JP4875661B2 - Aircraft soundness diagnosis apparatus and method, and program - Google Patents

Description

  The present invention relates to an aircraft soundness diagnosis apparatus, method, and program for performing aircraft soundness diagnosis.

  Airplanes operate in a wide variety of ways, such as flying at high speeds in the atmosphere for a long time on regular routes, and repeatedly taking off and landing several times a day. However, since the aircraft is always navigating while receiving all the loads, the fuselage, main wings, tails, etc. are prone to fatigue in proportion to the flight time, and the probability of distortion or cracking due to the load. Becomes higher. For this reason, the inspection and maintenance of the aircraft is performed in a regular cycle every operation and every flight time. In general, such inspection and maintenance is performed when the aircraft is parked on the ground.

Conventionally, damage and damage such as unevenness of surface and structure of aircraft, cracks, etc., macroscopic or microscopic visual inspection of skilled maintenance personnel, ultrasonic flaw detector, magnetic particle damage device, eddy current flaw detector Inspected by equipment such as X-ray inspection. This inspection is carried out at an airfield, a maintenance station, etc. after the aircraft is suspended for a certain period. Metal fatigue was managed simply by the time of flight and the number of takeoffs and landings.
JP 2005-241089 A

However, the above-described conventional diagnostic method has the following problems.
First, the structural inspection of the airframe is mainly performed by a spot-type inspection device, so that a long time is required for the inspection and the operational efficiency of the aircraft is lowered.
Second, in the maintenance stage on the ground, it is possible to detect a defect in the aircraft, but it cannot be detected during flight, and inspection data processing for diagnosing soundness in real time cannot be performed.
Third, it is impossible to ascertain the degree of impact on the aircraft against sudden events (hard landing, turbulence) during aircraft operation, and to ensure the safety of the aircraft immediately.
Fourth, it is desirable to detect failures as early as possible and to prepare for repairs before the aircraft arrives in order to shorten the ground residence time and protect the regular operation. Is possible.
Fifth, some parts of the airframe structure are inaccessible, and when using conventional inspection methods such as visual inspection, ultrasonic inspection, eddy current inspection, etc., the airframe must be disassembled. Become.
Sixth, it is impossible to evaluate the service life by individual management of the aircraft.

  The present invention has been made to solve the above-described problems, and can reduce the inspection time while parked and can also perform aircraft health diagnosis automatically and quantitatively. It is an object to provide a soundness diagnosis apparatus and method, and a program.

In order to solve the above problems, the present invention employs the following means.
The present invention relates to an aircraft health diagnostic apparatus for diagnosing the aircraft health using characteristic values created based on measurement data measured by a plurality of sensors provided in the aircraft, and measuring time When a plurality of characteristic values associated with the item are stored for each characteristic item and the characteristic values associated with the same measurement time are set as one data set, the data set is determined according to the flight status. The first storage means storing the identification information indicating the class classification to be stored and a plurality of characteristic values associated with the measurement time are stored for each characteristic item and associated with the same measurement time In the case where the characteristic value is one data set, the data set is provided with identification information indicating the class classification and constitutes the data set. Second storage means belonging to a predetermined reference range in which characteristic values of the specific characteristic items are defined, and a plurality of data sets used for diagnosis from the first storage means to extract data to be diagnosed A diagnostic file construction means for constructing a reference data file by constructing a file and extracting a plurality of the data sets to be compared with the diagnostic data file from the second storage means based on the class classification; and the diagnosis Based on the data file to be diagnosed and the reference data file constructed by the file construction unit, using a statistical calculation method, an index value calculating means for calculating a health index value indicating the health of the aircraft; and Evaluation means for evaluating the soundness of the aircraft based on the soundness index value calculated by the index value calculating means; and an evaluation by the evaluation means Providing health diagnosis apparatus for an aircraft and a notifying means for notifying the result.

In this way, since the health index value indicating the soundness of the aircraft is calculated using the diagnosis data file and the reference data file, quantitative evaluation is performed instead of qualitative evaluation based on experience and knowledge. It can be realized. Further, since the health index value is a value that takes into account the class classification given to each data set, it is possible to compare data acquired under the same situation. This makes it possible to more accurately evaluate the soundness of the aircraft. In addition, by performing the diagnosis in real time, it is possible to evaluate the soundness during the flight.
The characteristic item is an item representing characteristics of various parameters used for diagnosing the soundness of an aircraft. It is assumed that the user can arbitrarily set and select which characteristic item is used to diagnose the soundness.

  In the aircraft health diagnostic apparatus, the classification may be determined based on a flight process when a single flight pattern is divided into a plurality of processes.

  For example, the load or the like acting on each part of the aircraft shows different values in each process such as parking, traveling, and flying. Therefore, one flight pattern can be changed to “parking before takeoff”, “taxing before takeoff”, “takeoff”, “cruising”, “landing”, “taxing after landing”, “parking after landing”, etc. As described above, by dividing into a plurality of processes and determining the class classification according to this process, each data set can be appropriately classified according to its characteristics.

  In the aircraft health diagnostic apparatus, the classification may be determined according to an operation period or the number of flights.

  For example, characteristics such as loads acting on each part of the aircraft change according to the operation period and the number of flights. Therefore, by determining the class classification according to the operation period and the number of flights, it is possible to appropriately classify each data set according to its characteristics.

  In the aircraft health diagnostic apparatus, the classification may be determined according to a service route.

  For example, characteristics such as load acting on each part of the aircraft change according to the service route (change in airflow). Therefore, by determining the class classification according to the service route, each data set can be appropriately classified according to its characteristics.

  In the aircraft health diagnostic apparatus, the plurality of data sets stored in the second storage unit are characteristics related to the specific characteristic item among the plurality of data sets stored in the first storage unit. Only data sets belonging to a reference range in which values are set in advance may be extracted.

  For example, when the reference range is set to the normal range, a reference data file is constructed based on normal data, and the aircraft health is evaluated by comparison with this reference data file. It becomes possible to evaluate more accurately.

  In the aircraft health diagnostic apparatus, the index value calculation means obtains the characteristic distribution of the reference data file set by the diagnostic file construction means, obtains the characteristic distribution of the diagnostic data file, and The soundness index value may be calculated by quantitatively obtaining the distance at which the distribution is deviated.

  In this way, the characteristic distributions of each other are obtained, and the distance at which these distributions are separated is obtained quantitatively, so it is possible to quantitatively determine how far the diagnosis data is from the characteristic distribution of the reference data. It becomes possible to evaluate.

  In the aircraft health diagnostic apparatus, the health index value calculated by the index value calculation unit may be a Mahalanobis distance calculated using a Mahalanobis Taguchi method.

  The aircraft soundness diagnosis apparatus may further include a factor analysis unit that performs a factor analysis of an abnormality when the evaluation unit evaluates that an abnormality has occurred.

  In this way, by performing factor analysis, it is possible to quickly grasp which part is determined to be abnormal due to the cause. This makes it possible to respond quickly.

  In the aircraft health diagnostic apparatus, the maintenance time of the aircraft may be determined based on the transition of the health index value calculated by the index value calculation means.

  Thus, by determining the maintenance time of the aircraft based on the transition of the soundness index value, it becomes possible to carry out the maintenance at an appropriate time and to prevent a failure in advance.

  The present invention relates to an aircraft health diagnosis method for diagnosing the aircraft health using characteristic values created based on measurement data measured by a plurality of sensors provided in the aircraft, the measurement time being A process of creating a first data file in which a plurality of characteristic values associated with each item is stored for each characteristic item, and the characteristic values associated with the same measurement time in the first data file as one data set In addition, the data set belongs to a predetermined reference range in which identification values indicating class classifications determined according to flight status are given, and characteristic values related to specific characteristic items are defined in advance. The process of creating the second data file in which the characteristic value of each characteristic item is associated with the measurement time, and the same measurement time in the second data file When the associated characteristic values are set as one data set, a process of assigning identification information indicating the class classification to the data set, and a plurality of the data sets used for diagnosis are extracted from the first data file And constructing a diagnostic data file, and extracting a plurality of the data sets to be compared with the diagnostic data file from the second data file based on the class classification to construct a reference data file; Based on the constructed diagnostic data file and the reference data file, using a statistical calculation method, calculating a health index value indicating the health of the aircraft, and based on the health index value And providing a method for diagnosing aircraft health, comprising: a step of evaluating the soundness of the aircraft; and a step of notifying the evaluation result

  The present invention is an aircraft health diagnostic program used for diagnosing the aircraft health by using characteristic values created based on measurement data measured by a plurality of sensors provided in the aircraft. A process of creating a first data file in which a plurality of characteristic values associated with the measurement time are stored for each characteristic item, and the characteristic values associated with the same measurement time in the first data file In the case of a data set, a process for assigning identification information indicating the class classification determined according to the flight status to the data set, and a characteristic value related to a specific characteristic item within a predetermined reference range A process for creating a second data file in which the characteristic value of each characteristic item is associated with the measurement time, and the second data file When the characteristic values associated with the same measurement time are set as one data set, a plurality of processes for giving identification information indicating the class classification to the data set, and a plurality of data used for diagnosis from the first data file The data set is extracted to construct a diagnosis data file, and a plurality of the data sets to be compared with the diagnosis data file are extracted from the second data file based on the class classification to obtain a reference data file A process of calculating a health index value indicating the health of the aircraft using a statistical calculation method based on the constructed diagnostic data file and the reference data file, and the sound Based on the sex index value, a process for evaluating the soundness of the aircraft and a process for notifying the evaluation result are executed on a computer. To provide the soundness diagnosis program of aircraft for.

  Advantageous Effects of Invention According to the present invention, it is possible to shorten the inspection time while parked, and to obtain an effect that the soundness diagnosis of an aircraft can be performed automatically and quantitatively.

  DESCRIPTION OF EMBODIMENTS Embodiments of an aircraft soundness diagnosis apparatus and method, and a program according to the present invention will be described below with reference to the drawings.

[First Embodiment]
FIG. 1 is a block diagram showing a hardware configuration of an aircraft soundness diagnosis apparatus (hereinafter referred to as “health diagnosis apparatus”) according to the present embodiment.
As shown in FIG. 1, the soundness diagnosis apparatus 10 is a computer system (computer system), and includes a CPU (Central Processing Unit) 11, a RAM (Random Access).
By communicating with a main storage device 12 such as a memory, an auxiliary storage device 13 such as an HDD (Hard Disk Drive), an input device 14 such as a keyboard and a mouse, an output device 15 such as a monitor and a printer, and external devices. The communication device 16 and the like that exchange information are configured.
Various programs (for example, a soundness diagnosis program) are stored in the auxiliary storage device 13, and the CPU 11 reads out the programs from the auxiliary storage device 13 to the main storage device 12 and executes them to implement various processes.

  FIG. 2 is a functional block diagram showing the functions provided in the soundness diagnosis apparatus 10 in an expanded manner. As shown in FIG. 2, the soundness diagnosis apparatus 10 includes a measurement database 21, a flight database 22, a data generation unit 23, a class classification definition unit 24, a class classification unit 25, and a first storage unit (first storage unit). 1 storage unit) 26, normal data condition definition unit 27, normal data extraction unit 28, second storage unit (second storage unit) 29, diagnostic file construction unit (diagnosis file construction unit) 30, and index value A calculation unit (index value calculation unit) 31, an abnormality determination unit (evaluation unit) 32, and a notification unit (notification unit) 33 are provided.

In the measurement database 21, as shown in FIG. 3, a measurement data file is stored for each aircraft ID assigned to each aircraft. Each measurement data file stores a plurality of measurement data measured by various sensors installed on the aircraft body.
Examples of the measurement data include strain distribution measurement data measured by optical fiber sensors provided at a plurality of measurement locations set in the aircraft body, temperature measurement data measured by a temperature sensor, and the like. Each measurement data is associated with a measurement time and a measurement item in which the measurement data is measured. This measurement time functions as a linking parameter for associating various data with each other in a diagnostic data file creation process performed in the class classification unit 25 described later.

  In the flight database 22, a flight route data file indicating the operational state of the aircraft is stored for each aircraft ID. In the flight route data file, for example, flight logs such as altitude, ground speed, atmospheric pressure and the like are stored in association with their measurement times for each service route.

  Here, the various measurement data files stored in the measurement database 21 and the flight route data file stored in the flight database 22 are stored in a database mounted on the aircraft during the navigation of the aircraft, After the aircraft has landed on the ground, these data files stored in the database in the aircraft may be transferred to the measurement database 21 and the flight database 22 of the health diagnostic device 10. Thus, the method for obtaining data stored in the measurement database 21 and the flight database 22 is not particularly limited.

  When the time interval (hereinafter referred to as “sampling time”) of the measurement time of each measurement data stored in the measurement database 21 and the flight database 22 is not unified, the data generation unit 23 sets these sampling times. Perform a unified process. For example, the measurement data of each measurement item is reconstructed so as to have an interval of α minutes.

  For example, when the sampling time is sufficiently faster than α minutes, a representative value for α minutes is selected by a statistical method using all measurement data acquired during α minutes. For example, the representative value is represented by an average value and a standard deviation. By doing in this way, measurement data concerning each measurement item can be synchronously related at a common time interval.

In this way, various types of data with a uniform sampling time are output to the class classification unit 25.
The class classification unit 25 creates a diagnostic data file by integrating the measurement data file stored in the measurement database 21 and the flight route data file stored in the flight database 22 for each aircraft ID.

  FIG. 4 shows an example of an aircraft diagnosis data file identified by the aircraft ID: ABC001. As shown in FIG. 4, in the diagnosis data file, measurement data at each measurement time is associated with each measurement item. Here, the measurement item is defined as “characteristic item”, and the “measurement data” is defined as “characteristic value”. Specific items include, for example, “altitude”, “ground speed”, “atmospheric pressure”, “temperature”, “left main wing distortion”, “right main wing distortion”, and the like. In addition, the machine ID “ABC001” and model information “ABC” of the machine are also associated with the diagnosis data file.

Subsequently, the class classification unit 25 sets characteristic values associated with the same measurement time as one data set in the diagnosis data file, and adds identification information indicating the class classification to each data set.
Specifically, the class classification unit 25 determines which data set based on the class definition defined in the class classification definition unit 24, in other words, for each line of the diagnostic data file shown in FIG. Whether the data belongs to the class classification is classified, and a flag indicating the class classification is set in each data set.

  The “class classification” referred to here is a statistical diagnosis method such as the Mahalanobis Taguchi method (hereinafter referred to as “MT method”), which matches the reference range of a plurality of characteristic items defined by the class classification definition unit 24. Refers to the separation of data groups. As described above, the statistical diagnosis for identifying normality or abnormality between the same “class classification” has higher identification accuracy than the case of the whole data not classified.

In the present embodiment, the class classification is divided into three classifications including a first class classification, a second class classification, and a third class classification.
The first class classification relates to a “flight pattern” and is determined based on a flight process when one flight pattern is divided into a plurality of processes. For example, in the present embodiment, the first class classification is “1. Parking before takeoff”, “2. Taxiing before takeoff”, “3. Takeoff”, “4. Cruise (horizontal, turning)”, “5” .Landing "," 6. Taxiing after landing "," 7. Parking after landing ", classifying into 7 classes and identifying the first class classification indicating which class each data set belongs to Information 1 to 7 is added.

  In addition, the second class classification relates to “service route pattern”, such as “1. Japan-North America”, “2. Japan-Australia”, “3. Japan-Southeast Asia”, etc. Is attached, and identification information 1 to n of the second class classification indicating which service route corresponds to each data set is added.

Moreover, the third class classification relates to “operation period”, for example, “less than 1.3 years”, “2.3 years to less than 5 years”, “3.5 years to less than 6 years”, etc. As described above, identification numbers are assigned for each operation period, and identification information 1 to m of the third class classification indicating which operation period corresponds to each data set is added.
Thus, each identification number regarding three class classifications is added to each data set constituting the diagnostic data file according to the present embodiment.

  As described above, information in which classification related to each class classification is defined is stored in the class classification definition unit 24. The class classification unit 25 refers to the class classification stored in the class classification definition unit 24 and assigns an identification number for each class classification for each data set (for each row) of the diagnosis data file.

  The first storage unit 26 stores the diagnosis data file (first data file) classified by the class classification unit 25 (flagged). The diagnosed data file that has been classified and stored in the first storage unit 25 is handled as a “signal space” in the calculation process in the index value calculation unit 31.

  On the other hand, only the data set determined to be “normal” from the “past diagnosed data file” already acquired is extracted and stored in the second storage unit 29. This stored data is called a “normal data file”. Of course, at this stage, each data set of the normal data file stored in the second storage unit 29 is added with a class classification flag assigned by the class classification unit 25 of the preceding process. These normal data files stored in the second storage unit 29 are handled as “unit spaces” in the calculation processing in the index value calculation unit 31.

More specifically, the normal data condition definition unit 27 defines a determination condition for determining that the data is “normal data”, and the normal data extraction unit 28 first sets a data set that matches the determination condition. A normal data file is created by extracting from the diagnosis data file stored in the storage unit 26 and stored in the second storage unit 29.
Thus, by automating the normal data file creation process, the normal data file stored in the second storage unit 29 can be automatically updated.

As the determination conditions defined in the normal data condition definition unit 27, for example, after introduction of the aircraft or after a large inspection such as a periodic inspection including the structure, until a predetermined number of flights (for example, 100 flights) is exceeded. In the data set corresponding to the flight in which no abnormality such as foreign object collision, hard landing, hard turbulence, etc. occurred in the period of time or the predetermined operation period (for example, 3 months) Examples include conditions such as a data set corresponding to a flight that has been operated in accordance with a flight plan and for which an irregular report has not been generated during navigation.
The determination condition is an example, and is not limited to the above contents, and can be arbitrarily set and changed.

  In addition, each data set of diagnosis data data has been operated in accordance with information necessary for the determination, for example, a prior flight plan, in order to reliably determine whether or not the determination condition is met. And information indicating whether or not an irregular report has been generated.

  The diagnostic file construction unit 30 extracts a plurality of data sets to be diagnosed from the first storage unit 26 to construct a “diagnosis data file” (signal space) for diagnosis and stores it in the second storage unit 29. Referring to the classification classification identification number from the normal data file, extract only the optimal data set for the diagnosis of the “diagnostic data file” and build a “reference data file” (unit space) for diagnosis .

  For example, when the aircraft identified by the aircraft ID “ABC001” is set as a health diagnosis target (hereinafter referred to as “diagnosis target”), the diagnosis file construction unit 30 stores the diagnosis target stored in the first storage unit 26. A data set corresponding to the first class classification “1. Parking before take-off” is extracted from the data file to be diagnosed to construct a “diagnosis data file” for diagnosis and stored in the second storage unit 29 A data set corresponding to the first class classification “1. parking before take-off” is extracted from the normal data file to be diagnosed, and a “reference data file” (unit space) for diagnosis is constructed. Note that the extraction conditions of the data set that constitutes the reference data file can be arbitrarily determined as will be described later.

The index value calculation unit 31 uses the statistical diagnosis method based on the diagnostic “diagnostic data file” and “reference data file” created by the diagnostic file construction unit 30 to indicate the soundness of the aircraft. An index value is calculated.
Specifically, the index value calculation unit 31 normalizes the “diagnosis data file” and the “reference data file” created by the diagnosis file construction unit 30, and normalizes the “diagnosis data file” and the “reference data”. Each item distribution of the “file” is obtained, a state where the data distributions (groups) are deviated from each other is quantitatively obtained as a distance between the distributions, and the distance is treated as a soundness index value of soundness.

  More specifically, the MT method is used as the statistical diagnosis method of the index value calculation unit 30, and the Mahalanobis distance (hereinafter referred to as “the health index value” as one of the diagnostic output results obtained by the MT method). MD value) ”is calculated. A specific calculation method will be described later.

  The abnormality determination unit 32 compares the soundness index value calculated by the index value calculation unit 31 with a preset threshold value, and evaluates the soundness of the aircraft according to the comparison result. For example, when the soundness index value exceeds a threshold value, it is determined to be abnormal, and an abnormality determination signal is output to the notification unit 33.

  The notification unit 33 notifies the user of the occurrence of an abnormality by displaying the occurrence of the abnormality of the aircraft on a display or the like when an abnormality determination signal is input. In addition, instead of or in addition to the visual notification method, an abnormality may be notified by auditory notification, for example, a report sound. Thus, the notification method is not particularly limited.

Next, details of processing executed in each unit included in the soundness diagnosis apparatus 10 according to the present embodiment will be described in detail with reference to FIGS. 5 and 6. 2 are executed by the CPU 11 shown in FIG. 1 reading the health diagnosis program stored in the auxiliary storage device 13 into the main storage device 12 and executing it. It is realized.
Moreover, in the following description, a case where soundness diagnosis is performed by dividing each flight pattern process will be described as an example.

  First, the data generation unit 23 of the soundness diagnosis apparatus 10 unifies the sampling time of measurement data stored in the measurement database 21 and the flight database 22 (step SA1 in FIG. 5).

  Subsequently, the class classification unit 25 integrates the measurement data file stored in the measurement database 21 and the flight route data file stored in the flight database 22 for each aircraft ID, so that a diagnosis data file for each aircraft ID is obtained. Are respectively generated (step SA2 in FIG. 5). Further, in each diagnosis data file by machine ID, flags of identification numbers related to the first class classification to the third class classification are set for each data set associated with the same measurement time (FIG. 5). Step SA3).

  Accordingly, for example, each data set (each row) of each diagnosis data file includes, as a first class classification, “1. parking before take-off”, “2. taxiing before take-off”, “3. take-off”. , "4. Cruising (horizontal, turning)", "5. Landing", "6. Taxiing after landing", "7. Parking after landing" are flagged and the second class As a classification, any flag indicating a service route pattern such as “1. Japan-North America”, “2. Japan-Australia”, “3. Japan-Southeast Asia” is set. Furthermore, a flag relating to the operation period of the aircraft is set as the third class classification.

As described above, the plurality of diagnosis data files to which the flags are added for the first to third class classifications are stored in the first storage unit 26 (step SA4 in FIG. 5).
Next, the normal data extraction unit 28 extracts only normal data sets from the classified diagnostic data file stored in the first storage unit 26 to create a normal data file, and stores it in the second storage unit 29. Stored (step SA5 in FIG. 5). This normal data file is also created for each machine ID.

  Next, the diagnostic file construction unit 30 extracts all or a part of the data set of the diagnostic data file in the first storage unit 26 to construct a diagnostic data file for diagnosis, and the second storage unit 29 All or part of the data set of the normal data file is extracted to construct a normal data file for diagnosis (hereinafter referred to as “reference data file”) (step SA6 in FIG. 5).

  For example, in this embodiment, since the health diagnosis is performed for each class classification of the flight pattern, the diagnosis file construction unit 29 firstly selects the diagnosis target data file stored in the first storage unit 26. All data sets whose first class classification indicates “1. parking before take-off” are extracted, and a “diagnosis data file” for diagnosis is constructed.

  Further, the diagnostic file construction unit 29 extracts all data sets whose first class classification indicates “1. parking before take-off” from the second storage unit 29 and constructs a “reference data file”. At this time, a data set in which the first class classification indicates “1. parking before take-off” may be extracted from the normal data file to be diagnosed, and the second memory may be stored regardless of the aircraft ID, model, etc. A data set whose first class classification indicates “1. parking before take-off” may be extracted from all normal data files stored in the unit 29.

When the diagnostic “diagnosis data file” and “reference data file” are constructed in this way, the index value calculation unit 30 performs the calculation process of the health index value (step SA7 in FIG. 5).
The soundness index value calculation process will be described below with reference to FIG.

[Data standardization]
First, the index value calculation unit 31 executes a data normalization process (step SB1 in FIG. 6).
For example, if the measurement time number of the reference data file set in the diagnostic file construction unit 30 is i and the number of characteristic items is j, the reference data file forms a matrix of i rows and j columns.

The reason for standardizing the reference data is to treat characteristic values between different characteristic items (between measured physical quantities) fairly in statistical processing. Therefore, the characteristic value x _ij identified by each row and each column is normalized using the average value m _j and standard deviation σ _j calculated based on the following equations (1) and (2). The normalized value of the characteristic value x _ij is expressed as a positive standard value X _ij and is obtained by the following equation (3).
In the following description, as shown in FIG. 7, a description will be given assuming a data file of n rows and k columns.

Similarly, the index value calculation unit 31 normalizes the “diagnostic data file” by performing the same calculation as that of the reference data file. As the average value m _j and standard deviation σ _j used for normalization, the values of the “reference data file” calculated by the above formulas (1) and (2) are used. As a result, a characteristic standard value Y _ij obtained by standardizing each characteristic value y _ij of the “diagnosis data file” is calculated by the following equation (4).

  The index value calculation unit 31 reconstructs the respective data files by replacing the characteristic values of the “reference data file” and the “diagnostic data file” with the normalized characteristic standard values.

[Calculation of correlation matrix]
Next, the index value calculation unit 31 calculates the correlation matrix R = (r _ij ) using the characteristic standard value X _ij of the reference data file (step SB2 in FIG. 6). The correlation matrix R is derived using the following equation (5). The correlation matrix is a k-th order matrix whose diagonal component is 1.

Here, a specific description for obtaining the correlation matrix will be given. When there are k (k columns) types of characteristic items j of the characteristic standard value _Xij of the reference data file, the number of correlation combinations is k × k. As an example, if the number of characteristic items j in the reference data file is k = 200 types (columns), there are 200 × 200 = 40000 correlation combinations, which simultaneously become the characteristics of a 200 × 200 regular matrix. The diagonal components of the regular matrix have the property of inevitably being 1 because they are the correlations between the same characteristic items. Further, the correlation coefficient other than the diagonal line is r _pq = r _qp , and its value is symmetrically equal across the diagonal line.

[Calculation of inverse matrix of correlation matrix]
Subsequently, the index value calculation unit 31 calculates the inverse matrix A = R ⁻¹ of the correlation matrix R of the reference data file using the following equation (6) (step SB3 in FIG. 6).

[Calculation of Mahalanobis distance]
Next, the Mahalanobis distance D ² _i (hereinafter referred to as “MD value”) using the correlation inverse matrix A of the reference data file obtained by the above equation (6) and each characteristic standard value Y _ij of the diagnosis data file after normalization. (Refer to step SB4 in FIG. 6). The MD value D ² _i is calculated using the following equation (7).

  Here, k is the number of characteristic items of the “diagnostic data file”, that is, the number of columns, and the MD value is calculated for each data set (for each row) of the “diagnostic data file”.

Here, when calculating the MD value of a certain measurement time, the data line i (i is any one of 1 to n) corresponding to the certain measurement time of the standardized “diagnostic data file” is designated. , Y _i1 to Y _ik which are values of each column of the i row are substituted into the equation (7) for calculation. In this way, MD values are obtained for the number of rows of the “diagnostic data file”, in other words, the number of data sets.
In this way, when the MD value of “1. Parking before take-off” to be diagnosed is obtained, this calculation result is output to the abnormality determination unit 32. The abnormality determination unit 32 compares each input MD value D ² _i with a preset threshold value (a value that can be arbitrarily set, for example, 3), and the MD value D ² _i is greater than the threshold value. Is also larger (step SA8 in FIG. 5). As a result, when the MD value D ² _i greater than the threshold value is present in a predetermined ratio or more, an abnormal signal is output assuming that the machine to be diagnosed is in an abnormal state. As a result, the notification unit 33 notifies the user of the abnormality of the aircraft (step SA9 in FIG. 5).

On the other hand, if the MD value D ² _i exceeding the threshold is equal to or less than the predetermined ratio in step SA8, it is determined that the state of the main body is normal, and the process returns to step SA1 in FIG. 2. The diagnosis data file and the reference data file relating to “Taxing” are constructed by the diagnosis file construction unit 30, and using the constructed diagnosis data file and the reference data file, “2. A soundness diagnosis process related to “taxing” is performed.

  Similar processing is performed for “3. Takeoff” to “7. Parking after landing”, for example, and a diagnostic result as shown in FIG. 8 is obtained and displayed on the display unit. FIG. 8 is a diagram showing the MD value with respect to the operation time in one flight pattern of the aircraft to be diagnosed, with the operation time on the horizontal axis and the MD value on the vertical axis. In FIG. 8, the numbers in parentheses indicate the identification numbers of the first class classification. In FIG. 8, at the time of “4. cruise”, the MD value is equal to or greater than the threshold value, indicating that an abnormality has occurred.

  As described above, according to the aircraft soundness diagnosis apparatus, method, and program according to the present embodiment, the soundness of the aircraft is compared with the “reference data file” that is the actually measured value classified by class. Can be quantitatively determined by the MD value, and appropriate evaluation can be realized instead of qualitative evaluation based on experience and knowledge.

  In the above embodiment, a case has been described in which measurement data files of a plurality of aircraft are collected and the soundness of each aircraft is diagnosed using these data. However, the measurement data file and flight route data file of the own aircraft are described. It is good also as performing a soundness diagnosis using only. In this way, if only the data of the own device is used, the amount of data to be handled can be reduced, and real-time health diagnosis can be performed.

  In other words, if only the data of the own aircraft is used, it is not necessary to exchange data with other aircraft. For example, the aircraft health diagnostic device 10 can be mounted on the aircraft. By doing in this way, soundness diagnosis is performed in real time, and after landing, maintenance and the like can be performed promptly based on the soundness diagnosis result. Thereby, the maintenance time on the ground can be shortened, and the aircraft can be operated with high efficiency.

  Furthermore, since it is possible to diagnose the health during the flight, for example, when sudden damage is caused by a bird collision or turbulence, an abnormality can be detected at that time. Accordingly, it is possible to promptly carry out repairs after landing by notifying the relevant maintenance station, command post, etc. outside the aircraft by means of radio, satellite communication or the like during the flight.

  In the above-described embodiment, the normal data stored in the second storage unit 29 is created using the data stored in the first storage unit 26, but the normal data stored in the second storage unit 29 is used. The data is not limited to this example. For example, normal data calculated by predetermined simulation software or the like may be used.

  In the above embodiment, as shown in FIG. 5, the processing from step SA1 to step SA9 has been described as a series of processing. However, until the normal data file generation processing, that is, data is stored in the first storage unit 26. The processing to store and the processing to store data in the second storage unit 29 are treated as pre-processing necessary for diagnosing the soundness of the aircraft, and the processing after setting the reference data file and the data file to be diagnosed, specifically The processing from step SA6 to step SA9 in FIG. 5 may be handled as the main processing for diagnosing the soundness of the aircraft. The preprocessing and the main processing may have a time difference or may be realized by different computers.

  In the above description, the first class classification, which is the class classification related to the operation pattern, extracts the normal data identical to the diagnosis data file from the second storage unit 29 and constructs the “reference data file”. Instead, for example, normal data having the same model as the diagnosis target and having the same first class classification may be extracted from the second storage unit 29 to construct a “reference data file”.

  Furthermore, it is the same model as the diagnosis target, the number of flights is close to the diagnosis target, that is, the diagnosis target and the third class classification are the same, and the diagnosis target and the first class classification are the same. Only the data may be extracted from the second storage unit 29 to construct the reference data file. For example, as shown in FIG. 9, if the operation period (number of flights) is different, load accumulation increases, so the distortion characteristics of the aircraft change, and the reference space gradually moves (N = 3000). See reference space). Therefore, it is possible to further improve the accuracy of soundness diagnosis by extracting the reference data of other aircraft in the same or similar environment / state as the diagnosis target and constructing the reference data file.

  Moreover, it is the same model as the diagnosis target, the third class classification is the same, the first class classification is the same, and the service route is the same. In other words, the second class classification is Only the normal data that is the same may be extracted from the second storage unit 29 to construct the reference data file. Thus, the diagnosis accuracy can be further improved by taking the service route into consideration.

  Further, as described above, the reference data file may be constructed by extracting all normal data stored in the second storage unit 29 regardless of the class classification. Thus, by constructing the reference data file using normal data in any state, it is possible to diagnose the soundness of the target aircraft from a wide viewpoint in consideration of all models and operating conditions.

  As described above, in the present invention, normal data extraction conditions for constructing the reference data file can be arbitrarily set, and the first class classification, the second class classification, and the third class classification (model) are set. It can be set arbitrarily in consideration. In the present invention, since the classification is performed based on the operation process, service route, and operation period that cause the reference space to change, each data set can be divided into appropriate classes, and the soundness with high accuracy. Diagnosis can be realized.

  In the present embodiment, the class classification unit 25 generates the diagnosis data file. Instead, the data generation unit 23 generates the diagnosis data file and the class classification unit 25 generates the diagnosis data file. It is also possible to flag the classification of the diagnostic data file.

In this embodiment, the case where the normal data file is automatically generated by the normal data extraction unit 28 has been described. However, the method for generating the normal data file stored in the second storage unit 29 is not particularly limited.
For example, the operator may extract a data set that matches the determination condition from the diagnosis data file. In addition, based on the respective data stored in the flight database 22 and the measurement database 21 instead of the diagnosis data file stored in the first storage unit 26, the other simulation apparatus or the operator himself is normal. A data file may be created. That is, the second storage unit 29 only needs to store data necessary for generating a unit space in the Mahalanobis-Taguchi method, and the generation method is not particularly limited.

[Second Embodiment]
Next, a soundness diagnostic apparatus according to the second embodiment of the present invention will be described.
In the health diagnostic apparatus according to the first embodiment described above, when an abnormality is detected by the abnormality determination unit 31, it is quantitatively determined which characteristic item is involved in the abnormal state or is not involved. The need to be identified. The present embodiment has been proposed in view of the request.

  In the present embodiment, in the soundness diagnosis apparatus according to the first embodiment described above, when an abnormality is determined by the abnormality determination unit 32 as shown in FIG. 10, a factor analysis unit that performs a factor analysis of the abnormality (Factor analysis means) 50 is further provided. Hereinafter, regarding the soundness diagnosis apparatus of the present embodiment, description of points that are common to the first embodiment will be omitted, and “factor / effect analysis” that is a different point will be mainly described.

  “Factor effect analysis” of the present embodiment is, for example, a quantitative analysis of which characteristic items have an influence on the length of MD values, assuming that there are 200 characteristic items constituting diagnostic data. As shown in FIG. 11, the rank-up display is performed according to the magnitude of the SN ratio gain, which is the output value of the factor effect analysis.

  Hereinafter, a method for calculating the SN ratio gain of each characteristic item, which is an index value of the factor effect, will be described with reference to FIG. For convenience of explanation, FIG. 12 shows an example in which there are five types of characteristic items. Each of the five types of characteristic items is assigned to a two-level orthogonal table of “use ○” and “not use ×”, and the five types. The MD value calculation process (see FIG. 6) is performed under the respective conditions according to the 12 combination conditions (row No. of the orthogonal table) of “use ○” and “not use ×” of the characteristic items of 5 The MD value of each data set in the diagnostic data file consisting of individual characteristic items is calculated.

For example, when the number of data lines related to the measurement time interval determined to be abnormal in the diagnosis data file by the abnormality determination unit 32 is 2, each of the 12 characteristic items “Use ○”, The MD value is calculated according to the case of “not used ×”. As a result, 12 types × 2 MD values D ² (1) and D ² (2) are added as calculation results to the right end of the orthogonal table of FIG. From these MD values D ² (1) and D ² (2), the SN ratios η of 12 combinations are calculated using the following equation (8).

Here, n is the number of data (number of rows) of the factor effect target, and n = 2 in this example.
The results of the _twelve SN ratios η _{1 to} η 12 calculated by Expression (8) are added to the right end of the orthogonal table as shown in FIG. This completes the preparation for calculating the S / N ratio gain for each characteristic item (characteristic item 1 to characteristic item 5) of the factor effect.
Specifically, considering in conjunction with the characteristic items of the aircraft, for example, “characteristic item 1 = altitude”, “characteristic item 2 = ground speed”, “characteristic item 3 = atmospheric pressure”, “characteristic item 4 = temperature” It can be considered as “characteristic item 5 = left main wing distortion”.

  Hereinafter, the calculation of the S / N ratio and the factor analysis using the S / N ratio performed by the factor analysis unit 50 will be described in detail.

[Calculation of SN ratio]
As shown in the following equation, the SN ratio gains η _{c1 to} η _{c5 of the} characteristic items 1 to 5 obtained by the factor effect analysis in the abnormality diagnosis data file are the combinations (◯) of the characteristic items. This is the difference between the SN ratio and the SN ratio when the combination was not used (x).

  As a result, it can be determined that the characteristic value having a larger SN ratio gain is more likely to be involved in the abnormality. As the value of ηc (◯, x) to be substituted into the above formula, the average values of the SN ratio values in the auxiliary table list of FIG. 13 are used by using the calculated values in the orthogonal table of FIG.

[Factor analysis]
The factor analysis unit 50 can contribute to the cause of abnormality from a plurality of characteristic items in the diagnosis data file by quantifying the contribution rate of the factor effect to each characteristic item based on the gain of Expression (9). A characteristic item with high characteristics is selected, and the result of this factor effect is output to the notification unit 33. Thereby, the notification part 33 notifies the user of the analysis result of the factor analysis part 50.

  FIG. 11 is a diagram showing an example of a display screen showing a factor analysis result. It shows that a characteristic item having a large gain value can be a cause of occurrence of an abnormality detected this time.

  As described above, according to the aircraft health diagnosis apparatus and method and the program according to the present embodiment, when an abnormal state of the aircraft is detected, the characteristic items that are likely to cause the abnormality are displayed. It is possible to analyze and notify the analysis result to the user. Thereby, it is possible to promptly take an appropriate response to the occurrence of an abnormality.

  The factor analysis result obtained by the factor analysis unit 50 described above may be adopted for maintenance or after-sales service. As described above, by using the factor analysis result secondarily, it is possible to find a sign of abnormality, and thus it is possible to prevent occurrence of a serious abnormality such as replacement of a device in advance. As a result, it is possible to prevent a reduction in the operation efficiency of the aircraft due to the occurrence of an abnormality, and it is possible to reduce maintenance costs.

[Third Embodiment]
Next, an aircraft soundness diagnosis apparatus according to a third embodiment of the present invention will be described.
In each of the embodiments described above, a normal data file is used as the reference data file. In the present embodiment, instead of this, the above-described MD value is calculated by using an abnormal data file as the reference data file. And when this MD value is smaller than a predetermined threshold value, it determines with abnormality having generate | occur | produced.

  As described above, by using the abnormal data file as the reference data file, for example, by dividing the various abnormal / failure states in the unit space and calculating the MD value, the factor as in the second embodiment described above is obtained. Without performing the effect analysis, it is possible to easily specify which characteristic item diagnosis data indicates what kind of abnormality.

[Fourth Embodiment]
Next, an aircraft soundness diagnosis apparatus according to a fourth embodiment of the present invention will be described.
In the aircraft soundness diagnosis apparatus according to the present embodiment, the maintenance time and retirement time of the diagnosis target are determined based on the temporal transition of the MD value calculated by the index value calculation unit 31 in the first embodiment described above. . For example, as shown in FIG. 14, the horizontal axis represents the number of operations, the vertical axis represents the average MD value, and the temporal transition of the MD value is obtained for each aircraft ID. Thus, by taking the number of operations on the horizontal axis, it is possible to grasp the state of aging deterioration of each aircraft. In addition, a threshold value X indicating the overall inspection time and a threshold value Y indicating the retirement time are set, and the overall inspection is performed when the MD value exceeds the threshold value X, and the aircraft is disposed when the threshold value Y is exceeded. To do.

  Thus, by determining the maintenance time and the retirement time based on the MD value, it is possible to set an appropriate maintenance time or the like for each aircraft. In addition, since the way of receiving the load varies depending on the service route, the slope of the MD value varies depending on the service route. Therefore, for example, by applying to service routes that are windy and susceptible to load, the slope of the MD value increases abruptly. If it has decreased, the slope of the MD value is gentle, that is, it can be used efficiently until the aircraft reaches the end of its life by changing to a gentle route that is less susceptible to load. Become.

[Fifth Embodiment]
Next, an aircraft soundness diagnostic apparatus according to a fifth embodiment of the present invention will be described.
The aircraft soundness diagnosis apparatus according to this embodiment is different from the soundness diagnosis apparatuses according to the above-described embodiments in that the occurrence of soundness abnormality is detected without using the MT method.
Hereinafter, the soundness diagnosis apparatus according to the present embodiment will be described.

In the soundness diagnostic apparatus according to the present embodiment, a load measuring instrument that measures strain (load) is provided in a plurality of main parts of the airframe structure, and the ratio of the load measured by these load measuring instruments becomes a value in the initial stage. In comparison, when it changes more than a predetermined value, it is detected that an abnormality due to some damage has occurred in this part.
For example, as shown in FIG. 15, two target parts are set at the base part of the main wing, and the ratio of the load at these target parts is calculated. FIG. 16 shows a time series change of the load ratio A / B with respect to the operation time. The load ratio A / B at the beginning of the operation is set as a reference value, and it is determined that an abnormality has occurred when the load ratio has changed by a predetermined range or more from this reference.

  Further, as described above, when the occurrence of an abnormality is detected, the cause of the abnormality is specified using the MT method. For example, cracks, damage, peeling, etc. of the composite material can be considered as the cause of abnormality of the main wing. Therefore, such a damage form is artificially created, and load data at that time is acquired. Subsequently, using the data file composed of the load data relating to each damage mode acquired in this way, the “reference data file” in the MT method is constructed for each damage mode, and the actual load data is used. The above-mentioned “diagnostic data file” is constructed, and the MD value is obtained by comparing this diagnostic data file with the “reference data file” for each damage form. And in each damage form, the damage form when MD value shows the smallest value is specified as this damage form. Thus, the damage form can be easily specified by using the MD method.

  As mentioned above, although embodiment of this invention was explained in full detail with reference to drawings, the specific structure is not restricted to this embodiment, The design change etc. of the range which does not deviate from the summary of this invention are included.

It is the figure which showed the hardware constitutions of the soundness diagnostic apparatus of the aircraft which concerns on the 1st Embodiment of this invention. It is the functional block diagram which expanded and showed the function of the soundness diagnostic apparatus of the aircraft which concerns on the 1st Embodiment of this invention. It is a figure for demonstrating the measurement data file stored in a measurement database. It is the figure which showed an example of the diagnostic data file. It is the flowchart which showed the operation | movement flow of the soundness diagnostic apparatus of the aircraft in the 1st Embodiment of this invention. It is the flowchart which showed the procedure of the calculation process of a soundness index value. It is explanatory drawing for demonstrating each data used by the calculation process of Mahalanobis distance. It is the figure which showed an example of the evaluation result of a soundness index value. It is a figure for demonstrating the change of the reference space by the difference in an operation period. It is the functional block diagram which expanded and showed the function of the soundness diagnostic apparatus of the aircraft which concerns on the 2nd Embodiment of this invention. It is the figure which showed an example of the factor analysis result. It is the figure which showed an example of the 2-level orthogonal table used in the 2nd Embodiment of this invention. FIG. 13 is a diagram in which MD values and SN ratio gains are added to the two-level orthogonal table shown in FIG. 12. It is the figure which showed the time transition of MD value. It is the figure which showed an example of the object site | part which installs a load sensor. It is the figure shown about the abnormality detection which concerns on the soundness diagnostic apparatus of the aircraft which concerns on the 5th Embodiment of this invention.

Explanation of symbols

11 CPU
12 main storage device 13 auxiliary storage device 14 input device 15 output device 16 communication device 10 aircraft soundness diagnosis device 21 measurement database 22 flight database 23 data generation unit 24 class classification definition unit 25 class classification unit 26 first storage unit 27 normal Data condition definition unit 28 Normal data extraction unit 29 Second storage unit 30 Diagnostic file construction unit 31 Index value calculation unit 32 Abnormality determination unit 33 Notification unit 50 Factor analysis unit

Claims (11)

An aircraft health diagnostic apparatus for diagnosing the health of the aircraft using characteristic values created based on measurement data measured by a plurality of sensors provided in the aircraft,
When a plurality of characteristic values associated with the measurement time are stored for each characteristic item, and the characteristic values associated with the same measurement time are made into one data set, the data set corresponds to the flight status. First storage means to which identification information indicating the class classification determined in this manner is assigned and stored;
When a plurality of characteristic values associated with the measurement time are stored for each characteristic item, and the characteristic values associated with the same measurement time are set as one data set, the class classification is included in the data set. Second storage means to which identification information indicating is given, and a characteristic value of a specific characteristic item constituting the data set belongs to a predetermined reference range defined in advance,
A plurality of the data sets used for diagnosis are extracted from the first storage means to construct a diagnosis data file, and a plurality of the data sets to be compared with the diagnosis data file are stored in the class from the second storage means. A diagnostic file construction means for extracting and constructing a reference data file based on the classification;
Index value calculation means for calculating a health index value indicating the health of the aircraft using a statistical calculation method based on the diagnosed data file and the reference data file constructed by the diagnostic file construction unit; ,
Evaluation means for evaluating the soundness of the aircraft based on the soundness index value calculated by the index value calculation means;
An aircraft soundness diagnosis apparatus comprising: notification means for notifying an evaluation result by the evaluation means.   2. The aircraft health diagnosis device according to claim 1, wherein the class classification is determined based on a flight process when one flight pattern is divided into a plurality of processes.   The aircraft health diagnosis apparatus according to claim 1, wherein the classification is determined according to an operation period or the number of flights.   The aircraft health diagnosis apparatus according to claim 1, wherein the classification is determined according to a service route.   The plurality of data sets stored in the second storage means are reference ranges in which characteristic values relating to the specific characteristic items are preset among the plurality of data sets stored in the first storage means. The aircraft health diagnostic apparatus according to any one of claims 1 to 4, wherein only the data set belonging to is extracted.   The index value calculation means obtains the characteristic distribution of the reference data file set by the diagnostic file construction means, obtains the characteristic distribution of the diagnostic data file, and quantifies the distance between the characteristic distributions of each other The aircraft health diagnostic apparatus according to any one of claims 1 to 5, wherein the health index value is calculated by automatically obtaining the value.   The aircraft health diagnostic apparatus according to claim 6, wherein the health index value calculated by the index value calculation means is a Mahalanobis distance calculated using a Mahalanobis Taguchi method.   The aircraft soundness diagnosis apparatus according to any one of claims 1 to 7, further comprising factor analysis means for performing a factor analysis of an abnormality when the evaluation means evaluates that an abnormality has occurred.   The aircraft health diagnostic apparatus according to any one of claims 1 to 8, wherein an aircraft maintenance time is determined based on a transition of the health index value calculated by the index value calculation means. An aircraft health diagnostic method for diagnosing the aircraft health using characteristic values created based on measurement data measured by a plurality of sensors provided in the aircraft,
Creating a first data file in which a plurality of characteristic values associated with the measurement time are stored for each characteristic item;
In the first data file, when characteristic values associated with the same measurement time are set as one data set, identification information indicating class classification determined according to the flight status is given to the data set. When,
A process of creating a second data file in which characteristic values relating to specific characteristic items belong to a predetermined reference range that is defined in advance, and characteristic values of each characteristic item are associated with measurement time;
In the second data file, when characteristic values associated with the same measurement time are set as one data set, a process of giving identification information indicating the class classification to the data set;
A plurality of the data sets used for diagnosis are extracted from the first data file to construct a diagnosis data file, and a plurality of the data sets to be compared with the diagnosis data file from the second data file are defined in the class. Extracting based on the classification and building a reference data file,
A process of calculating a health index value indicating the health of the aircraft using a statistical calculation method based on the constructed diagnostic data file and the reference data file;
A process of evaluating the health of the aircraft based on the health index value;
A method for diagnosing aircraft health, comprising: notifying the evaluation result. An aircraft health diagnosis program used for diagnosing the health of the aircraft, using characteristic values created based on measurement data measured by a plurality of sensors provided in the aircraft,
Processing to create a first data file in which a plurality of characteristic values associated with the measurement time are stored for each characteristic item;
In the first data file, when the characteristic values associated with the same measurement time are set as one data set, a process for assigning identification information indicating a class classification determined according to the flight status to the data set When,
A process of creating a second data file in which the characteristic values related to a specific characteristic item belong to a predetermined reference range defined in advance and the characteristic values of each characteristic item are associated with the measurement time;
In the second data file, when the characteristic values associated with the same measurement time are set as one data set, a process of giving identification information indicating the class classification to the data set;
A plurality of the data sets used for diagnosis are extracted from the first data file to construct a diagnosis data file, and a plurality of the data sets to be compared with the diagnosis data file from the second data file are defined in the class. Extracting based on the classification and building a reference data file;
Based on the constructed diagnostic data file and the reference data file, a process of calculating a health index value indicating the health of the aircraft using a statistical calculation method;
A process for evaluating the soundness of the aircraft based on the soundness index value;
An aircraft health diagnosis program for causing a computer to execute the process of notifying the evaluation result.

Images