Research Of Multi-time Scale Energy Management Strategy For Microgrid With Uncertainty

As an important form of future smart grid,microgrid can integrate distributed renewable energy and improve the efficiency of renewable energy.Besides,it can increase the diversity of energy sources and enhance the stability of power system.Different types of distributed microsources have different operation modes and constraints,and their penetration in the microgrid brings large uncertainties.These uncertainties will bring large power deviation to the actual operation of microgrid,which pose a threat to its stable and economic operation,so it is important to carry out research on modelling and optimization of microgrid with uncertainties.In this thesis,the energy management and economic operation of microgrid are studied from two aspects of analysis modeling and optimization algorithm,which are applied to do with the uncertainty with the connection of distributed microsources and electric vehicle to the microgrid.The main work is as follows:(1)The operation characteristics of distributed microsource are analyzed,and the energy management model of microgrid is established.Distributed microsources bring greater uncertainty when they are connected to the microgrid,and the analysis of the uncertainty is a necessary step for energy management.Besides,the mathematical model of each distributed microsource and the energy management model of microgrid are established respectively.The model is based on the actual operational requirements of the microgrid and considers the energy flow and information flow within the system,which lays a solid foundation for energy management and economic operation.(2)A multi-time scale stochastic optimal scheduling strategy based on multiscenario modeling method is proposed.The Monte Carlo method is used to generate scenarios to describe the uncertainty of specific distributed microsources,and the chance constraints are introduced into the day-ahead scheduling stage to adjust the confidence interval of the scheduling decisions.The day-ahead robust scheduling decisions are obtained by taking the expected scene as the leading factor and ensures the normal operation under random scenarios.The rolling optimization strategy is introduced into the intraday scheduling stage to dynamically adjust the optimal dayahead power allocation.According to the periodically updated ultra-short-term forecast information,the unit output is obtained without changing the day-ahead scheduling decisions,and the power deviation is smoothed out through real-time interaction with the grid to realize the stable and economic operation of microgrid.(3)Because the probability distribution of uncertain variables is difficult to obtain accurately in the multi-time-scale stochastic optimization scheduling strategy,an interval method based multi-time-scale dynamic robust optimization strategy is proposed.According to the prediction error of uncertain variables,the forecast value of renewable energy and load demand are described as fluctuation interval,and the robust factors and optimization factors are introduced into adjusting the fluctuation interval of uncertain variables and the robustness of day-ahead scheduling results.The day-ahead optimal power allocation is dynamically adjusted in the intraday scheduling stage based on the rolling optimization strategy,and the real-time stable operation of the microgrid is realized by interacting with the power grid.By comparing the scheduling results of the two multi-time scale optimization strategies,the effectiveness in reducing the operation cost and practical engineering application are verified.