考虑风电不确定性的电热系统经济调度

随着风电等新能源大规模并网,其出力的不确定性给电力系统日前调度带来很大挑战。传统的研究方法多是假设风电功率预测误差服从某种概率分布,但风电功率预测误差受到多种因素影响,概率分布模型无法准确描述其特性。为此,采用基于神经网络的组合预测方法对风电功率误差进行建模,再将预测的风电误差加入到包含热电机组、火电机组、风电、储热装置和电锅炉的热电联合优化调度模型中,最后以实际的10机系统为例进行仿真,分析了风电预测误差对机组出力、风电消纳及调度成本的影响。结果表明,与传统方法相比,所建模型可减少机组燃煤成本与旋转备用成本,降低了经济调度成本,提高了风电消纳水平。

Economic Dispatch of Electrothermal System Considering Wind Power Uncertainty

With the large-scale grid connection of new energy sources such as wind power,the uncertainty of its output has brought great challenges to the day-ahead generation scheduling of power systems.The traditional research methods mostly assume that the wind power prediction error obeys a certain probability distribution,but the wind power prediction error is affected by many factors,and the probability distribution model cannot accurately describe its characteristics.In this paper,the combined prediction method based on neural network is used to model the wind power error.Then the predicted wind power error is added to the combined thermal power optimization scheduling model including thermal power unit,thermal power unit,wind power,heat storage device and electric boiler.Finally,the actual 10 unit system is simulated to analyze the impact of prediction error of wind power on unit output,wind power consumption and dispatching cost.Compared with the traditional method,the results show that this model can reduce the cost of coal-fired unit and spinning reserve,reduce the cost of economic scheduling,and improve the level of wind power consumption.