Optimal scheduling of flexible interconnected distribution network based on adaptive model predictive control

Aiming at the uncertainties of distributed generation and load in flexible interconnected distribu-tion network, an optimal scheduling method based on improved model predictive control is proposed. The model predictive control-based intraday optimal scheduling model of flexible interconnected distribution network is established. The adaptive dynamic weight method is used to deal with the multi-objective optimi-zation problem including comprehensive power supply cost and voltage deviation. The dynamic scenario generation and K-means clustering scenario reduction method are used to deal with the prediction errors of distributed generation and load in the prediction model. Aiming at the problem of partial constant domain parameters in the rolling optimization of classical model predictive control, a rolling optimization method based on adaptive adjustment of domain parameters is proposed. The effectiveness of the proposed optimal scheduling method is verified by a simulation example of a 33-bus system interconnected with four feeders. © 2020, Electric Power Automation Equipment Press. All right reserved.