随着传统化石能源面临枯竭的问题日益加剧,使用太阳能进行光伏发电成为世界各国能源结构调整的重要方向,如何进一步提高光伏发电功率的预测精度成为亟待解决的问题。为提高光伏功率短期预测的准确性和可靠性,提出一种耦合太阳辐射预报...随着传统化石能源面临枯竭的问题日益加剧,使用太阳能进行光伏发电成为世界各国能源结构调整的重要方向,如何进一步提高光伏发电功率的预测精度成为亟待解决的问题。为提高光伏功率短期预测的准确性和可靠性,提出一种耦合太阳辐射预报模式系统(weather research and forecasting model for solar energy,WRF-Solar)及辐照度订正的光伏短期预测模型,先使用WRF-Solar进行动力降尺度天气数值预报,得到包含辐照度等在内的未来气象因子,再利用随机森林对预报辐照度进行订正,在此基础上运用长短期神经网络、反向传播神经网络和逐步聚类分析建立光伏功率短期预测模型,利用某40 MW光伏电站的实际运行数据进行模型对比分析。结果表明,使用随机森林模型订正后的辐照度更接近真实值,平均绝对误差率下降了56.06个百分点;与另外2种模型预测结果对比发现,长短期神经网络模型预测效果最好,平均绝对百分比误差降低了4.13个百分点,说明组合模型能够进一步提高功率预测的精度。展开更多
Solar energy is a pivotal clean energy source in the transition to carbon neutrality from fossil fuels.However,the intermittent and stochastic characteristics of solar radiation pose challenges for accurate simulation...Solar energy is a pivotal clean energy source in the transition to carbon neutrality from fossil fuels.However,the intermittent and stochastic characteristics of solar radiation pose challenges for accurate simulation and prediction.Accurately simulating and predicting solar radiation and its variability are crucial for optimizing solar energy utilization.This study conducted simulation experiments using the WRF-Solar model from 25 June to 25 July 2022,to evaluate the accuracy and performance of the simulated solar radiation across China.The simulations covered the whole country with a grid spacing of 27 km and were compared with ground observation network data from the Chinese Ecosystem Research Network.The results indicated that WRF-Solar can accurately capture the spatiotemporal patterns of global horizontal irradiance over China,but there is still an overestimation of solar radiation,and the model underestimates the total cloud cover.The root-mean-square error ranged from 92.83 to 188.13 W m^(-2) and the mean bias(MB)ranged from 21.05 to 56.22 W m^(-2).The simulation showed the smallest MB at Lhasa on the Qinghai–Tibet Plateau,while the largest MB was observed in Southeast China.To enhance the accuracy of solar radiation simulation,the authors compared the Fast All-sky Radiation Model for Solar with the Rapid Radiative Transfer Model for General Circulation Models and found that the former provides better simulation.展开更多
采用中尺度气象模式WRF(Weather Research Forecast)对北京地区的太阳辐射进行了4个典型月的逐时预报试验,用南郊观象台的辐射观测数据对预报结果进行了对比分析和初步订正试验。结果表明:在现有模式条件下,5 km分辨率的短波辐射预报结...采用中尺度气象模式WRF(Weather Research Forecast)对北京地区的太阳辐射进行了4个典型月的逐时预报试验,用南郊观象台的辐射观测数据对预报结果进行了对比分析和初步订正试验。结果表明:在现有模式条件下,5 km分辨率的短波辐射预报结果和1 km分辨率预报结果无明显差别;WRF模式对太阳辐射的预报性能在晴天较好,多云天次之,在满云或阴雨天最差;通过误差分解发现,位相偏差、系统偏差及振幅偏差在各月对均方根误差的贡献有明显差异;针对模式预报结果的系统偏差和振幅偏差。经过简单的线性订正可以较明显地改进模式预报结果;双偏订正(DBC)法比线性回归(LR)法对预报误差的改进效果略明显;仅通过简单的线性订正,位相差很难消除,需要针对位相差研究新的订正方法。展开更多
太阳能光伏发电已成为仅次于水电和风能的第三大可再生能源,光伏发电受云量时空变化的影响较大,因此准确模拟云天太阳辐射的时空变化对电网安全运行至关重要。围绕如何减小中尺度气象模式的云初始场误差,进而改进云天的太阳辐射模拟这...太阳能光伏发电已成为仅次于水电和风能的第三大可再生能源,光伏发电受云量时空变化的影响较大,因此准确模拟云天太阳辐射的时空变化对电网安全运行至关重要。围绕如何减小中尺度气象模式的云初始场误差,进而改进云天的太阳辐射模拟这一关键科学问题,本文通过研究基于卫星资料同化的LAPS(Local Analysis Prediction System)多时间层三维云分析同化方法,改进三维云结构,并将LAPS模式输出结果作为WRF(Weather Research and Forecasting)模式的初始场,模拟了2008年1月及夏季(6~8月)北京地区的总云量和总辐射的时空分布,重点分析了多云和有降水天气过程总辐射的模拟改进效果及其原因。结果表明,同化前后的总云量模拟值与观测值的时间变化趋势基本一致,但大部分时次总云量的模拟值低于观测值;大部分多云及降水时段同化后总云量模拟值较接近于实测值。1月晴天、多云天以及夏季晴天同化前后总辐射模拟值与实测值的时间变化趋势较一致,但同化前后两者的相关性差异不明显;晴天条件下同化前后总辐射模拟值均低于实测值,1月多云条件下多数时段同化后总辐射模拟误差减小不明显,与总云量的改进效果不显著有关。夏季多云、有降水及6月典型降水三种天气条件下同化前后总辐射模拟值与观测值的相关性稍差,同化后两者的相关性较同化前有所改进,尤其是6月典型降水过程改进效果较明显;同化前总辐射模拟误差较大,而同化后误差显著减小,尤其是6月典型降水过程同化后均方根误差和平均相对误差较同化前分别减小了102.6 W m^(-2)和355.9%,最大相对误差减小更显著;同化后总辐射模拟误差小于同化前的比例高达75%,即大部分时刻同化后模拟误差小于同化前。多云和有降水天气过程总辐射模拟效果的显著改进与总云量的改进密切相关,即同化后总云量模拟值增加,云的反射和散射作用增强,导致模拟总辐射减小,即更接近于实测总辐射值。研究结果对于多云和降水天气条件下太阳辐射的模拟效果改进、太阳能资源客观评估以及光伏电站的发电量预测具有一定的科学和实际应用价值。展开更多
This paper investigates the accuracy of weather research and forecasting by improving coding for solar radiation forecasting for location in Dili Timor Leste. Weather Research and Forecasting (WRF) model version 3.9.1...This paper investigates the accuracy of weather research and forecasting by improving coding for solar radiation forecasting for location in Dili Timor Leste. Weather Research and Forecasting (WRF) model version 3.9.1 is used in this study for improvement purposes. The shortwave coding of WRF is used to improve in order to decrease error simulation. The importance of improving WRF coding at a specific region will reduce the bias and root mean square root when comparing to the observed data. This study uses high resolution based on the WRF modeling to stabilize the performance of forecasting. The decrease in error performance will be expected to enhance the value of renewable energy. The results show the root mean square error of the WRF default is 233 W/m<sup>2</sup> higher compared to 205 W/m<sup>2</sup> from the WRF improvement model. In addition, the Mean Bias Error (MBE) of the WRF default is obtained value 0.06 higher than 0.03 from the WRF improvement in rainy days. Meanwhile, on sunny days, the performance Root Mean Square Error (RMSE) of WRF default is 327 W/m<sup>2</sup> higher than 223 W/m<sup>2</sup> from the WRF improvement. The MBE of WRF improvement obtained 0.13 lower compared to 0.21 of WRF default coding. Finally, this study concludes that improving the shortwave code under the WRF model can decrease the error performance of the WRF simulation for local weather forecasting</span></span><span style="font-family:Verdana;">.展开更多
A study of a combination of Weather Research and Forecasting (WRF) model and Long Short Term Memory (LSTM) network for location in Dili Timor Leste is introduced in this paper. One calendar year’s results of solar ra...A study of a combination of Weather Research and Forecasting (WRF) model and Long Short Term Memory (LSTM) network for location in Dili Timor Leste is introduced in this paper. One calendar year’s results of solar radiation from January to December 2014 are used as input data to estimate future forecasting of solar radiation using the LSTM network for three months period. The WRF model version 3.9.1 is used to simulate one year’s solar radiation in horizontal resolution low scale for nesting domain 1</span><span style="font-family:""> </span><span style="font-family:Verdana;">×</span><span style="font-family:""> </span><span style="font-family:Verdana;">1 km. It is done by applying 6-hourly interval 1</span><span style="font-family:Verdana;">º</span><span style="font-family:""> </span><span style="font-family:""><span style="font-family:Verdana;">×</span><span style="font-family:Verdana;"> 1</span></span><span style="font-family:Verdana;">º</span><span style="font-family:""><span style="font-family:Verdana;"> NCEP FNL analysis data used as Global Forecast System (GFS). LSTM network is applied for forecasting in numerous learning problems for solar radiation forecasting. LSTM network uses two-layer LSTM architecture of 512 hidden neurons coupled with a dense output layer with linear as the model activation to predict with time steps are configured to 50 and the number of features is 1. The maximum epoch is set to 325 with batch size 300 and the validation split is 0.09. The results demonstrate that the combination of these two methods can successfully predict solar radiation where four error metrics of mean bias error (MBE), root mean square error (RMSE), normalized MBE (nMBE), and normalized RMSE (nRMSE) perform small error distribution and percentage in three months prediction where the error percentage is obtained below the 20% for nMBE and nRMSE. Meanwhile, the error distribution of RMSE is obtained below 200 W/m</span><sup><span style="font-family:Verdana;">2</span></sup><span style="font-family:Verdana;"> and maximum bias error is 0.07. Finally, the values of MBE, RMSE, nMBE, and nRMSE conclude that the good performance of the combination of two methods in this study can be applied to simulate any other weather variable for local necessary.展开更多
Study of comparison of solar power generation between the GridLAB-D tool and System Advisor Model (SAM) in Dili, Timor Leste is presented in this paper. Weather Research and Forecasting (WRF) model is used to simulate...Study of comparison of solar power generation between the GridLAB-D tool and System Advisor Model (SAM) in Dili, Timor Leste is presented in this paper. Weather Research and Forecasting (WRF) model is used to simulate solar radiation for one calendar year from January to December 2014 using six-hourly interval 1° × 1° NCEP FNL analysis data. The one calendar year results from the WRF model will be used as input data for GridLAB-D and SAM to estimate the solar power generation. GridLAB-D is an open-source and analysis tool designed to operate the distribution power systems with a high-performance algorithm. System Advisor Model version SAM 2017.9.5 is used to estimate solar power performance with Photovoltaics (PVWatts)-<span style="font-family:;" "=""> <span style="font-family:;" "="">Commercial Distributed model. This model is designed to analyze the performance and the financing of renewable energy for electricity generation. The results show the lowest solar radiation is 512 W/m<sup>2</sup> obtained in June with an average monthly power of 20.6 kW and 30.55 kW generated from the SAM model and the GridLAB-D simulator, respectively. Meanwhile, the highest solar radiation is 1100 W/m<sup>2</sup>, 1112 W/m<sup>2</sup>, 1046 W/m<sup>2</sup>, and 1077 W/m<sup>2</sup> obtained in October, November, December, and January with an average monthly power of 55.72 kW, 62.44 kW, 56.65 kW, and 56.97 kW from the SAM model, in the other hand, 48.89 kW, 51.31 kW, 55.51 kW, and 57.18 kW generated by the GridLAB-D. Finally, the results show that the performance of the GridLAB-D and the SAM model was quite good because both model precisely presented values are almost closest to each other. This study proposes that the results of solar output power from both methods, GridLAB-D and SAM can be used to design grid-connected or stand-alone electric power projects to increase the quality of electricity generation in Dili, Timor Leste.</span></span>展开更多
区域气候模式对高原降水模拟存在系统性高估。该文从热力学角度考虑了局地(小尺度)地形对陆气过程的影响,以提高区域气候模式高原降水模拟能力。通过在天气研究与预报(weather research and forecasting,WRF)模式陆面过程模块耦合地形...区域气候模式对高原降水模拟存在系统性高估。该文从热力学角度考虑了局地(小尺度)地形对陆气过程的影响,以提高区域气候模式高原降水模拟能力。通过在天气研究与预报(weather research and forecasting,WRF)模式陆面过程模块耦合地形热力效应方案,将二维地形太阳辐射方案改进为三维地形太阳辐射方案,根据山区太阳辐射理论对太阳辐射直射、辐射散射、地形反射过程进行改进,得到改进方案;基于改进方案在亚东河谷开展高分辨率WRF模式降水模拟研究。结果表明:采用改进方案模拟的太阳辐射、降水结果与实测结果更接近,能够降低河谷降水高估的现象。改进方案能够体现出更多地形变化对地表辐射分布的影响:日间,山体坡面由于接收更多辐射得到加热(0.09~0.20℃),产生上升气流,在坡面形成更强上坡风(0.02~0.08 m/s),将河谷水汽携带到坡面以上,水汽在坡面以上区域爬升并形成降水,导致河谷区域降水减少(-4.54~-3.34 mm)、坡面以上区域降水增多(0.59~2.82 mm)。该研究对提高区域气候模式对山区降水的模拟效果具有重要意义。展开更多
文摘随着传统化石能源面临枯竭的问题日益加剧,使用太阳能进行光伏发电成为世界各国能源结构调整的重要方向,如何进一步提高光伏发电功率的预测精度成为亟待解决的问题。为提高光伏功率短期预测的准确性和可靠性,提出一种耦合太阳辐射预报模式系统(weather research and forecasting model for solar energy,WRF-Solar)及辐照度订正的光伏短期预测模型,先使用WRF-Solar进行动力降尺度天气数值预报,得到包含辐照度等在内的未来气象因子,再利用随机森林对预报辐照度进行订正,在此基础上运用长短期神经网络、反向传播神经网络和逐步聚类分析建立光伏功率短期预测模型,利用某40 MW光伏电站的实际运行数据进行模型对比分析。结果表明,使用随机森林模型订正后的辐照度更接近真实值,平均绝对误差率下降了56.06个百分点;与另外2种模型预测结果对比发现,长短期神经网络模型预测效果最好,平均绝对百分比误差降低了4.13个百分点,说明组合模型能够进一步提高功率预测的精度。
基金supported by the National Natural Science Foundation of China[grant number 42175132]the National Key R&D Program[grant number 2020YFA0607802]the CAS Information Technology Program[grant number CAS-WX2021SF-0107-02]。
文摘Solar energy is a pivotal clean energy source in the transition to carbon neutrality from fossil fuels.However,the intermittent and stochastic characteristics of solar radiation pose challenges for accurate simulation and prediction.Accurately simulating and predicting solar radiation and its variability are crucial for optimizing solar energy utilization.This study conducted simulation experiments using the WRF-Solar model from 25 June to 25 July 2022,to evaluate the accuracy and performance of the simulated solar radiation across China.The simulations covered the whole country with a grid spacing of 27 km and were compared with ground observation network data from the Chinese Ecosystem Research Network.The results indicated that WRF-Solar can accurately capture the spatiotemporal patterns of global horizontal irradiance over China,but there is still an overestimation of solar radiation,and the model underestimates the total cloud cover.The root-mean-square error ranged from 92.83 to 188.13 W m^(-2) and the mean bias(MB)ranged from 21.05 to 56.22 W m^(-2).The simulation showed the smallest MB at Lhasa on the Qinghai–Tibet Plateau,while the largest MB was observed in Southeast China.To enhance the accuracy of solar radiation simulation,the authors compared the Fast All-sky Radiation Model for Solar with the Rapid Radiative Transfer Model for General Circulation Models and found that the former provides better simulation.
文摘采用中尺度气象模式WRF(Weather Research Forecast)对北京地区的太阳辐射进行了4个典型月的逐时预报试验,用南郊观象台的辐射观测数据对预报结果进行了对比分析和初步订正试验。结果表明:在现有模式条件下,5 km分辨率的短波辐射预报结果和1 km分辨率预报结果无明显差别;WRF模式对太阳辐射的预报性能在晴天较好,多云天次之,在满云或阴雨天最差;通过误差分解发现,位相偏差、系统偏差及振幅偏差在各月对均方根误差的贡献有明显差异;针对模式预报结果的系统偏差和振幅偏差。经过简单的线性订正可以较明显地改进模式预报结果;双偏订正(DBC)法比线性回归(LR)法对预报误差的改进效果略明显;仅通过简单的线性订正,位相差很难消除,需要针对位相差研究新的订正方法。
文摘太阳能光伏发电已成为仅次于水电和风能的第三大可再生能源,光伏发电受云量时空变化的影响较大,因此准确模拟云天太阳辐射的时空变化对电网安全运行至关重要。围绕如何减小中尺度气象模式的云初始场误差,进而改进云天的太阳辐射模拟这一关键科学问题,本文通过研究基于卫星资料同化的LAPS(Local Analysis Prediction System)多时间层三维云分析同化方法,改进三维云结构,并将LAPS模式输出结果作为WRF(Weather Research and Forecasting)模式的初始场,模拟了2008年1月及夏季(6~8月)北京地区的总云量和总辐射的时空分布,重点分析了多云和有降水天气过程总辐射的模拟改进效果及其原因。结果表明,同化前后的总云量模拟值与观测值的时间变化趋势基本一致,但大部分时次总云量的模拟值低于观测值;大部分多云及降水时段同化后总云量模拟值较接近于实测值。1月晴天、多云天以及夏季晴天同化前后总辐射模拟值与实测值的时间变化趋势较一致,但同化前后两者的相关性差异不明显;晴天条件下同化前后总辐射模拟值均低于实测值,1月多云条件下多数时段同化后总辐射模拟误差减小不明显,与总云量的改进效果不显著有关。夏季多云、有降水及6月典型降水三种天气条件下同化前后总辐射模拟值与观测值的相关性稍差,同化后两者的相关性较同化前有所改进,尤其是6月典型降水过程改进效果较明显;同化前总辐射模拟误差较大,而同化后误差显著减小,尤其是6月典型降水过程同化后均方根误差和平均相对误差较同化前分别减小了102.6 W m^(-2)和355.9%,最大相对误差减小更显著;同化后总辐射模拟误差小于同化前的比例高达75%,即大部分时刻同化后模拟误差小于同化前。多云和有降水天气过程总辐射模拟效果的显著改进与总云量的改进密切相关,即同化后总云量模拟值增加,云的反射和散射作用增强,导致模拟总辐射减小,即更接近于实测总辐射值。研究结果对于多云和降水天气条件下太阳辐射的模拟效果改进、太阳能资源客观评估以及光伏电站的发电量预测具有一定的科学和实际应用价值。
文摘This paper investigates the accuracy of weather research and forecasting by improving coding for solar radiation forecasting for location in Dili Timor Leste. Weather Research and Forecasting (WRF) model version 3.9.1 is used in this study for improvement purposes. The shortwave coding of WRF is used to improve in order to decrease error simulation. The importance of improving WRF coding at a specific region will reduce the bias and root mean square root when comparing to the observed data. This study uses high resolution based on the WRF modeling to stabilize the performance of forecasting. The decrease in error performance will be expected to enhance the value of renewable energy. The results show the root mean square error of the WRF default is 233 W/m<sup>2</sup> higher compared to 205 W/m<sup>2</sup> from the WRF improvement model. In addition, the Mean Bias Error (MBE) of the WRF default is obtained value 0.06 higher than 0.03 from the WRF improvement in rainy days. Meanwhile, on sunny days, the performance Root Mean Square Error (RMSE) of WRF default is 327 W/m<sup>2</sup> higher than 223 W/m<sup>2</sup> from the WRF improvement. The MBE of WRF improvement obtained 0.13 lower compared to 0.21 of WRF default coding. Finally, this study concludes that improving the shortwave code under the WRF model can decrease the error performance of the WRF simulation for local weather forecasting</span></span><span style="font-family:Verdana;">.
文摘A study of a combination of Weather Research and Forecasting (WRF) model and Long Short Term Memory (LSTM) network for location in Dili Timor Leste is introduced in this paper. One calendar year’s results of solar radiation from January to December 2014 are used as input data to estimate future forecasting of solar radiation using the LSTM network for three months period. The WRF model version 3.9.1 is used to simulate one year’s solar radiation in horizontal resolution low scale for nesting domain 1</span><span style="font-family:""> </span><span style="font-family:Verdana;">×</span><span style="font-family:""> </span><span style="font-family:Verdana;">1 km. It is done by applying 6-hourly interval 1</span><span style="font-family:Verdana;">º</span><span style="font-family:""> </span><span style="font-family:""><span style="font-family:Verdana;">×</span><span style="font-family:Verdana;"> 1</span></span><span style="font-family:Verdana;">º</span><span style="font-family:""><span style="font-family:Verdana;"> NCEP FNL analysis data used as Global Forecast System (GFS). LSTM network is applied for forecasting in numerous learning problems for solar radiation forecasting. LSTM network uses two-layer LSTM architecture of 512 hidden neurons coupled with a dense output layer with linear as the model activation to predict with time steps are configured to 50 and the number of features is 1. The maximum epoch is set to 325 with batch size 300 and the validation split is 0.09. The results demonstrate that the combination of these two methods can successfully predict solar radiation where four error metrics of mean bias error (MBE), root mean square error (RMSE), normalized MBE (nMBE), and normalized RMSE (nRMSE) perform small error distribution and percentage in three months prediction where the error percentage is obtained below the 20% for nMBE and nRMSE. Meanwhile, the error distribution of RMSE is obtained below 200 W/m</span><sup><span style="font-family:Verdana;">2</span></sup><span style="font-family:Verdana;"> and maximum bias error is 0.07. Finally, the values of MBE, RMSE, nMBE, and nRMSE conclude that the good performance of the combination of two methods in this study can be applied to simulate any other weather variable for local necessary.
文摘Study of comparison of solar power generation between the GridLAB-D tool and System Advisor Model (SAM) in Dili, Timor Leste is presented in this paper. Weather Research and Forecasting (WRF) model is used to simulate solar radiation for one calendar year from January to December 2014 using six-hourly interval 1° × 1° NCEP FNL analysis data. The one calendar year results from the WRF model will be used as input data for GridLAB-D and SAM to estimate the solar power generation. GridLAB-D is an open-source and analysis tool designed to operate the distribution power systems with a high-performance algorithm. System Advisor Model version SAM 2017.9.5 is used to estimate solar power performance with Photovoltaics (PVWatts)-<span style="font-family:;" "=""> <span style="font-family:;" "="">Commercial Distributed model. This model is designed to analyze the performance and the financing of renewable energy for electricity generation. The results show the lowest solar radiation is 512 W/m<sup>2</sup> obtained in June with an average monthly power of 20.6 kW and 30.55 kW generated from the SAM model and the GridLAB-D simulator, respectively. Meanwhile, the highest solar radiation is 1100 W/m<sup>2</sup>, 1112 W/m<sup>2</sup>, 1046 W/m<sup>2</sup>, and 1077 W/m<sup>2</sup> obtained in October, November, December, and January with an average monthly power of 55.72 kW, 62.44 kW, 56.65 kW, and 56.97 kW from the SAM model, in the other hand, 48.89 kW, 51.31 kW, 55.51 kW, and 57.18 kW generated by the GridLAB-D. Finally, the results show that the performance of the GridLAB-D and the SAM model was quite good because both model precisely presented values are almost closest to each other. This study proposes that the results of solar output power from both methods, GridLAB-D and SAM can be used to design grid-connected or stand-alone electric power projects to increase the quality of electricity generation in Dili, Timor Leste.</span></span>
文摘区域气候模式对高原降水模拟存在系统性高估。该文从热力学角度考虑了局地(小尺度)地形对陆气过程的影响,以提高区域气候模式高原降水模拟能力。通过在天气研究与预报(weather research and forecasting,WRF)模式陆面过程模块耦合地形热力效应方案,将二维地形太阳辐射方案改进为三维地形太阳辐射方案,根据山区太阳辐射理论对太阳辐射直射、辐射散射、地形反射过程进行改进,得到改进方案;基于改进方案在亚东河谷开展高分辨率WRF模式降水模拟研究。结果表明:采用改进方案模拟的太阳辐射、降水结果与实测结果更接近,能够降低河谷降水高估的现象。改进方案能够体现出更多地形变化对地表辐射分布的影响:日间,山体坡面由于接收更多辐射得到加热(0.09~0.20℃),产生上升气流,在坡面形成更强上坡风(0.02~0.08 m/s),将河谷水汽携带到坡面以上,水汽在坡面以上区域爬升并形成降水,导致河谷区域降水减少(-4.54~-3.34 mm)、坡面以上区域降水增多(0.59~2.82 mm)。该研究对提高区域气候模式对山区降水的模拟效果具有重要意义。
