Font Size: a A A

Research And Realization Of Voltage Sag Forecast And Warning System On Association Rules

Posted on:2015-07-11Degree:MasterType:Thesis
Country:ChinaCandidate:Y LuoFull Text:PDF
GTID:2272330431983014Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
With the increasing of precision equipment and sensitive users in modern power system, a higher requirement is put forward for power quality. The accidents which caused by voltage sags reach more than80%of the total, in various power quality problems. And the voltage sag events will cause huge economic loss almost every time. Therefore, the prediction and early warning of voltage sags become one of the most important topics of modern power quality field.Although the voltage sag event itself has certain randomness, the historical events shown that the happening of voltage sags existence inevitability. This article proposes a method of using history to predict the future, and applies it in forecasting voltage sag events. Firstly, on the basis of the existing large number of voltage sags in historical events, a few attributes which have the closest relation with the occurrence of voltage sag are selected and discretized. In the next place, strong association rules of voltage sags’ attributes are excavated, using the improved multidimensional association rules algorithm. Through recording the different support and confidence of association rules among voltage sags, a sag rule base is established with those values. Then, used Case-Based Reasoning and Rule-Based Reasoning to match the combination of attributes and the case in the rule base. Thus, the value of nearest neighbor distance is worked out by comparing the distance between the combination of attributes and each dip rules, its minimum value’s rule can be taken as the matching rules and predicting outcomes. Finally, the predicted results issue warnings immediately in four ways.In this paper, the design and implementation of voltage sags forecasting and early warning system was introduced in detail. This system was served as a subsystem of the intelligent information system of power quality, and it implemented the rule base regularly update, customizability prediction, multi-way combination of early warning and other functions. The interface which was simple and efficient showed the function of this system precisely. All efforts made in this paper have active and indicative meaning for the further research of voltage dips.
Keywords/Search Tags:Voltage Sags, Forecast and Warning, Association Rules, Rule-BasedReasoning
PDF Full Text Request
Related items