Using Machine Learning for Pump Diagnostics


Using simulation models to develop an ultra-precise pump, using simulations to generate training data for neural networks, use machine learning to diagnose pumps? If that makes you want to know more you should read the article that was jointly written by Controllab and Spark Holland. It was published last Friday in Mikroniek.

Spark wanted to use machine learning to diagnose its high-performance liquid-chromatography (UHPLC) pumps by analyzing. Controllab helped by training a neural network with artificially generated data sets. This technique was so succesful that it is now implemented in the diagnostics software of Spark. A joint patent has successfully been granted in the US and is now pending in Europe.

This technique is very useful for machines and applications with small sales volumes, i.e. where neural networks cannot be trained because of limited data. If you are interested in how machine learning can be used for your machines please click the link below to contact us.