Research On An On-line Spinning Reserve Rolling Dispatch Method For Accommodating Large-scale Wind Power

In recent years, the rapid development of wind power has become an important green energy in the 21 st century. As the proportion of wind power in the grid continues to increase, due to wind power has the characteristics of randomness, intermittent and volatility, meanwhile wind power is difficult to accurately predict, which has brought new challenges to the optimal allocation of spare capacity of wind power system on the temporal and spatial scales. Because of the wind power prediction errors in days, ultra short-term load forecast deviation, unit failure and other factors, so there is a big day forecast bias between days after the power forecast and day-ahead generation schedule, seriously affected the execution of traditional power generation plan. At the same time the current standby configuration with high reliability based on conventional power can not meet the needs of system reliability of wind power grid. Therefore, we need to scroll spare capacity optimization allocation strategy based on power deviation to better coordinate system reliability and economy, achieve the smooth transition of power generation plan from day-ahead to days.First, this paper introduces the factors that cause power deviation, which focuses on the causes of large wind power prediction error. The combination of normal distribution with Laplace distribution is taken to simulate the probability distribution of wind power forecast error, then confidence level is used to determine the top and bottom limitation of wind power.Subsequently, we use stochastic production simulation of power system with large-scale wind farms to strike the largest spare capacity. After discussing the advantages and disadvantages of three typical stochastic production simulation method, we determined the method used in this paper. First of all arrange the wind power which is taken as a fixed output for the extreme value of conventional power. Then arrange the unit with maximum output, the equivalent durative load duration curve is taken to strike the largest spare capacity of wind power system under a certain reliability index.Finally, On the basis of above, a strategy for on-line spinning reserve rolling generation dispatch is established for effectively accommodating large-scale wind power, a mathematic model for on-line rolling dispatch is also given. According to the relationship between power deviation and maximum spare capacity, the optimization model is established to realize the lowest purchase cost. Then the model is solved by improved particle swarm optimization algorithm to get the optimal spinning reserve capacity of the rolling plan for each period. The strategy not only can reduce the amount of abandoned wind, at the same time improve the economy and reliability of the system, has guidance of reserve capacity in each period.