Applying multiple types of demand response to optimal day‐ahead stochastic scheduling in the distribution network

Abstract

With the increasing power demand of end‐users, demand side management has been an important resource to been applied to the optimal day‐ahead scheduling of distribution network. This work deeply studies the mechanism of multi‐type demand response projects participating in load management. The authors particularly design the novel compensation mechanisms of interruptible load (IL) and transferable load (TL). The step‐wise price‐quantity offer package including basic response stage and elastic stages is devised for IL while the two‐dimensional alternating function based on transferable time and transferable capacity is set up to quantify the load shifting cost for TL. To model uncertainties of distributed generation output power and response behaviour of end‐users, a large number of stochastic scenarios are generated by the Monte Carlo method and Latin Hypercube Sampling and clustered into some typical scenarios by K‐means method. Then, the optimal day‐ahead stochastic scheduling model of the distribution network is proposed based on the obtained typical scenarios. In addition, three indexes including the contribution degree, confidence degree and strategy confidence degree are put forward to evaluate the rationality of the scheduling decision. Finally, the numerical results on the IEEE 33‐bus network verify the effectiveness of the proposed model.