Rockwell Collins, Inc.

Christopher A Hohensee

Daniel J Kaplan

Jason R Owen

Angelo J Joseph

Yeshani D WijesekaraGamachchige

Andrea P Gonzalez

Nels Waineo

Christopher Kistler

Yeshani D Wijesekara Gamachchige

Chadwick K J Harvey

Andrea Gonzalez

A predictive maintenance system is disclosed. The system includes a network of analog and digital sensors, each sensor configured for measuring telemetry data associated with temperature levels, voltage levels, current levels, and other analog or digital parameters. The system includes microprocessors for receiving the (digitized) analog and digital telemetry data, tabulating and timestamping the raw telemetry datasets. The microprocessors compress the raw data and reduce its dimensionality by generating principal component sets from the raw data based on scalar parameters corresponding to machine learning algorithms stored to memory, the principal component sets capturing a majority of variances within the raw data. The principal component sets are organized into data packets including identifiers for the relevant algorithms. The data packets are transmitted via real time networks for either onboard storage or ground-based analysis.

Predictive maintenance system using avionics ethernet networks

A system for time synchronization over redundant switch-based avionics networks is disclosed. The system includes a master or source clock for determining precise UTC timing information from received satellite signals and generating time marks based on the timing information. The source clock generates network-compatible timing messages and forwards the timing messages to network switches within the switch-based avionics networks. The network switches modify the timing information to account for switch-based delays and forward the modified timing messages to destination clocks in aircraft end systems. The end systems relay timing messages back to the source clock via the network switches, the timing information again modified by the network switches according to switch-based delays, and based on the precise timing information exchanged destination clocks in end systems throughout the switched network can precisely synchronize to the source clock.

Time synchronization over redundant and deterministic switch-based avionics networks

A predictive maintenance system for an aircraft-based power converter subsystem (PCS) continuously measures primary system parameters such as input and output currents and voltages, ambient and PCS component temperatures, and hold-up capacitance voltage. The predictive maintenance system updates the likely failure rate of the PCS based on the primary parameters and secondary parameters (e.g., power consumption, power efficiency, thermal resistance, hold-up storage capacitance) derived therefrom. Sensed primary parameters and derived secondary parameters, as well as the updated failure rate, are stored to memory in order to monitor their time derivatives and regression over the PCS lifespan. Should the updated failure rate ever reach or exceed an acceptable failure rate threshold, the predictive maintenance system alerts the aircraft control system, end user, or provider to service the PCS in advance of an unanticipated field failure.

Predictive maintenance system for aviation power supplies

Growing community of inventors

Palm Bay, FL, United States of America

Christopher A Hohensee