基于不同优化算法下风电预测精度对风水电协调影响研究

研究了持续法、ARIMA方法、改进BP神经网络3种不同的风电预测模型,在相同条件下,经实例仿真发现,改进BP神经网络模型的预测精度好于ARIMA预测模型,而ARIMA预测模型的预测精度好于持续法预测模型.基于上述3种不同的风电预测模型,建立了风-水发电联合协调运行的模型.采用遗传粒子群和混合粒子群2种不同的优化算法来研究风电预测精度对风-水电协调影响,通过仿真实例发现,风电预测模型精度越高,得到的理论值与实际值偏差越小;在考虑2种不同优化算法的情况下,遗传粒子群优化算法得到的数值与实际值偏差比混合粒子群大,同时基于不同风电预测模型下的误差也要比混合粒子群大.     

基于支持向量回归的PV/T组件温度实时预测

在光伏光热系统中,光伏板的发电效率与PV/T组件温度密切相关。实时、精确地预测PV/T组件温度,对优化控制决策、提高光伏板发电效率具有重要意义。文章利用支持向量回归(SVR)算法建立PV/T组件温度预测模型。为了提高该模型预测结果的精确度,采用网格搜索与交叉验证相结合的方法对SVR核函数参数g和惩罚因子c进行寻优;然后,结合实验平台的测量数据,划分训练集和预测集,并对原始数据进行归一化处理;最后,文章将基于SVR算法温度预测模型的预测结果与BP神经网络的预测结果进行对比。分析结果表明:基于SVR算法温度预测模型的预测值与实测值基本一致,该模型的预测精度和泛化性能均优于BP神经网络的预测结果。     

基于小波网络的次日太阳逐时总辐射预测技术研究

现有的地面太阳逐时总辐射预测模型的预测精度及泛化能力尚不能令人满意。利用小波神经网络在提升非线性函数影射能力方面的优势,以及递归网络的优良的动态性能,建立了对角递归小波BP网络（DRWBPN）模型,用以对次日地面太阳逐时总辐射进行精确预测。进一步提高预测精度的措施还包括将ASHRAE太阳辐射确定性模型的计算结果和经模糊化处理的气象预报中的云量信息加入到网络输入向量中,以充分利用已知可靠信息。采用分阶段训练网络的方法,提高了有限次数下的训练质量。太阳逐时总辐射预测实例及与其它典型模型预测结果的比较表明,提出的地面太阳逐时总辐射预测模型具有更高精度和实际可行性。     

基于改进RBF碟式太阳集热器出口气流温度预测

为了改善传统的数学建模预测集热器出口气流温度较为复杂的问题,采用RBF(径向基函数)神经网络建立温度预测模型。考虑到碟式太阳集热器出口气流温度影响因素较多,在大量历史实验数据基础上,分析其主要影响因素。为提高预测精度,提出一种自适应聚类算法对RBF神经网络进行改进,并利用Kennard-Stone选取方法(简称K-S法)的空间均匀分布原理提高训练样本质量。根据碟式太阳能光热系统的实测数据对所提的模型进行测试,结果表明,改进算法可进一步提高预测精度和效率,验证了改进算法的可行性和有效性。     

基于PCA-GA-ELM的光伏发电功率预测研究

针对目前光伏发电功率预测方法所存在的预测精度较低和不同天气类型适应性较弱的问题,提出一种利用主成分分析(PCA)和遗传算法(GA)改进极限学习机(ELM)的光伏发电功率预测模型(PCA-GA-ELM预测模型)。该模型的计算过程:首先,基于季节因素和天气类型等气象因素对于光伏发电系统的影响,在不同季节下建立了不同的子模型,并利用灰色关联分析法选取同种天气类型下的相似日;然后,利用PCA将多个原始输入变量降维成少数彼此独立的变量;最后,利用GA对ELM的初始权值和阈值进行寻优。此外,文章利用光伏电站的实际发电功率数据对预测模型进行验证。分析结果表明,PCA-GA-ELM预测模型具有较高的预测精度和较强的泛化能力。     

基于LSTM网络的光伏发电功率短期预测方法的研究

提高光伏发电功率预测结果的精度对电网规划和调度具有重要意义。基于前向神经网络或回归分析法的传统预测模型因缺乏历史记忆能力而导致自身鲁棒性较差、适应能力较弱。为了解决上述问题,文章提出了一种基于LSTM网络的光伏发电功率短期预测方法。在预处理过程中,文章先将天气类型依据日照晴朗指数量化为具体数值;然后,利用主成分分析法将与光伏发电功率相关性较高的多元数据序列进行降维,得到主成分数据序列;最后,建立基于LSTM网络的光伏发电功率短期预测模型,并将该模型的预测结果与BP网络预测模型和RNN网络预测模型的预测结果进行对比。模拟结果表明,基于LSTM网络的光伏发电功率短期预测模型能较好地反映时序数据的动态特性,预测精度较高,预测结果能够为电力调度部门提供可靠的数据支持。     

基于特征筛选与ANFIS-PSO的分布式光伏发电功率预测方法研究

短期分布式光伏发电功率预测对配电网调度计划的安排及优化具有重要意义。人工智能技术的进步为精细化分析光伏发电功率预测结果的影响因素以及提高光伏发电功率的预测精度提供了有效途径。文章提出一种基于特征筛选与ANFIS-PSO的分布式光伏发电功率预测方法。首先,基于随机森林中的增益情况,对影响分布式光伏发电系统的各项特征参数进行筛选;然后,通过自适应神经模糊推理算法对输入数据进行训练,并使用粒子群算法对ANFIS模型进行优化;接着,建立基于离线训练和在线预测的ANFIS-PSO分布式光伏发电功率预测模型;最后,利用北京某地分布式光伏发电系统的实际数据来验证模拟结果的准确性。     

基于GA-ELM神经网络的逐时太阳辐照量预测

摘要： 太阳能辐照量是影响光伏发电的主要因素,准确的太阳能辐照量预测对于光伏发电具有重要的作用。为提高预测模型对环境因素的敏感性与预测精度,提出基于遗传算法（genetic algorithm,GA）改进极限学习机（extreme learning machine,ELM）的太阳辐照量预测方法。首先,选取与太阳能辐照量相关的候选属性因素,确定输入变量;其次,以2009年到2014年与待预测日相同日期前后15 d范围内数据为训练集;再次,采用GA优化ELM的隐含层输入权值及偏置向量;最后,采用优化后的GA-ELM模型,开展逐时太阳辐照预测模型。实测算例表明,相较ELM、BP神经网络,新方法具有更高的预测精度,能够适应外界气象条件突变情况下的辐照预测需要。     

基于小波变换与Elman神经网络的短期风速组合预测

风速的准确预测对风电场发电系统的经济和安全运行有着重要的作用。为了克服风速随机性强的缺点,提高短期风速预测的精度,提出了一种将小波变换与Elman神经网络相结合的短期风速组合预测模型。该模型由小波预处理模块和神经网络预测模块组成。首先利用小波预处理模块将风速序列作多尺度分解,重构得到不同频段的子序列,然后利用Elman神经网络模块分别对其训练和预测。实际风速预测结果表明,与单一的Elman和ARMA法相比,该组合预测模型的预测精度有较大的改善,可以用于风电场短期风速的预测。     

基于Elman算法的光伏阵列的短期功率预测研究

根据青海某5 MW光伏电场的历史光伏发电功率数据和当地的气象预报信息,分析影响功率预测的主要气象因素。采用Elman神经网络算法,结合与预测日同日类型下整点时刻的气象数据和光伏输出功率数据,建立光伏发电短期功率预测模型。对不同日类型的光伏出力的预测结果表明,该短期预测模型具有较高的精度,有助于电网能量的调度,对电力系统的安全稳定运行有积极作用。通过与BP神经网络和非线性状态估计(NSET)算法对比研究表明,Elman神经网络具有更高的预测精度。     

基于改进递归深度信念网络的CSP电站短期出力预测

以预测CSP电站短期出力为目的,首先引入自适应思想对递归深度信念网络的训练算法进行改进,并建立直接法向辐射的短期预测模型。随后提出一种结合静态模型的CSP电站短期出力预测方法。最后进行性能检验,验证了改进递归深度信念网络的可行性,以及CSP电站短期出力预测方法的有效性。研究结果表明：建立的改进递归深度信念网络可提升预测准确性和收敛速度;提出的CSP电站短期出力预测方法可较为准确地预测其短期出力情况。     

Viability of a concentrated solar power system in a low sun belt prefecture

Concentrating solar power (CSP) is considered as a comparatively economical, more efficient, and large capacity type of renewable energy technology. However, CSP generation is found restricted only to high solar radiation belt and installed where high direct normal irradiance is available. This paper examines the viability of the adoption of the CSP system in a low sun belt region with a lower direct normal irradiance (DNI). Various critical analyses and plant economics have been evaluated with a lesser DNI state. The obtained results out of the designed system, subjected to low DNI are not found below par, but comparable to some extent with the performance results of such CSP plants at a higher DNI. The analysis indicates that incorporation of the thermal energy storage reduces the levelized cost of energy (LCOE) and augments the plant capacity factor. The capacity factor, the plant efficiency, and the LCOE are found to be 32.50%, 17.56%, and 0.1952 $/kWh, respectively.     

Using direct normal irradiance models and utility electrical loading to assess benefit of a concentrating solar power plant

The objective of this paper was to determine if three different direct normal irradiance (DNI) models were sufficiently accurate to determine if concentrating solar power (CSP) plants could meet the utility electrical load. DNI data were measured at three different laboratories in the United States and compared with DNI calculated by three DNI models. In addition, utility electrical loading data were obtained for all three locations. The DNI models evaluated were: the Direct Insolation Simulation Code (DISC), DIRINT, and DIRINDEX. On an annual solar insolation (e.g. kW h/m²) basis, the accuracy of the DNI models at all three locations was within: 7% (DISC), 5% (DIRINT), and 3% (DIRINDEX). During the three highest electrical loading months at the three locations, the monthly accuracy varied from: 0% to 16% (DISC), 0% to 9% (DIRINT), and 0% to 8% (DIRINDEX). At one location different pyranometers were used to measure GHI, and the most expensive pyranometers did not improve the DNI model monthly accuracy. In lieu of actually measuring DNI, using the DIRINT model was felt to be sufficient for assessing whether to build a CSP plant at one location, but use of either the DIRINT or DIRINDEX models was felt to be marginal for the other two locations due to errors in modeling DNI for utility peak electrical loading days – especially for partly cloudy days.     

Performance of parabolic through solar power plant under weather conditions of the Oujda city in Morocco

A numerical simulation of Concentrating Solar Power (CSP) plant based on an Organic Rankine Cycle (ORC) power generation unit integrated with parabolic trough collectors is carried out. For the study we refer to the Solar Electric Generating System VI (SEGS VI), installed in the Mojave desert-California (USA), whose solar field which is constituted by LS2 parabolic trough collectors and we consider the same plant implementation in the region of Oujda city (Morocco). To predict the energy performance, the simulations are carried out using TRNSYS 16 simulation program known for its modularity and flexibility and the external library known as the Solar Thermal Electric Components (STEC) library. The meteorological parameters including Direct Normal Irradiation (DNI), ambient temperature and other weather conditions are taken from meteorological year database provided by a high precision MHP station located in Mohamed Premier University. The obtained results show that the region of East offers great potential in general for implementing this type of plant. In fact, the value of 30 MWe is reached during the strongest sunshine day and the operating time can go from 7 AM until 19 PM for a summer day.     

Day-ahead resource forecasting for concentrated solar power integration

In this work, we validate and enhance previously proposed singe-input direct normal irradiance (DNI) models based on numerical weather prediction (NWP) for intra-week forecasts with over 200,000 hours of ground measurements for 8 locations. Short latency re-forecasting methods to enhance the deterministic forecast accuracies are presented and discussed. The basic forecast is applied to 15 additional locations in North America with satellite-derived DNI data. The basic model outperforms the persistence model at all 23 locations with a skill between 12.4% and 38.2%. The RMSE of the basic forecast is in the range of 204.9 W m⁻² to 309.9 W m⁻². The implementation of stochastic learning re-forecasting methods yields further reduction in error from 204.9 W m⁻² to 176.5 W m⁻². To a great extent, the errors are caused by inaccuracies in the NWP cloud prediction. Improved assessment of atmospheric turbidity has limited impact on reducing forecast errors. Our results suggest that NWP-based DNI forecasts are very capable of reducing power and net-load uncertainty introduced by concentrated solar power plants at all locations in North America. Operating reserves to balance uncertainty in day-ahead schedules can be reduced on average by an estimated 28.6% through the application of the basic forecast.     

Effects of atmospheric aerosol on the direct normal irradiance on the earth's surface

Direct normal irradiance (DNI) plays a key role on the quantity and rate of hydrogen production. The accurate calculation of DNI has very important significance for low-cost hydrogen economy and efficient utilization of solar energy. This study mainly takes account of the influence of atmospheric aerosol on DNI and the experimental tests. The main idea of this paper is: obtaining the distribution characteristics of aerosol particles in the atmosphere and the optical depth of aerosol spectrum based on inversion method of ground observation station data; calculating the attenuation coefficient of solar spectrum with classical Mie scattering theory and particle system radiation characteristics; calculating aerosol attenuation coefficient under full spectrum, namely the aerosol correction factor (defined as the ratio of the attenuation coefficient of aerosol atmosphere to standard atmosphere under full spectrum) with Planck model, Rosseland model and Planck–Rosseland model respectively; choosing with the theoretical calculation model of aerosol correction factor based on the solar spectrum radiation calculated by SMARTS software; verifying the accuracy of this theoretical model with experimental DNI in city Harbin. The results show that there is a good agreement with a minimum variation of 3.08% and a maximum variation of 9.97%.     

Evaluation of the WRF model solar irradiance forecasts in Andalusia (southern Spain)

In this work, we evaluate the reliability of three-days-ahead global horizontal irradiance (GHI) and direct normal irradiance (DNI) forecasts provided by the WRF mesoscale atmospheric model for Andalusia (southern Spain). GHI forecasts were produced directly by the model, while DNI forecasts were obtained based on a physical post-processing procedure using the WRF outputs and satellite retrievals. Hourly time resolution and 3 km spatial resolution estimates were tested against ground measurements collected at four radiometric stations along the years 2007 and 2008. The evaluation was carried out independently for different forecast horizons (1, 2 and 3 days ahead), the different seasons of the year and three different sky conditions: clear, cloudy and overcast. Results showed that the WRF model presents considerable skill in forecasting both GHI and DNI, overall, better than a trivial persistence model. Nevertheless, both MBE and RMSE values presented a marked dependence on the sky conditions and season of the year. Particularly, for 24 h lead time, the MBE of the forecasted GHI was 2% for clear-skies and 18% for cloudy conditions. However, the MBE of the forecasted DNI increased up to about 10% and 75% for clear and cloudy conditions, respectively. Regarding RMSE values, in the case of forecasted GHI, results ranged from below 10% under clear-skies to 50% for cloudy conditions. In the case of forecasted DNI, RMSE ranged from 20% to 100% for clear and cloudy skies, respectively. This proved the higher sensitivity of DNI to the sky conditions. In general, an increment of the MBE and RMSE values with the cloudiness was observed. This reflects a still limited ability of the WRF model to properly forecast cloudy conditions compared to clear skies. Nevertheless, the model was able to accurately forecast steep changes in the sky (cloudiness) conditions. Finally, WRF performed considerable better than the persistence model for clear skies both for GHI and DNI, with relative RMSE values about a half. However, for cloudy conditions, performance was similar.     

基于深度学习的翼型气动弹片流场参数预测

弹片是解决翼型流动分离的重要技术手段,合理的弹片参数对翼型表面压力分布尤为重要。基于数据驱动的深度学习方法与计算流体力学(Computational Fluid Dynamics, CFD)相结合,可快速有效地完成对复杂流场特征的识别与提取。本文提出一种基于卷积神经网络(Convolutional Neural Network, CNN)的翼型表面压力分布预测方法,通过提取流场的尾流速度、压力等流动特征构建翼型表面压力分布的预测模型。首先,通过数值模拟计算了8种不同抬起角度的NACA 0012弹片翼型的流场;其次,采用提取的流场数据建立CNN预测模型;最后,将预测值和CFD计算值进行对比。结果表明：基于CNN的预测模型对翼型表面压力系数分布有较高的预测精度,其中尾流速度模型在弹片抬起角度为15°时的预测均方根误差仅为0.1,说明尾流速度中包含丰富的流场信息。     

Analysis of CSP plants for the definition of energy policies: The influence on electricity cost of solar multiples,capacity factors and energy storage

The effect on the cost of electricity from concentrating solar power (CSP) plants of the solar multiple, the capacity factor and the storage capacity is studied. The interplay among these factors can be used to search for a minimal-cost objective that can serve as a technical criterion to guide in the design of economic incentives for CSP plants. The probability-density function of irradiation is used in conjunction with screening models to evaluate the performance characteristics and costs of concentrating solar power plants. Two technologies have been analyzed in this study: parabolic-trough and tower plants. The results provide information to define the optimal operational range as a function of the desired objective. Thus, it is possible to derive a technical criterion for the design of CSP plants which optimizes the solar electricity produced and its generation cost. The methodology is applied to Spain, and the analysis of the results shows that a solar energy production of 37 kWh/m²/year for tower plants and 66 kWh/m²/year for parabolic-trough ones define the approximate optimal working conditions for the mean DNI in Spain.     

基于SOM聚类和ECA的超短期光伏预测组合模型研究

为提高不同天气类型下光伏输出功率的预测精度,提出了一种基于注意力机制的超短期光伏预测组合模型。首先,通过皮尔逊相关系数分析,选取与光伏发电功率密切相关的关键气象因子,并对其进行逐月标准化,然后加权求和计算得到分类指标天气条件因子(Sky Condition Factor, SCF),以降低输入变量的维度,并消除季节对天气分类的干扰和众多气象因子之间的耦合关系。其次,通过自组织映射神经网络(Self Organizing Map, SOM)对SCF进行无监督聚类,划分出3种天气类型。然后,在3种天气类型下分别构建卷积神经网络(Convolutional Neural Network, CNN)预测模型,并引入高效通道注意力模块（Efficient Channel Attention, ECA）,自适应地为特征信息的多重通道分配相应的权重,使模型集中于重要的特征信息,提高模型的预测精度。采用历史实测数据进行仿真,结果表明：与〖JP2〗未引入ECA模块的预测模型相比,所提预测模型在3种天气类型下的预测准确度分别提高了1.006 1%,〖JP〗1.626 1%和1.610 4%,验证了该模型的有效性。