Aircraft Engine Neuromorphic Sensing for Low-Data Anomaly Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional synchronous sensors in aircraft engines generate large volumes of data at high frequencies, posing challenges in efficient analysis and storage, particularly on aircraft with limited data processing and storage capabilities.

Innovation Solution

Implementing neuromorphic sensors that asynchronously report changes in data characteristics, allowing the data acquisition system to activate additional sensors only when events warranting further data collection are detected, thereby reducing unnecessary data processing and storage during nominal operating conditions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional synchronous sensors are used to monitor aircraft engine parameters, then continuous monitoring coverage is achieved, but large volumes of data are generated requiring significant storage and processing resources

Engineering Contradiction:
Improvecontinuous monitoring coverageVSAvoiddata volume
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent extracts only the meaningful changes from the continuous sensor data stream. The neuromorphic sensor selectively reports only when parameter changes exceed a threshold, extracting the essential information while discarding redundant data points that occur during stable operating conditions.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system implements event-driven periodic sampling where additional sensors are activated only when changes are detected by the neuromorphic sensor. This creates an on-demand sampling regime that replaces continuous synchronous sampling, reducing data volume while maintaining monitoring reliability.

Inventive Principle:
Principle #19Periodic action

2Loss of information

If synchronous sensors report data at predetermined intervals, then complete data records are maintained, but unnecessary data is generated during nominal operating conditions

Engineering Contradiction:
Improvedata completenessVSAvoidprocessing energy
Core Design Contradiction:
Loss of informationVSLoss of energy

Solution Approach 1:

The system applies partial action by activating additional sensors only when necessary - specifically when the neuromorphic sensor detects a change exceeding the threshold. During nominal conditions, only the neuromorphic sensor operates, while during events, additional sensors are activated to capture complete data, thus avoiding excessive data collection during stable periods.

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The neuromorphic sensor performs preliminary monitoring of all parameters continuously with minimal resource usage. When it detects an event, it triggers the activation of additional sensors, ensuring that complete data is captured at the right moment without requiring all sensors to run continuously.

Inventive Principle:
Principle #10Preliminary action

3Reliability

If all sensors are activated continuously, then comprehensive data is collected for all parameters, but data storage requirements increase significantly

Engineering Contradiction:
Improveanomaly detection capabilityVSAvoiddata storage volume
Core Design Contradiction:
ReliabilityVSVolume of stationary object

Solution Approach 1:

The monitoring system is segmented into two operational modes: a low-power surveillance mode using only the neuromorphic sensor, and a high-data-collection mode using all sensors. This segmentation allows the system to maintain anomaly detection capability while dramatically reducing average data storage requirements by activating full sensor arrays only when anomalies are detected.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different sensors are activated with different qualities based on local conditions. The neuromorphic sensor operates continuously with high sensitivity to detect events, while other sensors are activated locally only when and where needed - specifically when event detection occurs - thus optimizing storage usage while maintaining detection reliability.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP4488781A1Data acquisiton for aircraft engines using neuromorphic sensors
Publication Date: 2025.01.08 RTX CORP
  • EP4488781A1 patent drawingFigure 1
  • EP4488781A1 patent drawingFigure 2
  • EP4488781A1 patent drawingFigure 3

AI summary

Embodiments of the present disclosure generally relate to aircraft engines (20) and, more particularly to data acquisition for aircraft engines (20) using neuromorphic sensors (206). In some embodiments, an event associated with the aircraft engine (20) may be identified based on a change in a data characteristic measured from a neuromorphic sensor (206) and, in response to identifying the event associated with the aircraft engine (20), at least one other sensor (206) coupled to the aircraft engine (20) may be activated. Other embodiments may be disclosed or claimed.