KR20190081844A - Method and apparatus for GENERATING VIRTUAL SENSOR DATA - Google Patents

Abstract

The present invention relates to a method for generating virtual sensor data for semiconductor manufacturing equipment in a computing device, comprising: a step of collecting sensor data used in determining a defect in a semiconductor manufacturing process from a plurality of sensors attached to semiconductor manufacturing equipment; a step of setting the equipment type of a plurality of virtual equipment, respectively based on recipe time of each of the plurality of virtual equipment corresponding to the semiconductor manufacturing equipment and the number of the virtual sensors attached to each of the plurality of virtual equipment; a step of deforming the sensor data based on each of the equipment types of the plurality of virtual equipment, and storing reference files including the deformed sensor data; and a step of controlling each of the plurality of virtual equipment to generate virtual sensor data having a predetermined pattern based on sub-reference files on the basis of the equipment type of the virtual equipment within the reference file. Therefore, the present invention can conduct accurate pattern matching.

Description

TECHNICAL FIELD [0001] The present invention relates to a method and apparatus for generating virtual sensor data,

The present invention relates to a method and apparatus for generating virtual sensor data.

One of the equipment-intensive and complex production systems, FAB (Fabrication), is one of the processes for producing semiconductors. It consists of several hundreds of unit processes, and it takes several months to complete all the processes for one lot . In the process, there may be a sudden stoppage due to equipment failure. It takes several days to restart the stopped production line, and the opportunity loss to it is several hundred billion won. Equipment failures are accompanied by changes in noise, vibration, pressure, temperature, etc. These changes are symptomatic before the equipment is abruptly stopped. To prevent loss, monitor the condition of the equipment before the failure do.

Therefore, there is a need for a technique for generating virtual sensor data similar to the sensor data generated according to an actual semiconductor process.

The computing device generates virtual sensor data having a predetermined pattern in each of the plurality of virtual equipments by using the recipe time of the sensor data and the plurality of virtual equipments collected in advance from the plurality of sensors attached to the semiconductor manufacturing equipment Can be controlled.

The computing device can perform more accurate pattern matching by setting the generation ratio of the normal data and the defect data in the virtual sensor data and using the virtual sensor data generated according to the generation ratio.

The computing device may generate virtual sensor data similar to sensor data generated in accordance with actual semiconductor processing.

According to one aspect, there is provided a method comprising: collecting sensor data used to determine defects in a semiconductor manufacturing process from a plurality of sensors attached to semiconductor manufacturing equipment; Setting a device type of each of the plurality of virtual machines based on a recipe time of each of the plurality of virtual machines corresponding to the semiconductor manufacturing equipment and a number of virtual sensors attached to each of the plurality of virtual machines; Modifying the sensor data based on the equipment type of each of the plurality of virtual devices and storing a reference file including the modified sensor data; And controlling each of the plurality of virtual devices to generate virtual sensor data having a predetermined pattern based on a sub-reference file based on a device type of the virtual device in the reference file. A method of generating hypothetical sensor data of a manufacturing equipment is provided.

According to another aspect, there is provided a method comprising: collecting sensor data used to determine a defect in a semiconductor manufacturing process from a plurality of sensors attached to a semiconductor manufacturing equipment; Setting a device type of each of the plurality of virtual machines based on a recipe time of each of the plurality of virtual machines corresponding to the semiconductor manufacturing equipment and a number of virtual sensors attached to each of the plurality of virtual machines; Modifying the sensor data based on the equipment type of each of the plurality of virtual devices and storing a reference file including the modified sensor data; And controlling to cause each of the plurality of virtual devices to generate virtual sensor data having a predetermined pattern based on a sub-reference file based on a device type of the virtual device in the reference file A computer program stored on a medium is provided.

The computing device can control to generate various types of virtual sensor data in each of the plurality of virtual devices.

The computing device can perform pattern matching using the virtual sensor data and can detect an error in the manufacturing process of the semiconductor sorting equipment according to the result of the pattern match.

By reducing the error between the sensor data generated in the actual semiconductor manufacturing process and the virtual sensor data, it is possible to minimize the cost and error generated in the semiconductor manufacturing process and improve the quality of the produced product.

This disclosure may be readily understood by reference to the following detailed description and the accompanying drawings, in which reference numerals refer to structural elements.
FIG. 1 is a diagram illustrating a process for performing pattern matching and error detection using virtual sensor data of a semiconductor manufacturing equipment in a computing device according to an exemplary embodiment. Referring to FIG.
2 is a flow chart illustrating a method for generating virtual sensor data of semiconductor manufacturing equipment at a computing device, in accordance with one embodiment.
3A is a diagram for explaining a process of modifying sensor data according to an interpolation method in a computing device according to an embodiment.
3B is a diagram for explaining a process of modifying sensor data according to an interval averaging method in a computing device according to an embodiment.
3C is a diagram for explaining a process of modifying sensor data in a computing device when the type of sensor data is switch data, according to an embodiment.
3D is a diagram for explaining a process of modifying sensor data in a computing device when the type of sensor data is stepwise data according to an embodiment.
3E is a diagram for explaining a process of modifying sensor data in a computing device when the type of sensor data is vibration data, according to an embodiment.
4 is a flow chart illustrating a method of controlling virtual device data in each of a plurality of virtual devices in a computing device, according to one embodiment.
5 illustrates virtual sensor data obtained using sensor data set as a representative of sensors attached to a virtual machine, according to one embodiment.

Although the terms used in this disclosure have taken into account the functions in this disclosure and have made possible general terms that are currently widely used, they may vary depending on the intent or circumstance of the person skilled in the art, the emergence of new technologies and the like. Also, in certain cases, there may be a term selected arbitrarily by the applicant, in which case the meaning thereof will be described in detail in the description of the corresponding invention. Accordingly, the terms used in this disclosure should be defined based on the meaning of the term rather than on the name of the term, and throughout the present disclosure.

When an element is referred to as "including" an element throughout the specification, it is to be understood that the element may include other elements as well, without departing from the spirit or scope of the present invention.

The terms first, second, etc. may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present disclosure, a first component can be named a second component, and similarly, a second component can also be named a first component. And / or < / RTI > includes any combination of a plurality of related listed items or any of a plurality of related listed items.

Throughout the specification, the "computing device" modifies the sensor data collected based on the equipment type of the virtual machine, and uses the reference file containing the modified sensor data to input data used to generate the virtual sensor data , Control, memory, calculation, and output of the data. For example, the "computing device" may be a special-purpose computer, a general-purpose computer, a supercomputer, a mainframe computer, a personal computer, And is not limited to these.

The "computing device" may be physically separate from the manufacturing equipment for producing the semiconductor, or may be provided in a form mounted on the equipment. Further, the "computing device" may be provided as a device independent of the sensor for sensing data according to the semiconductor process, or may be provided on a sensor.

Also, a " computing device "may be provided with a computer program that executes instructions that instruct to generate virtual sensor data of semiconductor manufacturing equipment.

On the other hand, a "computing device" may include a memory, a processor, a user interface, a communication interface, and a display, and may further include other general components. Specifically, the memory may store software or a computer program. For example, the memory may store data received from equipment that produces semiconductors. In addition, the memory may store instructions that modify sensor data collected based on the machine type of the virtual machine, and generate virtual sensor data using a reference file containing the modified sensor data. The processor may execute instructions stored in the memory to control the computing device to generate virtual sensor data for the semiconductor manufacturing equipment. The user interface can receive various commands from the user. The communication interface can perform communication with an external device according to various types of communication methods. The display may display a screen for setting the equipment type of each of the plurality of virtual equipments based on the recipe time of each of the plurality of virtual equipments and the number of virtual sensors attached to each of the plurality of virtual equipments. In addition, the display may display modified data based on the equipment type of each of the plurality of virtual devices. In addition, the display may display virtual sensor data generated in each of the plurality of virtual devices.

In the following, various operations or applications performed by a memory, a processor, a user interface, a communication interface, and a display of a computing device are described, including but not limited to a memory, a processor, a user interface, a communication interface, It is to be appreciated that those skilled in the art will readily appreciate that many modifications may be made to adapt a memory device, processor, user interface, communication interface, and display of a computing device, The scope of the rights is not limited by the name of the specific configuration or the physical / logical structure.

Hereinafter, embodiments will be described in detail with reference to the drawings.

FIG. 1 is a diagram illustrating a process for performing pattern matching and error detection using virtual sensor data of a semiconductor manufacturing equipment in a computing device according to an exemplary embodiment. Referring to FIG.

Referring to block 110, the computing device 10 may collect sensor data used to determine faults in the semiconductor manufacturing process from a plurality of sensors attached to the semiconductor manufacturing equipment.

Referring to block 120, the computing device 10 may control to generate virtual sensor data in each of the plurality of virtual machines. The process by which the computing device 10 controls virtual sensor data to be generated in each of a plurality of virtual machines will be described in detail with reference to FIG. 2 to FIG.

Referring to block 130, the computing device 10 may perform pattern matching using the virtual sensor data and may detect errors in the manufacturing process of the semiconductor breeding equipment, depending on the result of the pattern match.

On the other hand, FDC (Fault Detection and Classification) based monitoring system is a summary based on specification such as Minimum, Maximum, and Average of sensor data, An analysis model can be established based on data (raw trace data). However, analytical models require modeling management that is inefficient and requires expertise. In addition, an event that affects yield may not be detected due to an unmanaged analysis model.

In addition, the DFD (Dynamic Fault Detection) based monitoring system can detect errors by pattern matching all the raw traces of the sensor data, not the sections. Since the DFD-based monitoring system reads the low-trace data in real time and performs pattern matching, it may not be suitable for the FAB process which has a highly integrated work process and has a unique structural pattern of process variables.

Therefore, the computing device 10 can use the recipe time of the sensor data and the plurality of virtual equipments previously collected from the plurality of sensors attached to the semiconductor manufacturing equipment, It is possible to control to generate sensor data. The computing device 10 may apply the generated virtual sensor data to the FDC or DFD-based monitoring system to perform a pattern match to detect a process or equipment that has caused an error / defect.

2 is a flow chart illustrating a method for generating virtual sensor data of semiconductor manufacturing equipment at a computing device, in accordance with one embodiment.

At block 210, the computing device 10 may collect sensor data used to determine defects in the semiconductor manufacturing process from a plurality of sensors attached to the semiconductor breeding equipment. Specifically, the computing device 10 may acquire sensor data from a plurality of sensors and store the sensor data in a memory within the computing device 10. The computing device 10 may read the sensor data stored in the memory to generate the virtual sensor data. Here, the sensor data may be data corresponding to at least one of noise, vibration, pressure and temperature measured from a plurality of sensors, but is not limited thereto. In addition, the sensor data may be time-series data obtained according to the operation of the semiconductor manufacturing equipment.

At block 220, the computing device 10 may determine, based on the recipe time of each of the plurality of virtual machines corresponding to the semiconductor manufacturing equipment and the number of virtual sensors attached to each of the plurality of virtual machines, You can set the equipment type. The equipment type can vary depending on the recipe time and the number of virtual sensors. In addition, since the output intervals are different for each virtual sensor attached to each of the plurality of virtual devices, the number of data generated during the recipe time varies for each virtual device.

At block 230, the computing device 10 may transform the sensor data based on the equipment type of each of the plurality of virtual instruments. The computing device 10 may store a reference file that contains modified sensor data.

Specifically, the computing device 10 may compare the number of the first virtual sensor data to be generated during the first recipe time of the first virtual equipment with the number of the first sensor data corresponding to the first virtual equipment. The computing device 10 may apply the interpolation method or the interval averaging method to the first sensor data based on the comparison result to transform the first sensor data so that the number of the first virtual sensor data matches the number of the first sensor data have. The process of modifying sensor data by applying interpolation or interval averaging to sensor data is described in FIGS. 3A-3B.

In addition, when the type of the first virtual sensor data to be generated during the first recipe time of the first virtual equipment is at least one type of switch data, cascade data, and vibration data, The value of the first sensor data corresponding to the first virtual device can be mapped to the value of the first virtual sensor data according to the number of the sensor data. The process of modifying the sensor data according to the type of virtual sensor data is described in Figures 3C-3E.

At block 240, the computing device 10 may control to generate virtual sensor data having a predetermined pattern of each of the plurality of virtual devices based on a sub-reference file based on the device type of the virtual device in the reference file .

The computing device 10 can control so that defect data included in the first virtual sensor data to be generated during the first recipe time of the first virtual equipment is continuously generated at any point in time. Specifically, the computing device 10 can determine the generation ratio of the normal data and the defect data included in the first virtual sensor data. The computing device 10 can classify the defect data to be generated according to the generation ratio into a predetermined group and control so that the classified defect data is generated at any point in the first recipe time.

In addition, the computing device 10 may extract the first sub-reference data from the first sub-reference file corresponding to the first recipe time of the first virtual machine. The computing device 10 may transmit the first sub-reference data to the virtual sensor attached to the first virtual device, in accordance with the output interval of the virtual sensor attached to the first virtual device during the first recipe time. When the first recipe time has elapsed, the computing device 10 may transmit the second sub-reference data, which is different from the first sub-reference data, to the virtual sensor attached to the first virtual equipment. The computing device 10 may control to generate second virtual sensor data based on the second sub-reference data in the first virtual machine.

Specifically, once the first recipe time has elapsed, the computing device 10 may extract the second sub-reference data different from the first sub-reference data in the first sub-reference file. The computing device 10 may transmit the second sub-reference data to the virtual sensor attached to the first virtual device, in accordance with the output interval of the virtual sensor attached to the first virtual device during the first recipe time after the first recipe time have. Here, if the first recipe time elapses and the first sub-reference data is not replaced with the second sub-reference data, since the same virtual sensor data is generated every period of the first recipe time, One sub-reference data can be replaced with the second sub-reference data.

Also, the computing device 10 may extract sub-reference data representative of each of the virtual sensors attached to the first virtual device in the sub-reference file, and set each of the sub-reference data as an average. The computing device 10 can control to generate the first virtual sensor data in the first virtual device based on the normal distribution using the average of each of the sub reference data and the predetermined standard deviation corresponding to each of the sub reference data .

On the other hand, the computing device 10 can perform a method of generating virtual sensor data of a semiconductor manufacturing equipment, in a different order, instead of the block order shown in Fig. For example, the computing device 10 may perform a method of generating virtual sensor data of semiconductor manufacturing equipment in accordance with block 210, block 230, block 220 and block 240.

The way in which the computing device 10 controls each of the plurality of virtual devices to generate virtual sensor data is described in Figures 4-5.

3A is a diagram for explaining a process of modifying sensor data according to an interpolation method in a computing device according to an embodiment.

When the number of the first virtual sensor data to be generated during the first recipe time of the first virtual equipment is larger than the number of the first sensor data corresponding to the first virtual equipment, the computing device 10 transmits the first sensor data The scale of the first sensor data can be extended by interpolation so that the number of the first virtual sensor data matches the number of the first sensor data.

The graph 310 of FIG. 3A shows first sensor data corresponding to a first virtual machine. The number of the first sensor data corresponding to the first virtual equipment is ten. On the other hand, the number of first virtual sensor data to be generated during the first recipe time of the first virtual equipment is 28 pieces. Thus, as shown in graph 320 of FIG. 3A, the computing device 10 can modify the first sensor data to add two pieces of data between the values of the first piece of sensor data, resulting in a total of 28 pieces of data .

Specifically, the computing device 10 uses the following equations (1) to (3) when adding M variable data between the two data? And?.

On the other hand, when the total number of data before the interpolation method is N, and the number of data of M variables is added, the number of interpolation method total data is expressed by Equation (4).

3B is a diagram for explaining a process of modifying sensor data according to the averaging method in a computing device, according to an embodiment.

When the number of the first virtual sensor data to be generated during the first recipe time of the first virtual equipment is smaller than the number of the first sensor data corresponding to the first virtual equipment, , The computing device 10 applies the interval averaging method to the first sensor data to reduce the scale of the first sensor data so that the number of the first virtual sensor data matches the number of the first sensor data can do.

The graph 330 of FIG. 3B shows the first sensor data corresponding to the first virtual machine. The number of the first sensor data corresponding to the first virtual equipment is ten. On the other hand, the number of first virtual sensor data to be generated during the first recipe time of the first virtual equipment is four. Thus, as shown in graph 340 of FIG. 3B, the computing device 10 may set the interval of the first sensor data into three consecutive data. The computing device 10 may calculate the average of the values of the data of the set interval and modify the first sensor data to be a total of four sensor data. That is, the computing device 10 may perform the interval averaging to obtain an average for each set interval. The computing device 10 may calculate the interval average for the interval as shown in Equation (5).

로 줄어들 수 있다.Here, if the interval average is repeated, the total number of data is about

.

Meanwhile, the computing device 10 may apply the interpolation method and the interval averaging method to the first sensor data to adjust the scale of the first sensor data so that the number of the first virtual sensor data matches the number of the first sensor data have.

3C is a diagram for explaining a process of modifying sensor data in a computing device when the type of sensor data is switch data, according to an embodiment.

If the type of sensor data is switch data, there are only two values of data on-off (0, 1), so if interpolation is used for sensor data, unacceptable data such as 0.5 may be generated. Therefore, when adding M data between two consecutive data n, n + 1, the computing device 10 has a value equal to n if it is close to n in order and n + 1 if n + It can be set to have the same value. If the distance between n and n + 1 is the same, the computing device 10 may select either n or n + 1 so that the data generated in the center is set to the same value each time.

The graph 350 of FIG. 3C shows the first sensor data corresponding to the first virtual machine. Here, when the scale of the first sensor data needs to be expanded so that the number of the first virtual sensor data is equal to the number of the first sensor data, as shown in the graph 360 of FIG. 3C, 10) can add one data between n and n + 1.

3D is a diagram for explaining a process of modifying sensor data in a computing device when the type of sensor data is stepwise data according to an embodiment.

The graph 370 in FIG. 3D shows the first sensor data corresponding to the first virtual machine. Here, the type of the first sensor data is cascade data. The cascade data has a sudden change in value and can be repeated with a value changed more than a certain number of times. In this case, if interpolation is simply used for the first sensor data, the modified first sensor data can not recognize the sudden change in the value and can have an intermediate value. Therefore, the computing device 10 uses Equations (6) to (8) as follows when adding the M variable data between two consecutive data? And?.

If the Mth additional data calculated according to Equation (8) is close to?, The additional data is set to the same value as?, And if it is close to?, The additional data can be set to the same value as?. As shown in the graph 380 of FIG. 3D, the computing device 10 may modify the first sensor data by adding one piece of data between? And?.

3E is a diagram for explaining a process of modifying sensor data in a computing device when the type of sensor data is vibration data, according to an embodiment.

Graph 390 of FIG. 3e shows the first sensor data corresponding to the first virtual machine. Here, the type of the first sensor data is vibration data. Since the vibration data is a pair of two points n and n + 1, when the first sensor data is expanded, n, n + 1 can be increased by one pair. As shown in graph 395 of Figure 3e, computing device 10 may increment n, n + 1 pairs by setting n, n + 1 to a pair.

4 is a flow chart illustrating a method of controlling virtual device data in each of a plurality of virtual devices in a computing device, according to one embodiment.

At block 410, the computing device 10 may determine the rate of generation of normal data and defect data contained in the first virtual sensor data. The computing device 10 may generate virtual sensor data including normal data and defect data to perform pattern matching. Here, if the ratio of the defect data is a predetermined ratio or more, the defect data can be determined as normal. Therefore, the computing device 10 can set the generation ratio of the normal data and the defect data to a predetermined ratio or less. For example, the generation rate of the defect data can be set to 1% of the entire virtual sensor data.

On the other hand, since the probability that defect data will be regularly generated in an actual semiconductor manufacturing apparatus is low and the probability that defect data will not be generated only once but continuously is high, As shown in FIG. That is, the computing device 10 can classify the defect data to be generated according to the generation ratio into a predetermined group, and to control so that the classified defect data is generated at any time point of the first recipe time.

For example, if the generation ratio of defect data per 500 sensor data is set to 1%, a total of five defect data must be generated. The computing device 10 may divide the five defect data into two groups and generate the second defect data at an arbitrary point in time.

At block 420, the computing device 10 generates virtual sensor data having a predetermined pattern, each of the plurality of virtual machines changing sub-reference data based on a sub-reference file based on the type of virtual machine in the reference file .

For example, in the case where the virtual sensor data of the first virtual device among the plurality of virtual devices is generated, the computing device 10 calculates the first virtual reference data in the first sub- reference file corresponding to the first recipe time of the first virtual device, The sub reference data can be extracted. Here, the first sub-reference data may include normal data and defect data. The computing device 10 may transmit the first sub-reference data to the virtual sensor attached to the first virtual device, in accordance with the output interval of the virtual sensor attached to the first virtual device during the first recipe time. The computing device 10 may control to generate the first virtual sensor data based on the first sub-reference data in the first virtual machine.

Once the first recipe time has elapsed, the computing device 10 may maintain the pattern and cause the individual data to be generated differently. Specifically, the computing device 10 may transmit the second sub-reference data, which is different from the first sub-reference data, to the virtual sensor attached to the first virtual equipment. The computing device 10 may transmit the second sub-reference data to the virtual sensor attached to the first virtual device, in accordance with the output interval of the virtual sensor attached to the first virtual device.

At block 430, the computing device 10 determines the representative sub-reference data, and based on the normal distribution using the mean and standard deviation, controls each of the plurality of virtual devices to generate virtual sensor data having a predetermined pattern can do.

For example, in the case where the virtual sensor data of the first virtual device among the plurality of virtual devices is generated, the computing device 10 generates the sub-reference data representing the virtual sensor attached to the first virtual device Can be extracted to set each of the sub reference data as an average. The computing device 10 can control to generate the first virtual sensor data in the first virtual device based on the normal distribution using the average of each of the sub reference data and the predetermined standard deviation corresponding to each of the sub reference data . The computing device 10 may maintain the pattern and cause the individual data to be generated differently. The computing device 10 sets the representative sub-reference data as an average (mu) and determines a standard deviation (sigma) of a predetermined value corresponding to the sub-reference data. And the computing device 10 may calculate a new random variable u according to equation (9) for standardization.

In this case, the probability density function of u is expressed by Equation (10).

The computing device 10 may calculate any value generated within a certain range of the obtained probability density to again correspond to the average.

5 illustrates virtual sensor data obtained using sensor data set as a representative of sensors attached to a virtual machine, according to one embodiment.

Referring to 510 of FIG. 5, the virtual sensor data obtained using the normal distribution of the sub-reference data set by the computing device 10 as the representative of the sensor data of the virtual sensor is shown. By reducing the standard deviation to 2, the virtual sensor data can be closer to the actual sensor data.

The computing device 10 according to the present disclosure may include a processor, a memory for storing and executing program data, a permanent storage such as a disk drive, a communication port for communicating with an external device, a touch panel, a key, Or the like, and the like. Methods implemented with software modules or algorithms may be stored on a computer readable recording medium as computer readable codes or program instructions executable on the processor. Here, the computer-readable recording medium may be a magnetic storage medium such as a read-only memory (ROM), a random-access memory (RAM), a floppy disk, a hard disk, ), And a DVD (Digital Versatile Disc). The computer-readable recording medium may be distributed over networked computer systems so that computer readable code can be stored and executed in a distributed manner. The medium is readable by a computer, stored in a memory, and executable on a processor.

All documents, including open documents, patent applications, patents, and the like, cited in this disclosure, may be incorporated into this disclosure as if each cited document were individually and specifically incorporated or represented in its entirety as a whole .

For purposes of understanding of the present disclosure, reference is made to the preferred embodiments shown in the drawings and specific terminology is used to describe the embodiments of the present disclosure, but the present disclosure is not limited by the specific terminology, May include all components that are commonly conceivable in the ordinary artisan.

The present disclosure may be represented by functional block configurations and various processing steps. These functional blocks may be implemented in a wide variety of hardware and / or software configurations that perform particular functions. For example, the present disclosure may be directed to integrated circuit configurations, such as memory, processing, logic, look-up tables, etc., that may perform various functions by control of one or more microprocessors or other control devices Can be adopted. Similar to the components in the present disclosure may be implemented with software programming or software components, the present disclosure may be implemented as a combination of C, C ++, and / or C ++, including various algorithms implemented in data types, processes, routines, , Java (Java), assembler, and the like. Functional aspects may be implemented with algorithms running on one or more processors. In addition, the present disclosure may employ conventional techniques for electronic configuration, signal processing, and / or data processing, and the like. Terms such as "mechanism", "element", "means", "configuration" may be used broadly and are not limited to mechanical and physical configurations. The term may include the meaning of a series of routines of software in conjunction with a processor or the like.

The specific implementations described in this disclosure are, by way of example, not intended to limit the scope of the present disclosure in any way. For brevity of description, descriptions of conventional electronic configurations, control systems, software, and other functional aspects of such systems may be omitted. Also, the connections or connecting members of the lines between the components shown in the figures are illustrative of functional connections and / or physical or circuit connections, which may be replaced or additionally provided by a variety of functional connections, physical Connection, or circuit connections. Also, unless expressly stated to the contrary, such as "essential "," significant ", etc., may not be a necessary component for the application of this disclosure.

On the other hand, the embodiments relating to the method of generating virtual data described above are such that in the computing device 10, sensor data used for judging defects in the semiconductor manufacturing process from a plurality of sensors attached to the semiconductor manufacturing equipment Setting a device type of each of a plurality of virtual machines based on a recipe time of each of a plurality of virtual machines corresponding to semiconductor manufacturing equipment and a number of virtual sensors attached to each of the plurality of virtual machines, Based on a device type of each of the plurality of virtual devices, modifies the sensor data, stores a reference file containing the modified sensor data, and generates a plurality of virtual devices based on the device type of the virtual device in the reference file, Stored in the computer-readable storage medium so that each of the virtual devices of the virtual machine can generate virtual sensor data having a predetermined pattern. It may be provided in the form of revocation or a computer program forms.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. For example, it is to be understood that the techniques described may be performed in a different order than the described methods, and / or that components of the described systems, structures, devices, circuits, Lt; / RTI > or equivalents, even if it is replaced or replaced.

Therefore, the scope of the present disclosure should not be construed as being limited to the embodiments described, but should be determined by the appended claims, as well as the appended claims.

Claims (10)

Collecting sensor data used to determine defects in a semiconductor manufacturing process from a plurality of sensors attached to the semiconductor manufacturing equipment;
Setting a device type of each of the plurality of virtual machines based on a recipe time of each of the plurality of virtual machines corresponding to the semiconductor manufacturing equipment and a number of virtual sensors attached to each of the plurality of virtual machines;
Modifying the sensor data based on the equipment type of each of the plurality of virtual devices and storing a reference file including the modified sensor data; And
And controlling each of the plurality of virtual devices to generate virtual sensor data having a predetermined pattern based on a sub reference file based on the equipment type of the virtual equipment in the reference file. A method for generating virtual sensor data of an instrument. The method according to claim 1,
The sensor data includes:
And data corresponding to at least one of noise, vibration, pressure, and temperature measured from the plurality of sensors,
Wherein the virtual sensor data is time-series data acquired according to an operation of the semiconductor manufacturing equipment. The method according to claim 1,
Wherein modifying the sensor data based on the equipment type of each of the plurality of virtual devices comprises:
Comparing the number of first virtual sensor data to be generated during the first recipe time of the first virtual equipment with the number of first sensor data corresponding to the first virtual equipment; And
Modifying the first sensor data by applying an interpolation method or an interval average method to the first sensor data based on the comparison result so that the number of the first virtual sensor data matches the number of the first sensor data Wherein the virtual sensor data includes at least one of the following: The method according to claim 1,
Wherein modifying the sensor data based on the equipment type of each of the plurality of virtual devices comprises:
When the type of the first virtual sensor data to be generated during the first recipe time of the first virtual equipment is at least one type of the switch data, the stepped data, and the vibration data, depending on the number of the first virtual sensor data, And mapping the value of the first sensor data corresponding to the first virtual equipment to the value of the first virtual sensor data. The method according to claim 1,
Wherein the step of controlling each of the plurality of virtual devices to generate virtual sensor data having a predetermined pattern includes:
And controlling the defective data included in the first virtual sensor data to be generated during the first recipe time of the first virtual equipment to be continuously generated at a certain point in time. 6. The method of claim 5,
Wherein the step of controlling the defect data to be continuously generated at an arbitrary point in time includes:
Determining normal data included in the first virtual sensor data and a generation ratio of the defect data; And
Classifying the defect data to be generated into a predetermined group according to the generation ratio and controlling the classified defect data to be generated at the arbitrary point in time of the first recipe time How to. The method according to claim 1,
Wherein the step of controlling each of the plurality of virtual devices to generate virtual sensor data having a predetermined pattern based on a sub reference file based on a device type of the virtual device in the reference file,
Extracting first sub-reference data from a first sub-reference file corresponding to a first recipe time of the first virtual machine;
Reference data is transmitted to the virtual sensor attached to the first virtual equipment in accordance with the output interval of the virtual sensor attached to the first virtual equipment during the first recipe time, Controlling the first virtual sensor data based on the first sub-reference data to be generated; And
Reference data that is different from the first sub-reference data is transmitted to the virtual sensor attached to the first virtual device when the first recipe time elapses, and transmits the second sub-reference data to the second sub- And controlling the second virtual sensor data based on the second virtual sensor data to be generated. 8. The method of claim 7,
And transmitting the sub-reference data different from the first sub-reference data to the virtual sensor attached to the first virtual equipment when the first recipe time elapses,
Extracting the second sub-reference data different from the first sub-reference data in the first sub-reference file when the first recipe time has elapsed; And
Transferring the second sub-reference data to a virtual sensor attached to the first virtual device, in accordance with the output interval of the virtual sensor attached to the first virtual device for a first recipe time after the first recipe time; And generating a virtual sensor data. The method according to claim 1,
Wherein the step of controlling each of the plurality of virtual devices to generate virtual sensor data having a predetermined pattern based on a sub reference file based on a device type of the virtual device in the reference file,
Extracting sub-reference data representative of each virtual sensor attached to the first virtual device in the sub-reference file and setting each of the sub-reference data as an average; And
And controlling the first virtual sensor data to be generated in the first virtual equipment based on a normal distribution using an average of each of the sub reference data and a predetermined standard deviation corresponding to each of the sub reference data. A method for generating virtual sensor data. A computing device,
Collecting sensor data used to determine defects in a semiconductor manufacturing process from a plurality of sensors attached to the semiconductor manufacturing equipment;
Setting a device type of each of the plurality of virtual machines based on a recipe time of each of the plurality of virtual machines corresponding to the semiconductor manufacturing equipment and a number of virtual sensors attached to each of the plurality of virtual machines;
Modifying the sensor data based on the equipment type of each of the plurality of virtual devices and storing a reference file including the modified sensor data; And
Based on a sub-reference file based on a device type of a virtual device in the reference file, controlling each of the plurality of virtual devices to generate virtual sensor data having a predetermined pattern, ≪ / RTI >

Images