Supervising cathodic protected oil pipe nets with Cl-based methods

摘要

In this paper we present a tool, which supervises automatically a cathodic protected oil pipe net by Cl-based methods. To achieve this aim we use special pre-processing strategy of the DLS-Spectra and new kinds of neural clusters working with the principle of "Computing with Activities (C WA)". We will show that this strategy leads to a save detection of faults in the cathodic protection, trends of wear effects and sudden uprising of so far unknown sensor states. Furthermore we will show that metrological and urbane influences on the cathodic sensor net can be classified as noise contribution, which leads to a high reduction of the false alarm rate.