Current paper presents a features extraction method when analyzing the acoustic emission (AE) control data of the technical condition of construction facilities as a part of decision support in cyber physical monitoring system. A set of statistical parameters that describe local properties of AE time series in time and frequency domains has been proposed. Furthermore, a features calculation method based on utilizing sliding windows with overlays in two time scales was introduced. Presented method has been pilot-tested during the technical diagnostics of the oil tank containing defects of different hazard classes.
Technical condition estimation of the constructions is a relevant problem. In order to acquire comprehensive information of the testing object monitoring should be complex, providing effective and accurate estimate of the hazard class of the defects and forecasting its failure. Most of the current monitoring systems are based on acquiring and handling diagnostic via acoustic emission (AE) method. However, importantly, parameters of the acoustic emission propagated by defects depend on multiple factors such as type of the defect and its origin and the presence of noise on the testing object during data acquisition. In this regard, the problem of training the technical condition monitoring system is particularly important. In current work, we proposed a training method of monitoring systems for technical diagnostics of the constructions based on four subsequent stages: features extraction from AE data on two-time scales, features' dimensionality reduction, outliers detection and anomalies detection. Proposed method provides trained model for the detection of defects evolution in the building constructions. It has been tested on real constructions of the oil reservoir. The verification of the proposed method was provided by estimation of the accuracy metric of the trained model. Based on cross-validation, the mean error was 1.4 %. This confirms that proposed method can be effectively utilized as a part of technical condition monitoring system for more accurate forecasting hazard class of the defects and their evolution inside constructions.
Technical diagnostics of facilities is an urgent problem during its operation. An integral part of the implementation of diagnostic monitoring systems is the development of a decision support system (DSS) based on the analysis of acoustic emission (AE) diagnostic data and machine learning methods. A necessary condition for the application of machine learning methods in the development of DSS is the process of extracting diagnostic features from the AE signal. In the present work, an improved method is proposed for extracting diagnostic features from time series of AE signals. This includes two successive steps. At the first step, the frequency and frequency-time characteristics are calculated in a sliding window of short duration, which describe local changes in the shape and structure of single pulses. At the second step, the resulting matrix of informative features is aggregated by calculating statistical moments of various orders, which makes it possible to effectively detect long-term trends in the AE signal changes emitted by the defect. Verification of the proposed method was carried out on a full-scale control object of the oil tank RVS No. 3 ("NTEK LLC"). Based on the results obtained, a conclusion was made about the effectiveness of the proposed method in the development of diagnostic monitoring systems based on acoustic emission data.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.