太阳能光伏发电预报效果评价

基于2010年12月1日～2011年3月16日华中科技大学新能源研究中心光伏电站(114.E,30°N)的发电数据和同期的气象数据,建立了光伏发电量与气象因子间的相关关系,利用线性回归法、原理法、神经网络法预报光伏电站的发电量.检验结果表明,三种方法对逐时太阳能预报与实况的相关系数均通过信度为0.001的显著性检验,对发电量的预报均具有使用价值,其中线性回归法预报效果最优.     

定日镜自动控制系统设计应用

一前言 目前中国太阳能低温热利用已经走到了世界的前列,中国的发展模式得到了世界的关注和认可,然而我国的太阳能发电技术还只是处于起步阶段.太阳能发电的两种形式是太阳能光伏发电和太阳能热发电,太阳能热发电在国外有几十年的发展历史,在我国还只是一个新兴的产业.太阳能热发电的三种主要形式是塔式、槽式和碟式,其中塔式高温热发电具有能大规模推广,实现产业化的优点,发展前景比较广阔.     

南宁太阳能光伏发电开发及规划研究

介绍南宁市太阳能资源情况,以及在太阳能光伏发电方面的利用现状及存在的问题。通过分析,提出在"十三五"期间,南宁市建设集中式光伏发电项目、分布式光伏发电项目以及扶贫光伏发电项目的规划目标。最后,从节能减排和社会效益方面对规划的光伏发电项目进行了分析和测算。     

中国太阳能光伏发电产业现状及展望

通过对近年太阳能光伏发电和光伏制造产业(多晶硅、硅片、电池片和电池组)数据分析表明,我们西部地区以大型光伏电站为主,中东部地区以分布式光伏发电为主,特别是与建筑一体化的离网分布式光伏。受光伏发电的强劲牵引,我国光伏上游产业也日益壮大,已成为太阳能产品的出口大国。本文最后还对基于"互联网+"和当前的局限对太阳能光伏发电产业进行了展望。     

光伏发电预报资料采集处理子系统设计及实现

为实现资料采集处理子系统能为太阳能光伏发电预报系统的稳定运行提供基础数据保障,设计了资料采集处理子系统结构模块图、处理资料格式规范及业务数据流程,实现了内部业务流程与实际预报业务和信息流程的紧密结合,能将太阳能光伏预报系统所需资料定时存储到数据库中,井将原始文件备份到指定目录,在系统界面中可方便地查询各种入库资料,具有良好的可扩展性.     

光伏发电有望实现平价上网

太阳能发电分为光热发电和光伏发电。通常说的太阳能发电指的是太阳能光伏发电,简称"光电"。光伏发电是利用半导体界面的光生伏特效应而将光能直接转变为电能的一种技术。这种技术的关键元件是太阳能电池。太阳能电池经过串联后进行封装保护可形成大面积的太阳电池组件,再配合上功率控制器等部件就形成了光伏发电装置。     

我国初步开发完成“太阳能光伏发电预报系统”

受中国气象局委托,湖北、吉林、新疆、陕西、云南、宁夏等6省(区)气象部门初步开发完成"太阳能光伏发电预报系统",今年6月前该系统将在上述地区推广应用。项目牵头单位湖北省气象局介绍,"太阳能光伏发电预报系统"由资料采集处理、数据库、预报产品制作和预报产品Web发布等4个子系统组成,可以预报任何地点或光伏电站未来3天逐小时辐射量和发电量,并采用了     

太阳能发电研究综述

介绍了太阳能发电研究现状及其在未来能源供应中的重要性,通过对太阳能发电两个重要利用方式-光伏发电和光热发电研究分析,阐述了他们目前研究和应用进展及存在的技术问题,提出了应对措施,为太阳能发电研究指出方向,设计提供参考.     

分布式能源系统——太阳能发电简介

该文介绍太阳能资源情况，太阳能发电：热发电和光伏发电，它们当前国内外概况，并结合国情提出几条加快光伏发电建设的建议。     

光伏发电逆变器结构的优化探讨

太阳能光伏发电并网是光伏发电产业的必然趋势.目前,制约光伏发电快速发展的因素是多方面的,其中逆变器结构的优化问题是重要的因素之一,如何提高逆变器的效率成为当前太阳能光伏发电研究的重要内容之一.通过比较部分逆变器的性能、结构等特点,提出了对其结构优化设计的基本思路.     

PV power conversion and short-term forecasting in a tropical,densely-built environment in Singapore

With the substantial growth of solar photovoltaic installations worldwide, forecasting irradiance becomes a critical step in providing a reliable integration of solar electricity into electric power grids. In Singapore, the number of PV installation has increased with a growth rate of 70% over the past 6 years. Within the next decade, solar power could represent up to 20% of the instant power generation. Challenges for PV grid integration in Singapore arise from the high variability in cloud movements and irradiance patterns due to the tropical climate. For a thorough analysis and modeling of the impact of an increasing share of variable PV power on the electric power system, it is indispensable (i) to have an accurate conversion model from irradiance to solar power generation, and (ii) to carry out irradiance forecasting on various time scales. In this work, we demonstrate how common assumptions and simplifications in PV power conversion methods negatively affect the output estimates of PV systems power in a tropical and densely-built environment such as in Singapore. In the second part, we propose and test a novel hybrid model for short-term irradiance forecasting for short-term intervals. The hybrid model outperforms the persistence forecast and other common statistical methods.     

基于遗传算法优化神经网络的光伏电站短期功率预测

针对现有光伏发电预测的不足,基于遗传算法(GA)和神经网络(BP)算法构建光伏电站功率预测模型,并使用组合权重法遴选相似日对模型进行修正。采用新疆某光伏电站运行实例验证模型的有效性,并对比BP-GA模型与单一BP模型的预测误差。结果表明,BP-GA模型克服了传统单一BP模型的不足,具有较高的预测精度,可为光伏发电预测工程实践提供参考。     

神经网络短期光伏发电预测的应用研究进展

准确的太阳能发电功率短期预测是保证电力调度和大规模光伏并网的关键。该文对近年来光伏发电功率短期预测研究进展进行综述,并对影响光伏发电功率的各种气象因素进行相关性分析。针对用于光伏发电短期功率预测的人工神经网络模型和深度学习模型进行总结和评述。太阳辐照度是影响预测模型精度的主要气象参数。在光伏发电功率短期预测中,神经网络及其组合模型均表现出较好的预测精度,但组合模型整体上优于单一预测模型。     

基于神经网络和改进相似日的光伏电站功率预测

为提高光伏电站功率预测的准确率,提出了一种基于SOM神经网络与熵权法优化关联系数的相似日预测模型,利用麻城市某100MW光伏电站的气温、相对湿度、风速及国家气象站日照时数、总云量、低云量等气象要素,采用SOM神经网络推算出预测日的三个相似日,再利用熵权法优化关联系数确定三个相似日的系数求出相似日分辨率为15min的瞬时功率,作为BP神经网络输入对光伏电站进行短期功率预测,并通过与其他四种预测模型的对比分析评估其性能。结果表明,模型的月相对均方根误差、月平均绝对百分比误差分别为5.88%、3.03%,与效果最佳的原理法模型误差接近;基于熵权法优化的关联系数和云量数据的加入对预测准确率有较大提高;模型预测准确率较高,抗扰动能力较强,可集合至本部门开发的预测系统运用到实际中。     

Online 24-h solar power forecasting based on weather type classification using artificial neural network

Power forecasting is an important factor for planning the operations of photovoltaic (PV) system. This paper presents an advanced statistical method for solar power forecasting based on artificial intelligence techniques. The method requires as input past power measurements and meteorological forecasts of solar irradiance, relative humidity and temperature at the site of the photovoltaic power system. A self-organized map (SOM) is trained to classify the local weather type of 24 h ahead provided by the online meteorological services. A unique feature of the method is that following a preliminary weather type classification, the neural networks can be well trained to improve the forecast accuracy. The proposed method is suitable for operational planning of transmission system operator, i.e. forecasting horizon of 24 h ahead and for PV power system operators trading in electricity markets. Application of the forecasting method on the power production of an actual PV power system shows the validity of the method.     

基于时空图卷积神经网络的光伏发电功率超短期预测方法

为了解决传统光伏电站超短期功率预测方法不能同时准确提取发电功率的时间和空间特征的问题,提出一种基于时空图卷积神经网络的光伏发电功率超短期预测方法。针对同一区域内的多个光伏电站,首先对电站进行图建模,利用图卷积网络（GCN）与门控线性单元（GLU）提取发电功率的时空特征。利用提取到的时空特征信息以及区域内光伏电站的历史发电功率数据训练预测模型,最终实现对多个光伏电站发电功率超短期预测。实验结果表明,该方法能够将超短期功率预测均方根误差减小至1.122%,对工作人员根据实际情况进行电网的调度管理具有重要意义。     

A new integrated single‐diode solar cell model for photovoltaic power prediction with experimental validation under real outdoor conditions

The output power prediction by a photovoltaic (PV) system is an important research area for which different techniques have been used. Solar cell modeling is one of the most used methods for power prediction, the accuracy of which strongly depends on the selection of cell parameters. In this study, a new integrated single‐diode solar cell model based on three, four, and five solar cell parameters is developed for the prediction of PV power generation. The experimental validation of the predicted results is done under outdoor climatic conditions for an Indian location. The predicted power by three models is found close to measured values within 4.29% to 4.76% accuracy range. The comparative power estimation analysis by these models shows that the three‐parameter model gives higher accuracy for low solar irradiance values <150 W/m², the four‐parameter model in the range of 150 to 500 W/m², and the five‐parameter model for >500 W/m². The present model is also compared with other models in literature and is found to be more accurate with less percentage error. The overall results also show that the power produced depends on temperature and solar radiation levels at a particular location. Thus, single solar cell model developed can be used with sufficient accuracy for power forecast of PV systems for any location worldwide. The follow‐up research areas are also identified.     

Online short-term solar power forecasting

This paper describes a new approach to online forecasting of power production from PV systems. The method is suited to online forecasting in many applications and in this paper it is used to predict hourly values of solar power for horizons of up to 36 h. The data used is 15-min observations of solar power from 21 PV systems located on rooftops in a small village in Denmark. The suggested method is a two-stage method where first a statistical normalization of the solar power is obtained using a clear sky model. The clear sky model is found using statistical smoothing techniques. Then forecasts of the normalized solar power are calculated using adaptive linear time series models. Both autoregressive (AR) and AR with exogenous input (ARX) models are evaluated, where the latter takes numerical weather predictions (NWPs) as input. The results indicate that for forecasts up to 2 h ahead the most important input is the available observations of solar power, while for longer horizons NWPs are the most important input. A root mean square error improvement of around 35% is achieved by the ARX model compared to a proposed reference model.     

基于CA-SE-GA-BP的光伏发电功率预测

针对光伏发电功率预测精度低的问题,以澳大利亚爱丽丝泉地区某200kW的光伏电站为例,选用遗传算法(GA)优化BP神经网络,采用相关性分析法(CA)确定太阳辐照度、温度、湿度为影响光伏发电功率的主要因子,结合经样本熵(SE)量化的天气类型作为模型输入量,提出CA-SE-GA-BP神经网络的光伏发电功率预测模型。结果表明,多云天气下CA-SE-GA-BP神经网络均方根误差、平均绝对百分比误差分别为4.48%、2.27%,晴天、雾霾、雨天三种天气类型下的预测误差也基本上不超过10%,相较于SE-GA-BP、CA-GA-BP、GA-BP神经网络,CA-SE-GA-BP神经网络预测误差降低,为解决光伏系统发电功率预测提供了一种高效准确可行的方法。     

基于数据挖掘和遗传小波神经网络的光伏电站发电量预测

为了解决光伏发电预测不确定性问题,进一步提高光伏电站发电量的预测精度。提出一种基于数据挖掘和遗传小波神经网络的光伏电站发电混合预测模型,利用K均值聚类算法对历史数据进行分类,并对传统BP神经网络进行改进。以BP神经网络为基础,引入小波分析构建小波神经网络,同时利用遗传算法对网络的初始参数进行全局寻优得到最优参数,利用交叉熵函数对学习规则进行改进。改进后的网络模型既具有小波分析的良好的局部时域和频域特性,又具有全局搜索能力,可增大跳出局部最优的可能性,同时拥有更快的收敛能力和稳定性。实验结果验证了该算法的有效性。