基于马尔科夫链风速修正的风电功率预测研究

发布时间：2018-06-10 06:20

本文选题：风电功率预测 + 风速修正　；参考：《内蒙古大学》2017年硕士论文

【摘要】：随着全球经济的飞速发展,对支撑经济发展的能源需求量越来越大。由于可利用的传统化石能源量有限,能源供需不平衡的矛盾限制着世界各国经济的发展,使得能源短缺成为整个人类社会面临的问题。化石燃料的使用引起的环境污染问题越来越严重,逐渐成为当今人类社会共同面临的巨大挑战。风能是太阳能的一种转化形式,是一种可再生的清洁能源,而风力发电是风能利用的主要形式。风力发电主要受风速和空气密度的影响,风速的波动性和间歇性使风电场的输出功率也具有波动性和间歇性,所以大规模风电并网会对电网的日常稳定运行以及电网系统安全和电能质量均造成严重影响。如果能够准确的预测出风电功率,便可以有效地改善电力系统运行可靠性,减少风电并网对电力系统的不利影响,并且可以更大限度的利用风力资源。本文提出基于马尔科夫链的风速误差修正方法,进行风电功率间接预测。在基于NWP风速预测值与SCADA风速值的风速误差序列的基础上进行马尔科夫链风速误差修正。首先对风速误差序列进行模糊C均值状态划分,统计出初始状态中各状态的初始概率分布,然后建立各状态之间的转移概率矩阵,根据转移概率矩阵预测出下一时刻风速误差修正值,进而预测出风速误差修正值,最终得到修正后的风速值。根据风速功率曲线分别对风速修正前和风速修正后的风电功率进行预测。为了对提出的马尔科夫链风速误差修正方法可行性验证,对不同数量的风速样本点进行马尔科夫链风速误差修正,并对修正结果的误差进行分析,结果证实马尔科夫链风速误差修正模型可以提高风速预测精度。进行基于马尔科夫链风速修正的风电功率预测,采用常用的绝对误差、均方根误差、平均绝对误差、平均绝对百分比误差四种误差评价指标对功率预测的结果进行分析评价,基于马尔科夫链风速修正的风电功率预测的误差与风速修正前的功率预测误差相比都有所下降,结果证实基于马尔科夫链风速修正后的风电功率预测精度高于未修正风速的功率预测精度。因此本文提出的马尔科夫链风速误差修正可以提高风电功率预测精度。
[Abstract]:With the rapid development of the global economy, the energy demand for supporting economic development is increasing. Due to the limited amount of traditional fossil energy available, the contradiction between the imbalance of energy supply and demand restricts the economic development of countries in the world, which makes the energy shortage become a problem facing the whole human society. The environmental pollution caused by the use of fossil fuels is becoming more and more serious. Wind energy is a conversion form of solar energy, is a renewable clean energy, and wind power is the main form of wind energy utilization. Wind power generation is mainly affected by wind speed and air density. The fluctuation and intermittence of wind speed make the output power of wind farm also fluctuate and intermittent. Therefore, large-scale wind power grid connection will have a serious impact on the daily stable operation of the grid, as well as the security and power quality of the power system. If the wind power can be accurately predicted, it can effectively improve the reliability of the power system, reduce the adverse effects of wind power grid connection on the power system, and make greater use of wind power resources. In this paper, an error correction method of wind speed based on Markov chain is proposed to predict wind power indirectly. Based on the wind speed error series of NWP and SCADA, the Markov chain wind speed error correction is carried out. First, the fuzzy C-means state is used to divide the wind speed error series, and the initial probability distribution of each state in the initial state is calculated, and then the transition probability matrix among the states is established. According to the transfer probability matrix, the correction value of wind speed error at the next moment is predicted, and the revised wind speed value is finally obtained. According to the wind speed power curve, the wind power before and after the wind speed correction are predicted. In order to verify the feasibility of the proposed Markov chain wind speed error correction method, the Markov chain wind speed error correction is carried out on different wind speed sample points, and the error of the correction result is analyzed. The results show that the modified model of Markov chain wind speed error can improve the accuracy of wind speed prediction. Wind power prediction based on Markov chain wind speed correction is carried out. The results of power prediction are analyzed and evaluated by four kinds of error evaluation indexes, such as absolute error, root mean square error and average absolute percentage error. The error of wind power prediction based on Markov chain wind speed correction is lower than that before wind speed correction. The results show that the prediction accuracy of wind power based on modified Markov chain wind speed is higher than that of uncorrected wind speed. Therefore, the Markov chain wind speed error correction proposed in this paper can improve the prediction accuracy of wind power.
【学位授予单位】：内蒙古大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM614

【参考文献】