Multi-objective optimization dispatching of a micro-grid considering uncertainty in wind power forecasting

Abstract

This paper presents a new multi-objective optimization dispatching method to optimize the output power of distributed generators of a micro-grid considering uncertainty in wind power forecasting with the aims of minimizing the operational cost and pollutant emission. To deal with uncertainty in wind power forecasting, an adaptive concept of confidence interval (ACCI) is proposed and probabilistic wind power interval forecasting model based on deterministic wind power forecasting and ACCI is developed to address this problem. Then, a comprehensive formulation with different operational constraints is applied to construct the multi-objective optimization dispatching model of a micro-grid considering uncertainty in wind power forecasting, and two-step solution methodology based on chaos and sinusoidal mapping multi-objective optimization bat algorithm (CSMOBA) and fuzzy theory set is employed to solve the multi-objective optimization dispatching problem. Eventually, simulation results demonstrate the proposed ACCI method can work well and obtain appropriate confidence level at each dispatching hour, and the proposed CSMOBA algorithm outperforms multi-objective optimization bat algorithm and well-known non-dominated sorting genetic algorithm II when applied in multi-objective optimization dispatching of micro-grid.

Keywords

• Wind power interval forecasting
• Least square support vector machine
• Confidence interval
• Multi-objective bat algorithm
• Micro-grids