Neural clouds for monitoring of complex systems

Abstract

-Condition monitoring is an important and challenging task actual for many areas of industry, medicine and economics. Nowadays it is necessary to provide on-line monitoring of the complex systems status, e.g. the steel production, in order to avoid faults, breakdowns or wrong diagnostics. In the present paper a novel machine learning method for the automated condition monitoring is presented. Neural Clouds (NC) is a novel data encapsulation method, which provides a confidence measure regarding classification of the complex system conditions. The presented adaptive algorithm requires only the data which corresponds to the normal system conditions, which is typically available. At the same time the fault related data acquisition is expensive and fault modeling is not always possible, especially in case one is dealing with steel production, power stations operation, human health condition or critical phenomena in financial markets. These real word applications are also presented in the paper.

Application of adaptive classification of tensotremorograms for revealing the pathological states of human motor control system

“…One should also take into account the other method, which can be used instead of the ABC method, namely Neural Clouds (NC) [6]. Let us consider this method and compare it with the ABC method.…”

“…According to [5,6], the optimum number of clusters is determined as a point of saturation of the dependence of the clusters number on the number of algorithm iterations. The point, at which the saturation occurs (see Fig.…”

“…Then preprocessed data should be processed with the advanced K-means method (AKM). Advanced K-means is a K-means [6] which is being performed iteratively, increasing by one the initial number of centers with each iteration. After some iterations one can see the saturation of the resulting number of clusters, (see Fig.…”

“…Previously ABC was applied to the classification of other kinds of time series [6][7][8]. The purpose of the present work is the adaptation of ABC method for classification of TTG signals and its implementing in diagnostics.…”

Data-driven modelling for flood defence structure analysis

“…A combination of this approach with finite element Virtual Dike model was presented in (Pyayt et al 2011b). One-side classification approach based on Neural Clouds (Lang et al 2008) was used for dike behaviour assessment.…”