Abstract: A method for predictive maintenance through automatic detection of a pump anomaly is provided. The method according to some embodiments may include receiving a plurality of sensing values of two or more categories, among a plurality of categories, from a plurality of sensors provided in a first pump; inputting a feature representing each of sensing values, selected among the plurality of sensing values, to a first anomaly prediction model that is machine-learned in advance; determining whether a future anomaly of the first pump is predicted to occur, by using data output from the first anomaly prediction model; and providing alarm information based on a determination that the future anomaly is predicted to occur, wherein the first pump belongs to a first pump model group, among a plurality of pump model groups, matched with the first anomaly prediction model.

Abstract: A display device according to an embodiment includes a substrate including a main area including a display area and an inorganic encapsulation area surrounding the display area, and a sub-area extending from the main area, a circuit layer including a semiconductor layer, conductive layers, and insulating layers disposed between the semiconductive layer and the conductive layers, respectively, a light emitting element layer disposed on the circuit layer and including a light emitting element disposed in the display area, and an encapsulation layer covering the light emitting element layer in the display area and including inorganic encapsulation films and an organic encapsulation film disposed between the inorganic encapsulation films, wherein the circuit layer includes a first line extending from the sub-area to the display area through the inorganic encapsulation area, and the first line is covered by an inorganic insulating layer which constitutes the circuit layer.

Abstract: An organic light-emitting display device comprises a first thin-film transistor disposed on a substrate; and a second thin-film transistor disposed on the substrate and spaced apart from the first thin-film transistor. The first thin-film transistor comprises a first semiconductor layer, a first conductive layer disposed on the first semiconductor layer and that overlaps the first semiconductor layer, and a first insulating layer disposed between the first semiconductor layer and the first conductive layer. The second thin-film transistor comprises a second semiconductor layer, and a second conductive layer disposed on the second semiconductor layer and that overlaps the second semiconductor layer.

Abstract: A display device includes a plurality of first constant voltage lines extended along a first direction and arranged along a second direction crossing the first direction, a plurality of second constant voltage lines extended along the second direction and arranged along the first direction, a first overlap area in which the plurality of first constant voltage lines and the plurality of second constant voltage lines overlap and contact each other, and a second overlap area in which the plurality of first constant voltage lines and the plurality of second constant voltage lines overlap and are insulated from each other.

Abstract: An organic light-emitting display device comprises a first thin-film transistor disposed on a substrate; and a second thin-film transistor disposed on the substrate and spaced apart from the first thin-film transistor. The first thin-film transistor comprises a first semiconductor layer, a first conductive layer disposed on the first semiconductor layer and that overlaps the first semiconductor layer, and a first insulating layer disposed between the first semiconductor layer and the first conductive layer. The second thin-film transistor comprises a second semiconductor layer, and a second conductive layer disposed on the second semiconductor layer and that overlaps the second semiconductor layer.

Abstract: A multi-machine and performance based continuous production planning global optimization scheduling method and device are disclosed. The device for multi-machine and performance-based continuous production planning global optimization scheduling may include a data pre-processing module configured to generate a multi-machine list on performable multiple machines through pre-processing scheduling target data, a scheduling module configured to generate a scheduling result by setting a processing time for a machine related to a single machine list, wherein the single machine list is selected from the multi-machine list, and a data post-processing module configured to store the scheduling result in a database.

Abstract: Disclosed herein is an image processing method and an image processing system for deep learning. The image processing method includes converting image data including at least one figure image into a vector image by changing a data format, encrypting at least one first attribute value for a position of the figure image in the vector image according to a selected encryption scheme, constructing a de-identified image for the image data by using a second attribute value which is changed by the encryption, and transferring the de-identified image to a cloud server in which a deep learning model is managed and processing deep learning operations of the de-identified image by the cloud server.

Abstract: The present invention relates to a substrate heating apparatus. More specifically, the present invention relates to a substrate heating apparatus including a first heating element located in an inner region of the substrate heating apparatus, a second heating element located in an outer region, and a third heating element supplying current to the second heating element passing through the inner region, wherein the diameter of a wire constituting the third heating element is thicker than the diameter of a wire constituting the second heating element, thereby inhibiting the generation of an overheating region by the heating of the third heating element.

Abstract: Disclosed herein is an image processing method and an image processing system for deep learning. The image processing method includes converting image data including at least one figure image into a vector image by changing a data format, encrypting at least one first attribute value for a position of the figure image in the vector image according to a selected encryption scheme, constructing a de-identified image for the image data by using a second attribute value which is changed by the encryption, and transferring the de-identified image to a cloud server in which a deep learning model is managed and processing deep learning operations of the de-identified image by the cloud server.

Abstract: A method of load balancing by multiple cores in a multi-core-based load balancing apparatus comparing arriving packets with a signature is provided, and comprises first load-balancing first packets arriving on the multiple cores during a first period based on an arrival rate of the first packets, identifying a signature for the comparison, analyzing the first packets, determining at least one service type of the first packets, estimating a mean deep packet inspection (DPI) time corresponding to the determined at least one service type of the first packets, generating a load balancing rule using the estimated average DPI time, and second load-balancing second packets arriving on the multiple cores during a second period using the generated load balancing rule.

Abstract: In the present invention, when a circular lid body is opened or closed by a hinge-coupling structure at an eccentric location out of a circle, the circular lid body can be opened or closed without interference between an edge protrusion part of the circular lid body and a circular edge support frame of a support frame body. Therefore, the present invention is useful in that the present invention has overcome innumerable interferences on arcs through a curve insertion scheme of a circular structure, that is, a curve insertion scheme by a rotational inclination angle ? of the edge protrusion part and an operation model view of an S-shaped hinge curve part.

Abstract: The present invention relates to a substrate heating apparatus. More specifically, the present invention relates to a substrate heating apparatus including a first heating element located in an inner region of the substrate heating apparatus, a second heating element located in an outer region, and a third heating element supplying current to the second heating element passing through the inner region, wherein the diameter of a wire constituting the third heating element is thicker than the diameter of a wire constituting the second heating element, thereby inhibiting the generation of an overheating region by the heating of the third heating element.

Abstract: A method of load balancing by multiple cores in a multi-core-based load balancing apparatus comparing arriving packets with a signature is provided, and comprises first load-balancing first packets arriving on the multiple cores during a first period based on an arrival rate of the first packets, identifying a signature for the comparison, analyzing the first packets, determining at least one service type of the first packets, estimating a mean deep packet inspection (DPI) time corresponding to the determined at least one service type of the first packets, generating a load balancing rule using the estimated average DPI time, and second load-balancing second packets arriving on the multiple cores during a second period using the generated load balancing rule.

Abstract: Provided is a refrigerator including: a storage compartment in which a plurality of accommodation spaces are disposed; a detection unit that detects a storage container accommodated in the plurality of accommodation spaces; a display unit that displays accommodation information regarding the plurality of accommodation spaces; and a controller that controls the display unit so as to display the accommodation spaces in which a new storage container may be accommodated, based on a result of detection of the detection unit.

Abstract: A character string pattern matching method for detecting the presence of at least one of N (N is a natural number equal to or greater than 2) patterns in specific text shifts a detection location across text by a maximum shift length using single-byte character-based layered SHIFT tables, thereby increasing a pattern matching speed as compared with the prior art pattern matching algorithms.

Abstract: A semiconductor laser diode capable of improving reliability and mass-productivity is disclosed. The semiconductor laser diode comprises a first clad layer; a first optical guide layer disposed on the first clad layer; an active layer disposed on the first optical guide layer; a second optical guide layer disposed on the active layer; and a second clad layer disposed on the second optical guide layer, having a greater band gap energy than the second optical guide layer, the band gap energy decreasing as being farther from the second optical guide layer.

Abstract: A character string pattern matching method for detecting the presence of at least one of N (N is a natural number equal to or greater than 2) patterns in specific text shifts a detection location across text by a maximum shift length using single-byte character-based layered SHIFT tables, thereby increasing a pattern matching speed as compared with the prior art pattern matching algorithms.

Abstract: A semiconductor laser diode having a graded interlayer is provided. The semiconductor laser diode has the graded interlayer between an active layer composed of InGaN and an electron blocking layer (EBL) composed of AlGaN. The graded interlayer is composed of InxAlyGa1-x-yN(0?x?0.2, 0?y?0.5) and is formed by grading a composition of group III materials. Accordingly, the active layer and the p-EBL have a reduced difference in rigidities and lattice parameters, and an abrupt gradient of an energy band and generation of a strain can be avoided in an interface between the active layer and the p-EBL. Since a crack can be prevented from being generated along the interface between the active layer and the p-EBL when a cleavage facet is formed, characteristics of the semiconductor laser diode can be improved.

Abstract: A semiconductor laser diode capable of improving reliability and mass-productivity is disclosed. The semiconductor laser diode comprises a first clad layer; a first optical guide layer disposed on the first clad layer; an active layer disposed on the first optical guide layer; a second optical guide layer disposed on the active layer; and a second clad layer disposed on the second optical guide layer, having a greater band gap energy than the second optical guide layer, the band gap energy decreasing as being farther from the second optical guide layer.

Abstract: A semiconductor laser diode having a graded interlayer is provided. The semiconductor laser diode has the graded interlayer between an active layer composed of InGaN and an electron blocking layer (EBL) composed of AlGaN. The graded interlayer is composed of InxAlyGa1-x-yN(0?x?0.2, 0?y?0.5) and is formed by grading a composition of group III materials. Accordingly, the active layer and the p-EBL have a reduced difference in rigidities and lattice parameters, and an abrupt gradient of an energy band and generation of a strain can be avoided in an interface between the active layer and the p-EBL.