基于深度学习的气象要素预测方法

针对气象预测内容繁多且影响因素多样的问题,提出了一种基于长短时记忆(LSTM)的气象预测方法.方法能够对繁杂的气象数据进行自动预处理,提取相应的特征信息.通过神经网络的前向训练、长短时记忆反馈学习,经过多隐藏层地自主训练,对能见度、温度、露点、风速、风向以及压力气象信息实现准确预测.通过实验以及与经典机器学习预测方法的...     

Estimating net radiation at surface using artificial neural networks: a new approach

This study describes the results of artificial neural network (ANN) models to estimate net radiation (R _n), at surface. Three ANN models were developed based on meteorological data such as wind velocity and direction, surface and air temperature, relative humidity, and soil moisture and temperature. A comparison has been made between the R _n estimates provided by the neural models and two linear models (LM) that need solar incoming shortwave radiation measurements as input parameter. Both ANN and LM results were tested against in situ measured R _n. For the LM ones, the estimations showed a root mean square error (RMSE) between 34.10 and 39.48?W?m^?2 and correlation coefficient (R ²) between 0.96 and 0.97 considering both the developing and the testing phases of calculations. The estimates obtained by the ANN models showed RMSEs between 6.54 and 48.75?W?m^?2 and R ² between 0.92 and 0.98 considering both the training and the testing phases. The ANN estimates are shown to be similar or even better, in some cases, than those given by the LMs. According to the authors?? knowledge, the use of ANNs to estimate R _n has not been discussed earlier, and based on the results obtained, it represents a formidable potential tool for R _n prediction using commonly measured meteorological parameters.     

Estimating daily pan evaporation using artificial neural network in a semi-arid environment

The objective of this study was to test an artificial neural network (ANN) for estimating the evaporation from pan (E _Pan) as a function of air temperature data in the Safiabad Agricultural Research Center (SARC) located in Khuzestan plain in the southwest of Iran. The ANNs (multilayer perceptron type) were trained to estimate E _Pan as a function of the maximum and minimum air temperature and extraterrestrial radiation. The data used in the network training were obtained from a historical series (1996–2001) of daily climatic data collected in weather station of SARC. The empirical Hargreaves equation (HG) is also considered for the comparison. The HG equation calibrated for converting grass evapotranspiration to open water evaporation by applying the same data used for neural network training. Two historical series (2002–2003) were utilized to test the network and for comparison between the ANN and calibrated Hargreaves method. The results show that both empirical and neural network methods provided closer agreement with the measured values (R ²?>?0.88 and RMSE?<?1.2 mm day^?1), but the ANN method gave better estimates than the calibrated Hargreaves method.     

模块化模糊神经网络的数值预报产品释用预报研究

综合应用预报量自身时间序列的拓展,数值预报产品和模块化模糊神经网络方法,进行了一种新的数值预报产品释用预报研究。将这种新方法与常规的数值预报产品完全预报(PP)方法进行了对比试验。结果表明,这种模块化模糊神经网络数值预报产品释用预报方法比PP预报方法的预报精度显著提高。并且,通过对预报模型“过拟合”现象的研究发现,这种模块化模糊神经网络的数值预报产品释用预报模型具有很好的泛化性能。     

基于1DCNN和LSTM的单站逐时气温预报方法

针对海量气象观测数据间存在大量的物理噪声、与气温无关的冗余特征以及时间相关性,提出了一种将一维卷积神经网络(1DCNN)和长短期记忆神经网络(LSTM)相结合的多信息融合气温预报方法。首先,运用差分法对气象观测数据进行预处理,得到平稳时间序列数据;其次,运用1DCNN提取与气温变化相关的特征变量作为神经网络模型的输入变量;最后,运用1DCNN和LSTM构建多信息融合气温预报模型1DCNNLSTM,并以云南省昆明市历史气象观测数据为例,与传统的LSTM、1DCNN和反向传播神经网络(BP)对未来24小时的逐时气温预报进行了比较研究。研究结果表明,1DCNN-LSTM的均方根误差(RMSE)相较于LSTM、1DCNN和BP最大降低了5.221%、19.350%和9.253%,平均绝对误差(MAE)最大降低了4.419%、17.520%和8.089%。为气温的精准预报提供了参考依据。     

Comparative analysis of regression and artificial neural network models for wind speed prediction

In this study, wind speed was modeled by linear regression (LR), nonlinear regression (NLR) and artificial neural network (ANN) methods. A three-layer feedforward artificial neural network structure was constructed and a backpropagation algorithm was used for the training of ANNs. To get a successful simulation, firstly, the correlation coefficients between all of the meteorological variables (wind speed, ambient temperature, atmospheric pressure, relative humidity and rainfall) were calculated taking two variables in turn for each calculation. All independent variables were added to the simple regression model. Then, the method of stepwise multiple regression was applied for the selection of the “best” regression equation (model). Thus, the best independent variables were selected for the LR and NLR models and also used in the input layer of the ANN. The results obtained by all methods were compared to each other. Finally, the ANN method was found to provide better performance than the LR and NLR methods.     

Relative importance of parameters affecting wind speed prediction using artificial neural networks

In traditional artificial neural networks (ANN) models, the relative importance of the individual meteorological input variables is often overlooked. A case study is presented in this paper to model monthly wind speed values using meteorological data (air pressure, air temperature, relative humidity, and precipitation), where the study also includes an estimate of the relative importance of these variables. Recorded monthly mean data are available at a gauging site in Tabriz, Azerbaijan, Iran, for the period from 2000 to 2005, gauged in the city at the outskirt of alluvial funneling mountains with an established microclimatic conditions and a diurnal wind regime. This provides a sufficiently severe test for the ANN model with a good predictive capability of 1 year of lead time but without any direct approach to refer the predicted results to local microclimatic conditions. A method is used in this paper to calculate the relative importance of each meteorological input parameters affecting wind speed, showing that air pressure and precipitation are the most and least influential parameters with approximate values of 40 and 10 %, respectively. This gained knowledge corresponds to the local knowledge of the microclimatic and geomorphologic conditions surrounding Tabriz.     

高血压病发病率预报的人工神经网络方法研究

通过统计分析，选取影响银川地区高血压病发病率的主要气象因素，将其作为输入变量经多层前馈型神经网络的BP(Back Propagation)算法进行学习训练，建立了疾病发病率的人工神经网络(ANN，Artifical Neural Net)预报模型。结果表明：该方法计算简便、误差较小，为疾病发病率预报提供了一种新的预报方法。     

Prediction of wind properties in urban environments using artificial neural network

Artificial neural network (ANN) modeling has been performed to predict turbulent boundary layer characteristics for rough terrain based on experimental tests conducted in a boundary-layer wind tunnel to simulate atmospheric boundary layer using passive roughness devices such as spires, barriers, roughness elements on the floor, and slots in the extended test section. Different configurations of passive devices assisted to simulate urban terrains. A part of the wind tunnel test results are used as training sets for the ANN, and the other part of the test results are used to compare the prediction results of the ANN. Two ANN models have been developed in this study. The first one has been used to predict mean velocity, turbulence intensity, and model length scale factor. Results show that ANN is an efficient, accurate, and robust modeling procedure to predict turbulent characteristics of wind. In particular, it was found that the ANN-predicted wind mean velocities are within 4.7%, turbulence intensities are within 6.2%, and model length scale factors are within 3.8% of the actual measured values. In addition, another ANN model has been developed to predict instantaneous velocities that enables calculating the power spectral density of longitudinal velocity fluctuations. Results show that the predicted power spectra are in a good agreement with the power spectra obtained from measured instantaneous velocities.     

Prediction of soil temperature using regression and artificial neural network models

In this study, monthly soil temperature was modeled by linear regression (LR), nonlinear regression (NLR) and artificial neural network (ANN) methods. The soil temperature and other meteorological parameters, which have been taken from Adana meteorological station, were observed between the years of 2000 and 2007 by the Turkish State Meteorological Service (TSMS). The soil temperatures were measured at depths of 5, 10, 20, 50 and 100 cm below the ground level. A three-layer feed-forward ANN structure was constructed and a back-propagation algorithm was used for the training of ANNs. In order to get a successful simulation, the correlation coefficients between all of the meteorological variables (soil temperature, atmospheric temperature, atmospheric pressure, relative humidity, wind speed, rainfall, global solar radiation and sunshine duration) were calculated taking them two by two. First, all independent variables were split into two time periods such as cold and warm seasons. They were added to the enter regression model. Then, the method of stepwise multiple regression was applied for the selection of the “best” regression equation (model). Thus, the best independent variables were selected for the LR and NLR models and they were also used in the input layer of the ANN method. Results of these methods were compared to each other. Finally, the ANN method was found to provide better performance than the LR and NLR methods.     

Hourly air temperature driven using multi-layer perceptron and radial basis function networks in arid and semi-arid regions

This study employed two artificial neural network (ANN) models, including multi-layer perceptron (MLP) and radial basis function (RBF), as data-driven methods of hourly air temperature at three meteorological stations in Fars province, Iran. MLP was optimized using the Levenberg–Marquardt (MLP_LM) training algorithm with a tangent sigmoid transfer function. Both time series (TS) and randomized (RZ) data were used for training and testing of ANNs. Daily maximum and minimum air temperatures (MM) and antecedent daily maximum and minimum air temperatures (AMM) constituted the input for ANNs. The ANN models were evaluated using the root mean square error (RMSE), the coefficient of determination (R ²) and the mean absolute error. The use of AMM led to a more accurate estimation of hourly temperature compared with the use of MM. The MLP-ANN seemed to have a higher estimation efficiency than the RBF ANN. Furthermore, the ANN testing using randomized data showed more accurate estimation. The RMSE values for MLP with RZ data using daily maximum and minimum air temperatures for testing phase were equal to 1.2°C, 1.8°C, and 1.7°C, respectively, at Arsanjan, Bajgah, and Kooshkak stations. The results of this study showed that hourly air temperature driven using ANNs (proposed models) had less error than the empirical equation.     

基于统计降尺度方法的夏季降水精细化预报

利用1999—2009年安徽省淮河以南地区60个县市站夏季逐日降水资料和安庆市探空站逐日资料,研究了中低层不同风向配置下局地降水与大尺度降水场之间的关系,以3种不同预报对象及相应的预报因子分别采用神经网络和线性回归方法设计6种预报模型对观测资料进行逼近和优化,从而实现空间降尺度.分析对比6种预报模型46站逐日降水量的拟合和预报效果,结果表明:采取相同的预报对象及预报因子的BP神经网络模型在拟合和预报效果上均好于线性回归模型,可见夏季降水场之间以非线性相关为主;神经网络模型预报结果同常用的Cressman插值预报相比,能很好地反映出降水的基本分布及局地特征;预报对象为单站降水序列的神经网络模型在以平原、河流为主要地形的区域预报效果较好,预报对象为REOF主成分的神经网络模型则在山地和丘陵地形区域预报效果较好.     

TIMESERIES PREDICTION MODEL CONSISTING OF ARTIFICIAL NEURAL NETWORK AND GENETIC ALGORITHM

The paper introduces the basic concept and flow diagram of genetic algorithm (GA) and themerits and demerits of artificial neural network (ANN) as a timeseries prediction model andthereupon developed is a new model with ANN and GA in combination. Eventually, calculationsare presented with the results and model examined.     

A Short-Term Climate Prediction Model Based on a Modular Fuzzy Neural Network

In terms of the modular fuzzy neural network (MFNN) combining fuzzy c-mean (FCM) cluster and single-layer neural network, a short-term climate prediction model is developed. It is found from modeling results that the MFNN model for short-term climate prediction has advantages of simple structure, no hidden layer and stable network parameters because of the assembling of sound functions of the selfadaptive learning, association and fuzzy information processing of fuzzy mathematics and neural network methods. The case computational results of Guangxi flood season (JJA) rainfall show that the mean absolute error (MAE) and mean relative error (MRE) of the prediction during 1998 2002 are 68.8 mm and 9.78%, and in comparison with the regression method, under the conditions of the same predictors and period they are 97.8 mm and 12.28% respectively. Furthermore, it is also found from the stability analysis of the modular model that the change of the prediction results of independent samples with training times in the stably convergent interval of the model is less than 1.3 mm. The obvious oscillation phenomenon of prediction results with training times, such as in the common back-propagation neural network (BPNN) model, does not occur, indicating a better practical application potential of the MFNN model.     

TIMESERIES PREDICTION MODEL CONSISTING OF ARTIFICIAL NEURAL NETWORK AND GENETIC ALGORITHM*

The paper introduces the basic concept and flow diagram of genetic algorithm(GA) and the merits and demerits of artificial neural network(ANN) as a timeseries prediction model and thereupon developed is a new model with ANN and GA in combination. Eventually, calculations are presented with the results and model examined.     

时问序列动态学习率神经网络模型及其初步试验

神经网络在气象上的应用往往是采用固定学习率的BP算法建模,学习过程易出现振荡现象和网络存在冗余连接等缺陷,基于此对神经网络进行了改进。利用时间序列分析方法对样本数据进行处理,用改进后的神经网络对时间序列样本数据进行训练预测,创建了时间序列动态学习率神经网络模型。最后用库车县1997—2007年四季的平均气温值作样本数据进行训练,其训练精度和拟合度都达到很高的标准,用该模型预测了库车县2008年的气温。通过实例证明这个模型在气象预测领域有一定的实用价值。     

Estimation of dew point temperature using neuro-fuzzy and neural network techniques

This study investigates the ability of two different artificial neural network (ANN) models, generalized regression neural networks model (GRNNM) and Kohonen self-organizing feature maps neural networks model (KSOFM), and two different adaptive neural fuzzy inference system (ANFIS) models, ANFIS model with sub-clustering identification (ANFIS-SC) and ANFIS model with grid partitioning identification (ANFIS-GP), for estimating daily dew point temperature. The climatic data that consisted of 8 years of daily records of air temperature, sunshine hours, wind speed, saturation vapor pressure, relative humidity, and dew point temperature from three weather stations, Daego, Pohang, and Ulsan, in South Korea were used in the study. The estimates of ANN and ANFIS models were compared according to the three different statistics, root mean square errors, mean absolute errors, and determination coefficient. Comparison results revealed that the ANFIS-SC, ANFIS-GP, and GRNNM models showed almost the same accuracy and they performed better than the KSOFM model. Results also indicated that the sunshine hours, wind speed, and saturation vapor pressure have little effect on dew point temperature. It was found that the dew point temperature could be successfully estimated by using T _mean and R _H variables.     

All India summer monsoon rainfall prediction using an artificial neural network

The prediction of Indian summer monsoon rainfall (ISMR) on a seasonal time scales has been attempted by various research groups using different techniques including artificial neural networks. The prediction of ISMR on monthly and seasonal time scales is not only scientifically challenging but is also important for planning and devising agricultural strategies. This article describes the artificial neural network (ANN) technique with error- back-propagation algorithm to provide prediction (hindcast) of ISMR on monthly and seasonal time scales. The ANN technique is applied to the five time series of June, July, August, September monthly means and seasonal mean (June + July + August + September) rainfall from 1871 to 1994 based on Parthasarathy data set. The previous five years values from all the five time-series were used to train the ANN to predict for the next year. The details of the models used are discussed. Various statistics are calculated to examine the performance of the models and it is found that the models could be used as a forecasting tool on seasonal and monthly time scales. It is observed by various researchers that with the passage of time the relationships between various predictors and Indian monsoon are changing, leading to changes in monsoon predictability. This issue is discussed and it is found that the monsoon system inherently has a decadal scale variation in predictability. Received: 13 March 1999 / Accepted: 31 August 1999     

深度学习方法在干旱预测中的应用

使用标准化降水蒸散指数作为农业干旱的监测指标,以站点气象要素和大尺度环流要素为驱动变量,建立干旱预测模型,分析评价传统的整合移动平均自回归(ARIMA)时间序列模型以及不同深度神经网络模型(DNN)的预测效果.结果 表明:DNN模型的总体预测能力优于ARIMA模型;同基于长短期记忆网络(LSTM)提出的传统LSTM预测...