NDE encompasses many sensor modalities and engineering capabilities.
NDE encompasses many sensor modalities and engineering capabilities. We combine experience and experimental capability to optimize for each application using models and simulations to support experimental design, to better understand signal/material interactions, and to create test specimens that represent fielded components and anticipated flaws. By incorporating automation throughout the inspection process, we increase NDE reliability, repeatability, and throughput and reduce cost.

Nondestructive evaluation and mechanical testing

Our Nondestructive Evaluation (NDE) researchers provide expertise in development, enhancement, and integration of advanced NDE and characterization technologies for aerospace materials and their properties.

Our researchers bring expertise in materials characterization, processing, mechanical testing, and systems design to every application. In addition, our researchers have extensive experience in NDE-related areas including robotics and automation, statistical and inversion methods, image and signal processing, as well as modeling and simulations. Our perspective on NDE methods is particularly unique since our researchers and laboratory include state-of-the-art destructive characterization capabilities including mechanical testing (high temperature and vacuum) and metallographic capability that can be used to correlate NDI data with actual flaw parameters.

UDRI’s NDE Engineering experience in robotics and automation provides the capability to enhance existing experimental systems, as well as design the next generation of automated systems for laboratory, industrial or depot use. Our experience in image and signal processing includes a variety of data formats obtained from NDE sensors and imaging systems. This broad experience base provides our customers with a research team that can assist with identifying technological advances in optical imaging and sensor signal processing that are applicable for NDE sensing. Our experience in modeling and simulations of NDE measurements and related behavior provides a research team that understands the benefits and limitations provided by model-based experimentation, as well as how experiments can be best used to validate model predictions. UDRI’s experience with statistical methods ensures that our NDE research focuses on quantitative outputs and imparts confidence in those results. All of these supporting areas of expertise provide an NDE solution that is flexible and responsive to future NDE needs and innovation requirements.

Capabilities

Nondestructive Evaluation

  • Eddy current
  • X-ray computed tomography
  • Thermography
  • Ultrasound/phased array
  • Resonance methods
  • Metallography/optical microscopy
  • Surface profiling/roughness

Mechanical Testing

  • Monotonic, fatigue, sustained loading
  • Up to 1500⁰C
  • Vacuum testing to 10-9 torr
  • Thermal gradients

Integration Support

  • Automation
  • Systems development
  • Statistical methods
  • Machine learning
  • Image/signal processing

UDRI Developed Software

Probability of Detection (POD) Software v4.5

UDRI's nondestructive evaluation engineering experts have been developing software for the the analysis of probability of detection (POD) data since the 1990s. Over the past three decades our POD analysis software has become a critical tool for both UDRI and our external customers that has generated countless POD estimates for engineers managing small- and large-scale POD studies and technicians performing individual inspections.

This latest version, developed in collaboration with the Air Force Research Laboratory (AFRL), introduces modern and validated statistical approaches to POD analysis.

New to POD v4.5

  • New Logistic Regression Technique: Firth's Bias-Reduced
  • New Confidence Interval Techniques: Modified Wald, Likelihood Ratio confidence interval, and Modified Likelihood Ratio Confidence Interval
  • New Sampling Methods: Random and Ranked Set
  • New Transformation Method: Box-Cox

Key Features

  • Complete analysis by following the simple wizard style interface guide
  • Import data quickly from multiple Excel files using Copy & Paste
  • Preview all imported data before creating your analyses
  • Export any chart to an image file
  • Export both data and charts to Excel files to easily share with others
  • Track multiple analyses in a single POD project file
  • Provides both full analysis and quick analysis options
  • Active feedback to guide adherence to statistical assumptions
  • Perform hit/miss or signal response analysis
  • New Confidence Interval Techniques: Modified Wald, Likelihood Ratio confidence interval, and Modified Likelihood Ratio Confidence Interval
  • New Sampling Methods: Random and Ranked Set
  • New Transformation Method: Box-Cox
POD Software v4.5 screenshot
A Hat vs. A Analysis

POD v4.5 includes 4 side charts to help understand and visualize crack size responses while performing POD analyses.

 

POD Software v4.5 screenshot
Hit Miss Analysis

POD v4.5 supports multiple combinations of confidence interval types, sampling types and POD models to improve analysis for more challenging data sets.