The Research On Expert System For Transmission Line Cing Thickness Based On On-Line Monitoring

Towers falling down and wires broken often occurred by excess icing load on highvoltage transmission lines at home and abroad. Single in early2008, extreme ice raindestroyed struck the power system infrastructures in Southern China, electrical powerinfrastructure suffered tremendous damage and the widespread occurrence of towers collapseand wires broken, which were to security and stability of power grid. Due to complexity ofphysical process of icing on overhead transmission lines, which was affected by the changingclimate and the weather, the season, the terrain, the altitude, the alignment and the wire type,it was difficult to run status assessment and pre-warning by mathematical and physical model.Freezing icing is the most damaging kind. It is useful for to on-line monitoring offreezing icing state. It was an important issue to assess and pre-warn icing on transmissionlines for the security and stability of power system. Combined with online monitoring, fuzzymathematics and the expert system, a new method was build to assess objectively andscientifically the state and ensures the security of power system. Firstly, research backgroundand significance of online monitoring of transmission lines icing were reviewed, as well asdomestic and international research profiles. The main work of this article was introduced.Secondly, the overall framework of online monitoring system of transmission lines icing andselection requirements of various sensors in harsh conditions were introduced. A new wayabout image recognition of ice thickness on transmission lines based on online monitoringsystem was proposed and could be compared with calculation model about the ice thicknessabase on the mechanical. Thirdly, it was difficult to obtain many parameters existing icingprediction model and a way to build a combined gray neural network model was constructedto predict fast the trend of icing, based on historical data through remote online monitoringsystem. Then, knowledge about the expert system and fuzzy theory were introduced. Theknowledge base of fuzzy rules about the assessment of icing thickness was build to achieveintelligent state assessment. At last, the development platform and the main framework ofexpert system were introduced and the main functions also were illustrated.