An Equivalent Aggregated Model of Large-Scale Flexible Loads for Load Scheduling

Abstract

The popularity of smart applications enables the large-scale integration of flexible loads to power systems, which poses a considerable challenge to system scheduling. Focusing on scheduling optimization with large-scale flexible loads, this paper proposes the equivalent aggregated method for flexible loads, which converts a large number of flexible loads into a few equivalent models to participate in system scheduling. In order to establish the equivalent model, flexible loads are grouped based on their parameters such that loads with the same or similar parameters are clustered into the same group. Then, the equivalent model for each group is established, a proof of the equivalence relationship between the original model and the equivalent model is provided, and the upper bound of equivalence deviations is estimated. The whole equivalent aggregated model is expressed as the sum of equivalent models of all groups. Afterwards, a new approach of applying the equivalent aggregated model to system scheduling is proposed. Case studies show that the equivalent aggregated model can effectively schedule large-scale flexible loads to shave the peak and fill the valley, with small equivalent deviations and fast calculation speed. This validates the proposed model and demonstrates its promising applications to large-scale load scheduling.