Aircraft Structural Load Estimation via Neural Network Impact Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current systems for detecting hard landings in aircraft are inefficient due to the need for additional sensors, high costs, complex integration, and inability to accurately measure stresses at specific locations, leading to unnecessary maintenance and inspections.

Innovation Solution

A method using neural networks to estimate load criteria on structural components by determining the instant of impact and calculating load criteria from measured parameters, allowing for precise stress estimation without additional sensors, and providing threshold comparisons for inspection decisions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If specific sensors are incorporated to measure loads at various locations, then measurement precision is improved, but device complexity and cost increase

Engineering Contradiction:
Improveload measurement precisionVSAvoidsensor integration complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent uses a neural network to create a virtual model (copy) of the structural component's stress state based on data from conventional sensors. Instead of physically measuring stress at multiple locations with specialized sensors, the system copies the behavior of stress sensors through software modeling, eliminating the need for complex physical sensor integration while maintaining measurement capability

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent replaces the mechanical sensor system with a computational model. The neural network substitutes physical stress sensors by processing data from conventional acceleration and load sensors to calculate stress values, thereby eliminating the need for complex mechanical sensor integration in inaccessible regions

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Device complexity

If conventional sensors are used alone, then device complexity is reduced, but measurement precision of structural stresses deteriorates

Engineering Contradiction:
Improvesensor system complexityVSAvoidstructural stress measurement precision
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent introduces a neural network as an intermediary between conventional sensors and stress measurement. The neural network acts as a mediator that transforms data from simple acceleration and load sensors into accurate stress estimates, enabling precise structural stress measurement without requiring complex sensor systems

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent transforms the type of parameters being measured by using the neural network to convert raw sensor data (acceleration, load) into derived parameters (stress values at specific locations). This parameter transformation allows conventional sensors to provide precise stress measurement information through computational processing

Inventive Principle:
Principle #35Parameter changes

3Productivity

If peak values of measured parameters are combined, then productivity is improved, but measurement precision deteriorates due to timing mismatches

Engineering Contradiction:
Improvehard landing detection speedVSAvoidstress estimation accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent applies preliminary action by training the neural network offline before actual use. During training, the system learns the correct temporal relationships between different sensor parameters and stress responses. This pre-learning ensures that during real-time operation, the network can accurately combine parameters without timing mismatches, maintaining both speed and precision

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP2279466B1Estimating a loading criterion for the load experienced by a structural component of an aircraft, and assistance with detection of a so-called hard landing using such a criterion
Publication Date: 2018.04.18 AIRBUS OPERATIONS (SAS)
  • EP2279466B1 patent drawingFigure 1~2
  • EP2279466B1 patent drawingFigure 3~5
  • EP2279466B1 patent drawingFigure 6~7

AI summary

The present invention relates to methods and systems for estimating a loading criterion relating to the load (C) experienced by a structural component (100, 102, 104) of an aircraft (10), and assistance with detecting a so-called “hard” landing. These methods involve measuring (50) parameters of said aircraft and calculating (54) at least one loading criterion for the loading of said structural component using at least one neural network (36) receiving said parameters as input. Assistance with detecting a hard landing then requires the determining (72, 76) of a time (t0, t-i) of impact (14, 16) of said aircraft on a landing strip (12) from said measured parameters, then estimating (74, 78) a plurality of said parameters at said determined time of impact so as to calculate (54) the at least one loading criterion relating to the loading of the structural component.