基于统计方法的商都风电场功率预测应用研究

发布时间：2018-07-10 06:36

本文选题：统计方法 + 功率预测　；参考：《华北电力大学》2016年硕士论文

【摘要】：随着风力发电发展越来越快,在电网中比重也越来越大,但风电的间歇性、随机性等特点使得其对电力系统的安全、运行和调度有很大的挑战。我国具有丰富的风能资源,风电产业的发展具有良好的资源基础,不仅是年平均风速高,风功率密度大,风能开发利用还有很大的发展空间。此外,很长一段时间的单一的能源结构,新能源和传统能源发展滞后和环境污染等问题,已经严重阻碍了经济和社会的可持续发展。提高风电场功率预测是解决这些问题的有效途径之一。本文通过对风电场功率特性进行深入研究,完成了以下几个方面的工作:第一,研究了天气条件对风电场功率预测产生的影响,提出了四种建立风电场功率预测模型的方法,其中非线性方法最为理想。第二,介绍了多种风电功率预测技术方法,并且着重研究了统计方法中的支持向量机算法和人工神经方法。第三,设计了基于统计方法的风电场功率预测系统,其中详细的介绍了预测系统的模块设计,结合支持向量机(Support Vector Machine,SVM)算法和神经网络(Neural Network)预测风电功率,最后对模型和预测效果进行分析。本文以风电场输出功率预测为研究对象,提出了基于统计方法的风电场功率预测方法,通过进行试验数据分析,取得了较为显著的效果。
[Abstract]:With the rapid development of wind power generation, the proportion of wind power in the power network is increasing. However, the intermittent and stochastic characteristics of wind power make it a great challenge to the security, operation and dispatch of power system. China is rich in wind energy resources, and the development of wind power industry has a good resource base, not only the annual average wind speed is high, wind power density is large, wind energy development and utilization also have a lot of room for development. In addition, for a long time, the single energy structure, the lagging development of new and traditional energy sources and environmental pollution have seriously hindered the sustainable development of economy and society. Improving wind farm power prediction is one of the effective ways to solve these problems. In this paper, the power characteristics of wind farm are studied in depth, and the following work is accomplished: first, the influence of weather conditions on wind farm power prediction is studied, and four methods to establish wind farm power prediction model are proposed. The nonlinear method is the most ideal. Secondly, several methods of wind power prediction are introduced, and support vector machine (SVM) and artificial neural method (Ann) are studied. Thirdly, a wind farm power forecasting system based on statistical method is designed, in which the module design of the forecasting system is introduced in detail, combining the support Vector Machine (SVM) algorithm and the neural network (NN) to predict the wind power. Finally, the model and prediction results are analyzed. This paper takes wind farm output power prediction as the research object, and puts forward a wind farm power prediction method based on statistical method. Through the analysis of test data, remarkable results have been obtained.
【学位授予单位】：华北电力大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TM614

【相似文献】