Aircraft Structure Diagnostics Using Displacement Sensors and Neural Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for diagnosing damage or defects in aircraft structures are laborious and imprecise, failing to reflect real-time physical and mechanical conditions, and require complex mathematical modeling.
Innovation Solution
A method utilizing displacement sensors and neural networks to correlate real-time displacement data with predicted displacements, allowing for continuous estimation of a structure's condition without the need for a physical/mathematical model, enabling on-board diagnostics during service or missions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If historical reconstruction and fatigue estimation methods are used to diagnose damage, then diagnostic coverage is provided, but measurement precision and real-time accuracy deteriorate
Solution Approach 1:
The patent replaces traditional mechanical/mathematical modeling approaches (finite element models, load calculations) with a neural network-based system. The neural network learns the relationship between sensor inputs and structural state directly from training data, substituting complex mechanical calculations with a data-driven model that provides real-time diagnostic accuracy without requiring detailed knowledge of loading conditions or material properties.
Solution Approach 2:
The patent implements a preliminary training phase where the neural network is trained offline using historical data and finite element models. This preliminary action allows the system to learn the complex relationships between sensor measurements and structural damage patterns before actual deployment, enabling real-time diagnostics without requiring complex calculations during operation.
2Measurement precision
If complex mathematical models and finite element analysis are used, then diagnostic accuracy is improved, but device complexity and computational requirements worsen
Solution Approach 1:
The patent extracts the essential diagnostic functionality from complex finite element models by training a neural network on representative data. The neural network captures the critical damage detection patterns without requiring the full computational infrastructure of finite element analysis during operation, effectively separating the complex modeling phase (training) from the simple execution phase (real-time diagnosis).
Solution Approach 2:
The patent creates a simplified copy of the complex structural system through the neural network. Instead of directly analyzing the actual structure using complex models, the neural network learns a representation of the structure's behavior from training data and uses this copied knowledge for rapid real-time diagnosis, avoiding the computational burden of repeated finite element analyses.
3Measurement precision
If real-time monitoring with multiple sensors is implemented, then measurement precision improves, but device complexity and cost worsen
Solution Approach 1:
The patent makes the sensor system universal by training the neural network to diagnose multiple types of damage (delamination, impact damage, matrix cracking) and evaluate the structural state of different aircraft components using the same sensor arrangement and processing methodology. This multi-functional capability reduces the need for specialized diagnostic systems for different damage types.
Solution Approach 2:
The patent enables the structure to self-diagnose by using embedded sensors and the neural network system that automatically processes sensor data and identifies damage. The system continuously monitors its own structural health without requiring external inspection, providing real-time self-assessment of the structural condition.
Data Source
AI summary
A method for performing diagnostics of a structure subject to loads, in particular an aircraft structure, is described, said method being implemented by means of an arrangement of sensors located at relevant points of the structure and corresponding neural networks, and comprising: training the neural network in order to establish an associative relationship between the local displacement of the structure in a subset of relevant points and the local displacement of the structure in at least one residual relevant point; detecting the local displacement of the structure in a plurality of relevant points under operating conditions; estimating the local displacement of the structure in at least one residual relevant point by means of the associated neural network on the basis of the pre-established associated relationship; and comparing the local displacement of the estimated structure with the detected local displacement at the residual relevant point.