KR101988596B1 - Apparatus for Monitoring Production Equipment and Driving Method Thereof - Google Patents

Abstract

The present invention relates to a production equipment monitoring apparatus and a driving method thereof. According to an embodiment of the present invention, the production equipment monitoring apparatus comprises: a storage unit storing reference data for determining whether an event occurs wherein the event includes at least access to an authorized person and an occurrence of an impact in designated production equipment; and a control unit receiving and analyzing a photographed image of the production equipment from a photographing apparatus installed in the periphery of the production equipment, and determining whether the access to the authorized person and the occurrence of the impact occur based on the analyzed result and the stored reference data.

Description

Technical Field [0001] The present invention relates to a production equipment monitoring apparatus and a driving method thereof,

The present invention relates to a production equipment monitoring apparatus and a driving method thereof, and more particularly, to a production equipment monitoring apparatus and a method of operating the same, And a driving method of the apparatus.

Generally, semiconductors are manufactured through processes such as diffusion, photo, etching, and the like. In each process, wafers are processed using various facilities. Wafers are typically manufactured by moving along a manufacturing line, where the quality and productivity of the semiconductor is greatly influenced by the state of the equipment during processing of the wafer. If there is an error in the semiconductor manufacturing equipment, if the continuous process is performed, the yield will be lowered and the cost will also be considerably lost. Accordingly, various methods have been proposed for quickly detecting faults in the wafer fabrication facility.

Most importantly, the importance of the exposure process during the semiconductor processing process is dominant and has a direct impact on the production yield. As with other production plants, semiconductor manufacturing plants also need to operate 24 hours a day, so it is necessary to monitor the equipment. If production continues because of erroneous data, it can cause tremendous losses, and the loss will be equally damaging to the semiconductor industry, which is responsible for the economy of a country as well as the damage of the company. Most of the semiconductor production equipment currently applied to the semiconductor production line is operated by the operator or the equipment person through direct auditory or visual alarm such as buzzer or signal tower when the operation is abnormal during operation. So that it can be recognized.

However, in such a case, if the operator or the person in charge of the equipment is not in the vicinity of the equipment, it is impossible to immediately recognize the occurrence of the equipment abnormality, so that the maintenance of the equipment is delayed and the factors hindering the yield and productivity are still not solved.

Korean Patent Registration No. 10-0596438 (Jun. 27, 2006) Korean Patent Registration No. 10-1779827 (Jul. Korean Patent Registration No. 10-0839071 (Jun. 11, 2008) Korean Registered Patent No. 10-1465500 (Apr. 20, 2014) Korean Patent Laid-Open Publication No. 10-2016-0130671 (Nov.

Embodiments of the present invention provide a production equipment monitoring apparatus and a method of driving the same that monitor the exposure equipment and the like which have a large proportion in the semiconductor production process and directly affect the production yield, There is a purpose.

The production equipment monitoring apparatus according to an embodiment of the present invention includes a storage unit that stores reference data for determining whether an event occurs, which includes at least an access to an authorized person and an occurrence of an impact on a designated production equipment, And a control unit for receiving and analyzing the photographed image of the production equipment from an installed photographing apparatus, and determining whether the event of the access to the authorized party and the occurrence of the shock occurrence is generated based on the analyzed result and the stored reference data.

The production equipment monitoring device may further include a communication interface unit for communicating with an external device that performs remote control of the production equipment and the control unit may control the communication interface unit to notify the external device of the occurrence of an event when an event occurs have.

The control unit may perform deep learning using the analysis result of the photographed image to set reference data including the event, and may set and store the priority of the set reference data.

The control unit may further use sensing data of a sensor provided in the periphery of the production equipment to determine the occurrence of the impact, and use the sensing data for the deep learning.

The controller acquires the captured image of the section in which the event has occurred and stores the acquired captured image and the additional information in the storage unit and stores the time information provided by the user for searching the captured image and the stored additional information And the acquired photographed image can be displayed on the screen on the basis of the acquired photographed image.

In addition, a method of driving a production equipment monitoring apparatus according to an embodiment of the present invention is a method of operating a production equipment monitoring apparatus including a storage unit and a control unit, wherein the storage unit accesses an authorized person Storing the reference data for determining whether or not an event including at least an event is included; and a control unit for receiving and analyzing an image taken by the production equipment from an image pickup apparatus installed in the periphery of the production equipment, And determining whether the event of the access to the authorized user and the occurrence of the impact occurrence is generated based on the stored reference data.

The method for driving the production equipment monitoring apparatus may further include a step in which the communication interface unit communicates with an external device that performs remote control of the production equipment and the control unit controls the communication interface unit to generate an event You can notify the fact.

The storing step may include performing deep running using the analysis result of the captured image to set reference data including the event, and setting and storing a priority of the set reference data.

The driving method of the production equipment monitoring apparatus may further include a step of using the sensing data of the sensor provided in the periphery of the production equipment to determine the occurrence of the impact and using the sensing data for the deep learning .

The method of driving the production equipment monitoring apparatus may further include the steps of acquiring a photographed image of an interval in which an event has occurred and storing the acquired photographed image and the additional information in the storage unit, And displaying the captured image on a screen based on the stored additional information.

According to the embodiment of the present invention, even if a problem occurs in a semiconductor manufacturing site, measures against mistakes and malfunctions can be quickly dealt with.

In addition, in the case of high-priced equipment, it is possible to prevent problems that can be controlled by personnel other than the authorized administrator, and to quickly cope with the problem of the process which may be caused by an external impact of the equipment.

1 shows a production equipment monitoring system according to an embodiment of the present invention,
2 is a diagram illustrating a production equipment monitoring system according to another embodiment of the present invention;
Figure 3 is a block diagram illustrating the detailed structure of the monitoring device of Figures 1 and 2,
4 is an exemplary view of a screen implemented in the display unit of FIG. 3, and FIG.
5 is a flowchart illustrating a driving process of a monitoring apparatus according to an embodiment of the present invention.

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

FIG. 1 is a diagram showing a production equipment monitoring system according to an embodiment of the present invention, and FIG. 2 is a diagram showing a production equipment monitoring system according to another embodiment of the present invention.

1 and 2, a production equipment monitoring system 90 according to an embodiment of the present invention includes a photographing apparatus 100, a user apparatus 105, a monitoring apparatus 110, a communication network 120, And includes some or all of device 130.

Here, "including some or all of" means that the communication network 120 is omitted and the monitoring apparatus 110 and the server apparatus 130 perform a direct (e.g., P2P) ) May be configured to be integrated into a network device in the communication network 120, and the like are described as including all of them in order to facilitate a sufficient understanding of the present invention.

According to the embodiment of the present invention, the production equipment 97 may include various equipment installed in the production line A. In particular, in the case of a semiconductor process, the production equipment 97 can be installed in a clean room. The production equipment 97 according to the embodiment of the present invention may include exposure equipment and ASML equipment. In the case of exposure equipment, it occupies the largest portion in the semiconductor process, and since it is expensive, it needs to be strictly controlled. Exposure equipment may refer to a chamber that simply applies a photoresist (hereinafter referred to as a photo) on a wafer, applies a mask, and then exposes it to light. However, an apparatus that performs all the above steps is referred to as an exposure apparatus can do. More precisely, it can be referred to as a lithography equipment. In other words, the process of applying a photo, applying a mask, exposing it to light, and etching are all performed.

ASML equipment, which occupies a 64% share of the lithography equipment industry, occupies a large portion of the semiconductor FAB (Fabrication), which may require monitoring monitoring specific to ASML equipment. Since it is an expensive equipment, problems may occur when the authorized administrator, that is, other than the authorized person of the access control, causes a problem in the process when an external shock is applied to the equipment. In the embodiment of the present invention, a surveillance monitoring system is constructed by combining surveillance monitoring and surveillance cameras together to supplement such portions.

According to the embodiment of the present invention, the production equipment 97 may be equipped with a sensor, more precisely a sensor module, on one side. The sensor module may transmit the sensed value sensed by the sensor to the monitoring device 110 by communication with the photographing device 100 or directly to the monitoring device 110. To this end, the sensor module may include a basic control unit such as a CPU and a memory, and may further include a communication module. Communication modules can be used for short-range communication modules such as Bluetooth, Wi-Fi and Zig-bee in consideration of cost. In addition, the sensor may include an impact sensor or a vibration sensor.

In the case of using the sensing signal of the sensor, the monitoring apparatus 110 may analyze the image of the photographing apparatus 100 to determine whether it is an external shock or utilize the sensing data to more accurately determine whether an external shock has occurred . For example, the monitoring device 110 may construct big data by using sensing data in addition to the photographed image for precise sensing, and may analyze big data through a deep learning operation of the AI. May be useful.

In addition, various sensors may be provided directly or indirectly to the production equipment 97. That is, the sensors may be installed directly or in the vicinity of the production equipment 97. In the embodiment of the present invention, a humidity sensor or a temperature sensor for sensing various environments inside the production line A, or a temperature sensor for sensing the environment inside the production line A in order to prevent the external environment from affecting the expensive production equipment 97, In order to measure it, an illuminometer such as a lux meter may be installed and used. Thereby preventing an accident through the state of an operator or an administrator who may indirectly influence the production equipment 97. These various data will be more useful for deep running.

Suppose, for example, that a licensor is allowed to access the production equipment 97, but occasionally there was an intermittent occurrence of an error in a particular environment. In this case, a screen for notifying the manager of the specific situation may be displayed on the screen of the production equipment 97, and the UX / UI may be supplemented. In other words, when a manager judges that he / she is drowsy or has a bad physical condition at a certain time, and when he / she judges that he / she approaches the production equipment (97) , And the connection of the equipment is cut off. Thus, an accident caused by a human body can be prevented in advance. Of course, in addition to preventing button manipulation, it may be possible to inform people around by notifying them through broadcast announcements.

Furthermore, the production equipment 97 may communicate with the monitoring device 110 as in FIG. For example, in the case of remote control from the user device 105, the control device 110 receives the control command and transmits the control command to the production equipment 97, and the production equipment 97 can be controlled by remote control will be.

The photographing apparatus 100 may include, for example, a CCTV camera or an IP camera, and may include a stationary camera and a PTZ camera capable of remote control. PTZ cameras can be pan, tilt and zoom functions. The photographing apparatus 100 according to the embodiment of the present invention can be installed in the vicinity of the production equipment 97 in the clean room installed in the production line A or in the semiconductor process to photograph the production equipment 97, for example.

The photographing apparatus 100 photographs the circumstance of the production equipment 97 and provides a photographed image to the monitoring apparatus 110. The photographing apparatus 100 may transmit the photographed image of the production equipment 97 to the monitoring apparatus 110 through wired / wireless communication. Of course, wired communication may be more preferable in order to secure a stable photographed image. The photographing apparatus 100 may include a communication module, and may analyze the photographed image during the transmission of the photographed image to provide analysis results. In this case, the photographing apparatus 100 can mount an image analysis module therein. The system design of the photographing apparatus 100 and the monitoring apparatus 110 may be variously performed according to the designer's intention.

In the embodiment of the present invention, two CCTV cameras are combined as the photographing apparatus 100, and two CCTV cameras can be supplied with power through POE (Power over Ethernet) of the monitoring apparatus 110 such as a PC, Is configured to be reproducible in the monitoring device (110).

When the sensing device 100 is set or designed to receive sensing data (or sensing signals) of various types of sensors provided in the production equipment 97 and provide the sensed data to the monitoring device 110, Sensor or sensor module.

2, the user accesses the monitoring device 110 through a user device 105 such as a wearable device or smart phone possessed by the user, such as Galaxy * language of the S company, That is, the screen of the photographing apparatus 100 can be controlled. The user device 105 may connect to the monitoring device 110 via a network such as the Internet protocol and thereby control the production equipment 97. For this, the user device 105 and the monitoring device 110 may be operated based on a KVM (Keyboard Video Mouse) system. Here, since the monitoring device 110 is an interface with the user, the KVM may be an interface by voice command, of course. The KVM system is installed in the monitoring apparatus 110, and the user apparatus 105 can perform a control operation using a program (e.g., an application) linked to the KVM system. Accordingly, the user device 105 can freely view and control the real time process screen anywhere. That is, the embodiments of the present invention create a direct virtual KVM environment or a KVM-based virtualization environment to create the same conditions as when the KVM external device is connected.

The monitoring device 110 may be connected to the production equipment 97 as shown in FIG. The monitoring device 110 may include a computer (PC) or a separately manufactured kiosk. Of course, the monitoring device 110 may also function as an image distributor. In other words, the photographed image is transmitted to the server device 130, and the control is performed in the user device 105 for remote control of the production equipment 97 or the like. Thus, the monitoring device 110 may be referred to as a distribution device or a relay device.

The monitoring device 110 performs a real-time monitoring operation in order to quickly cope with a mistake or a malfunction even if a problem occurs in a semiconductor manufacturing site. For example, the monitoring device 110 has a function of detecting when a singular point occurs in a monitoring screen of a semiconductor equipment process that is proceeding in real time, so that if the user can not monitor real time through screen recording, Replay the screen to show the screen again. Furthermore, the surroundings of the production equipment (97) are monitored in real time through photographed images provided by the CCTV cameras around the production equipment (97), and the time records in which the singular points of the production equipment (97) are generated through the CCTV camera recording function are matched , And can trace and analyze the cause.

The monitoring apparatus 110 receives a real-time image of the production equipment 97, that is, a photographed image, from the photographing apparatus 100 in a hardware (H / W) manner. In addition, a program capable of monitoring the production equipment 97, in particular the exposure equipment of ASML, is installed in the monitoring device 110 as software (S / W). In the embodiment of the present invention, it can be called an ASML monitoring program in that it is a program for monitoring equipment of a specific manufacturer. The monitoring device 110 may store and provide a program (or an application) capable of providing a user UX / UI to the user device 105 of FIG. 2 possessed by the manager or the like who manages the production process. In addition, the monitoring apparatus 110 may have a recording system computing function for receiving a photographed image, that is, a video signal, provided by the photographing apparatus 100 or the production equipment 97. In order to receive a video signal, the monitoring device 110 may include a video capture card. A variety of digital and analog signals can be processed according to the type of the video signal. In an embodiment of the present invention, a video capture card capable of receiving a digital signal can be provided.

In other words, the monitoring device 110 stores the video signal of the production equipment 97 of the ASML company in real time, and when the process progress screen changes due to another change from the normal screen, that is, a user or an unexpected event If it is detected, it can be regarded as an event and can be saved in the form of a list. To this end, the monitoring apparatus 110 may classify event events through a deep learning method using the accumulated data through an image search algorithm for analyzing image records to be stored. In addition, the user can quickly respond to the event occurrence by prioritizing and classifying the importance of the event. Recorded event notifications can be used to mark the time of the problem and quickly access the history. In addition, the monitoring device 110 may provide a desired screen by designating not only an event that occurred but also a time desired by the user.

For example, the monitoring device 110 may perform a data building operation and perform a search operation based on the data building operation. In other words, when the system is constructed, the monitoring apparatus 110 may use the sensing data of the sensor in the course of learning of the photographed image received from the photographing apparatus 100, And acquires a still image. In other words, since 30 still images are received in one second, a still image of a stable unit frame is obtained based on the analysis result through image analysis. It can be set as a reference image of the production equipment 97, that is, reference data. When the setting of the reference data is completed, the monitoring device 110 may compare the received reference image with the received image based on the set reference data to determine whether the production equipment 97 is abnormal. For example, when a specific person is photographed together, the analysis result of the person image can be compared with previously stored permission person information to determine whether or not the person is an authorized access right person. If an event is generated in the interval of a specific image in this process, a plurality of unit frame images corresponding to a predetermined time interval are obtained, and additional information such as date and time information is matched Can be stored. Accordingly, the manager can easily search for events generated in the corresponding date or time zone by simply inputting the date or time information when searching for a specific image. As described above, as the data is accumulated, the stored data is deeply run, thereby enabling more accurate and precise operation.

In addition, the monitoring device 110 directly captures and records the monitoring screen of the ASML production equipment 97. And outputs the output signal to the PC and the equipment monitor constituting the monitoring device 110, so that the function of the image distributor can be performed. The monitoring device 110 recognizes the screen and captures a specific change when it occurs, thereby providing an alarm function. In addition, it can perform digital video recording function to record images of CCTV cameras and the like. In addition to simply viewing the screen, KVM can be implemented to control the screen of the device with a keyboard mouse. You can view the record of the event through the alarm function that occurred, select partial time by selecting the desired time, edit the image, and you can also make it as educational sample.

In summary, the monitoring device 110 analyzes the image to be reproduced and provides an alarm function. By analyzing the video signal of the semiconductor equipment, the pattern of a specific motion is defined as an error or an event, and the user is notified so that the time when the uniqueness (or uniqueness) of the process screen of the equipment occurs can be known. In addition, the user will be able to accurately analyze the problem, prevent or educate the user by grasping the time that has occurred. It saves the pattern of the normal screen and informs the event as a case when the result is a sample and a movement other than the usual pattern occurs. Image processing algorithms are implemented to reduce the range of error, and screen images can be provided through the screen compression. When such image processing data is accumulated, it can be efficiently managed by the deep learning method.

The communication network 120 includes both wired and wireless communication networks. A wired / wireless Internet network may be used as the communication network 120 or may be interlocked. Here, the wired network includes an Internet network such as a cable network or a public switched telephone network (PSTN). The wireless communication network includes CDMA, WCDMA, GSM, Evolved Packet Core (EPC), Long Term Evolution (LTE), and Wibro network It is meant to include. Of course, the communication network 120 according to the embodiment of the present invention is not limited to this, but may be used as an access network of a next generation mobile communication system to be implemented in future, for example, a cloud computing network and a 5G network under a cloud computing environment. For example, when the communication network 120 is a wired communication network, the access point in the communication network 120 can access the exchange of the telephone exchange or the like. In case of a wireless communication network, the access point connects to the SGSN or GGSN (Gateway GPRS Support Node) Or may be connected to various repeaters such as Base Station Transmission (BTS), NodeB, and e-NodeB to process data.

The communication network 120 may include an access point. The access point includes a small base station such as a femto or pico base station, which is installed in a large number of buildings. Here, the femto or pico base station is classified according to the maximum number of the classification monitoring apparatuses 110 of the small base stations. Of course, the access point includes a monitoring device 110 and a short-range communication module for performing short-range communication such as ZigBee and Wi-Fi. The access point may use TCP / IP or RTSP (Real-Time Streaming Protocol) for wireless communication. In this case, the short-range communication is performed by various standards such as RF (Radio Frequency) and UWB (Ultra Wide Band) communication such as Bluetooth, Zigbee, IrDA, UHF and VHF . Accordingly, the access point can extract the location of the data packet, specify the best communication path to the extracted location, and forward the data packet to the next device, e.g., the server device 130, along the designated communication path. The access point can share a plurality of lines in a general network environment, and includes, for example, a router, a repeater, a repeater, and the like.

The server device 130 includes a DB 130a. In FIG. 2, the server device 130 may simply serve as a DVR (Digital Video Recorder). The server device 130 may classify and store captured images or video signals provided by the monitoring device 110, and may store various data in addition to the captured images. Also, the server device 130 may include a backup program for recording the monitoring program. The server device 130 can be configured on a Linux basis, which is more reliable than a Windows-based PC and is stable in operating the server. Of course, it is not limited specifically to Linux. In order to provide a more stable system according to an embodiment of the present invention, data may be stored in the DB 130a through the server device 130. [

Modernized production plants are being replaced by robots on behalf of humans. Over time, a monitoring system 90 will be needed in areas where it is hard for a person to physically control, and the monitoring system 90 will fill the deficit. It will be possible to improve the working environment more efficiently by installing it in a semiconductor production equipment which is not easy to access. It is not necessarily limited to making a specialized product of a particular company equipment. ASML company products, as well as equipment and control environments that require monitoring. Through the embodiments of the present invention, the video monitoring system 90 may be variously applied to the semiconductor industry. In addition, if semiconductor equipment companies fill in areas that they do not care about, they will become a major equipment investment product that can develop as a partnership with semiconductor equipment manufacturers to increase semiconductor production yield.

FIG. 3 is a block diagram illustrating the detailed structure of the monitoring apparatus of FIG. 1 and FIG. 2, and FIG. 4 is an exemplary view of a screen implemented in the display unit of FIG.

1 and 2 according to an embodiment of the present invention includes a communication interface unit 300, a control unit 310, a display unit 320, a storage unit 330, A device state analysis unit 340, and a state notification unit 350. [0041]

Here, "including some or all" means that some components such as the display unit 320 and the status notification unit 350 are omitted so that the monitoring apparatus 110 is configured or some configuration such as the equipment status analysis unit 340 Elements may be integrated and configured in other components such as the control unit 310, and the like are described as including all in order to facilitate a sufficient understanding of the invention.

The communication interface unit 300 communicates with the production equipment 97, the photographing apparatus 100, the user apparatus 105, and the server apparatus 130. In the case of FIG. 2, the communication interface unit 300 can communicate with each other through an intranet, which is an internal network. In this process, the communication interface unit 300 may perform various operations, and may typically perform operations such as compressing, transmitting, or restoring an image or video signal. The system will be constructed according to various specifications for each equipment and device.

The communication interface unit 300 may provide a service screen as shown in FIG. 4 so that the production equipment 97 can be remotely controlled when the user apparatus 105 is requested under the control of the control unit 310. Of course, if the user device 105 is a mobile device, a non-web-based application-based screen may be provided. Accordingly, if there is a remote control of the user, the control command can be transmitted to the production equipment 97 accordingly.

The communication interface unit 300 receives the photographed image provided from the photographing apparatus 100 and provides the photographed image to the control unit 310 and analyzes the photographed image and the photographed image received under the control of the control unit 310 To the server device 130. [0050]

The control unit 310 includes a communication interface unit 300, a display unit 320, a storage unit 330, an equipment status analysis unit 340, and a status notification unit 350 of the monitoring device 110 of FIG. And performs an overall control operation. For example, if there is a request for remote control from the user device 105, the control unit 310 provides a service screen for remote control. In addition, the control unit 310 provides the equipment state analysis unit 340 with a photographed image from the photographing apparatus 100 to determine whether an event has occurred. Of course, the control unit 310 may receive sensed data, which can determine an external impact of the production equipment 97, in addition to the photographed image, and provide the sensed data to the equipment state analysis unit 340.

The control unit 310 classifies various data provided by the equipment state analysis unit 340, for example, an image of an interval in which an event occurs, or an image edited by a request of the user device 105, Lt; / RTI > The data stored in the storage unit 330 may be periodically backed up and stored in the server unit 130.

In addition, the control unit 310 may display various data stored in the storage unit 330 on the screen of the display unit 320 in the format as shown in FIG. That is, if the equipment state analysis unit 340 generates data for screen display, the data is inserted into the designated format as shown in FIG. 4 and displayed on the screen. And, when the specific item displayed on the screen is selected by the user, the related image is reproduced on the screen.

In other words, when a manager or the like requests a specific operation from the screen of the display unit 320, the control unit 310 controls the display unit 320 to operate in response to the request, and when there is a request for a specific operation such as editing And may operate in conjunction with the equipment state analysis unit 340 to respond to the corresponding operation. Of course, since the operation in the display unit 320 can also be performed through the user apparatus 105, the control unit 310 will operate correspondingly thereto.

The display unit 320 may include a touch screen type monitor for interfacing with a user. The display unit 320 may display information on the screen in the form as shown in FIG. 4, and may be divided into a video playback region 400 and an event information display region 410. An icon display unit 401 may be displayed on one side of the image reproduction area 400. The icon display unit 401 displays various function icons for editing the image to be reproduced. In the event information display area 410, additional information 411 and 413 such as an image of a section in which an event occurs and a time of occurrence can be displayed. Therefore, when an administrator selects a specific event image, for example, the corresponding image can be reproduced in the image reproduction area 400. [

In other words, the display unit 320 may display a menu screen for performing operations according to an embodiment of the present invention. This enables various monitoring and control of the production equipment 97. If the control unit 310 determines that the user is not the access permission user of the monitoring device 110, the control unit 310 can perform the locking operation on the display screen displayed on the display unit 320. [ In other words, the touch of the screen will not react. Alternatively, in the case where the user is not the access permission authority, the status notification unit 350 may notify the user of the status, but may display a warning message on the screen of the display unit 320. Especially, as mentioned above, in case of a manager who frequently makes a mistake at a specific time even if a permission is granted, a variety of screen control methods can be applied to alert the user.

In addition, the display unit 320 may further perform user authentication to determine whether the user is a true authorized user. Accordingly, a screen displayed on the display unit 320 can perform an activation operation. In other words, the first screen is set to lock, and when a touch (eg, resistance or capacitive method) is detected, user authentication is requested through it. When the authentication is completed, the lock state is released according to the result, so that the touch is performed.

In the embodiment of the present invention, it is shown that an interface with a user is performed through a screen of a display unit 320 implemented by a touch screen method. However, it is possible to further use a keyboard, a mouse, etc. as a user interface part and further include a sound acquisition section such as a microphone for using a voice command as an operation command, a voice recognition section for recognizing the acquired voice, have. For example, user authentication may be performed by speech recognition.

The storage unit 330 may classify and store various data processed by the control unit 310 and may output the stored data at the request of the control unit 310. [ In addition, the storage unit 330 may further store user information to be used for authentication.

The equipment state analysis unit 340 analyzes the photographed image to determine whether an event has occurred in the production equipment 97, and can use the sensing data of the sensor at this time. Typically, unlicensed access to the production equipment (97) may be attempted to operate the equipment or to determine the impact of the production equipment (97). For example, the equipment state analysis unit 340 may determine whether the user is authorized to use the singularity information by searching for a user characteristic of the user accessing the production equipment 97, for example, a singularity in the image. Typically, face recognition can be used. It is determined whether or not the user is an authorized person through a deep learning operation for accurate face recognition. Although not shown separately in the drawing, the equipment status analyzer 340 may use an authentication method using biometrics such as fingerprint recognition, iris recognition, and the like. When the fingerprint recognition is used as a singular point, the information about the uniqueness of the authorized person can be stored and used.

In addition, the equipment state analysis unit 340 obtains a still image of the production equipment 97 in a normal state, that is, in a state in which the production equipment 97 is installed on the ground, and uses it as a reference image, It is possible to determine whether or not an external shock of the motor 97 is caused. Accordingly, an algorithm for measuring the tilt can be performed. In addition, in the course of this process, the equipment condition analyzing unit 340 can accurately determine whether an external impact has occurred by further using a sensor such as an impact sensor provided in the production equipment 97. It is also possible to judge whether or not the camera is impacted by using only sensing data instead of a method of comparing images. For example, when a shock occurs on a production line due to a fire or the like in a nearby building, the impact sensor may detect such an impact, and the equipment state analysis unit 340 may detect the impact of the production equipment 97 It is possible to judge whether or not it has an influence on the process or manufacture of the product. In other words, if the production equipment 97 is to be continuously operated in any situation, if the process conditions are changed by the impact, then the other process values are changed based on the changed process conditions. For example, if the tilting of the production equipment 97 occurs and the exposure to the wafer is shifted to one side, it will be typical that the wafer is further moved to one side so as to conform to it.

In addition, the equipment state analysis unit 340 can set priorities for various event occurrences and inform the state notification unit 350 or display specific markers on the screen. For example, as seen in FIG. 4, a particular item is to perform a blinking operation. Or you could mark the word "urgent" around the item. According to embodiments of the present invention, priorities can be set according to the magnitude of damage that has a significant impact on production. For example, if external impacts have a profound impact through deep running, this is the top priority. This priority may be automatically changed by the deep running operation. For example, if there is a sudden increase in the number of events caused by an administrator's operation mistake while the priority is reflected in the external impact, the priority will be placed in the manager's operation mistake. Thus, the priority setting by the deep learning is variable.

Of course, the equipment state analysis unit 340 may also provide a search function. In other words, when the administrator selects a specific time, event occurrence events in the corresponding time zone are retrieved as a list and displayed on the screen. The captured image of the section in which the event has occurred is acquired and stored together with the additional information so that the additional information such as the date and time can be used in the search.

In addition, the apparatus state analyzing unit 340 may perform various operations based on commands of a user or an administrator provided through the user apparatus 105 or the display unit 320, for example, . You can also save edited images and extract them separately and use them where you need them.

The state notification unit 350 may include a speaker for outputting sound or sound. It is possible to notify the unauthorized person when the unauthorized person accesses the production equipment (97), and to inform the outside of the production equipment (97) when various situations are detected.

Meanwhile, as another embodiment of the present invention, the control unit 310 of FIG. 3 may include a CPU and a memory. The CPU may include a control circuit for performing a control function, an operation unit (ALU) for computing information, and a command interpreting unit for interpreting the machine language. Of course, the instruction interpretation unit interprets the input binary bit information as information for performing an operation. The control circuit can be regarded as performing an operation according to the interpretation instruction of the instruction interpretation unit. The memory may include RAM (RAM). Accordingly, in the initial operation of the monitoring device 110, the CPU can copy the program stored in the equipment state analysis unit 340 to the memory, load the program, and execute the program. do.

5 is a flowchart illustrating a driving process of a monitoring apparatus according to an embodiment of the present invention.

1, a monitoring apparatus 110 according to an exemplary embodiment of the present invention includes a monitoring unit 110 for determining whether or not an event occurs, which includes at least an access to the authorized production equipment 97 and an occurrence of an impact The reference data is stored (S500). Here, the reference data may be a reference image, but the reference data for determining whether an impact has occurred may be bit information generated in a binary bit format. In case of using the impact sensor, binary bit information can be usefully used.

The monitoring device 110 receives and analyzes the photographed image of the production equipment 97 from the photographing device 100 installed in the vicinity of the production equipment 97 and generates an event based on the analysis result and the stored reference data (S510). Of course, when the sensor is constructed, the sensing data can be further received. The monitoring device 110 may periodically monitor the inclination of the production equipment 97 to detect whether an external impact has occurred or not. The characteristics of the user or the manager approaching the production equipment 97 may be determined, .

As mentioned above, the impact can be judged through sensing data, and the accuracy of measurement can be increased by combining them. If it is difficult to grasp whether or not the user is an authorized person through the photographed image, it is necessary to additionally request the user authentication in the locked state, and then release the lock when the authentication is completed. I will be able to do it

In addition to the above, various operations related to the monitoring device 110 may be possible.

While the present invention has been described in connection with what is presently considered to be the most practical and preferred embodiment, it is to be understood that the invention is not limited to the disclosed embodiments. That is, within the scope of the present invention, all of the components may be selectively coupled to one or more of them. In addition, although all of the components may be implemented as one independent hardware, some or all of the components may be selectively combined to perform a part or all of the functions in one or a plurality of hardware. As shown in FIG. The codes and code segments constituting the computer program may be easily deduced by those skilled in the art. Such a computer program may be stored in a non-transitory computer readable medium readable by a computer, readable and executed by a computer, thereby implementing an embodiment of the present invention.

Here, the non-transitory readable recording medium is not a medium for storing data for a short time such as a register, a cache, a memory, etc., but means a medium which semi-permanently stores data and can be read by a device . Specifically, the above-described programs can be stored in non-volatile readable recording media such as CD, DVD, hard disk, Blu-ray disk, USB, memory card, ROM, and the like.

While the invention has been described in connection with what is presently considered to be practical exemplary embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the present invention.

97: Production equipment 100: Photographic equipment
105: User device 110: Monitoring device
120: communication network 130: server device
300: communication interface unit 310:
320: display unit 330: storage unit
340: equipment state analysis unit 350: state notification unit

Claims (6)

1. A production equipment monitoring apparatus installed in a production line where semiconductor processing is performed and connected to a server apparatus for storing photographed images,
A storage unit for storing reference data for determining whether an event including at least an authorized party access permitted to operate the equipment to a designated production equipment installed in a clean room of the production line and an occurrence of an impact of the production equipment part; And
The control unit receives and analyzes an image of the production equipment from a photographing apparatus fixedly installed in the periphery of the production equipment in the clean room, analyzes the result of the analysis, And a control unit for determining whether or not an event has occurred,
Wherein,
Determining whether the production equipment is operated in order to determine whether or not the authorized person is approaching, and if it is difficult to grasp whether the authorized person is the authorized person through the photographed image, After the authentication is completed, the lock is released to allow the operation to be performed,
The method includes classifying event events by deep learning using analysis results of the captured images, setting a priority of importance of the event events in the classified event events to the reference data, Notifies the occurrence of the event but automatically changes the set priority by a deep learning operation,
When it is judged that the user is allowed to access the production equipment and the equipment is operated even when the equipment is operated at a specific time, Warnings A production equipment monitoring device that displays messages on the screen or disconnects equipment. The method according to claim 1,
And a communication interface for communicating with an external device that performs remote control of the production equipment,
Wherein the control unit controls the communication interface unit to notify the external device of the occurrence of an event when an event occurs. delete The method according to claim 1,
Wherein the control unit further uses sensing data of a sensor provided in the periphery of the production equipment to determine the occurrence of the impact and uses the sensing data for the deep running. The method according to claim 1,
The controller acquires the captured image of the section in which the event has occurred and stores the acquired captured image and the additional information in the storage unit and stores the time information provided by the user in searching for the captured image and the stored additional information And a production equipment monitoring device for displaying the acquired photographed image on the screen on the basis of the obtained image. A method of driving a production equipment monitoring apparatus, which is installed in a production line on which a semiconductor process is performed, is connected to a server apparatus for storing photographed images, and includes a storage unit and a control unit,
Storing the reference data for determining whether or not an event including at least an authorized party access permitted to operate the equipment to a designated production equipment installed in a clean room of the production line and an occurrence of an impact of the production equipment is stored ; And
Wherein the controller receives and analyzes an image of the production equipment from a photographing apparatus fixedly installed in the periphery of the production equipment in the clean room and analyzes the photographed image of the production equipment based on the analysis result and the stored reference data, And determining whether or not the user-
Wherein,
Determining whether the production equipment is operated in order to determine whether or not the authorized person is approaching, and if it is difficult to grasp whether the authorized person is the authorized person through the photographed image, After the authentication is completed, the lock is released to allow the operation to be performed,
The method according to claim 1, further comprising: classifying the event event by deep learning using the analysis result of the captured image, setting a priority of the importance of the event event to the reference data, The priority is automatically changed by the deep learning operation,
When it is judged that the user is allowed to access the production equipment and the equipment is operated even when the equipment is operated at a specific time, A method of driving a production equipment monitoring device that displays a warning message on the screen or disconnects the equipment.

Images