Research On Economic Dispatch Of Power System Containing Wind Power And Energy Storage System

Wind energy has attracted more attention in recent years because of its pollutionfree and low-cost characteristics.Wind power has strong uncertainty and volatility,the former leads to large forecast errors,and the latter leads to wind power ramp events.Therefore,large-scale wind power integration affects the safe and stable operation of power system,which causes wind curtailment and load shedding.Energy storage system has been widely used in the field of promoting new energy accommodation in recent years because of its good ability of peak load shifting and reducing imbalance power supply.But at present,the cost of allocation and utilization of energy storage system is relatively high,so it needs more reasonable coordinated utilization.In addition,in order to improve wind power accommodation,the dispatch model of power system containing wind power becomes more and more complex,which puts forward higher requirements for optimization algorithms.Therefore,how to establish a reasonable dispatch model to promote wind power accommodation,how to solve the optimization model quickly and accurately has become a research hot in this field.Based on this,the research contents of this thesis are as follows:(1)An Improved Radial Movement Optimization(IRMO)algorithm is proposed to solve the optimization problem of power system economic dispatch model.The algorithm combines the idea of population mutation in genetic algorithm,and mutates a part of particles randomly in the iteration process,so as to improve the diversity of population and avoid from falling into local optimum.The concave parabolic non-linear decline strategy for inertia weight is introduced to improve the search accuracy in the middle and later stages of the optimization and to find the global optimal solution more easily.The global optimization ability of the algorithm is verified by test functions and dispatch model.(2)In order to solve the problem of wind power forecast error,a method of wind power forecast error estimation method based on probabilistic optimal factor features extraction is proposed.Through the statistics of historical data and the analysis of the factors related to forecast error,the method of extracting the probabilistic optimal factor features is obtained,and the extracted factor features are integrated with the weight of the probabilistic optimal correlation coefficients to obtain the estimation indicator of forecast error.A multi-time scale rolling dispatch model for power system based on real-time error compensation is established.In real-time compensation section,the output of thermal power units is revised according to the forecast error estimation indicator,and energy storage system is used for further compensation,so as to promote wind power accommodation and reduce the cost of power generation.The accuracy of the forecast error estimation method and the validity of the model are verified in case study.(3)To cope with wind power ramp events,a wind power ramp event forecast method based on Variational Mode Decomposition-Long Short Term Memory(VMDLSTM)is proposed.VMD is introduced to decompose wind power signals into three constituent modes,and LSTM is used to deeply learn and forecast each constituent mode.A thermal/storage coordinated strategy and dispatch model are established to deal with wind power ramp events.By coordinating the thermal power units in advance,it has sufficient adjustable reserve capacity during wind power ramping.By using the energy storage system to compensate in advance,the wind power ramp rate is reduced to the unit ramp rate.It can promote the wind power accommodation and ensure the safe operation of power system under the condition of wind power ramp events.At the same time,it can reduce the requirement for power and capacity of energy storage system,so as to save costs.The validity of the forecast method,coordinated strategy and dispatch model is verified in case study.