数据融合技术在风电企业信息系统中的应用研究

发布时间：2018-03-30 19:10

本文选题：数据融合　切入点：风电预测　出处：《华北电力大学》2017年硕士论文

【摘要】：科学技术飞速发展的同时,能源消耗也在迅速增长,从而导致化石能源日益枯竭。因此,风能作为一种清洁能源,得到广泛的开发利用。在我们大规模利用风能的同时,也出现了很多问题,比如在风电并网时,由于风能的不稳定性对整个电网系统的稳定运行造成冲击。目前为止,风能的不稳定性在很大程度上限制了风电的发展。风电企业信息系统中最为重要的一部分就是风电功率预测,因为风能的不稳定性,就导致风电的波动性和间歇性以及反调峰特性,实现风电功率的准确预测,电网就可以根据预测功率,制定合理的功率调度计划,保证电网的稳定性,同时也可以最大效率的利用风能,避免了因风电功率不稳定和调度问题从而导致风电不能物尽其用。合理的风电企业信息系统可以协助电网完成功率调度问题,同时促进了风能的利用率,减少了化石能源的消耗。本文研究了数据融合技术,从数据融合技术的定义、分类和数据融合方法三个方面进行研究。主要研究了像素级融合、特征层融合和决策层融合。还详细介绍了数据融合中的贝叶斯方法、D_S证据推理、人工神经网络和模糊理论算法等经典方法。此外本文还从风电功率预测方法和风电功率预测类型等方面认真研究了风电功率预测技术。本文通过对比现有的风电功率预测技术,结合数据融合技术,本文设计了基于数据融合的风电功率预测算法。该算法分为两个阶段,第一阶段通过贴近度进行筛选、清洗,最后将数据进行融合。第二个阶段通过四步:建立神经网络,制定模糊控制规则,设计模糊推理,最后将输出变量还原,得到风电功率的预测值。该算法实时性较好,且排除了不精确数据的误差。最后本文进行了风电企业信息系统的设计研究,该系统主要应用与风电功率的预测,从而满足本地的管理需求和并网需要。主要从系统总体的设计思想,需求分析和系统结构与开发框架进行分析研究。通过将数据融合算法应用到系统中完成了风电企业信息系统的设计,并进行了界面展示。本文将数据融合技术应用于风电数据处理,解决风电企业信息化中的风电预测问题,这应用了数据融合技术可以全面的去除冗余数据,综合多样信息,得到正确结果的特点,方便风电企业构建合适的风电企业信息系统,解决风电并网造成的电网不稳定问题。
[Abstract]:With the rapid development of science and technology, energy consumption is also increasing rapidly, which leads to the depletion of fossil energy. Therefore, wind energy, as a clean energy, has been widely developed and utilized. There are also a lot of problems. For example, when wind power is connected to the grid, the instability of wind power has an impact on the stability of the entire power system. So far, The instability of wind energy limits the development of wind power to a large extent. The most important part of wind power enterprise information system is wind power prediction, because of the instability of wind energy. As a result of wind power fluctuation, intermittence and backpeak-shaving characteristics, and accurate prediction of wind power, the power grid can make a reasonable power dispatching plan according to the predicted power and ensure the stability of the power grid. At the same time, wind energy can be used efficiently to avoid the problem of wind power instability and dispatch, which leads to wind power not being able to make the best use of it. A reasonable wind power enterprise information system can help the power grid to complete the power dispatching problem. At the same time, it promotes the utilization of wind energy and reduces the consumption of fossil energy. In this paper, the data fusion technology is studied, including the definition, classification and data fusion methods of data fusion technology. Feature level fusion and decision level fusion. The Bayesian method in data fusion is also introduced in detail. The classical methods such as artificial neural network and fuzzy theory algorithm are also studied in this paper. In addition, wind power prediction methods and types of wind power prediction are carefully studied in this paper. Combined with data fusion technology, this paper designs a wind power prediction algorithm based on data fusion. The algorithm is divided into two stages. Finally, the data is fused. In the second stage, the neural network is established, the fuzzy control rules are formulated, the fuzzy reasoning is designed, and the output variables are restored to get the predictive value of the wind power. Finally, this paper studies the design of wind power enterprise information system, which is mainly used to predict wind power. In order to meet the local management needs and network connection needs. Mainly from the overall design ideas of the system, Through the application of data fusion algorithm to the system, the design of wind power enterprise information system is completed. The data fusion technology is applied to wind power data processing to solve the wind power forecasting problem in wind power enterprise informatization. This data fusion technology can remove redundant data comprehensively and synthesize various information. It is convenient for wind power enterprises to build appropriate wind power enterprise information system and solve the problem of power grid instability caused by wind power grid connection.
【学位授予单位】：华北电力大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP202;TP311.52;TM614

【参考文献】