人工神经元网在新能源发电系统中的应用.doc

摘 要 随着传统化石能源的枯竭和环境污染问题的日益严重，新能源发电越来越受到世界各国的关注。目前，新能源发电系统因为其非线性、间歇性、不稳定性等特点，在发电、控制、并网、检修等环节中存在一系列问题，人工神经元网络为解决这些问题提供了一个新的途径。 本文首先以BP神经网络和RBF神经网络为例说明了人工神经元网络的基本概念，分析了新能源发电系统的类型特点并介绍了光伏发电和风力发电的工艺流程。 接着，选取风力发电功率预测作为案例研究了人工神经元网络在新能源发电功率预测中的应用，在分析风力发电功率的影响因素的基础上，建立了基于BP神经网络的风电功率预测的基本模型并进行了仿真实验，表明BP神经网络可以进行风电功率预测并具有较高精度。 之后，研究了人工神经元网络在新能源发电智能控制中的典型应用，在概述了人工神经网络控制的基础上，分别给出了基于BP神经网络的变桨距风机控制和基于神经网络的太阳能跟踪光敏传感控制，验证了人工神经元网络在新能源发电控制中的可行性。 最后，研究人工神经元网络在新能源发电故障诊断中的应用，介绍了故障诊断与检测的理论知识，总结了基于人工神经网络的故障诊断方法并进行了优缺点分析，并给出了基于小波包和BP神经网络风机齿轮箱故障诊断模型。 关键词：人工神经元网络，新能源发电，功率预测，智能控制，故障诊断 ABSTRACT With the depletion and environmental pollution problems of traditional fossil fuels increasingly serious, new energy generation more and more attention around the world. Currently, new energy generation systems because of its nonlinear, intermittent, instability, etc., in power generation, control, and there is a range of issues network, maintenance and other aspects of artificial neural networks to address these issues to provide a new way. Firstly, with BP neural network and RBF artificial neural network example to illustrate the basic concepts of neural networks to analyze the characteristics of the types of new energy generation systems and describes the photovoltaic and wind power generation process. Then, select the wind power prediction of artificial neural networks in new energy power generation forecasting as a case on the basis of analysis of influencing factors of wind power generation on the establishment of a basic model based on BP neural network of wind power prediction and simulation experiments show that the BP neural network can be wind power prediction and has high accuracy. After study the typical applications of artificial neural networks in the new generation of intelligent control of energy, in an overview of the basic controls on artificial neural networks, were given away fan control based solar pitch and BP neural network based on neural networks Tracking photosensitive senso