基于灰色关联分析和最小二乘支持向量机的光伏功率预测算法的研究

发布时间：2019-06-10 18:25

【摘要】：随着全球能源危机和环境污染的日益严峻,开发利用新能源和可再生能源已成为世界各国解决能源问题的主要方法,光伏发电系统也由此得到迅速的发展。但是,光伏并网发电系统的发电功率由于受气象因素的影响,其发电功率具有间歇性和波动性特点,为了减轻其对电网的冲击,保证电网的稳定性,有必要对光伏系统的发电功率进行准确的预测。在阅读大量国内外文献的基础上,本文从两个不同方面对光伏并网的发电功率预测进行了研究。通过分析影响光伏并网发电功率的相关因素和数据挖掘技术,从大量数据中筛选出与预测时段具有相似气象特征的数据序列,采用灰色关联分析理论对光伏发电功率进行预测。选取辐照度、温度、湿度主要影响因子作为最小二乘支持向量机预测模型的输入变量,提前24h预测光伏系统的输出功率。深入分析灰色关联分析方法和最小二乘支持向量机方法各自的优缺点,结合光伏发电系统的特性,提出并联灰色关联小二乘支持向量机预测方法和串联灰色关联小二乘支持向量机预测方法。本文通过天津大学光伏并网发电监控系统,获取光伏并网的相关数据,分季节建立了四种预测模型,分别对晴天、多云、雨天和雾霾天进行了预测。实验结果表明串并联灰色关联小二乘支持向量机预测方法比单独的灰色关联分析方法和最小二乘支持向量机方法的预测精度高,其中串联灰色关联小二乘支持向量机模型的输出值最接近光伏系统输出功率的实际值。
[Abstract]:With the increasing severity of the global energy crisis and environmental pollution, the development and utilization of new and renewable energy has become the main method to solve the energy problem in the world, and photovoltaic power generation system has been developed rapidly. However, due to the influence of meteorological factors, the power generation power of photovoltaic grid-connected power generation system is intermittent and fluctuating, in order to reduce its impact on the power grid and ensure the stability of the power grid. It is necessary to predict the power generation power of photovoltaic system accurately. On the basis of reading a large number of domestic and foreign literature, this paper studies the power generation power prediction of photovoltaic grid-connected from two different aspects. By analyzing the related factors and data mining technology that affect the power of photovoltaic grid-connected power generation, the data sequences with similar meteorological characteristics to the predicted period are selected from a large number of data, and the grey relational analysis theory is used to predict the power of photovoltaic power generation. The main influencing factors of irradiance, temperature and humidity are selected as the input variables of the least square support vector machine prediction model, and the output power of photovoltaic system is predicted 24 hours in advance. The advantages and disadvantages of grey relational analysis method and least square support vector machine method are deeply analyzed, and the characteristics of photovoltaic power generation system are combined with the characteristics of photovoltaic power generation system. The prediction methods of parallel grey correlation least square support vector machine and series grey correlation least square support vector machine are proposed. In this paper, through the photovoltaic grid-connected power generation monitoring system of Tianjin University, the relevant data of photovoltaic grid-connected power generation are obtained, and four prediction models are established by season to predict sunny, cloudy, rainy and haze days, respectively. The experimental results show that the prediction accuracy of serial and parallel grey correlation least squares support vector machine prediction method is higher than that of single grey correlation analysis method and least square support vector machine method. Among them, the output value of series grey correlation least square support vector machine model is the closest to the actual value of the output power of photovoltaic system.
【学位授予单位】：天津大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP18;TM615

【参考文献】