Industrial Big Data Analysis

Signal processing and machine learning have been two of the most used and matured fields of data science. They are highly interdisciplinary fields building upon ideas from many different kinds of fields such as artificial intelligence, optimization theory, information theory, statistics, control theory, and many other disciplines of science, engineering, and mathematics. Because of their implementation in a wide range of applications, machine learning has covered almost every scientific domain, which has brought great impact on the science and society. However, the majority of the presented and used algorithms do not design to fit with big data. On the other hand, big data is increasing its impact on all fields of science and technology. It

The massive volumes of data that we are currently dealing with has required scientists to rethink storage and processing paradigms in order to develop the tools needed to properly analyze it. For example, most of the existing manufacturing and production plants consists of sensors, process control systems, network devices and protocols practically supporting limited amount of data volume transmission. In addition, wireless sensor networks, which attracts lots of attention in industries have limited bound with and power sources. Consequently, these limitations should take into account in order to provide some suitable solutions for these

This is because Industrial Big Data is generated by automated equipment and processes, where the environment and operations are more controlled and human involvement is reduced to minimum. Although machines are more connected and networked, it is necessary pre-process the data before actually analyzing it. For example, due to communication issues and multiple sources, data from the system might be discrete and asynchronous and the data should be pre-processed to make them complete, continuous and synchronized. Moreover, since variables usually possess clear physical meanings, data integrity is of vital importance to the development of the analytical system. Low-quality data or incorrect recordings will alter the relationship between different variables and will have a catastrophic impact on the estimation accuracy. The last but not least, we are more interested in finding the physical root cause behind features extracted from the phenomena in Industrial Big Data analytics instead of capturing the phenomena