Design of a data acquisition system to be used in fault diagnosis

In Machine learning, the availability of reliable datasets to be used by training algorithms is a widely posed problem. In this perspective, this work represents a design of a data acquisition system that allows the collection of data from a real world industrial machine (Direct Current motor machines). The goal of this data collection is the construction of a fault diagnosing tool by using heterogeneous data. Those heterogeneous data are collected from different types of sensors measuring different types of variables that are directly related to the industrial system. Owing to this data collection, one can build machine learning models such as Bayesian networks, Artificial Neural Networks, etc. Those models can be used in fault detection, diagnosis and prognosis.