DiagFit
Published by
Amiral Technologies
Features
Supervised and unsupervised learning
When no historical data is available, DiagFit learns over machines in a normal working condition. We predict equipment failures without historical failure data.
Automatic generation of features or state of health indicators
This is an algorithmic innovation that transforms a time series into a set of discrete, rich and discriminant features quick to calculate. Specificity of this transformation resides in the fact that it preserves information contained in the time series and reveals transitory phenomenon that express the equipment state of health.
Generic industrial predictive failure algorithms
DiagFit provides a set of Machine Learning-based models specifically designed to answer the following industrial problems:
Detection of defects and failure prediction
Detection of near end-of-life of an equipment
Estimation of the remaining useful life of an equipment
Supervised and unsupervised learning
When no historical data is available, DiagFit learns over machines in a normal working condition. We predict equipment failures without historical failure data.
Automatic generation of features or state of health indicators
This is an algorithmic innovation that transforms a time series into a set of discrete, rich and discriminant features quick to calculate. Specificity of this transformation resides in the fact that it preserves information contained in the time series and reveals transitory phenomenon that express the equipment state of health.
Generic industrial predictive failure algorithms
DiagFit provides a set of Machine Learning-based models specifically designed to answer the following industrial problems:
Detection of defects and failure prediction
Detection of near end-of-life of an equipment
Estimation of the remaining useful life of an equipment