Diagnostic device, diagnostic system, and diagnostic method for process processing device

Abstract

The purpose of the present invention is to provide a technique capable of taking an effective countermeasure by suppressing the occurrence of false alarms with respect to components whose states are measured by a plurality of sensors or components of a group of processing devices in diagnosis of degradation of processing devices. Therefore, this diagnostic device for a process processing device uses a component of the process processing device as a maintenance target, and is provided with: a characteristic quantity change calculation unit that calculates the change in the characteristic quantity of each of sensor waveform data for sensor items of a plurality of sensors (state sensor groups) that measure the state of the component on the basis of the characteristic quantity of each sensor waveform data; calculating the deterioration degree corresponding to the change index; and a component degradation degree calculation unit that calculates component degradation degrees corresponding to the components in a one-to-one manner using the plurality of degradation degrees.

Description

Technical Field

[0001] The present invention relates to a diagnostic device, a diagnostic system and a diagnostic method for a process processing device for processing semiconductor wafers. Background Art

[0002] A process processing device is a device that performs process processing for forming fine shapes on semiconductor wafers. In a process processing device, maintenance operations such as replacement and cleaning of components to be maintained are usually performed regularly based on the number of wafers processed, etc. However, due to the degradation of components accompanied by changes over time and the accumulation of reaction byproducts corresponding to the usage method, unplanned maintenance operations may occur. In order to reduce the non-operating time caused by unplanned maintenance, it is necessary to monitor the degradation state of the components in sequence and take initial countermeasures such as cleaning and component replacement according to the degradation state.

[0003] In order to implement such an initial countermeasure, in the diagnostic device of the process treatment device, first, the sensor waveform data of the time series signal composed of multiple sensor items sequentially acquired from multiple state sensors attached to the components of the process treatment device in each process treatment is used to calculate the feature quantity representing the feature of the sensor waveform data (the data item of the sensor waveform data is called sensor item). Furthermore, based on the feature quantity, the degradation degree representing the degradation state of the component is estimated according to the deviation from the normal state, and compared with the pre-set reporting threshold value to determine whether maintenance is required and issue an alarm report.

[0004] Regarding the diagnostic processing of process processing equipment, for example, International Publication No. 2018 / 061842 (Patent Document 1) states that "the abnormality detection device applies statistical modeling to the summary value that summarizes the observed values, thereby inferring the state after the noise is removed from the summary value, and generating a predicted value of the summary value that predicts one step ahead based on the inference. The abnormality detection device detects whether the monitored object device has an abnormality based on the predicted value." In addition, Japanese Patent Publication No. 2012-9064 (Patent Document 2) states that "a learning-type process abnormality diagnosis device that achieves abnormality diagnosis performance suitable for actual process monitoring while accurately detecting abnormalities." These prior arts are so-called individual methods that perform statistical processing on each sensor waveform data separately and perform degradation estimation by comparing the predicted value with the current value.

[0005] Prior Art Literature

[0006] Patent Literature

[0007] Patent Document 1: International Publication No. 2018 / 061842

[0008] Patent Document 2: Japanese Patent Application Publication No. 2012-9064 Summary of the invention

[0009] -Problems that the invention aims to solve-

[0010] However, in the prior art, it is considered that in the diagnosis of the process equipment, false reports on the diagnosis of degradation occur due to the following factors, and it is difficult to take effective measures based on the diagnosis results.

[0011] First, in a process device, since the status of the device components also changes according to the processing history, there may be a false report caused by the initial status difference between the components in the process device group. In the individual diagnosis method such as Patent Documents 1 and 2, there is no problem recognition corresponding to the deviation between the components of the process device group.

[0012] Second, false reports may be caused by the setting of the reporting threshold. When the threshold is set based on the operator's past knowledge, the occurrence of false reports depends on the operator's experience. In addition, sometimes the threshold is set by setting a management benchmark such as 3σ for the deviation of the characteristic quantity of each sensor of each process treatment device under the normal state. In this case, since the process treatment device is equipped with a variety of sensors, they interfere with each other through control or are affected by the device state, so it is rare for the output of all sensor items to be stable. Therefore, when diagnosing deviations from the normal state for a variety of sensor items, false reports may occur frequently. In addition, when the threshold is set based on the sensor value when the degradation phenomenon occurs, the setting itself is difficult because the frequency of the degradation phenomenon is relatively low. In individual diagnosis methods such as Patent Documents 1 and 2, there is no recognition of the issues related to the setting of thresholds for degradation states corresponding to differences in devices and a variety of sensor items.

[0013] Third, since the multiple sensors attached to the components of the process treatment device interfere with each other, in the method of performing diagnosis based on the characteristic quantity calculated by each sensor as in the individual methods of Patent Documents 1 and 2, the diagnostic result of each sensor may not always be directly related to the degradation state of the component, which may result in false reporting.

[0014] Therefore, an object of the present invention is to provide a technique for suppressing false reports generated in components whose states are measured by a plurality of sensors or in each component of a process processing device group in the diagnosis of deterioration of a process processing device, and enabling effective countermeasures to be taken.

[0015] -Methods used to solve the problem-

[0016] In order to solve the above-mentioned problems, one of the representative diagnostic devices of a process treatment device of the present invention is a diagnostic device for a process treatment device that takes a component of the process treatment device as a maintenance object, and the diagnostic device for the process treatment device comprises: a characteristic quantity change calculation unit, which calculates a degradation degree corresponding to a change index based on a characteristic quantity of each sensor waveform data of a sensor item of a plurality of sensors (a state sensor group) for measuring the state of the component; and a component degradation degree calculation unit, which uses a plurality of the degradation degrees to calculate a component degradation degree that corresponds one-to-one to the component.

[0017] -Effects of the Invention-

[0018] According to the present invention, it is possible to provide a technique for suppressing false reports generated in components whose states are measured by a plurality of sensors or in each component of a process processing device group in the diagnosis of deterioration of a process processing device, and to take effective measures.

[0019] Problems, structures, and effects other than those described above will become more apparent from the following description of the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is an overall structural diagram of a process treatment device and a diagnostic device of the process treatment device.

[0021] Figure 2 It is a diagram showing the relationship between the component degradation degree and the positioning of the device common threshold and the like.

[0022] Figure 3 This is a flowchart showing an example of the procedure of the processing in the computer 30 .

[0023] Figure 4 This is a flowchart showing an example of the procedure of the processing in the server 50 .

[0024] Figure 5 This is a diagram for explaining an example of a method for setting a threshold value common to devices.

[0025] Figure 6 2 is a diagram showing an example of a display screen of the diagnosis result display unit 512 . DETAILED DESCRIPTION

[0026] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In addition, in all the drawings for describing the embodiments, the same reference numerals are given to the same parts in principle, and their repeated descriptions are omitted.

[0027] (Processing equipment)

[0028] Figure 1 It is an overall structural diagram of the process treatment device and the diagnostic device of the process treatment device. Figure 1 As shown, the process processing device A (10), process processing device B (11), etc. constituting the process processing device group 1 in this embodiment process a wafer (sample 101) according to a pre-set process processing condition. In addition, each process processing device has a component of the same type, and each component is attached with a state sensor group 102 composed of a plurality of sensors for measuring the state of the component. The state sensor group 102 can obtain the measured value of the sensor value (for example, temperature, pressure) during processing or idle as sensor waveform data. As an example of a process processing device, a component, and a state sensor, a plasma processing device, a microwave generating unit, a current / voltage sensor, etc. are listed respectively.

[0029] (Diagnostic device)

[0030] like Figure 1 As shown, the diagnostic device 2 includes a computer group (computer 30, computer 40, ...), which is composed of an execution unit 31 that performs acquisition of sensor waveform data corresponding to each process treatment device of the process treatment device group 1 and a storage unit 32 that stores information required for the execution unit's processing. Furthermore, the diagnostic device 2 includes a server 50, which is composed of an analysis unit 51 and a storage unit 52, wherein the analysis unit 51 performs the calculation conditions of the execution unit 31, the setting of the device common threshold, and the display of the diagnostic results, and the storage unit 52 stores the information required for the analysis unit 51's processing.

[0031] The process treatment device group 1 is directly connected to the computer group (computer 30, computer 40, ...), or connected to the computer group (computer 30, computer 40, ...) via a network. In addition, the computer group (computer 30, computer 40, ...) and the server 50 are connected via a network. As a result, each computer can perform calculations using sensor waveform data acquired from each process treatment device at high speed through the execution unit 31. In addition, the server 50 can set calculation conditions and thresholds across the process treatment device group 1, analyze and display diagnostic results. However, the connection between the computer group and the server 50 is not necessarily limited to the network, and can also be directly connected in the processing device.

[0032] (computer)

[0033] The execution unit 31 is composed of a processor such as a CPU or a GPU, and as described later, includes a pre-processing unit 310 , a feature value change calculation unit 311 , a diagnosis / report unit 312 , a feature value calculation unit 313 , and a component degradation degree calculation unit 314 .

[0034] The storage unit 32 is composed of a memory, a hard disk, and the like, and includes a sensor waveform storage unit 320 and a calculation result storage unit 321 as described later.

[0035] (server)

[0036] The analyzing unit 51 is constituted by a processor such as a CPU or a GPU, and includes a calculation condition setting unit 510 , a device common threshold setting unit 511 , and a diagnosis result display unit 512 , as described later.

[0037] The storage unit 52 is composed of a memory, a hard disk, and the like, and includes a diagnosis result storage unit 520 and a maintenance period information storage unit 521 as described later.

[0038] Figure 2 This is a relationship diagram showing the positioning of component degradation degree and common threshold value of the device. The process device has one or more components, and each component is attached with multiple sensors (state sensor group). And one or more feature quantities are calculated for the sensor items of each sensor, and one or more degradation degree is calculated for each feature quantity. In this way, based on multiple degradation degrees obtained from one component, the component degradation degree corresponding to the component one-to-one is calculated by calculation.

[0039] In addition, a device common threshold value serving as a reference for degradation detection is set based on the component degradation degree of the same type of components included in a plurality of process apparatuses (process apparatus group).

[0040] (Processing in a computer)

[0041] An example of the diagnosis process for the deterioration state of each component of the process apparatus 10 performed by the computer 30 will be described. Figure 3 30 is a flowchart showing an example of the sequence of processing in the computer 30. Figure 3 The process flow for each diagnosis cycle is shown. The diagnosis cycle may be set, for example, for each process or for each fixed time interval.

[0042] First, the computer 30 acquires sensor waveform data within the diagnosis period from the sensor waveform storage unit 320. The sensor waveform data is structured such that sensor values ​​are stored in a time series of processing time, time, etc. for a plurality of sensor items. Here, in the pre-processing unit 310, one sensor item is selected from the sensor waveform data (S1).

[0043] Next, the sensor waveform data of the selected sensor item is preprocessed in the preprocessing unit 310 (S2). The preprocessing method is arbitrary, for example, the process ID used in the diagnosis is specified using the process ID as information for identifying the process conditions, and the sensor waveform data associated with the process ID is extracted. In addition, the processing time interval used in the diagnosis is extracted, and the missing value is removed and the data is standardized to facilitate processing.

[0044] Next, the feature quantity calculation unit 313 uses the pre-processed sensor waveform data to calculate the feature quantity representing the feature of the sensor waveform data (S3). Figure 2 As shown, for each sensor item, one or more feature quantities are calculated. The feature quantity can use statistics such as the average value and standard deviation of the sensor waveform data. For example, in a plasma processing device, for the waveform data of the current value, the average value within a specific time interval can be used as a feature quantity. It is possible to calculate a feature quantity with a high probability of being associated with degradation by comparing it with the domain knowledge of the component, or to calculate multiple feature quantities.

[0045] Next, the characteristic quantity change calculation unit 311 determines whether the initial state period has been exceeded (S4). The initial state period refers to the time interval from immediately after the maintenance work of the component to a certain period.

[0046] If the diagnostic cycle at the time point of determination has not exceeded the initial state interval, the pre-processing unit 310 determines whether the characteristic quantity calculation has been completed for all sensor items of the characteristic quantity calculation object (S6). If completed, the processing in the diagnostic cycle ends. For S6, if the calculation is not completed, the process moves to the selection of other sensor items and continues the processing.

[0047] In S4, when the initial state interval is exceeded, the characteristic quantity change calculation unit 311 calculates the degradation degree of the change amount from the initial state for each calculated characteristic quantity (S5). For each characteristic quantity, the degradation degree is calculated using a plurality of change indicators. Examples of change indicators include the difference or ratio between the average value of the characteristic quantity in the initial state interval and the characteristic quantity in the diagnosis period, the Mahalanobis distance between the characteristic quantity in the initial state interval and the characteristic quantity in the diagnosis period, and the like. In this way, in each process treatment device of the process treatment device group 1, the degradation degree is defined based on the change amount from the initial state, thereby enabling comparison of the degradation degree between process treatment devices. In addition, since the change indicator to be compared depends on the sensor item, a plurality of change indicators are defined, and a plurality of degradation degrees are calculated based on the change amount. For example, it is assumed that if the change amount from the initial state is a significant physical quantity, the difference is effective as a change indicator, and if the rate of change is a significant physical quantity, the ratio is effective as a change indicator. In addition, it is assumed that even in the initial state, if the characteristic quantity has a large deviation, the Mahalanobis distance taking the deviation into consideration is also effective.

[0048] Next, the process proceeds to whether the calculation of the degradation degree for all sensor items is completed in the component degradation degree calculation unit 314 ( S7 ). If not completed, the same process is repeated for other sensor items. If completed, the process proceeds to S8 .

[0049] In S8, in the component degradation degree calculation unit 314, the multiple degradation degrees calculated for each component are integrated into an index of the component degradation degree. In addition, the contribution of each degradation degree to the component degradation degree is calculated together. The calculation conditions of the component degradation degree are calculated in the calculation condition setting unit 510, so an example of the calculation method will be described later. It is assumed that in the degradation of the components of the process treatment device, multiple degradation factors overlap, or even a single degradation factor, its degradation state will affect multiple sensor items. By utilizing a comprehensive method of the component degradation degree calculated based on each degradation degree, it is possible to perform calculation and diagnosis of the degradation degree that takes into account the interaction between multiple sensor items.

[0050] Next, in the diagnosis / reporting unit 312 , it is determined that the calculation of the component degradation degree is completed for all the components to be diagnosed, and the calculation is repeated until the calculation is completed ( S9 ).

[0051] Finally, in the diagnosis / reporting unit 312, the device common threshold value set in advance for each component by the device common threshold value setting unit 511 is compared with the value calculated by calculating the component degradation degree in the diagnosis cycle, and when the device common threshold value is exceeded, an alarm report is notified to the server 50 (S10). A series of calculation results calculated by the computer 30 are stored in the calculation result storage unit 321.

[0052] (Processing in the server)

[0053] An example of a process performed by the server 50 to set the calculation conditions and the device common threshold in the computer 30 will be described. Figure 4 1 is a flowchart showing an example of the flow of processing in the server 50. The flow of processing related to setting of the calculation conditions and the device common threshold value may be performed at a timing when it is necessary to set / update the value of new calculation conditions and the device common threshold value.

[0054] First, in the calculation condition setting unit 510, for the component of interest, the calculation result of the degree of degradation in the process treatment device group 1 and the maintenance period information (T1) are obtained from the calculation result storage unit 321 and the maintenance period information storage unit 521, respectively. The maintenance period information includes historical information such as the period and content of maintenance performed in accordance with the degradation of the components of each process treatment device.

[0055] Next, in the calculation condition setting unit 510, for the process treatment device of interest, the degradation degree (T2) within the maintenance interval is extracted based on the time series trend of the degradation degree calculated in the computer 30 for each diagnostic cycle and based on the maintenance period information. As a result, the left end of the time series trend of the degradation degree becomes the initial state, and the right end represents the degradation state. Since multiple degradation degrees are calculated in the computer 30, there are also multiple time series trends of the degradation degree. In addition, for process treatment devices that do not contain maintenance period information, that is, new process treatment devices, and process treatment devices that have been in use since maintenance (no degradation phenomenon has occurred), the degradation degree trend from the initial state to the latest calculation time point is extracted. Since the occurrence of degradation phenomena is relatively rare, by making use of the data of process treatment devices that do not contain maintenance period information, it is also effective to suppress false reports.

[0056] Next, in the calculation condition setting unit 510, the component degradation degree is calculated using the time series trends of multiple degradation degrees (T3). There are many methods for calculating the component degradation degree. For example, a model for calculating the component degradation degree can be constructed by applying a machine learning technique including a decision tree structure. When constructing the model, each degradation degree trend is used as an explanatory variable and the degradation period is used as a target variable to obtain a model, and the parameters of the model are optimized according to the data. From the perspective of narrowing the range of effective feature quantities for each sensor item, a machine learning technique with a feature quantity selection function can also be used, such as Figure 2 As shown, based on selecting one degradation degree for each sensor item, the component degradation degree is calculated. In addition, from the perspective of the explanatory nature of the diagnosis result, it is desirable to set a model that can calculate the contribution of each degradation degree to the component degradation degree. In addition, multiple models can also be constructed.

[0057] The device common threshold setting unit 511 determines that the calculation is completed for all the process treatment devices to be diagnosed, and repeatedly performs the process from T2 to T3 for the process treatment devices to calculate the component degradation degree of the component of interest ( T4 ).

[0058] Next, the device common threshold value setting unit 511 sets a threshold value common to the process device group 1 for each component degradation degree ( T5 ). Figure 5 FIG. 1 is a diagram for explaining an example of a method for setting a common threshold value for each device. The state in which the time series trend of the component degradation degree is calculated over the process device group 1 is shown in FIG. Figure 5 . Scan the set thresholds in detail, and set a common threshold between devices so that the prediction accuracy index of the expected degradation period becomes the maximum. In order to suppress false reports, as a prediction accuracy index, for example, a recurrence rate under conditions where the suitability rate is considered to be above a reference value is used. In the case where multiple component degradation degree calculation models are constructed in T3, the component degradation degree calculation model used in the diagnosis of the computer 30 is selected based on the prediction accuracy index. In this way, by setting a common threshold for the devices so as to optimize the prediction accuracy index in a comprehensive manner, false reports caused by individual factors of the devices can be suppressed.

[0059] Finally, in the calculation condition setting unit 510 , the selected component degradation degree calculation model (calculation condition) and the inter-device common threshold are set as conditions used in the diagnosis of the computer 30 ( T6 ).

[0060] Next, in the calculation condition setting unit 510 , it is determined that the setting is completed for all the components to be diagnosed, and the process from T1 to T6 is repeated ( T7 ), and the process ends.

[0061] Figure 6is a diagram showing an example of a display screen of the diagnosis result display unit 512. Figure 6 As shown in (a), for all components of the process treatment device group 1 to be diagnosed, the time series trend of component degradation and the common threshold value between devices can be viewed at a glance. As a result, it is possible to understand at a glance which component of which process treatment device is close to a degraded state. In addition, if an error of a process treatment device with an unknown cause occurs, by giving priority to the components close to a degraded state as maintenance work objects, effective countermeasures can be taken.

[0062] Furthermore, if Figure 6 As shown in (b), by specifying the component ID, the contribution of each degradation degree (degradation degree calculation condition) to the component degradation degree can be confirmed. This serves as auxiliary information for factor estimation based on comparison with component domain knowledge, and effective measures can be taken.

[0063] The above embodiment is described, but the present invention is not limited to the above embodiment, and various changes can be made without departing from the scope of the present invention. For example, when it is difficult to install the server 50, the computer 30 can also be configured to execute part of the processing of the server 50.

[0064] The following describes the embodiments that may be the contents of the present invention, but is not limited thereto.

[0065] (Method 1)

[0066] A diagnostic device for a process treatment device, which takes a component of the process treatment device as a maintenance object, is characterized in that it comprises: a characteristic quantity change calculation unit, which calculates a degradation degree corresponding to a change index based on a characteristic quantity of each sensor waveform data of a sensor item corresponding to a plurality of sensors (a state sensor group) for measuring the state of the component; and a component degradation degree calculation unit, which uses a plurality of the degradation degrees to calculate a component degradation degree corresponding one-to-one to the component.

[0067] (Method 2)

[0068] The diagnostic device for a process treatment device described in Embodiment 1 is characterized in that the diagnostic device for the process treatment device comprises: a diagnosis / reporting unit that performs diagnosis / reporting by comparing a device common threshold set between components of a process treatment device group and the component degradation degree.

[0069] (Method 3)

[0070] In the diagnostic device of the process treatment device described in method 1 or 2, it is characterized in that, in the characteristic quantity change calculation unit, as the change index, at least one of the difference between the characteristic quantity within the diagnostic period and the average value of the characteristic quantity within the initial state interval of the component (hereinafter referred to as "initial characteristic quantity"), the ratio of the characteristic quantity within the diagnostic period to the average value of the initial characteristic quantity, or the Mahalanobis distance between the characteristic quantity within the diagnostic period and the average value of the initial characteristic quantity is used.

[0071] (Method 4)

[0072] In the diagnostic device for a process treatment device according to any one of aspects 1 to 3, the component degradation degree calculation unit calculates the component degradation degree as a continuous probability value using a method of applying a regression model including a decision tree structure.

[0073] (Method 5)

[0074] In the diagnostic device for a process treatment device described in method 2, it is characterized in that the diagnostic device for the process treatment device comprises: a device common threshold setting unit, which searches in detail for the prediction accuracy index of the degradation period according to the time series trend of the component degradation degree in each component of the process treatment device group, thereby setting the device common threshold.

[0075] (Method 6)

[0076] In the diagnostic device for a process treatment device described in method 5, it is characterized in that, in the device common threshold setting unit, the suitability rate and the reproducibility rate are used as the prediction accuracy indicators, and the device common threshold is set so that the suitability rate is above the reference value and the reproducibility rate is the maximum.

[0077] (Method 7)

[0078] In the diagnostic device of a process treatment device described in any one of methods 2, 5, and 6, it is characterized in that the diagnostic device of the process treatment device comprises: a diagnostic result display unit, which displays the time series trend of the component degradation degree in each component of the process treatment device group and the contribution of multiple degradation degrees used in the calculation of the component degradation degree.

[0079] (Method 8)

[0080] A diagnostic system for a process treatment device, which takes a component of the process treatment device as a maintenance object, is characterized in that it comprises: a characteristic quantity change calculation unit, which calculates a degradation degree corresponding to a change index based on a characteristic quantity of each sensor waveform data of a sensor item of a plurality of sensors (a state sensor group) for measuring the state of the component; and a component degradation degree calculation unit, which uses a plurality of the degradation degrees to calculate a component degradation degree corresponding one-to-one to the component.

[0081] (Method 9)

[0082] The diagnostic system for a process treatment device described in method 8 is characterized in that the diagnostic system for the process treatment device comprises: a diagnosis / reporting unit that performs diagnosis / reporting by comparing a device common threshold set between components of the process treatment device group and the component degradation degree.

[0083] (Method 10)

[0084] A diagnostic method for a process treatment device, which takes a component of the process treatment device as a maintenance object, is characterized by comprising the following steps: in a characteristic quantity change calculation unit, a degradation degree corresponding to a change index is calculated based on the characteristic quantity of each sensor waveform data of a sensor item of a plurality of sensors (a state sensor group) for measuring the state of the component; and in a component degradation degree calculation unit, a plurality of the degradation degrees are used to calculate a component degradation degree corresponding one-to-one to the component.

[0085] (Method 11)

[0086] In the diagnostic method for a process treatment device described in method 10, it is characterized in that the diagnostic method for the process treatment device has the following steps: in the diagnosis / reporting unit, diagnosis / reporting is performed by comparing the device common threshold set between each component of the process treatment device group and the degradation degree of the component.

[0087] Description of Reference Numerals

[0088] 1...processing device group, 2...diagnostic device, 10, 11...processing device

[0089] 30, 40...computer, 31...execution unit, 32, 52...storage unit, 50...server

[0090] 51 ... analysis unit, 310 ... pre-processing unit, 311 ... characteristic quantity change calculation unit, 312 ... diagnosis / report unit

[0091] 313 ... characteristic quantity calculation unit, 314 ... component degradation degree calculation unit

[0092] 510 ... operation condition setting unit, 511 ... device common threshold value setting unit, 512 ... diagnosis result display unit

[0093] 520 ... a diagnosis result storage unit, 521 ... a maintenance period information storage unit.

Claims

1. A diagnostic device for a process treatment device, taking a component of the process treatment device as a maintenance object, characterized in that: have: a feature quantity change calculation unit that calculates a degree of degradation corresponding to a change index based on a feature quantity of each sensor waveform data of a sensor item of a state sensor group, which is a plurality of sensors for measuring the state of the component; and The component degradation degree calculation unit calculates a component degradation degree corresponding one-to-one to the component using the plurality of degradation degrees.

2. The diagnostic device for a process treatment device according to claim 1, characterized in that: The diagnostic device for process processing equipment includes a diagnostic / reporting unit configured to perform diagnosis / reporting by comparing a device common threshold value set between components of a process processing equipment group with the component degradation degree.

3. The diagnostic device for a process treatment device according to claim 1 or 2, characterized in that: In the characteristic quantity change calculation unit, as the change index, at least one of the difference between the characteristic quantity within the diagnostic period and the characteristic quantity within the initial state interval of the component, i.e., the average value of the "initial characteristic quantity", the ratio of the characteristic quantity within the diagnostic period to the average value of the initial characteristic quantity, or the Mahalanobis distance between the characteristic quantity within the diagnostic period and the average value of the initial characteristic quantity is used.

4. The diagnostic device for a process treatment device according to claim 1 or 2, characterized in that: The component degradation degree calculation unit calculates the component degradation degree as a continuous probability value using a method of applying a regression model including a decision tree structure.

5. The diagnostic device for a process treatment device according to claim 2, characterized in that: The process equipment diagnostic device includes: an equipment common threshold setting unit that exhaustively searches for a prediction accuracy index of a degradation period based on a time series trend of component degradation degrees in each component of the process equipment group, thereby setting the equipment common threshold.

6. The diagnostic device for a process treatment device according to claim 5, characterized in that: The device common threshold setting unit sets the device common threshold using a fit rate and a reproducibility rate as the prediction accuracy index so that the fit rate is equal to or greater than a reference value and the reproducibility rate is maximized.

7. The diagnostic device for a process treatment device according to any one of claims 2, 5 and 6, characterized in that: The diagnostic device for the process treatment apparatus includes a diagnostic result display unit that displays a time series trend of component degradation degrees in each component of the process treatment apparatus group and contribution degrees of a plurality of the degradation degrees used in calculation of the component degradation degrees.

8. A diagnostic system for a process treatment device, wherein components of the process treatment device are taken as maintenance objects, characterized in that: have: a feature quantity change calculation unit that calculates a degree of degradation corresponding to a change index based on a feature quantity of each sensor waveform data of a sensor item of a state sensor group, which is a plurality of sensors for measuring the state of the component; and The component degradation degree calculation unit calculates a component degradation degree corresponding one-to-one to the component using the plurality of degradation degrees.

9. The diagnostic system for a process treatment device according to claim 8, characterized in that: The diagnostic system for process equipment includes a diagnostic / reporting unit configured to perform diagnosis / reporting by comparing a device common threshold value set between components of a process equipment group with the component degradation degree.

10. A diagnostic method for a process treatment device, wherein a component of the process treatment device is taken as a maintenance object, characterized in that: The steps are as follows: The characteristic quantity change calculation unit calculates a degradation degree corresponding to the change index based on the characteristic quantity of each sensor waveform data of the sensor items of the state sensor group, which are a plurality of sensors for measuring the state of the component; and In the component degradation degree calculation unit, the component degradation degree corresponding to the component one-to-one is calculated using the plurality of degradation degrees.

11. The diagnostic method for a process treatment device according to claim 10, characterized in that: The diagnostic method of the process treatment device comprises the following steps: The diagnosis / reporting unit performs diagnosis / reporting by comparing a device common threshold value set among the components of the process device group with the component degradation degree.

Images