Development of a Diagnostic Data Fusion Model of the Electrical Equipment at Industrial Enterprises

Abstract

Continuous monitoring and diagnostics of the equipment technical condition are needed to improve reliability, to prevent possible failures, to ensure the service life extension of electrical equipment (EE) at industrial enterprises. In the present work, the authors suggest to use a diagnostic data fusion model developed for the EE technical condition diagnosis. To test the model, a scenario for searching the EE failure state was made and implemented. A diagnostic data fusion model is necessary to process the increasing amount of information produced by various EEs for subsequent analysis. The proposed data fusion model uses the levels of the Joint Directors of Laboratories (JDL) model, Data Mining technology, the ontology storage and EE diagnostics and prediction models and methods based on probabilistic statistical methods and soft computing methods. A detailed description of a fault detection model for EE at an oil company is considered. The developed diagnostic data fusion model will make it possible to identify EE faulty states and failures, as well as to increase the efficiency of making diagnostic decisions under the conditions of heterogeneous data obtained from a lot of EEs.