CN109522944B - Multi-dimensional rapid classification method for faults of spacecraft - Google Patents

Abstract

A multi-dimensional rapid classification method for faults of a spacecraft relates to the field of space safety; the method comprises the following steps: step one, establishing a keyword retrieval catalog of fault classification; step two, acquiring fault information of the aircraft; step three, performing fragmentation processing on the fault information of the aircraft according to the keyword retrieval catalog of the fault classification; step four, establishing a fault label; fifthly, correspondingly labeling the aircraft fault information subjected to fragmentation in the third step; step six, establishing an aircraft fault standard library, and storing the aircraft fault information subjected to labeling processing in the aircraft fault standard library; step seven, keyword retrieval query is carried out from the aircraft fault standard library to obtain the required fault information; the method is suitable for rapid classification and analysis of faults of various spacecrafts in the future, and has important significance for researching fault rules and rescue modes of the future aircrafts.

Description

A multi-dimensional rapid classification method for space vehicle faults

technical field

The invention relates to the field of space safety, in particular to a multi-dimensional rapid classification method for spacecraft faults.

Background technique

The orbital rescue system performs function recovery rescue for damaged satellites. The Orbital Rescue System has a wide range of military-civilian dual-use potentials. It is located in an area where national defense, military and commercial aerospace highly overlap. It is an important support for the development of military-civilian integration. It has important development potential and Value.

The world's major aerospace powers have carried out relevant research on fast-response launch vehicles and on-orbit service technologies. my country has been in the follow-up stage in space satellite rescue and other technologies. The gap is mainly reflected in the late start, less flight verification, and low technology maturity. There is no report on the classification method of space faulty aircraft. The proposal and application of the classification method of space faulty aircraft is of great significance for the use, rescue and rational allocation of space resources of space vehicles in the future.

Contents of the invention

The purpose of the present invention is to overcome the above-mentioned deficiencies of the prior art, to provide a multi-dimensional rapid classification method for spacecraft faults, which is applicable to the rapid classification and analysis of faults of various spacecraft in the future, and is important for studying future aircraft fault laws and rescue modes. significance.

Above-mentioned purpose of the present invention is achieved by following technical scheme:

A multi-dimensional rapid classification method for spacecraft faults, comprising the following steps:

Step 1, establishing a keyword retrieval directory for fault classification;

Step 2. Obtain the fault information of the aircraft;

Step 3, carry out fragmentation processing of the fault information of the aircraft according to the keyword retrieval directory of the fault classification;

Step 4, create a fault label;

Step 5. Corresponding labeling processing of the aircraft fault information after fragmentation processing in step 3;

Step 6, establish aircraft failure standard library, and store the aircraft failure information after tagging in the aircraft failure standard library;

Step 7. When analyzing aircraft faults, the fault information can be obtained by performing keyword search and query from the aircraft fault standard library.

In the above-mentioned multi-dimensional rapid classification method for space vehicle faults, in the first step, the fault categories in the fault classification keyword retrieval catalog include control system faults, telemetry control and communication system faults, power supply system faults, and power system faults , data management failure, thermal control system failure and payload failure.

In the above-mentioned method for multi-dimensional rapid classification of space vehicle failures, in the first step, the keywords of control system failures include attitude control failures, control computers, orbit control failures, sensors, navigation and reaction wheels; telemetry control and communication The keywords of system failure include telemetry and control, GPS, transponder and communication link; the keywords of power system failure include power regulator, power supply and distribution, power charging and discharging, power controller, battery pack, power lower computer and solar Battery array; keywords for power system failures are power actuators; keywords for data management failures are data management storage and data processors; keywords for thermal control system failures are platform external thermal control and stand-alone thermal control; payload failures are The keyword is camera.

In the above-mentioned multi-dimensional rapid classification method for spacecraft faults, in the second step, the acquisition source of aircraft fault information is an external ground measurement and control station.

In the above-mentioned multi-dimensional rapid classification method for spacecraft faults, the fault information of the aircraft is a comprehensive text in which all fault category information is arranged out of sequence.

In the above-mentioned multi-dimensional rapid classification method for spacecraft faults, the specific method for fragmenting the fault information of the aircraft is as follows:

S1: According to the keyword search directory of the fault classification, the fault information of the aircraft is classified according to the fault category;

S2: Summarize the fault information of the same fault category to form fault category fragments;

S3: Sort the failure category fragments in the order of control system failure, telemetry control and communication system failure, power supply system failure, power system failure, data management failure, thermal control system failure and payload failure.

In the above-mentioned multi-dimensional rapid classification method for spacecraft faults, in step 4, the categories and keywords of the fault label are consistent with the categories and keywords in the keyword retrieval directory for fault classification.

Compared with the prior art, the present invention has the following advantages:

(1) The present invention is specially aimed at the difficult difficult problem of fault aircraft statistics and classification, and establishes a space aircraft fault database; the method first classifies the space aircraft fault modes, and forms faults with satellite orbit height, functional use, platform type, and sub-systems Establish links with multi-dimensional information such as aircraft, and solve the difficulties brought about by the large amount of aircraft historical fault data, various types of information, and dynamic data updates, which have brought about statistical analysis;

(2) The present invention is specifically aimed at the difficult problem of statistics and classification of faulty aircraft, and proposes a multi-dimensional classification method that can be used for rapid analysis of large data of space vehicle faults, and uses the satellite fault database as the training sample of the algorithm to form abnormal parameters that can be analyzed The fault classification network structure is used for fast statistical analysis of space vehicle fault data;

(3) With the expansion of the database, the present invention can be further upgraded to adapt to new types of aircraft faults, and can also improve the accuracy of fault classification, and can provide data support for rapid fault analysis and classification, product quality improvement and other work.

Description of drawings

Fig. 1 is the flow chart of rapid classification method of the present invention;

Fig. 2 is a schematic diagram of the retrieval directory of fault classification keywords in the present invention.

detailed description

Below in conjunction with accompanying drawing and specific embodiment the present invention will be described in further detail:

The present invention overcomes the shortcomings of existing scientific research achievements, counts the recorded faults of space vehicles in history, builds a database and completes the research on reasonable labeling and rapid classification methods, and proposes a method for quickly analyzing and judging space vehicle failures in the future.

As shown in Figure 1, it is a flow chart of the rapid classification method. It can be seen from the figure that a multi-dimensional rapid classification method for spacecraft faults includes the following steps:

Step 1. Establish a keyword retrieval directory for fault classification; the fault categories in the fault classification keyword retrieval directory include control system faults, telemetry control and communication system faults, power system faults, power system faults, data management faults, thermal control system faults and Payload failure. The keywords of control system failure include attitude control failure, control computer, orbit control failure, sensor, navigation and reaction wheel; the keywords of telemetry control and communication system failure include telemetry control, GPS, transponder and communication link; power system The keywords of failure include power conditioner, power supply and distribution, power charging and discharging, power controller, battery pack, power lower computer and solar array; the keyword of power system failure is power actuator; the keyword of data management failure The memory and data processor are managed for data; the keywords of thermal control system failure are platform external thermal control and stand-alone thermal control; the keywords of payload failure are camera.

Step 2. Obtain the fault information of the aircraft; the source of the acquisition is an external ground measurement and control station; the fault information of the aircraft is a comprehensive text of all fault category information arranged in no order. Statistics have been made on thousands of aircraft failures (including satellites and spacecraft) in the past 10 years. The failure systems include telemetry and control systems, control systems, power systems, power systems, etc., to obtain more comprehensive space vehicle failure information under current technical conditions.

Step 3. According to the keyword search directory of the fault classification, the fault information of the aircraft is fragmented; the specific method of fragmenting the fault information of the aircraft is as follows:

S1: According to the keyword search directory of the fault classification, the fault information of the aircraft is classified according to the fault category;

S2: Summarize the fault information of the same fault category to form fault category fragments;

S3: Sort the failure category fragments in the order of control system failure, telemetry control and communication system failure, power supply system failure, power system failure, data management failure, thermal control system failure and payload failure.

Step 4: Create a fault label; the categories and keywords of the fault label are consistent with the categories and keywords in the keyword retrieval directory for fault classification.

Step 5. Corresponding labeling processing of the aircraft fault information after fragmentation processing in step 3;

Step 6, establish aircraft failure standard library, and store the aircraft failure information after tagging in the aircraft failure standard library;

Step 7. When analyzing and researching the faults of the aircraft, the fault information can be obtained by performing keyword search and query from the aircraft fault standard database by category or keyword. Statistical and analysis of the actual failures of the spacecraft on-orbit can reveal the law of on-orbit failures and guide the spacecraft to take preventive or corrective measures. Based on the collection of on-orbit failures of spacecraft, the article divides the categories of failures and summarizes the trend of failures.

And with the expansion of the database, it can be further upgraded to adapt to new types of aircraft faults, and can also improve the accuracy of fault classification. In the future, it can provide data support for rapid fault analysis and classification, product quality improvement and other work. By establishing the aircraft fault database and analyzing the cause and effect knowledge of the fault, it is possible to statistically analyze the faults that may be caused by the abnormal parameters of the future spacecraft and the abnormal phenomena.

Design a multi-dimensional rapid classification label network structure for spacecraft faults, and refine it to form a rapid classification tool for spacecraft fault systems. By inputting fault conditions, based on the researched database, scientific causal knowledge, and empirical knowledge of the overall impact coefficient, it can quickly form a pair of faults. Judgment of fault conditions.

The content that is not described in detail in the description of the present invention belongs to the well-known technology of those skilled in the art.

Claims (3)

1. A multi-dimensional rapid classification method for faults of a spacecraft is characterized by comprising the following steps: the method comprises the following steps:

step one, establishing a keyword retrieval catalog of fault classification;

step two, acquiring fault information of the aircraft;

step three, performing fragmentation processing on the fault information of the aircraft according to the keyword retrieval catalog of the fault classification;

step four, establishing a fault label;

fifthly, correspondingly labeling the aircraft fault information subjected to fragmentation in the third step;

step six, establishing an aircraft fault standard library, and storing the aircraft fault information subjected to labeling processing in the aircraft fault standard library;

step seven, when the aircraft fault is analyzed, keyword retrieval query is carried out from the aircraft fault standard library to obtain fault information;

in the first step, the fault categories in the fault classification keyword retrieval directory comprise control system faults, remote control and communication system faults, power system faults, data management faults, thermal control system faults and payload faults;

in the first step, keywords of the control system faults comprise attitude control faults, a control computer, rail control faults, a sensor, navigation and reaction wheels; the keywords of the fault of the telemetry and control system and the communication system comprise telemetry and control, GPS, a transponder and a communication link; the key words of the power system faults comprise a power supply regulator, a power supply and distribution device, a power supply charge and discharge device, a power supply controller, a storage battery pack, a power supply lower computer and a solar cell array; the key word of the power system fault is a power executing mechanism; the key words of the data management faults are a data management memory and a data processor; the keywords of the fault of the thermal control system are the thermal control outside the platform and the single machine thermal control; the key to the payload failure is the camera;

in the second step, the aircraft fault information is acquired from an external ground measurement and control station;

the specific method for fragmenting the fault information of the aircraft comprises the following steps:

s1: carrying out fault classification on the fault information of the aircraft according to fault categories according to the keyword retrieval catalog of the fault classification;

s2: collecting fault information of the same fault type to form fault type fragments;

s3: and sorting the fault category fragments according to the sequence of control system faults, remote control and communication system faults, power system faults, data management faults, thermal control system faults and payload faults.

2. The multi-dimensional rapid classification method of spacecraft faults according to claim 1, characterized in that: the fault information of the aircraft is a comprehensive character in which all fault category information is arranged in a non-sequence mode.

3. The multi-dimensional rapid classification method for the faults of the spacecraft according to claim 2, characterized in that: in the fourth step, the category and the keyword of the fault label are consistent with those in the keyword retrieval catalog of the fault classification.

Images