Predictive Control of Regional Flexible Load Cluster Based on Mixed Logical Dynamic Method

Uncertainty of renewable energy sources (RES) has brought about complexity in distribution network operation scheduling and rescheduling. In this paper, flexible load cluster composed of regional customers’ controllable electrical and thermal load is considered to deal with regional RES power fluctuation. Firstly, mixed logical dynamic (MLD) models of electric vehicles (EV), water heater (EWH) and inverter air-conditioner (IAC) are established to present accurate dynamic characteristics. According to MLD models, day-ahead optimal scheduling model and rolling optimization model of flexible load cluster are studied via model predictive control (MPC) method. Finally, the methodology is employed to a regional flexible load cluster including 100 residents, the result shows that the proposed models enables flexible loads in cluster to follow the most economical operation schedule, and to adaptively adjust power demand as well to compensate power fluctuation of RES without affecting residents’ comfort.