基于风电功率概率预测的日前电力平衡优化方法.docVIP

PAGE / NUMPAGES 基于风电功率概率预测的日前电力平衡优化方法 Day-ahead Power balance Optimization Methd Based on Probabilistic Forecasting of Wind Power Abstract: The installed capacity of wind power in China is increasing year by year, and the impact of wind power output on the balance of power output has become more and more important. This paper proposes a Day-ahead power balance optimization method based on wind power probability prediction. First, collect historical data of wind farms in different wind resource regions in the power grid, and establish a power prediction history database; secondly, based on the wind power point prediction results, estimate the prediction error probability distribution through the working condition identification and the kernel density function; finally, generate the whole network wind power. The power bandwidth prediction results are included in the plan in advance according to the change in load during the day. The analysis results based on the actual power grid show that the proposed method can save the reserve capacity and wind adjustment margin of the power grid, and improve the power consumption of the power grid while ensuring economy and safety. Key words: wind power, probabilistic forecasting, working condition identification, nuclear density estimation 摘要：我国风电装机容量逐年递增，风电出力对日前出力平衡安排的影响越来越大。本文提出一种基于风电功率概率预测的日前电力平衡优化方法。首先，收集电网中不同风资源区域风电场的历史数据，建立功率预测历史数据库；其次，基于风电功率点预测结果，通过工况识别和核密度函数估计预测误差概率分布；最后，生成全网风电功率带宽预测结果，按照一天中负荷的变化情况将预测结果纳入日前计划。基于实际电网的算例分析结果表明，本文方法能够节省电网的备用容量和调风裕度，在保障经济性和安全性的同时提高电网对风电的消纳能力。 关键词：风电功率，概率预测，工况识别，核密度估计 DOI：10.13335/j.1000-3673.pst.2014.01.论文序号（小五号Times New Roman字体加黑） 0 引言 截止2017年年底，中国新累计并网装机容量达到1.64亿千瓦，占全部发电装机容量的9.2%[1]。随着电网风电装机逐年增长，使新能源预测准确性在电网日前出力平衡和日内实时调整中的影响日益明显[2]。如何持续提升新能源功率预测水平，以及如何将预测结果更好的纳入日前平衡安排，是现阶段亟需持续开展的两项重点工作。 目前，电网调度机构主要采用修正风电场站上报的场站短期功率预测结果（点预测）纳入日前功率平衡计划。由于针对风电场的数值天气预报精度低导致风电功率短期预测误差较大，因此有必要对风电功率预测的不确定性进行估计，以便将风电短期预测功率合理修正后纳入日前平衡安排。 风电功率概率预测是指根据气象数据、风电功率历史预测数据与实测数据，针对风电场功率的不确定性建立预测模型，对未来风电功率的波动区间和分布函数进行预测[3]。