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.     

人工神经网络与遗传算法结合的时间序列预测模式

介绍了遗传算法的基本概念和流程，阐述了人工神经网络作为时间序列预测模式的可行性和不足之处，并提出了人工网络与遗传算法相结合的时间序列预测模式，最后给出了该算法的计算结果，并对结果和模式作了讨论。     

ARTIFICIAL NEURAL NETWORK CORRECTION OF ROUGH ERRORS OF OBSERVATIONS*

In the context of tower measured radiation datasets.following the correction principle meeting a diagnostic equation in data quality control and in terms of a technique for model construction on data and ANN (artificial neural network) retrieval for BP correction of radiation measurements with rough errors available,a BP model is presented.Evidence suggests that the developed model works well and is superior to a convenient multivariate linear regression model,indicating its wide applications.     

ARTIFICIAL NEURAL NETWORK CORRECTION OF ROUGH ERRORS OF OBSERVATIONS

In the context of tower measured radiation datasets.following the correction principle meetinga diagnostic equation in data quality control and in terms of a technique for model construction ondata and ANN(artificial neural network)retrieval for BP correction of radiation measurementswith rough errors available,a BP model is presented.Evidence suggests that the developed modelworks well and is superior to a convenient multivariate linear regression model,indicating its wideapplications.     

NEURAL NETWORK BP MODEL APPROXIMATION AND PREDICTION OF COMPLICATED WEATHER SYSTEMS*

An artificial neural network BP model and its revised algorithm are used to approximate quite successfully a Lorenz chaotic dynamic system and the mapping relation is established between the indices of Southern Oscillation and equatorial zonal wind and lagged equatorial eastern Pacific sea surface temperature(SST) in the context of NCEP/NCAR data,and thereby a model is prepared.The constructed net model shows fairly high fit precision and feasible prediction accuracy,thus making itself of some usefulness to the prognosis of intricate weather systems.     

用遗传算法优化小波神经网络的结构

用遗传算法确定小波神经网络中输入层单元数和隐含层单元数,同时采用梯度法计算小波神经网络中的权系数、伸缩和平移系数,从而达到优化小波神经网络的结构的目的。     

遗传算法进化设计BP神经网络气象预报建模研究

利用遗传算法进化设计神经网络的结构和连接权,并针对遗传算法局部调节能力比较弱的问题,采用从进化后的神经网络中用训练样本再次寻优的方法,建立神经网络气象预报模型,该方法克服了神经网络极易陷入局部解和遗传算法局部调节能力比较弱的问题,以广西的月降水量进行实例分析,计算结果表明该方法预报精度高、而且稳定。     

STUDY ON MIXED MODEL OF NEURAL NETWORK FOR FARMLAND FLOOD/DROUGHT PREDICTION

The paper concerns a flood/drought prediction model involving the continuation of time seriesof a predictand and the physical factors influencing the change of predictand.Attempt is made toconstruct the model by the neural network scheme for the nonlinear mapping relation based onmulti-input and single output.The model is found of steadily higher predictive accuracy by testingthe output from one and multiple stepwise predictions against observations and comparing theresults to those from a traditional statistical model.     

STUDY ON MIXED MODEL OF NEURAL NETWORK FOR FARMLAND FLOOD/DROUGHT PREDICTION*

The paper concerns a flood/drought prediction model involving the continuation of time series of a predictand and the physical factors influencing the change of predictand.Attempt is made to construct the model by the neural network scheme for the nonlinear mapping relation based on multi-input and single output.The model is found of steadily higher predictive accuracy by testing the output from one and multiple stepwise predictions against observations and comparing the results to those from a traditional statistical model.     

NEURAL NETWORK BP MODEL APPROXIMATION AND PREDICTION OF COMPLICATED WEATHER SYSTEMS

An artificial neural network BP model and its revised algorithm are used to approximate quitesuccessfully a Lorenz chaotic dynamic system and the mapping relation is established between theindices of Southern Oscillation and equatorial zonal wind and lagged equatorial eastern Pacific seasurface temperature(SST) in the context of NCEP/NCAR data,and thereby a model is prepared.The constructed net model shows fairly high fit precision and feasible prediction accuracy,thusmaking itself of some usefulness to the prognosis of intricate weather systems.     

热带气旋路径人工神经元预报方法对比试验研究

分别对具有动量项BP、LM、RBF人工神经网络建立36、48、60、72小时的热带气旋路径预测模型,各用100个独立样本进行预测检验,分析了网络"学习好,预报差"的原因,解决这一问题的关键是选择合适的网络结构参数、相应的学习算法和合适的预报因子,并总结了合理应用人工神经网络建立预测模型的经验.针对人工神经网络模型不具有自动选取因子的功能,给实际应用造成困难,提出了基于RBF的逐步选取因子的算法,并进行了对比试验,表明该方法具有较高的实用价值.     

基于BP神经网络与遗传算法反演大气温湿廓线

为提高地基微波辐射计大气探测精度,融合BP神经网络与遗传算法,研究0~10 km大气温湿度廓线。首先,结合数据特征,基于数值模拟技术,建立一套TP/WVP-3000型号地基微波辐射计的一级数据质量控制和订正模型。然后,为减小训练样本代表性误差对模型反演精度的影响,利用遗传算法优化训练样本数据,建立一套精度更高的神经网络大气温湿度反演模型。最后,利用构建的反演模型,开展大气温湿度反演试验,结合探空资料和微波辐射计二级产品,评价反演模型精度。研究结果表明:(1)经过质量控制后的实测数据与模拟数据之间的相关性有显著提升;(2)经过质量控制与订正后建立的神经网络模型对比原微波辐射计二级产品的反演精度有一定提升,温度提升6.77%,湿度提升20.11%;(3)经过遗传算法优化后的训练样本所建立的神经网络反演模型对比原微波辐射计二级产品反演精度有进一步的提升,温度提升10.21%,湿度提升23.75%,反演结果与该地区同类型研究结果相比有着较大提升。      

人工神经网络方法在夏季降水预报中的应用

在夏季雨型预报中引进了人工神经网络方法。首先,根据雨型与前期（冬季）环流和海温的关系,从前期冬季资料场中找预报因子;然后,用人工神经网络方法对我国夏季的雨型进行模拟预报,以前４０年资料做训练样本,让网络在一定的学习规则下进行学习,最后得到一种分类预报模型。经对１９９２～１９９６年夏季雨型做独立试报,结果与实况基本相符。     

人工神经网络预报模型的过拟合研究

针对神经网络方法在预报建模中存在的“过拟合”(overfitting)现象和提高泛化性能 (generalizationcapability)问题 ,提出了采用主成分分析构造神经网络低维学习矩阵的预报建模方法。研究结果表明 ,这种新的神经网络预报建模方法 ,通过浓缩预报信息 ,降维去噪 ,使得神经网络的预报建模不需要进行适宜隐节点数的最优网络结构试验 ,没有“过拟合”现象 ,并且与传统的神经网络预报建模方法及逐步回归预报模型相比泛化能力有显著提高     

运用天气雷达识别强对流天气的人工神经网络方法

选取对强对流天气反应较好的雷达回波参数和天气因子作为输入信息单元，应用Ｂ－Ｐ神经网络方法可对强对流天气和雷雨天气非常理想又迅捷地识别出来。     

影响广西的热带气旋年频数的BP神经网络预测模型

对影响广西的热带气旋(TC)年频数与大气环流的关系进行分析表明,TC年频数与全球范围大气环流异常有密切关系,特别是春季南半球中高纬度环流异常和低纬越赤道气流异常.利用相关分析从春季全球大气环流场中选择初选预报因子,然后对初选预报因子作EOF展开构造综合预报因子,运用BP神经网络方法建立TC年频数预报模型,并对所建立的模型进行独立样本试验.结果表明,该预报模型对历史样本拟合精度高,试报效果优于传统的逐步回归模型,可在汛期预测业务中应用.     

B—P算法在青海省降雨分区分级预报中的应用

选用０８：００（北京时）常规资料和数据预报产品作为预报输入资料，将关键区要素场作贝雪夫多项式展开，取车贝雪夫系数作为预报因子，由Ｂ－Ｐ神经元网络进行训练，分别建立了全省各片未来２４ｈ内有，无≥５，１０和２５ｍｍ降水天气过程的人工神经元降水分区预报系统，从学习结果看，历史概括率均达９５％以上，１９９４和１９９５年６，７，８三个月的业务试报，效果较好。     

DOWNSCALING FORECAST OF MONTHLY PRECIPITATION OVER GUANGXI BASED ON BP NEURAL NETWORK MODEL

Variables fields such as enstrophy, meridional-wind and zonal-wind variables are derived from monthly 500 hPa geopotential height anomalous fields. In this work, we select original predictors from monthly 500-hPa geopotential height anomalous fields and their variables in June of 1958 - 2001, and determine comprehensive predictors by conducting empirical orthogonal function (EOF) respectively with the original predictors. A downscaling forecast model based on the back propagation (BP) neural network is built by use of the comprehensive predictors to predict the monthly precipitation in June over Guangxi with the monthly dynamic extended range forecast products. For comparison, we also build another BP neural network model with the same predictands by using the former comprehensive predictors selected from 500-hPa geopotential height anomalous fields in May to December of 1957 - 2000 and January to April of 1958 - 2001. The two models are tested and results show that the precision of superposition of the downscaling model is better than that of the one based on former comprehensive predictors, but the prediction accuracy of the downscaling model depends on the output of monthly dynamic extended range forecast.     

STUDY ON CANONICAL AUTOREGRESSION PREDICTION OF METEOROLOGICAL ELEMENT FIELDS*

Through extension of canonical correlation to the analysis of meteorological element fields (MEF), a concept from combination of canonical autocorrelation with canonical autoregression (CAR) is developed for short-term climatic prediction of MEFs with a formulated scheme. Experimental results suggest that the scheme is of encouraging usefulness to a weak persistence MEF,i.e., rainfall field and, in particular, to a strong persistance one like a SST field.     

基于人工神经网络的集成预报方法研究和比较

用人工神经网络方法对同一预报量的各个子预报方程进行集成预报研究,并以同样的子预报方程进行回归、平均和加权预报集成。对神经网络集成预报模型与各个子预报方程及其它集成预报方法进行了对比分析研究。结果表明,人工神经网络方法所构造的集成预报模型不仅对历史样本的拟合精度比各个子预报方法及其它集成预报方法更好,独立样本的试验预报结果也显示出更好的预报准确性。并且,采用神经网络方法进行预报集成,可以避免以往集成预报方法难以确定权重系数的困难