Flying Through a Turbine Blade

Turbine blades are complicated and delicate pieces of engineering. Each blade has cooling channels that allow air to flow through and exit the porous shell of the blade, maintaining a stable temperature for the assembly. Due to their nature, each turbine blade requires a series of rigorous testing procedures in order to verify their structural integrity. Visual inspection is simple matter but capturing the integral structure is no easy task.

This is where neutron imaging can play a vital role. Neutron imaging is the process of making an image with neutrons. The resulting image is based on the neutron attenuation properties of the imaged object. The resulting images have much in common with industrial X-ray images, but since the image is based on neutron attenuating properties instead of X-ray attenuation properties, some things easily visible with neutron imaging may be very challenging or impossible to see with X-ray imaging techniques

Researchers from Oak Ridge National Laboratory (ORNL)’s Spallation Neutron Source (SNS) have recently utilized neutron imaging to examine the cooling channels and other inner workings of Inconel 718 turbine blades. Watch the video.

Historical Primer on the Algorithms, and Techniques that are shaping our world

giants of machine learningLast year in April I attended a good course of machine learning which pushed my self learning to a new level. The course also introduced the language R and I have loved dipping with machine learning since then though most of it is done in python.

We had an excellent trainer with more than 20 years of experience in the field. I loved his 101 course in statistics as the prelude to the course and the practical examples that the trainer brought up while teaching us the mechanics of perceptrons, ridge regression etc.

What the course lacked was an historical perspective on machine learning and how it all came about as we now know it.

Found this excellent video that provides that perspective.

From the first meeting in summer of 1956 to the many A.I winters followed by the eventual emergence of deep learning, the below video provides an excellent historical primer on the algorithms, and techniques that are shaping our world.


What can Software Teams Learn From Aviation?

This pycon talk combines my two passions, aviation and software development. Grab a cold coffee and watch this very interesting talk. Just a little over half an hour, if you have any interest in any of this two stream, you will like the talk.

What can Python-based software teams learn from aviation? Why should software always fail hard? What’s wrong with too many error logs? And why are ops people already like pilots? Learn all this, and about planes, too.

If in hurry, you can download the PDF of the presentation.