Controlling Random Test Kurtosis

Back to: Random Testing

Previous lessons introduced the statistics of kurtosis. Kurtosis is particularly important in random vibration control because it points to the signal’s most damaging peaks.

A normal, or Gaussian, distribution has a kurtosis of 3, where the highest peak accelerations are approximately ±3 times the average. Most vibration controllers assume a Gaussian distribution of data. However, real-world data are not always Gaussian, and a traditional random profile may under-test a product. Engineers want to replicate their product’s operational environment in the laboratory as closely as possible. How can they make their random vibration tests more realistic?

The Solution: Kurtosion

Kurtosion® is a patented method of random kurtosis control. According to the article, When Should I Use Random Kurtosis Control?, engineers should use random kurtosis control when:

• The random vibration test profile is missing peak acceleration levels found in the operational environment
• The engineer wants to break a product quickly without using more energy

Webinar (30:38)

If you’d prefer a briefer introduction, watch the following quick tip video (6:28).

If you are interested in taking a deeper dive into the technical details of kurtosis, please refer to the following technical papers.

The Third Dimension of Random Vibration Control: Learn about the development of kurtosis control in random testing.

Getting Kurtosis in the Resonances: Review a series of tests that validate the Kurtosion method.

Using Kurtosion to Accelerate Structural Life Testing: Learn how kurtosis control can accelerate random vibration tests.