Rotor Load Determination via Low-Rank Matrix Data Reconstruction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for determining helicopter rotor loads and motion are inadequate in accurately reconstructing and validating sensor data, particularly in the presence of missing or corrupted data during wireless transmission, which hinders usage-based maintenance, structural health monitoring, and individual blade control.

Innovation Solution

The system employs numerical analysis for low-rank matrices using principal component pursuit, matrix completion, and nuclear-norm regularized multivariate linear regression to reconstruct and correct sensor data, leveraging the periodic and correlated nature of rotor system data to isolate faults and provide accurate load and motion information.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If wireless transmission is used for sensor data collection, then ease of operation and data collection efficiency are improved, but data loss and corruption occur during transmission

Engineering Contradiction:
Improvesensor data collectionVSAvoiddata transmission
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The patent introduces an intermediary validation system that receives sensor data through wireless transmission, validates its integrity using correlation analysis and periodicity checks, and requests retransmission when data corruption or loss is detected. This intermediary layer protects the reliability of data transmission while maintaining the ease of wireless data collection.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If more sensors are deployed for comprehensive rotor monitoring, then measurement precision and monitoring accuracy are improved, but device complexity and data processing burden increase

Engineering Contradiction:
Improve rotor loads and motionVSAvoidsensor system
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent merges data from multiple sensors into a unified analysis framework that exploits the inherent correlations and periodicities in rotor system data. By combining sensor readings and analyzing them collectively using validation algorithms, the system achieves high measurement precision without proportionally increasing device complexity, as the same validation infrastructure serves all sensors.

Inventive Principle:
Principle #5Merging (Combining)

3Measurement precision

If real-time validation and reconstruction algorithms are implemented, then data accuracy and fault detection capability are improved, but computational requirements and processing time increase

Engineering Contradiction:
Improvesensor dataVSAvoidcomputational processing
Core Design Contradiction:
Measurement precisionVSPower

Solution Approach 1:

The patent implements partial validation by focusing computational resources on detecting and correcting the most critical data issues (corruption and loss) using targeted algorithms that check for periodicity and correlations. Rather than performing exhaustive analysis on every data point, the system applies validation selectively to maintain data accuracy while managing computational power consumption.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP2662741B1System and method of determining rotor loads and motion
Publication Date: 2020.01.22 SIKORSKY AIRCRAFT CORP
  • EP2662741B1 patent drawingFigure 1
  • EP2662741B1 patent drawingFigure 2
  • EP2662741B1 patent drawingFigure 3

AI summary

A system for reconstructing sensor data in a rotor system (1) that comprises a rotating component (10) of the rotor system, a plurality of sensors (12) in the rotating component (10) to sense at least one of loads and motion characteristics in the rotating component (10) and to generate sensor data, and an analysis unit (15) to generate reconstructed sensor data from the sensor data using numerical analysis for low-rank matrices.