Early Warning Model of Electricity Consumption Fluctuation Risk Based on Data Mining and Time Series
In order to obtain accurate early warning results of electricity consumption fluctuation risk,and ensure the stability of power system operation,a electricity consumption fluctuation risk early warning model based on data mining and time series is studied.The fuzzy C-means clustering algorithm is used to divide the historical electricity consumption data into fluctuating e-lectricity consumption data and non-fluctuating electricity consumption data.For users whose electricity consumption fluctu-ates,according to the historical electricity consumption data,the prediction method based on time series is adopted to predict their electricity consumption data in the future by taking into account the long-term trend,seasonal defecation and irregular changes.This paper builds a risk early warning model,compares the predicted electricity consumption data at the future time point with its upper and lower baselines,obtains the trend of electricity consumption data,and judges whether the electricity consumption may fluctuate in advance.If it does,the system gives an early warning.The experimental results show that the model can accurately classify the user electricity consumption data and obtain accurate electricity consumption prediction re-sults,and the early warning accuracy is higher than 97.5％.
data miningtime serieselectricity consumption fluctuationrisk warningclustering algorithmfluctuation influ-encing factors
万方数据
维普
