The Boeing Company

John M Boggio

Eric Lee Nicks

Alexander Stephen Burch

Kenneth D Bouvier

Stephen P Jewett

Liessman E Sturlaugson

Jeffrey Alan Schmitz

Clinton E Thompson

Examples for relational time-series classification method for corrosion maintenance dispatch. A method may involve generating a self-organizing map for each parameter that corresponds to a group of aircraft. The self-organizing map organizes time-series data associated with a parameter into a two-dimensional representation. The method may further involve, for each map, identifying polar coordinates that represent a location of a particular node, in a grid of nodes within the self-organizing map, that is located closest to the time-series data organized in the two-dimensional representation. The method may then involve providing polar coordinates identified from each self-organizing map, along with indications of corrosion levels corresponding to the aircraft, as inputs into a classification model to train the classification model to determine trends between the parameters. The classification model is configured to subsequently use the one or more trends collectively to estimate a corrosion level for a particular aircraft.

Relational time-series classification method and system for corrosion maintenance dispatch

A vehicle fault detection system including at least one sensor configured for coupling with a vehicle system, a vehicle control module coupled to the at least one sensor, and being configured to receive at least one time series of numerical sensor data from the at least one sensor, at least one of the at least one time series of numerical sensor data corresponds to a respective system parameter of the vehicle system being monitored, generate a graphical representation for the at least one time series of numerical sensor data to form an analysis image of at least one system parameter, and detect anomalous behavior of a component of the vehicle system based on the analysis image, and a user interface coupled to the vehicle control module, the user interface being configured to present to an operator an indication of the anomalous behavior for the component of the vehicle system.

Vehicle fault detection system and method utilizing graphically converted temporal data

An apparatus for maintenance of an aircraft is provided. The apparatus accesses flight data for a flight of the aircraft and predicts and thereby produces a predicted state of the aircraft for a next flight of the aircraft based on the flight data. The apparatus deploys a similarity detection model to determine similarities between the flight data for the flight and a training set of flight data for previous flights of the aircraft. The similarities determined by the similarity detection model are values of independent variables of a classification model. The apparatus deploys the classification model to predict and thereby produce a predicted state of the aircraft for the next flight of the aircraft based on the values of the independent variables, and generates an alert when the predicted state of the aircraft is a fault state.

Maintenance of an aircraft via similarity detection and modeling

A vehicle fault detection system including a vehicle control module coupled to a plurality of vehicle system sensors configured to detect respective time series of data corresponding to a component parameter, the vehicle control module determines, based on domain knowledge obtained by the vehicle control module from only the respective time series of data, an existence of one or more relationships between the plurality of component parameters, and identifies an anomaly in the respective time series of data and at least a vehicle component to be serviced based only on the domain knowledge and the one or more relationships between the plurality of component parameters. An indication of the anomaly in the respective time series of data and an identification of the vehicle component to be serviced are to be presented as a graphical representation that includes a strength of the one or more relationships between the component parameters.

Unsupervised multivariate relational fault detection system for a vehicle and method therefor

Techniques are described for analyzing text. Embodiments tokenize a plurality of documents into a plurality of sets of terms. An average top dimension weight corresponding to the plurality of documents is calculated based on performing singular value decomposition (SVD) factorization for a plurality of dimension counts. An average inverse top dimension top term ranking for the plurality of documents is further calculated based on the SVD factorization for the plurality of dimension counts. A number of dimensions is determined based on the average top dimension weight and the average inverse top dimension top term ranking. A topic model is built for the plurality of documents based on the number of dimensions. The topic model is adapted to identify patterns of terms that correspond to semantic topics in at least the plurality of documents.

Dimension optimization in singular value decomposition-based topic models

An apparatus for predicting structural degradation and performing maintenance induction for a plurality of in-service aircraft is provided. The apparatus identifies maintenance requirements for the plurality of in-service aircraft and receives fatigue metric values on historical usage and structural health condition of the plurality of in-service aircraft. The apparatus predicts levels of structural degradation to the plurality of in-service aircraft based on the fatigue metric values and assigns maintenance priorities to the plurality of in-service aircraft based on the levels of structural degradation. The apparatus generates an instruction to route an aircraft of the plurality of in-service aircraft to a maintenance facility for maintenance based on a maintenance priority of the maintenance priorities assigned to the aircraft.

Maintenance induction for aircraft

A vehicle includes a first system, a first sensor, a second sensor, and a processor. The first sensor is configured to generate first sensor data indicative of an operational status of the first system. The second sensor is configured to generate second sensor data associated with an adaptive performance metric adjustment criterion. The processor is configured to, in response to determining that the second sensor data satisfies the adaptive performance metric adjustment criterion, determine, based on the second sensor data, an adaptive performance metric to be compared to the first sensor data to identify anomalous behavior of the first system. The processor is further configured to compare the first sensor data to the adaptive performance metric and to generate a maintenance alert in response to determining that the first sensor data fails to satisfy the adaptive performance metric.

Vehicle anomalous behavior detection

A vehicle fault detection system including a plurality of sensors coupled to a vehicle system and configured to detect respective time series of data, where each time series of data corresponds to a component parameter of a plurality of component parameters, a vehicle control module coupled to the plurality of sensors, the vehicle control module being configured to determine an existence of one or more relationships between the plurality of component parameters, and identify anomalies in the respective time series of data based on the one or more relationships between the plurality of component parameters, and a user interface coupled to the vehicle control module, the user interface being configured to present to an operator an indication of the anomalies in the respective time series of data.

Techniques are described for optimizing a number of dimensions for performing a singular value decomposition (SVD) factorization. Embodiments tokenize each of a plurality of documents into a respective set of terms. For each of a plurality of dimension counts, embodiments perform the SVD factorization to determine a respective plurality of dimensions, the respective plurality of dimensions corresponding to the dimension count, determine, for each of the plurality of documents, a respective set of dimension weights for each of the plurality of dimensions, calculate an average top dimension weight across the sets of dimension weights for the plurality of documents and calculate an average inverse top dimension top term ranking across the sets of dimension weights for the plurality of documents. An optimal number of dimensions is calculated, based on the average top dimension weight and the average inverse top dimension top term ranking.

A method of performing maintenance. Data is received from a plurality of physical sensors that measure parameters of parts of the platform, wherein a data point is missing from the data. A plurality of different polynomial interpolation values is calculated for the missing data point using a plurality of different polynomial interpolation formulas. A relative strength for each of the plurality of different polynomial interpolation values is calculated, wherein relative strengths are determined. A best polynomial interpolation value is selected based on a highest strength of the relative strengths. The best polynomial interpolation value is substituted for the data point that is missing from the data, wherein filled data is created. A status of the parts is determined based on the filled data, wherein an analysis is formed. Responsive to the analysis indicating that maintenance is due for a part in the parts, causing maintenance to be performed.

Methods and devices for adaptive autonomous polynomial interpolation of time series data

A method of evaluating health of a control system with fluctuating parameters is provided. The method may include receiving session data including actual parameters and target parameters of the control system, segmenting the session data into one or more data segments, calculating a segment score for each data segment based on at least a linear regression correlation coefficient and an average parameter difference, generating a session score based on a combination of the segment scores, and generating a notification indicative of a comparison between the session score and a predefined threshold.

Growing community of inventors

Columbia, IL, United States of America

John M Boggio