Research On Optimal Dispatch Of Power Systems Incorporating Large-scale Renewable Energy

With the conflict between economic development and benefits of resources and environment has become increasingly prominent, the application of renewable energy in the power industry has been paid more attention for sustainable use and less polluting to environment.The power industry containing renewable energy represented by wind power and photovoltaic power has gradually become industrialized and scale, which eases the load and security pressures on the power systems effectively. However, the renewable energy generations increase the uncertainties in power system and affect the stability of operation due to the randomness and uncontrollability of their output. Therefore, it is necessary to adjust traditional dispatch model, in order to meet the operation requirements of power system containing large-scale renewable energy. In this paper, the optimal dispatch of power system containing large-scale renewable energy is studied.Firstly, the paper analyzes the output characteristics of wind power and photovoltaic power generations and establishes the probability models of their combined output by using convolution based on the probability distribution of independent output.Then, the prediction interval and scenario prediction of power are used to describe the predictive information of wind power and photovoltaic power so as to take the impact of the volatility of power output on generation and spinning reserve plan into consideration.Considering the characteristics of the probability distribution of combined output,power prediction interval is calculated by dichotomy the which is efficient in searching.This paper formulates an optimal dispatch model considering the costs of generation and spinning reserve based on the uncertainty information of wind and photovoltaic power prediction obtained by prediction interval of power. The confidence probability is used to control the system operation risk. Taking the10-units system for example, the simulation example solved by GAMS/Cplex proves the validity of the model.In the dispatch model based on scenario prediction, the typical scenarios of combined wind and photovoltaic power output are generated by latin hypercube sampling and reduction technology. On the basis, this paper defines the risk indices of spinning reserve which measure the tension level of system spinning reserve in volatility scenarios to control the risk of system. The reliability and economy of system operation depend on the selection of risk indices and scenarios’ number. Scheduling department can take it as a basis to formulate reasonable arrangements of the generation and spinning reserve for the units in power system.Then, the paper compares the optimization results of the above two models and analyses the optimization effect achieved on economy and reliability. For dispatch model based on interval prediction information, its economy and reliability depend on the confidence probability. In order to ensure the sufficient spinning reserve, the value of confidence probability is large, which leads to resulting in a decline in economy. On the consideration of scenario prediction, it can guarantee the reliable operation of the system in the event of typical scenarios with a large probability. The economy of this dispatch plan is better, but cannot reflect the extreme cases with little probability. Finally, the interruptible load and energy storage systems are introduced. The effect of the two measures on improving the operation of power system containing large-scale renewable energy is analyzed through the simulation results. It is helpful to make a more reasonable and comprehensive scheduling scheme.