二型模糊逻辑系统在风电功率预测中的应用

发布时间：2018-03-08 15:29

本文选题：区间二型模糊逻辑系统　切入点：主成分分析　出处：《兰州交通大学》2017年硕士论文　论文类型：学位论文

【摘要】：随着全球能源需求的日益增高和化石燃料的急剧下降,风力发电技术在能源领域得到了高度重视,其中,准确的风电功率预测是风力发电大规模开发利用的有效手段之一。精确、可靠的风电功率预测对于优化电网运行的成本和改进电力系统的可靠性极其重要,其中,短期风电功率预测对电力系统安全稳定的运行、电网的调度以及提前安排风电机组的维护有着重要的意义,是新能源领域非常重要的研究方向之一。目前的风电功率的预测方法通常可分为物理方法、统计方法、空间相关性预测方法、组合预测方法等。更先进的现代统计方法,如神经网络、SVM等,能从过去的时间序列中描述出输入与输出的非线性联系,已在风电功率超短期或短期预测中取得成功的应用。二型FLS(Fuzzy Logic Systems,模糊逻辑系统)作为一种强有力的时间序列建模方法,已被成功应用于混沌时间序列预测、风速预测、电力负荷预测、交通流预测中,具有很好的应用潜力。考虑到风电功率数据的随机性与间歇性,以及区间二型FLS方法在预测领域中的成功应用,它理应是风电功率预测的有力工具之一。进一步,在二型FLS的基础上,通过PCA(Principal Component Analysis,主成分分析)对输入降维,从而达到避免“规则爆炸”的难题。本文主要研究内容如下:(1)研究PCA、二型模糊集的基本原理与算法实现,同时进一步研究二型FLS的组成以及各个组成部分的算法实现。(2)研究区间二型模糊逻辑系统的建模问题,基于BP算法进行参数的调整,并应用SVD-QR算法进行一定程度的规则约简。建立了二型非单值区间二型FLS的多步预测模型,并且通过提前20、40及60min的风电功率预测证明了方法的可行性与有效性。(3)考虑二型FLS的“规则爆炸”问题,再通过对PCA方法的研究,将其结合二型FLS,提出基于PCA方法与一型非单值区间二型FLS及PCA方法与二型非单值区间二型FLS相结合的预测方法。(4)为了验证所提出方法的有效性,将本文的不同方法应用于不同地区短期风电功率例中,在同等条件下,可看出本文方法的预测精度高于支持向量机和一型模糊逻辑方法。同时,模型的模糊规则数少,较好地解决了模糊模型的规则“爆炸”问题,这使得PCA-区间二型FLS方法在风电功率预测领域具有很好的应用潜力。
[Abstract]:With the increasing global energy demand and the sharp decline of fossil fuels, wind power technology has received great attention in the field of energy, among which, Accurate wind power prediction is one of the effective methods for large-scale development and utilization of wind power generation. Accurate and reliable wind power prediction is very important for optimizing the operation cost of power network and improving the reliability of power system. Short-term wind power prediction is of great significance to the safe and stable operation of power system, the dispatch of power grid and the arrangement of maintenance of wind turbine units in advance. It is one of the most important research directions in the field of new energy. The current wind power prediction methods are usually divided into physical methods, statistical methods, spatial correlation prediction methods, combined forecasting methods, and more advanced modern statistical methods. Neural networks such as SVM can describe the nonlinear relationship between input and output from past time series. FLS(Fuzzy Logic systems (fuzzy logic system), as a powerful modeling method of time series, has been successfully applied to the prediction of chaotic time series and wind speed. Power load forecasting and traffic flow forecasting have good application potential. Considering the randomness and intermittency of wind power data, and the successful application of interval type 2 FLS method in forecasting field, It is supposed to be one of the powerful tools for wind power prediction. Further, on the basis of type 2 FLS, the input dimension is reduced by PCA(Principal Component Analysis (PCA). In order to avoid the problem of "rule explosion", the main contents of this paper are as follows: 1) the basic principle and algorithm realization of PCA, type 2 fuzzy set are studied. At the same time, the composition of type 2 FLS and the algorithm realization of each component are studied. (2) the modeling problem of interval type 2 fuzzy logic system is studied, and the parameters are adjusted based on BP algorithm. The SVD-QR algorithm is used to reduce the rules to a certain extent. The multistep prediction model of the second type non-single-valued interval second-type FLS is established. The feasibility and effectiveness of the method are proved by the wind power prediction of 20 ~ 40 and 60 minutes in advance. The "regular explosion" problem of type 2 FLS is considered, and then the study of PCA method is carried out. In order to verify the effectiveness of the proposed method, a prediction method based on the combination of PCA method and non-single-valued interval second-type FLS and PCA method and second-type non-single-valued interval second-type FLS is proposed. The different methods in this paper are applied to short-term wind power examples in different regions. Under the same conditions, it can be seen that the prediction accuracy of this method is higher than that of support vector machine and fuzzy logic method. At the same time, there are few fuzzy rules in the model. The problem of regular "explosion" of fuzzy model is solved well, which makes PCA-interval type 2 FLS method have good application potential in wind power prediction field.
【学位授予单位】：兰州交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM614

【参考文献】