Multi-objective dynamic economic emission dispatch considering the risk of wind power-electric vehicle interaction using teaching-learning-based optimization algorithm

Abstract

As the traditional energy systems dominated by fossil fuels transform into renewable energy, more and more renewable energy, such as wind power, are being integrated into the power system. However, due to the strong randomness of wind power, it will extremely affect the security of the power grid. Electric vehicles (EVs) can mitigate the instability of wind power through their energy storage capabilities, but the charging/discharging behavior of large-scale EVs is intensely unstable. Therefore, to assess the uncertainty of the interaction between wind power and EV (IWEv), a dynamic economic emission dispatch (DEED) model based on the conditional value at risk (CVaR) of IWEv (DEED_R-IWEv) is proposed considering the pollution emission, fuel cost and the system operation risk. Besides, some practical constraints such as the power balance, the residual power of the EVs, travel constraints for EVs owners, charging and discharging power, and climbing rate are included in DEED_R-IWEv. To obtain a satisfactory dispatching solution, a self-adaptive multi-mode teaching-learning-based optimization (SaMmTLBO) algorithm is proposed. In SaMmTLBO, the teaching factors in the teacher phase is improved to enhance the diversity, and then a parameter self-adaptive mechanism and multi-model learning are developed in learn phase to increase the efficiency. Finally, the feasibility and effectiveness of the proposed model and algorithm are validated on a 10-unit system.