Methods of Multi-Parameter Diagnostics of Electric Equipment Condition within On-line Monitoring Systems

Provision of power equipment with on-line monitoring systems is a relevant and highly sought task. Designing up-to-date intelligent diagnostic systems requires development of methods for multi-parameter diagnostics (MPD) of the facility condition based on the total diagnostic features. As exemplified in a high-voltage oil-filled transformer, a list of jobs needed for generation of an effective diagnostic complex has been compiled. Trends for concentration of gases dissolved in transformer oil of the ladle furnace have been extrapolated. Time of matching the threshold values determining the facility conditions by these parameters has been forecast. Experiments have proved that a stationary on-line monitoring system enables disclosure of developing defects. At this, the use of stationary systems as fault indicators only is non-efficient both in the technical and economical terms. The paper substantiates the list of task to be solved when developing methods for MPD in the on-line systems. It is shown that condition diagnostics is most efficient at a complex estimate of independent and cross-correlated signals. The authors highlight possibility of extended application of the developed methods for diagnostics of any electro-technical systems, which data may be submitted as trends.