基于用户用电行为预测的需求侧负荷调控

Demand-side load regulation based on residential power consumption behavior prediction

以风电出力消纳为应用场景,提出了一种基于时间序列分析的用户用电行为预测方法和需求侧负荷调控策略。通过分析用户在工作日和非工作日的负荷特性,采用双周期差分自回归移动平均(auto regressive integrated moving average,ARIMA)模型预测各个时间段内用户的可转移负荷序列。以居民负荷与风力发电出力的差异为目标函数建立调控模型,并依据负荷可转移时间约束以及负荷序列与风力发电序列的相关性实现用户负荷的有序调控。实验结果表明该算法具有较好的新能源消纳效果。

A residential power consumption behavior prediction approach was proposed based on time series analysis,and then a load regulation strategy was also proposed for wind power consumption.By analyzing the characteristics of residential load on weekdays and non-weekdays,the double-season ARIMA model was applied to predict the transferable load sequence of users in different time slots.Taking the difference between the electricity load and wind power output as the objective function,a load regulation model was established and solved according to the constraints on load transferable time and the correlation between the load series and wind power generation series.The experimental results verify the effectiveness of the proposed method.