基于小波包分解和布谷鸟算法的最小二乘支持向量机风速预测模型的研究与应用 Alternative Title Research and Application on LSSVM Wind Speed Forecasting Model based on WPT and CS Algorithm 魏翔 Thesis Advisor 魏翔 2015-05-24 Degree Grantor 兰州大学 Place of Conferral 兰州 Degree Name 硕士 Keyword 混合风速预测模型 数据预处理 参数优化 人工智能算法 Abstract 风能是一种最具规模发展潜力的清洁可再生能源,因为它不仅没有燃料问题,也不会产生辐射和空气污染, 而且随着风能设施日趋进步，其生产成本也大量降低，在某些地点，风力发电成本已低于其它发电方式的成本。而风速预测是风能资源评估和电网规划的关键。然而, 由于风能的间歇性和不稳定性等因素，准确预测风速成为一项艰巨的任务。传统方法总是直接预测原始数据,而忽略原始数据的预处理,因此, 预测方法的稳定性有时是没有保证的。本文提出了一个新的基于数据预处理和人工智能算法的混合预测方法,提出的混合方法包含三个部分：首先对原始的风速数据进行数据预处理，然后建立初始的LSSVM模型并利用布谷鸟搜索算法对模型进行参数寻优，最后用优化好的LSSVM模型对风速数据进行预测，通过以上步骤以保证风速预测准确性和稳定性。此外，本文提出的模型以中国山东省蓬莱虎山发电场每10分钟平均风速数据为例进行了实证研究，并和BP神经网络、小波神经网络以及粒子群优化的LSSVM模型的风速预测结果进行了对比，最后还进行了假设检验。风速预测和假设检验的结果表明,该混合方法不仅简单而且可以有效地提高风速预测的准确性。 Other Abstract Wind power generation is one of the most scale development potential of clean and renewable energy utilization because it not only hasn’t fuel problem, but also won't produce radiation and air pollution. With wind power facilities are getting better and better and reducing production cost, in the appropriate locations, the cost of wind power is lower than other power. Wind speed is the key to both the calculation of wind energy resources assessment content and wind power generation. However, accurately forecast the wind speed became a difficult task due to the intermittent and instability of wind. General methods always forecast the raw data directly, while ignoring the processing of raw data, therefore, stability of forecasting method may sometimes be not guaranteed. In this paper, a new hybrid forecasting method based on data preprocessing and artificial intelligence algorithm has been proposed, and the proposed hybrid method is constituted by three parts: data preprocessing, parameter optimization, and artificial intelligence algorithm, hence, a good predicting outcomes with accuracy and stability will be obtained. This model has been validated by an empirical study that forecasting wind speed with an average wind speed data series collected every 10 minutes and these data are collected from the Shandong province of China. The results of empirical study and hypothesis test show that this hybrid method not only is simple but also can effectively increase the accuracy of wind speed forecasting. URL 查看原文 Language 中文 Document Type 学位论文 Identifier https://ir.lzu.edu.cn/handle/262010/225096 Collection 数学与统计学院 Recommended CitationGB/T 7714 魏翔. 基于小波包分解和布谷鸟算法的最小二乘支持向量机风速预测模型的研究与应用[D]. 兰州. 兰州大学,2015.
 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 [魏翔]'s Articles Baidu academic Similar articles in Baidu academic [魏翔]'s Articles Bing Scholar Similar articles in Bing Scholar [魏翔]'s Articles Terms of Use No data! Social Bookmark/Share
No comment.