基于卡尔曼滤波与ARIMA混合算法的研究及其在风功率光伏功率预测系统中的应用 Alternative Title The hybrid statistical methods research based on Kalman filter and ARIMA in wind power and photovoltaic power forecasting system 苏仲岳 Thesis Advisor 王建州 2013-05-26 Degree Grantor 兰州大学 Place of Conferral 兰州 Degree Name 硕士 Keyword 风功率及光伏功率 稳定性 自适应ARIMAX模型 数据挖掘 Abstract 本文首先建立了基于统计方法与数值模式相结合的短期预报系统，该预报系统是基于统计方法与物理方法相结合的混合方法。在此研究的基础之上本文提出了一种自适应的自回归滑动平均结合外生变量（ARIMAX）的超短期预测模型。该模型使用统计方法，将物理预测结果作为外生变量，有效的提高了超短期预测的预测精度。其自适应模型参数经由混沌粒子群优化算法（CPSO）进行了优化，经过预测实例对比分析，该模型比ARMAX，ARIMAX及神经网络模型（NN）有更好的预测表现。 风力发电及光伏发电由于其对于经济及环境的重要影响，已经成为了世界上最有发展前途的可再生资源。由于风电本身的不稳定性，风功率与光伏功率预测系统的研究及如何提高功率预测精度是一个困难问题。结合统计方法在风功率及光伏功率预测领域的应用背景，本文给出了该预报系统在风功率及光伏功率预测应用的实例，就张掖地区为例详细分析了该预报系统的稳定性及预报误差。结果证明该预报系统能够稳定预测未来72小时内的功率信息，且预报误差在可接受的范围之内。在诸多统计方法应用于风功率及光伏功率预测领域的情形下，本文以功率预测为重点，讨论了进一步提高预测精度的研究方向，即挖掘历史预报数据中的信息，并将其有效的提高给预测模型，本文以风向为例给出了一种数据挖掘的实例。 Other Abstract In this paper, a short-term forecasting system based on the combination of numerical models and statistical methods is established. Combining statistical methods in the field of wind power and photovoltaic power forecast background, this article gives examples of statistical applications in the wind power and photovoltaic power forecasting. On this basis, this research paper proposes an adaptive autoregressive moving average with exogenous variables (ARIMAX) short-term forecasting model. The model uses statistical methods, as physical prediction results an exogenous variable, effectively improve the prediction accuracy of the Immediate -short-term forecasting. Adaptive model parameters are optimized by chaotic particle swarm optimization algorithm (CPSO). Compared with the original Auto-regressive Moving Average with exogenous variables (ARMAX), ARIMAX, adaptive ARMAX and Neural Networks (NNs) models, the proposed adaptive ARIMAX model performs best. Model performances of original ARMAX and ARIMAX have been improved significantly by using adaptive parameters. Wind energy and photovoltaic energy have become the most promising alternative resources, mainly due to its significant benefits to economy and environment. For wind power and photovoltaic power generation, prediction techniques, especially prediction accuracy, are critical. This paper focused on this meaningful and important subject. As an example in Zhangue, a detailed analysis of the stability and forecast accuracy are presented which prove that the forecast system reached the requirements. In the applications of many statistical methods used in the field of wind power and photovoltaic power prediction, this paper focus on the direction of research to further improve the prediction accuracy which is mining historical forecast data. As wind direction for example, this paper gives an example of a data mining. URL 查看原文 Language 中文 Document Type 学位论文 Identifier https://ir.lzu.edu.cn/handle/262010/225146 Collection 数学与统计学院 Recommended CitationGB/T 7714 苏仲岳. 基于卡尔曼滤波与ARIMA混合算法的研究及其在风功率光伏功率预测系统中的应用[D]. 兰州. 兰州大学,2013.
 Files in This Item: There are no files associated with this item.
 Related Services Recommend this item Bookmark Usage statistics Export to Endnote Altmetrics Score Google Scholar Similar articles in Google Scholar [苏仲岳]'s Articles Baidu academic Similar articles in Baidu academic [苏仲岳]'s Articles Bing Scholar Similar articles in Bing Scholar [苏仲岳]'s Articles Terms of Use No data! Social Bookmark/Share
No comment.