index value calculation unit 204 and the determination unit 205 constitute a comparison and determination unit that determines the state of the target actuator by comparing the reference probability distribution model p _s with the target probability distribution model p _o , which is a probability distribution model of the target actuator. The determination unit 205 outputs information indicating the state (normal state or abnormal state) of the target actuator to the output unit 206.

The output unit 206 outputs information indicating the state of the target actuator determined by the determination unit 205 to the outside of the state determination unit 200. For example, when the state of the target actuator is normal, the output unit 206 may output a signal indicating that the actuator is normal (hereinafter referred to as a "normal signal") to another element (functional module) in the control device 100 or to the outside of the control device 100. Also, when the state of the target actuator is abnormal, the output unit 206 may output a signal indicating that the actuator is abnormal (hereinafter referred to as an "abnormal signal") to another element in the control device 100 or to the outside of the control device 100. When the abnormal signal is output to another element in the control device 100, the control device 100 may perform control different from the normal control performed before the actuator is determined to be abnormal. For example, when the abnormal signal is output, the control device 100 may perform control to display (not shown) that the target actuator is abnormal (to notify the operator). Alternatively, the control device 100 may temporarily stop processing by the thermal processing unit U2 that contains the target actuator that has become abnormal, or processing in the coating and developing device 2.

The control device 100 including the state determination unit 200 is composed of one or more control computers. For example, the control device 100 has a circuit 120 shown in FIG. 7. The circuit 120 has one or more processors 121, a memory 122, a storage 123, a timer 125, and an input/output port 124. The storage 123 has a storage medium readable by a computer, such as a hard disk. The storage medium stores a program for causing the control device 100 (state determination unit 200) to execute the state determination procedure described below. The storage medium may be a removable medium such as a non-volatile semiconductor memory, a magnetic disk, or an optical disk. The memory 122 temporarily stores the program loaded from the storage medium of the storage 123 and the calculation results by the processor 121. The processor 121 configures each of the above-mentioned functional modules by executing the above-mentioned programs in cooperation with the memory 122. The input/output port 124 inputs and outputs electrical signals between the actuators 33, 45, and 47 according to instructions from the processor 121. The timer 125 measures the elapsed time, for example, by counting a reference pulse at a fixed interval.

The hardware configuration of the control device 100 is not necessarily limited to configuring each functional module by a program. For example, each functional module of the control device 100 may be configured by a dedicated logic circuit or an ASIC (Application Specific Integrated Circuit) that integrates these. The control device 100 may include a control computer that controls the substrate processing, and a status determination computer (status determination device) that is separate from the control computer.

[Status Judgment Procedure]
Next, a procedure for determining the state of the target actuator executed by the control device 100 (state determination unit 200) will be described with reference to Fig. 8. Fig. 8 shows a flowchart illustrating the state determination procedure as a state determination method.

In this state determination procedure, the control device 100 first executes steps S01 and S02. In step S01, for example, the acquisition unit 201 acquires the operation time T as a feature for each of the multiple actuators. In step S02, for example, the acquisition unit 201 determines whether a predetermined period (a period that is a unit of acquisition of the feature) has elapsed. Until the predetermined period has elapsed, the control device 100 repeats the processing of steps S01 and S02. As a result, for the multiple operations of each actuator executed within the predetermined period (for example, period "0": see FIG. 5), each feature (information on the operation time T) is acquired. Here, a case is illustrated in which the acquisition units (predetermined periods) of the feature for each of the multiple actuators are the same, but the acquisition units (predetermined periods) of the feature may be different from each other. Then, when it is determined in step S02 that the predetermined period has elapsed, the acquisition unit 201 outputs the acquired information on the operation time T to the probability distribution calculation unit 202.

Next, the control device 100 executes step S03. In step S03, for example, the probability distribution calculation unit 202 calculates a probability distribution model p _i for each of the multiple actuators based on multiple operation times T included in the period "0". The probability distribution calculation unit 202 may calculate the probability distribution model p _i based on the input variable X shown in the above formula (1). In this case, "k" in formula (1) may be set to the same value for each of the multiple actuators. The probability distribution calculation unit 202 outputs information indicating the probability distribution model p _i to the reference distribution calculation unit 203.

Next, the control device 100 executes step S04. In step S04, for example, the reference distribution calculation unit 203 calculates a reference probability distribution model p _s based on the probability distribution model p _i of each of the multiple actuators. The reference distribution calculation unit 203 may obtain an average value of parameters (mean u _i and variance σ _i ² ) included in the probability distribution model p _i, and set the average value as a parameter of the reference probability distribution model p _s . The reference distribution calculation unit 203 outputs information indicating the probability distribution model p _i to the index value calculation unit 204.

Next, the control device 100 executes step S05. In step S05, for example, the index value calculation unit 204 compares the target probability distribution model p _o of the target actuator with the reference probability distribution model p _s calculated by the reference distribution calculation unit 203. The index value calculation unit 204 may compare the target probability distribution model p _o with the reference probability distribution model p _s to calculate the distance D between these probability distribution models as an index value. The index value calculation unit 204 outputs information related to the index value to the determination unit 205.

Next, the control device 100 executes step S06. In step S06, for example, the determination unit 205 determines whether the index value calculated by the index value calculation unit 204 is greater than the threshold value Th1. The determination unit 205 may output information indicating the determination result of whether the index value is greater than the threshold value Th1 to the output unit 206.

If it is determined in step S06 that the index value is greater than the threshold value Th1 (step S06: YES), the control device 100 executes step S07. In step S07, for example, the output unit 206 outputs an abnormality signal indicating that the target actuator is in an abnormal state to the outside of the state determination unit 200. Note that, if it is determined in step S06 that the index value is equal to or less than the threshold value Th1 (step S06: NO), the output unit 206 may also output a normal signal indicating that the target actuator is in a normal state. This completes a series of state determination processes for one predetermined period. Note that, if there are multiple target actuators, the control device 100 executes the processes of steps S05 to S07 for each target actuator. Thereafter, the control device 100 repeats the processes of steps S01 to S07 for each predetermined period.

[Action]
In the state determination procedure and state determination unit 200 (state determination device) according to the above embodiment, the state of the target actuator is determined based on the probability distribution model p _i of each of the multiple actuators. When the state of the target actuator deviates from the normal state, a fluctuation occurs in the operation time T, which is a feature amount related to the operation. For this reason, the target probability distribution model p _o for the target actuator among the probability distribution models p _i of each of the multiple actuators changes. It is possible to determine the state of the target actuator by capturing the change in the target probability distribution model p _o of the target actuator based on the probability distribution models p _i of the multiple actuators. For example, the state of the target actuator can be determined by capturing the deviation between the target probability distribution model p _o and the probability distribution model for actuators other than the target actuator. That is, a state determination method and a state determination device suitable for determining the state of an actuator are provided.

In conventional state detection including failure prediction, a failure of an equipment is predicted based on historical data on the operation of the equipment and a failure pattern updated based on the historical data of the failed equipment. That is, in this conventional state detection method, a failure pattern when an actual failure occurs is stored in advance, and a failure is predicted (the state of the equipment is detected) by comparing the historical data of the equipment during normal operation with the failure pattern. In this method, various failure patterns, including patterns with a low occurrence rate, must be stored (acquired) in advance, so preparation is cumbersome. Furthermore, there is a risk that a failure pattern that is not stored and is not seen for the first time cannot be detected. Therefore, there is room for improvement in the method of evaluating the state of the actuator. In contrast, in the state determination method and state determination device according to the present embodiment, the state of the target actuator is determined by utilizing a probability distribution model based on the fact that the operation time T varies due to an abnormality in the actuator and the operation time T. Therefore, there is no need to collect features related to the operation when the actuator deviates from the normal state in advance, and the state of the actuator can be detected even if the state deviates from the normal state is seen for the first time.

In the above embodiment, the determination includes determining a reference probability distribution model p _s that serves as a reference for determination from the probability distribution model p _i of each of the multiple actuators, and determining the state of the target actuator by comparing the reference probability distribution model p _s with a target probability distribution model p _o that is a probability distribution model of the target actuator. In this case, by creating a reference probability distribution model p _s that serves as a reference for state determination from the probability distribution model p _i of each of the multiple actuators and using this for state determination, it becomes possible to easily determine the state of the target actuator.

In the above embodiment, the reference probability distribution model p _s is obtained by calculating the average of the probability distribution models p _i of each of the multiple actuators. In this case, by using the average of the probability distribution models p _i of each of the multiple actuators, it is possible to obtain a reference that is more suitable for state determination.

In the above embodiment, the determination is to determine whether the target actuator is abnormal. The above state determination method further includes outputting a signal indicating that the target actuator is abnormal when it is determined that the target actuator is abnormal. In this case, when it is determined that the target actuator is in an abnormal state, it is possible to execute a process that is different from the process when the target actuator is in a normal state.

The actuators 33, 45, and 47 described in the above embodiment are provided in the same heat treatment unit U2, but the actuators are different from each other. In this way, the actuators used to determine the operation of the target actuator may have different actuators. The actuators may be included in a substrate processing apparatus and provided in a single processing unit that performs a predetermined process. If the actuators have different actuators and different mounting positions, their operation times T are different from each other even in a normal state. Therefore, even if the target actuator approaches an abnormal state, there is a risk that the fluctuation in the operation time T caused by the approach to the abnormal state cannot be captured. In the above configuration, the state is determined based on a probability distribution model regarding the actuator operation time T, so that even if the actuators have different actuators and different mounting positions, the state of the target actuator can be determined based on the deviation between the probability distribution models.

Each of the multiple actuators is an air cylinder driven by air pressure. As the air cylinder moves away from a normal state (approaching an abnormal state or entering an abnormal state), the operation time T for the operation of the air cylinder moves away from the value in the normal state. For this reason, the probability distribution model based on the operation time T of the target air cylinder as the target actuator deviates from the probability distribution models of the other air cylinders, making it possible to determine the state of the target air cylinder.

[Modification]
Although the embodiments have been described above, the present disclosure is not necessarily limited to the above-described embodiments, and various modifications are possible without departing from the gist of the present disclosure.

Instead of determining whether the target actuator is in an abnormal state based on the index value, the control device 100 (determination unit 205) may determine whether the target actuator is approaching an abnormal state. FIG. 9 shows a graph for explaining an example of a method for determining whether the target actuator is approaching an abnormal state. For example, as shown in FIG. 9, the index value (distance D) gradually increases toward threshold value Th1 in the process in which the target actuator transitions from a normal state to an abnormal state. For this reason, the determination unit 205 may set a threshold value Th2 smaller than threshold value Th1 and determine whether the index value is greater than threshold value Th2 to determine whether the target actuator is approaching an abnormal state.

When the determination unit 205 determines that the target actuator is approaching an abnormal state, the output unit 206 may output a signal indicating that the target actuator is approaching an abnormal state (hereinafter referred to as an "abnormal proximity signal") to other elements of the control device 100. When the abnormal proximity signal is output to other elements in the control device 100, the control device 100 may perform control different from the normal control before it is determined that the target actuator is approaching an abnormal state. For example, when the abnormal proximity signal is output, the control device 100 may perform control to display on a display (not shown) that the target actuator is approaching an abnormal state (to notify the operator). Alternatively, the control device 100 may temporarily stop processing by the thermal processing unit U2 that includes the target actuator approaching an abnormal state, or processing in the coating and developing device 2. When the index value exceeds the threshold value Th2, the determination unit 205 may calculate a predicted line L1 (for example, an approximation line based on the least squares method) based on multiple index values in the rising period up to the threshold value Th2 (graph C1), and predict the period (grace period La) in which the predicted line L1 reaches the threshold value Th1. The output unit 206 may output information regarding the grace period La until the predicted line L1 reaches the threshold value Th1.

In the above method, determining the state of the target actuator means determining whether the target actuator is approaching an abnormal state. The above state determination method includes outputting a signal indicating that the target actuator is approaching an abnormal state when it is determined that the target actuator is approaching an abnormal state. In this case, when it is determined that the target actuator is approaching an abnormal state, it becomes possible to execute processing that is different from that executed when the target actuator is in a normal state.

The control device 100 (state determination unit 200) may determine the state of the target actuator based on the skewness S (absolute value) or kurtosis P, which represent the characteristics of the histogram of the target probability distribution model p _o as a normal distribution ratio. For example, the index value calculation unit 204 may calculate, as the index value, any of the corrected distances D1 to D3 obtained by performing the calculations shown in the following equations (7) to (9) on the distance D.

In this method, judging the state of the target actuator includes judging based on the skewness or kurtosis of the target probability distribution model p _o as well. In this case, since the skewness or kurtosis also changes due to fluctuations in the operation time T of the target actuator, by correcting the distance D between the probability distributions with the skewness or kurtosis, the deviation between the target probability distribution model p _o and the reference probability distribution model p _s is more emphasized. As a result, it becomes possible to more reliably judge the state of the target actuator.

The multiple actuators for which feature quantities are to be acquired and for which a probability distribution model is to be calculated may be actuators included in different heat processing units U2, but which have a common operating target. For example, the probability distribution models of the multiple actuators 33 included in the multiple heat processing units U2 may be calculated, and one of the multiple actuators 33 may be set as the target actuator.

In this example, the actuators have a common target for operation. The actuators are included in the substrate processing apparatus and are provided in different thermal processing units U2 that perform a common process. In this case, when the operation of the target actuator deviates from the normal state, a deviation occurs with respect to other probability distribution models, making it possible to determine the state of the target actuator.

The state determination unit 200 (state determination device) may be housed in a housing separate from the control device 100, and may be configured as a computer (circuit) separate from the control device 100. The state determination unit 200 may calculate the probability distribution model p _i for a plurality of actuators included in a plurality of coating/developing devices 2.

The actuator's operating target (function) is not limited to the above-mentioned examples. The actuator may be a device provided in the coating unit U1 or other processing unit that performs substrate processing, and that moves a member back and forth in the horizontal or vertical direction.

The feature quantity related to the operation of the actuator used in the calculation of the probability distribution model p _i is not limited to the operation time T. For example, the (average) operation speed of the actuator, or the acceleration and torque as a resultant force against the inertia and sliding resistance of the object to be operated can be used as the feature quantity related to the operation of the actuator. The acquisition unit 201 may acquire the feature quantity for each operation by performing statistical processing such as peak extraction or averaging on the feature quantity continuously obtained during one operation of the actuator. In the above embodiment, the acquisition unit 201 acquires the feature quantity related to the operation of the actuator by calculating the feature quantity based on the information related to the operation of the actuator, but the acquisition method is not limited to this. The acquisition unit 201 may acquire the feature quantity by obtaining the information indicating the feature quantity related to the operation of the actuator from the operation detection unit that detects the information.

The substrate to be processed in the coating/developing apparatus 2 is not limited to a semiconductor wafer, but may be, for example, a glass substrate, a mask substrate, or a flat panel display (FPD).

2...coating/developing device, 33, 45, 47...actuator, 100...control device, 200...state determination unit, 201...acquisition unit, 202...probability distribution calculation unit, 203...reference distribution calculation unit, 204...index value calculation unit, 205...determination unit, 206...output unit, U2...heat treatment unit.

Claims (13)

1. A state determination method for determining a state of a target actuator included in a plurality of actuators mounted in a substrate processing apparatus, comprising:
the plurality of actuators include the target actuator and actuators other than the target actuator,
acquiring a feature amount relating to an operation of each of the actuators;
calculating a probability distribution model for each of the actuators based on the feature amount;
determining a state of the target actuator included in the plurality of actuators based on the probability distribution model related to the feature amount of each of the plurality of actuators;
A state determination method comprising: The state determination method according to claim 1, wherein the characteristic amount is the operating speed of an actuator. The state determination method according to claim 1, wherein the characteristic quantities are acceleration and torque related to the operation of the actuator. The determining step comprises:
determining a reference probability distribution model serving as a reference for judgment from the probability distribution model related to the feature amount of each of the plurality of actuators;
The state determination method according to any one of claims 1 to 3, further comprising: determining a state of the target actuator by comparing the reference probability distribution model with a target probability distribution model, the target probability distribution model being the probability distribution model related to the feature quantity of the target actuator. The state determination method according to claim 4, wherein the reference probability distribution model is obtained by calculating an average of the probability distribution models related to the feature quantities of each of the actuators. The state determination method according to any one of claims 1 to 5, wherein the determining step includes determining the state of the target actuator based on the skewness or kurtosis of a probability distribution model related to the feature quantity of the target actuator. The determining step determines whether the target actuator is abnormal,
7. The state determination method according to claim 1, further comprising: when it is determined that the target actuator is abnormal, outputting a signal indicating that the target actuator is abnormal. The determining step determines whether the target actuator is approaching an abnormal state,
The state determination method according to any one of claims 1 to 6, further comprising: when it is determined that the target actuator is approaching an abnormal state, outputting a signal indicating that the target actuator is approaching an abnormal state. The actuators are operated on a common target.
9. The state determination method according to claim 1, wherein the plurality of actuators are included in the substrate processing apparatus and are provided in different processing units that perform a common process. The actuators have different operation targets,
9. The state determination method according to claim 1, wherein the plurality of actuators are provided in a single processing unit included in the substrate processing apparatus and performing a predetermined process. The state determination method according to any one of claims 1 to 10, wherein each of the actuators is an air cylinder driven by air pressure. A state determination device that determines a state of a target actuator included in a plurality of actuators mounted in a substrate processing apparatus, comprising:
the plurality of actuators include the target actuator and actuators other than the target actuator,
an acquisition unit that acquires a feature amount related to an operation of each of the actuators;
a probability distribution calculation unit that calculates a probability distribution model for each of the actuators based on the feature amount;
a determination unit that determines a state of the target actuator included in the plurality of actuators based on the probability distribution model related to the feature amount of each of the plurality of actuators;
A state determination device including: A computer-readable storage medium storing a program for causing an apparatus to execute the state determination method according to any one of claims 1 to 11.

Images