KR102698821B1 - Aircraft health management method and apparatus thereof - Google Patents

Abstract

The present invention provides an aircraft health management method including a process of collecting aircraft operation data from an aircraft in operation, a process of analyzing the aircraft operation data to select abnormal data, and a process of managing risk factors using the abnormal data, and an aircraft health management device applied thereto, wherein an aircraft health management method and device are provided that can stably manage the health of an aircraft by predictively evaluating risk factors of the aircraft for safe operation of the aircraft.

Description

{AIRCRAFT HEALTH MANAGEMENT METHOD AND APPARATUS THEREOF}

The present invention relates to an aircraft health management method and device, and more specifically, to an aircraft health management method and device capable of stably managing the health of an aircraft by predictively evaluating risk factors of the aircraft.

The main causes of aircraft accidents are aircraft defects, human factors, and natural disasters. In order to prevent aircraft accidents caused by aircraft defects, aircraft must maintain airworthiness.

In the past, regular maintenance work was performed on aircraft before and after the aircraft was put into service, according to maintenance programs approved by the airworthiness authorities, and through this, all systems of the aircraft were strictly managed for damage. This is called preventive health management.

Meanwhile, according to the cause analysis method of aircraft accidents, major accidents are more often caused by simultaneous or consecutive small combinations that do not directly affect airworthiness rather than by a single large defect.

However, there is a limit to finding all defects in an aircraft using a preventive health management method. In addition, there is a problem in which defects that did not occur during maintenance work occur during operation. Therefore, there is a problem in which the aircraft cannot strictly maintain airworthiness during operation.

The technology that serves as the background for the present invention is published in the following patent documents.

KR 10-2020-0008550 A

The present invention provides an aircraft health management method and device capable of predictively evaluating risk factors of an aircraft for safe operation of the aircraft and stably managing the health of the aircraft.

An aircraft health management method according to an embodiment of the present invention includes: a process of collecting aircraft operation data from an aircraft in operation; a process of analyzing the aircraft operation data to select abnormal data; and a process of managing risk factors using the abnormal data.

The process of collecting the above aircraft operation data may include a process of collecting status data from the aircraft; and a process of collecting environmental data from the aircraft.

The process of collecting the above aircraft operation data may further include a process of collecting destruction data from the aircraft.

The process of selecting the above abnormal data may include a process of dividing the aircraft operation data into aircraft operation data in the form of images and aircraft operation data in the form of messages; and a process of analyzing the divided aircraft operation data in different ways according to each form.

The process of dividing the above aircraft operation data into aircraft operation data in image form and aircraft operation data in message form may include a process of dividing the status data into aircraft operation data in image form and aircraft operation data in message form according to a data format of the status data; and a process of dividing the environment data into aircraft operation data in image form and aircraft operation data in message form according to a data format of the environment data.

The process of analyzing the above-mentioned classified aircraft operation data in different ways according to its form may include a process of analyzing the aircraft operation data in the form of an image as it is to determine whether there is an abnormal pattern; and a process of verifying and then analyzing the aircraft operation data in the form of a message to determine whether there is an abnormal pattern.

The process of directly analyzing the aircraft operation data in the form of an image and determining whether there is an abnormal pattern may include: a process of detecting an abnormal pattern from the aircraft operation data in the form of an image using pre-learned artificial intelligence; a process of determining aircraft operation data in the form of an image in which an abnormal pattern is detected as abnormal and labeling it with an abnormal-image label; and a process of determining aircraft operation data in the form of an image in which an abnormal pattern is not detected as normal and labeling it with a normal-image label.

The process of analyzing the aircraft operation data in the form of the above message to determine whether there is an abnormal pattern after verifying it may include: a process of performing a security integrity verification on the aircraft operation data in the form of the above message to determine the aircraft operation data in the form of the message in which no security integrity defect occurs as verified aircraft operation data in the form of the message; a process of detecting an abnormal pattern from the verified aircraft operation data in the form of the message using pre-learned artificial intelligence; a process of determining the aircraft operation data in the form of the message in which an abnormal pattern is detected as abnormal and labeling it with an abnormal-message label; and a process of determining the aircraft operation data in the form of the message in which no abnormal pattern is detected as normal and labeling it with a normal-message label.

The process of selecting the above abnormal data may include a process of selecting the above destruction data as abnormal data.

The process of managing risk factors using the above abnormal data may include a process of creating the above abnormal data as a file; and a process of outputting the created file to an external maintenance system of the aircraft to create a management history database.

An aircraft health management device according to an embodiment of the present invention is an aircraft health management device for managing the health of an aircraft in operation in real time, the device including: a collection unit installed in the aircraft and collecting aircraft operation data from the aircraft; a selection unit connected to the collection unit and analyzing the aircraft operation data to select abnormal data; and a management unit connected to the selection unit and using the abnormal data to manage risk factors for the aircraft.

The above collection unit may include a first collector that collects status data from the aircraft as the aircraft operation data; a second collector that collects environmental data from the aircraft as the aircraft operation data; and a third collector that collects destruction data from the aircraft as the aircraft operation data.

The above selection unit may include a data classifier that classifies the status data and the environment data into aircraft operation data in the form of an image and aircraft operation data in the form of a message; an image classifier that analyzes the aircraft operation data in the form of an image using pre-learned artificial intelligence to determine whether there is an abnormal pattern; a message verifier that verifies the aircraft operation data in the form of a message; and a message discriminator that analyzes the aircraft operation data in the form of a verified message using pre-learned artificial intelligence to determine whether there is an abnormal pattern.

The above selection unit may further include a data adder for adding the destruction data to the abnormal data.

The above management unit may include a file generator for generating the abnormal data as a file; and an output device for outputting a file of the abnormal data to an external maintenance system of the aircraft.

According to an embodiment of the present invention, aircraft operation data can be collected from an aircraft in operation, abnormal data can be selected by analyzing the collected aircraft operation data, and risk factors can be managed using the abnormal data. Therefore, risk factors of the aircraft can be evaluated predictively in an automated manner, and the health of the aircraft can be stably managed. From this, safe operation of the aircraft can be guaranteed.

Figure 1 is a conceptual diagram of an aircraft health management device according to an embodiment of the present invention.
Figure 2 is a flowchart of an aircraft health management method according to an embodiment of the present invention.

Hereinafter, with reference to the attached drawings, embodiments of the present invention will be described in detail. However, the present invention is not limited to the embodiments disclosed below, and may be implemented in various different forms. The embodiments of the present invention are provided only to ensure that the disclosure of the present invention is complete, and to fully inform those skilled in the art of the scope of the invention. In order to explain the embodiments of the present invention, the drawings may be exaggerated, and the same reference numerals in the drawings represent the same elements.

The present invention relates to an aircraft health management method and device. Hereinafter, an embodiment of the present invention will be described in detail by exemplifying a case where the aircraft health management method and device are applied to predictive health management of an aircraft. Of course, the aircraft health management method and device according to an embodiment of the present invention can also be applied to predictive health management of various types of vehicles. For example, the aircraft health management method and device according to an embodiment of the present invention can also be applied to predictive health management of a ship.

Before explaining an aircraft health management method according to an embodiment of the present invention, an aircraft health management device according to an embodiment of the present invention will be explained.

Figure 1 is a conceptual diagram of an aircraft health management device according to an embodiment of the present invention.

Referring to FIG. 1, an aircraft health management device according to an embodiment of the present invention is for managing the health of an aircraft in operation in real time, and includes a collection unit (100) installed in an aircraft (10) for collecting aircraft operation data from the aircraft (10), a selection unit (200) connected to the collection unit (100) for analyzing the aircraft operation data and selecting abnormal data, and a management unit (300) connected to the selection unit (200) for managing risk factors for the aircraft (10) using the abnormal data.

An aircraft in operation may mean an aircraft from the time the aircraft first begins to move for the purpose of takeoff until the time the flight ends and the aircraft finally comes to a stop, after which the aircraft's engines are stopped.

In other words, an aircraft in operation can mean an aircraft from the time it is maintained according to the maintenance work schedule for preventive maintenance before takeoff to the time it is maintained according to the maintenance work schedule for preventive maintenance after landing.

The collection unit (100) serves to collect aircraft operation data from the aircraft (10). The collection unit (100) can be installed in the aircraft. The collection unit (100) can include a first collector that collects status data from the aircraft (10) as aircraft operation data, a second collector that collects environmental data from the aircraft (10) as aircraft operation data, and a third collector that collects destruction data from the aircraft (10) as aircraft operation data.

Status data represents the status of the aircraft (10) and may include data on attitude, movement, flight speed, etc. In addition, environmental data represents the environment of the aircraft (10) and may include data on altitude, internal and external air pressure, wind speed, humidity, temperature, etc. of the aircraft (10). In addition, destruction data may include change data on the structure of components of the aircraft (10) (also referred to as damage data).

The components of the aircraft (10) may include, for example, the airframe of the aircraft (10) and various mechanical devices and aviation equipment mounted thereon. The airframe of the aircraft (10) may include a fuselage, wings, etc., and the mechanical devices may include landing gear, propulsion devices, etc. The aviation equipment may be system equipment constituting each system of the aircraft (10). Meanwhile, since the aircraft (10) has a fail-safe structure applied as a type of safety design, the ongoing flight mission can be maintained even if the structure of the components of the aircraft (10) changes.

The collection unit (100) may include various sensors for collecting aircraft operation data from the aircraft (10). The various sensors may collect status data, environmental data, and destruction data in the form of images or messages.

The selection unit (200) analyzes aircraft operation data to select abnormal data. The selection unit (200) may be connected to the collection unit (100). In addition, the selection unit (200) may be equipped with pre-learned artificial intelligence, such as a pre-learned machine learning model. The selection unit (200) may include a data classifier that divides status data and environmental data into aircraft operation data in the form of images and aircraft operation data in the form of messages, an image discriminator that analyzes aircraft operation data in the form of images using pre-learned artificial intelligence to determine whether there is an abnormal pattern, a message verifier that verifies aircraft operation data in the form of messages, and a message discriminator that analyzes aircraft operation data in the form of verified messages using pre-learned artificial intelligence to determine whether there is an abnormal pattern. In addition, the selection unit (200) may further include a data adder that adds destruction data to the abnormal data.

The management unit (300) manages risk factors for the aircraft (10) using abnormal data. The management unit (300) may be connected to the selection unit (200) and the maintenance system (30) of the control center (20). To this end, the management unit (300) may include a file generator that generates abnormal data as a file, and an output device that outputs the file of abnormal data to the maintenance system (30) outside the aircraft (10).

Meanwhile, in order to avoid duplication of explanation, the specific operations of the collection unit (100), the selection unit (200), and the management unit (300) will be described together with the description of the aircraft health management method according to the embodiment of the present invention. Meanwhile, the aircraft health management method according to the embodiment of the present invention may also be referred to as an artificial intelligence algorithm-based predictive aircraft health management method.

Hereinafter, with reference to FIGS. 1 and 2, an aircraft health management method according to an embodiment of the present invention will be described.

Referring to FIGS. 1 and 2, an aircraft health management method according to an embodiment of the present invention includes a process of collecting aircraft operation data from an aircraft in operation (S100), a process of analyzing the aircraft operation data to select abnormal data (S200), and a process of managing risk factors using the abnormal data (S300).

At this time, the aircraft health management method can be performed by the aforementioned health management device according to the embodiment of the present invention. Therefore, the following will explain an embodiment by exemplifying that the aircraft health management method is performed by the aircraft health management device according to the embodiment of the present invention. However, the invention is not limited thereto, and the aircraft health management method may be performed by health management devices of various configurations. That is, the contents described below can be similarly or identically applied even when the aircraft health management method is performed by health management devices of various configurations.

First, in the collection unit (100), aircraft operation data is collected from an aircraft in operation (S100). More specifically, the aircraft operation data may include status data, environmental data, and destruction data. In addition, the collection unit (100) may collect status data from the aircraft (10) and environmental data from the aircraft (10). In addition, destruction data may be collected from the aircraft (10).

In collecting status data, various sensors equipped in the collection unit (100) can be used to detect the status of the aircraft (10), such as attitude, motion, and flight speed, and the detection results can be collected as status data of the aircraft (10), such as attitude data, motion data, and flight speed data. At this time, the attitude of the aircraft (10) can include a rolling angle, a yaw angle, a pitching angle, and the like, and the motion can include takeoff, cruising, ascent, descent, acceleration, deceleration, turning, and landing. In collecting environmental data, various sensors equipped in the collection unit (10) can be used to detect the environment of the aircraft (10), such as altitude, air pressure inside and outside the aircraft (10), wind speed, humidity, and temperature, and the results can be collected as environmental data of the aircraft (10), such as altitude data, air pressure data, wind speed data, humidity data, and temperature data. In collecting destruction data, each component of the aircraft (10) can be monitored using various sensors provided in the collection unit (100), and the monitoring results of the area where deformation occurs can be collected as destruction data. At this time, each data collected by the collection unit (100) can include data in the form of images and data in the form of messages.

Next, the aircraft operation data is analyzed in the selection unit (200) to select abnormal data (S200). That is, the aircraft operation data can be divided into aircraft operation data in the form of images and aircraft operation data in the form of messages, and the divided aircraft operation data can be analyzed in different ways according to each form.

At this time, depending on the data format of the status data, the status data can be divided into aircraft operation data in the form of images and aircraft operation data in the form of messages. In addition, depending on the data format of the environmental data, the environmental data can be divided into aircraft operation data in the form of images and aircraft operation data in the form of messages.

At this time, the data format may include the CPL (Continuous Parameter logging) format and the QAR (Quick Access Recorder) format. The CPL format is created based on the ARINC 767 data format, which improves the data storage constraints of ARINC 717. Compared to ARINC 717, it can store various data such as voice and image, and can process large amounts of data by supporting 32 bits. The QAR format is created based on the ARINC 717 data format, and can be composed of flight storage data defined in the Flight Operational Quality Assurance (FOQA) program (storing approximately 1,500 parameters) and additional data defined by the airline (up to 2,700 variables can be added).

That is, the data in the CPL format among the status data and the data in the CPL format among the environmental data can be distinguished as aircraft operation data in the form of images. And the data in the QAR format among the status data and the data in the QAR format among the environmental data can be distinguished as aircraft operation data in the form of messages. Of course, the types of data formats applied to aircraft operation data can be diverse. That is, the CPL (Continuous Parameter logging) format and the QAR (Quick Access Recorder) format are examples for illustratively explaining the data format of environmental data, and are not intended to limit the types.

In addition, when analyzing the classified aircraft operation data in different ways according to their types, the presence or absence of an abnormal pattern can be determined by directly analyzing the aircraft operation data in the form of an image. To this end, by using pre-learned artificial intelligence, an abnormal pattern can be detected from the aircraft operation data in the form of an image, the aircraft operation data in the form of an image in which an abnormal pattern is detected can be determined as abnormal, and an abnormal-image label can be labeled, and the aircraft operation data in the form of an image in which an abnormal pattern is not detected can be determined as normal, and a normal-image label can be labeled.

In addition, in analyzing the classified aircraft operation data in different ways according to their types, the presence or absence of an abnormal pattern can be determined by verifying and analyzing the message-type aircraft operation data. More specifically, the security integrity verification can be performed on the message-type aircraft operation data, so that the message-type aircraft operation data in which no security integrity defect occurs can be determined as verified message-type aircraft operation data, and an abnormal pattern can be detected from the verified message-type aircraft operation data using pre-learned artificial intelligence, and the message-type aircraft operation data in which an abnormal pattern is detected can be determined as abnormal and labeled with an abnormal-message label, and the message-type aircraft operation data in which no abnormal pattern is detected can be determined as normal and labeled with a normal-message label.

Meanwhile, in selecting abnormal data, destruction data can be selected as abnormal data. For example, an aircraft may have a failsafe structure in an important part of the aircraft. At this time, each failsafe structure can be monitored by a sensor that can monitor each change. In addition, each sensor can detect a change when a change occurs in each failsafe structure and generate destruction data. Accordingly, the generation of destruction data means that a change has occurred in the corresponding failsafe structure of the aircraft. At this time, a change in the failsafe structure means that the failsafe structure is damaged. Therefore, destruction data can be selected as abnormal data as soon as it occurs without the process of determination.

In analyzing aircraft operation data in the sorting unit (200) to sort out abnormal data, the sorting unit (200) may store a history of failures that occurred in the corresponding aircraft (10) for a certain period of time. Here, the history of failures may be a history of failures of the aircraft (10) collected in advance by performing preventive health management for the aircraft (10). In addition, the characteristics of the history of failures may be diagrammed using artificial intelligence installed in the sorting unit (300), and this may be utilized to sort out abnormal data. In addition, major factors and weights for each factor may be derived from each data sorted out as abnormal data, and the correlation between these and defects that may occur in the aircraft may be analyzed. In addition, the degree of failure risk of the aircraft part related to the corresponding data may be analyzed according to the analyzed correlation. The analyzed results may be labeled together with the corresponding data and utilized for predictive health management of the aircraft. Accordingly, high-quality data may be generated.

Next, the management unit (300) manages risk factors using abnormal data (S300). For example, in managing risk factors using abnormal data, the abnormal data can be generated as a file. Here, the file can have various formats as long as it is in a format that can be viewed by an operator at the control center. In addition, the generated file can be output to the outside of the aircraft (10), for example, the maintenance system (30) of the control center (20), to generate a management history database, and a predictive health management database server can be constructed therefrom. Accordingly, in an embodiment of the present invention, a predictive health management system can be constructed that includes the aforementioned aircraft health management device according to an embodiment of the present invention and a predictive health management database server in which the management history database generated therefrom is stored. Therefore, by utilizing the predictive health management system according to an embodiment of the present invention, even while the aircraft is in operation, an operator, such as an aircraft maintenance engineer, can open the management history database in real time to check whether an abnormal data file is generated in the aircraft in operation, and thereby predictively manage the health of the aircraft.

The above embodiments of the present invention are for the purpose of explaining the present invention, and are not intended to limit the present invention. It should be noted that the configurations and methods disclosed in the above embodiments of the present invention may be combined or crossed with each other and combined and modified in various forms, and that modifications made thereby may also be considered within the scope of the present invention. That is, the present invention may be implemented in various different forms within the scope of the claims and equivalent technical ideas thereof, and those skilled in the technical field to which the present invention pertains will be able to understand that various embodiments are possible within the scope of the technical ideas of the present invention.

100: Collection Department
200: Selection Department
300: Management Department

Claims (15)

The process of collecting aircraft operational data from aircraft in operation;
The process of analyzing the above aircraft operation data and selecting abnormal data; and
A process for managing risk factors using the above abnormal data; including;
The process of collecting the above aircraft operation data is as follows:
The process of collecting status data from the above aircraft; and
A process for collecting environmental data from the above aircraft;
The process of selecting the above abnormal data is:
The process of dividing the above aircraft operation data into aircraft operation data in image form and aircraft operation data in message form; and
An aircraft health management method comprising: a process of analyzing the separated aircraft operating data in different ways according to their types; delete In claim 1,
The process of collecting the above aircraft operation data is as follows:
A method for managing aircraft health, further comprising: a process for collecting destruction data from said aircraft. delete In claim 1 or claim 3,
The process of dividing the above aircraft operation data into aircraft operation data in image form and aircraft operation data in message form is as follows.
A process of dividing the above status data into aircraft operation data in image form and aircraft operation data in message form according to the data format of the above status data; and
An aircraft health management method, comprising: a process of dividing the environmental data into aircraft operation data in image form and aircraft operation data in message form according to the data format of the environmental data. In claim 5,
The process of analyzing the above-mentioned aircraft operation data in different ways according to its type is as follows:
The process of analyzing the aircraft operation data in the form of the above video to determine whether there is an abnormal pattern; and
An aircraft health management method, comprising: a process of verifying and analyzing aircraft operation data in the form of the above message to determine whether there is an abnormal pattern; In claim 6,
The process of analyzing the aircraft operation data in the form of the above video to determine whether there is an abnormal pattern is as follows:
A process of detecting abnormal patterns from aircraft operation data in the form of images using pre-learned artificial intelligence;
The process of determining aircraft operation data in the form of images in which abnormal patterns are detected as abnormal and labeling the abnormal-image label; and
An aircraft health management method, comprising: a process of determining aircraft operation data in the form of images in which no abnormal patterns are detected as normal and labeling a normal-image label; In claim 6,
The process of verifying and analyzing aircraft operation data in the form of the above message to determine whether there are any abnormal patterns is as follows:
A process of performing security integrity verification on aircraft operation data in the form of the above message to determine aircraft operation data in the form of a message in which no security integrity defect occurs as aircraft operation data in the form of a verified message;
The process of detecting abnormal patterns from aircraft operation data in the form of verified messages using pre-learned artificial intelligence;
The process of determining aircraft operation data in the form of messages in which abnormal patterns are detected as abnormal and labeling them with abnormal message labels; and
An aircraft health management method, comprising: a process of determining aircraft operational data in the form of messages in which no abnormal patterns are detected as normal and labeling them with a normal-message label; In claim 3,
The process of selecting the above abnormal data is:
An aircraft health management method, comprising: a process of selecting the above destruction data as abnormal data. In claim 1 or claim 3,
The process of managing risk factors using the above abnormal data is as follows:
The process of creating the above abnormal data as a file; and
An aircraft health management method, comprising: a process of generating a management history database by outputting the generated file to an external maintenance system of the aircraft; As an aircraft health management device for managing the health of an aircraft in operation in real time,
A collection unit installed in the above aircraft and collecting aircraft operation data from the above aircraft;
A selection unit connected to the above collection unit and analyzing the aircraft operation data to select abnormal data; and
A management unit connected to the above selection unit and managing risk factors for the aircraft using the above abnormal data;
The above collection unit,
A first collector for collecting status data from the aircraft as the aircraft operation data;
A second collector for collecting environmental data from the aircraft as the aircraft operation data; and
A third collector for collecting destruction data from the aircraft as the aircraft operation data;
The above selection section,
A data separator that separates the above status data and the above environmental data into aircraft operation data in image form and aircraft operation data in message form;
An image discriminator that uses pre-learned artificial intelligence to analyze aircraft operation data in the form of images to determine the presence or absence of abnormal patterns;
A message verifier for verifying aircraft operation data in the form of the above message; and
An aircraft health management device including a message discriminator that analyzes aircraft operation data in the form of verified messages using pre-learned artificial intelligence to determine whether there is an abnormal pattern. delete delete In claim 11,
The above selection section,
An aircraft health management device further comprising a data adder for adding the above destruction data to the above abnormal data. In claim 11,
The above management department,
A file generator for generating the above abnormal data as a file; and
An aircraft health management device including an output device that outputs a file of the above abnormal data to an external maintenance system of the aircraft.

Images