基于蚁群算法优化神经网络的包头地区中长期电力负荷预测

发布时间：2018-06-11 16:25

本文选题：电力负荷预测 + BP神经网络　；参考：《华北电力大学》2013年硕士论文

【摘要】：电力负荷预测是电力系统安全经济规划、调度、设计研究的必要和前提。准确的负荷预测，不仅能够帮助电力部门有效合理的制定发配电计划，还能够帮助电力部门减少能源浪费，从而降低发电成本，提高企业的经济效益和社会效益。中长期电力负荷预测是指5年左右或10年以上的负荷预测工作，它是电网规划的重要基础，其预测精度的好坏直接影响到电网规划工作的优劣，因此，电力负荷预测精度的提高是目前众多研究者努力探寻的目标。目前，用于电力负荷预测的方法有很多种，，且每种预测方法都有自己的适用范围，很难应用于全部的预测情况。因此，我们必须根据具体的负荷预测特点，找出对应的预测方法。这就要求我们对负荷特性等方面进行充分的认识和分析，并且能够总结出影响电力负荷预测的主要影响因素，最后再按照根据具体的负荷预测特性选择合适的预测方法，对历史数据进行计算处理，最终获得理想的预测精度。本文系统的对目前用于电力负荷预测的方法进行了总结分析，并概述了当前电力负荷预测的特性、原理及步骤。在对电力负荷特性的规律及影响电力负荷预测的各种因素充分分析基础上，本文提出了基于蚁群算法优化BP神经网络的电力负荷预测模型；在分别对蚁群算法和BP神经网络充分介绍认识后，根据包头地区的电力负荷预测的实际情况，建立了基于蚁群算法优化BP神经网络的电力负荷预测模型，并利用相关分析法确定出网络的输入向量和输出向量。最后，本文利用蚁群算法优化BP神经网络预测模型对包头地区的电力负荷进行预测分析，并与优化前BP神经网络模型的预测值进行对比分析，结果表明：蚁群优化后的BP神经网络的预测精度要高于优化前BP神经网络模型的预测精度，说明蚁群算法优化的BP神经网络预测模型更加适合于该地区的电力负荷预测工作。
[Abstract]:Power load forecasting is the necessity and premise of power system safety and economic planning, dispatching and design research. Accurate load forecasting can not only help the power sector to make the distribution plan effectively and reasonably, but also can help the power sector to reduce the energy waste, thus reduce the cost of power generation, and improve the economic and social benefits of the enterprise. Medium-and long-term power load forecasting refers to the work of load forecasting about 5 years or more. It is an important foundation of power network planning, and the quality of forecasting accuracy directly affects the quality of power network planning. At present, there are many methods used in power load forecasting, and each method has its own scope of application, so it is difficult to be applied to all forecasting situations. Therefore, we must find out the corresponding forecasting methods according to the specific characteristics of load forecasting. This requires us to have a full understanding and analysis of the load characteristics, and to be able to sum up the main factors that affect the power load forecasting, and finally select the appropriate forecasting method according to the specific load forecasting characteristics. In this paper, the methods used in power load forecasting are summarized and analyzed systematically, and the characteristics, principles and steps of current power load forecasting are summarized. On the basis of analyzing the law of power load characteristic and various factors influencing power load forecasting, this paper puts forward a power load forecasting model based on ant colony algorithm optimization BP neural network. After introducing the ant colony algorithm and BP neural network respectively, according to the actual situation of power load forecasting in Baotou area, a power load forecasting model based on ant colony algorithm optimization BP neural network is established. The input vector and output vector of the network are determined by correlation analysis. Finally, the paper uses ant colony algorithm to optimize BP neural network forecasting model to forecast the power load in Baotou area. The results show that the prediction accuracy of BP neural network after ant colony optimization is higher than that of BP neural network model before optimization. It shows that the BP neural network forecasting model optimized by ant colony algorithm is more suitable for power load forecasting in this area.
【学位授予单位】：华北电力大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F426.61;TP18

【参考文献】