基于小波—灰色模型的电力负荷预测研究

发布时间：2018-11-25 14:41

【摘要】：电力负荷预测是电力系统规划投资的直接依据,新上发电机组和电网改扩建项目的建设必要性、建设规模、接入系统方式和建设时序的安排,都是由电力负荷预测结果决定的。随着电力系统各项技术的不断发展,负荷预测的应用越来越广泛,对预测结果的精度要求也越来越高。特别是在投资控制严格的电网规划情况下,以及日益严格的电力需求侧管理情况下,负荷预测精度成为其它工作的绝对前提。如何在尽量不增加计算量的情况下,提高负荷预测的精确度,同时还能在短期、中长期负荷预测中具有普遍适用的效果,是本文研究的主要内容。本文首先对电力负荷的特点、电力负荷预测特点和影响因素做了简单介绍,并对电力负荷预测的经典方法及主要现代方法进行了简单概括。在总结小波分析理论和灰色系统理论的基础上,对两个理论的应用情况进行了简单分析。其次,本文在前期分析的基础上,提出将小波模型与灰色模型相结合的建模思想。以小波变换的优点和算法上的易实现性为要点,完成对小波变换的推导,得到小波分解函数。在介绍生成数据意义和具体算法的基础上,完成GM(1,1)灰色模型迭代公式进行推导,得到建模序列,并对灰色模型的可行性进行判定检验,形成负荷预测的基本条件。最后,本文以历城区电力负荷预测为实例,对小波-灰色联合预测模型进行验证和分析。通过对历城区电网现状及地区用电需求的介绍,掌握历城区电力负荷的基本特点。并以历城区"十一五""十二五"规划报告数据为对象,采用MATLAB工具对预测算法进行仿真试验。通过分析仿真结果,将试验结果同报告中真实数据进行比对,得出预测的误差。试验结果表明,在临近原始序列的节点上,预测结果误差可以稳定在5%以下,而且当原始序列质量相当时,预测结果并不受原始数据量的影响。
[Abstract]:Power load forecasting is the direct basis of power system planning and investment. The construction necessity, construction scale, access system mode and construction time series of new generation units and power grid reconstruction and extension projects are all determined by the results of power load forecasting. With the development of power system technology, the application of load forecasting is more and more extensive, and the precision of forecasting results is more and more high. Especially in the case of strict investment control of power network planning and increasingly stringent power demand side management the accuracy of load forecasting becomes the absolute premise of other work. How to improve the accuracy of load forecasting without increasing the amount of calculation as far as possible, and at the same time, it is the main content of this paper to study how to improve the accuracy of load forecasting, and at the same time, it can have universal effect in the short and long term load forecasting. In this paper, the characteristics of power load, the characteristics of power load forecasting and the influencing factors are briefly introduced, and the classical methods and main modern methods of power load forecasting are briefly summarized. On the basis of summing up the wavelet analysis theory and the grey system theory, the application of the two theories is simply analyzed. Secondly, based on the previous analysis, this paper puts forward the idea of combining wavelet model with grey model. Based on the advantages of wavelet transform and the easiness of the algorithm, the wavelet transform is derived and the wavelet decomposition function is obtained. On the basis of introducing the significance and algorithm of generating data, the iterative formula of GM (1 / 1) grey model is deduced, the modeling sequence is obtained, and the feasibility of the grey model is tested to form the basic condition of load forecasting. Finally, the wavelet-grey combined forecasting model is verified and analyzed with the example of power load forecasting in Licheng district. The basic characteristics of power load in Licheng District are grasped by introducing the current situation and local demand of electricity in Liicheng District. Based on the data of the 11th Five-Year Plan and the 12th Five-Year Plan of Licheng District, the MATLAB tool is used to simulate the prediction algorithm. By analyzing the simulation results, the experimental results are compared with the real data in the report, and the prediction errors are obtained. The experimental results show that the error of the prediction results is less than 5% at the nodes near the original sequence, and when the quality of the original sequences is equal, the prediction results are not affected by the original data amount.
【学位授予单位】：山东大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM715

【相似文献】