基于时间序列SVR模型的玉米价格预测研究

研究旨在运用时间序列和支持向量回归(support vector regression,SVR)理论,建立短期玉米价格预测模型,为玉米市场监测预警提供技术支持与决策依据。本研究根据玉米价格序列波动的非线性特征,以河南省2010年1月1日—2019年6月30日的月度数据为研究对象,结合混沌性时间序列与SVR理论,研究一种短期玉米价格预测模型。通过相空间重构的方法对价格序列进行处理,利用重构后的数据训练模型,运用网格交叉验证(GridSearchCV)对研究模型进行优化。得到一种基于时间序列支持向量回归的玉米价格预测模型。将研究模型的预测结果进行对比分析,结果显示,研究模型的预测值更贴近真实值,其平均相对误差(MRE)和均方根误差(RMSE)分别为0.006和20.57,优于传统支持向量回归模型的预测结果。研究模型可以为玉米价格预测提供方法依据,且相较于传统预测方法在精度方面具有优势。     

季节性模型在杨凌示范区气温预报中的应用

运用季节性时间序列模型的理论和方法,对杨凌示范区张家岗1935—1945年月平均气温建立季节性模型,并利用所求得的模型对1946年各月平均气温进行预报。结果表明：1946年1月,2月和12月的月平均温度预报值的相对误差较大,同年3月—11月预报值的相对误差较小,其原因之一是中国北方每年1月,2月和12月温度较低,较小的波动就可能使相对误差很大,而3月—11月温度较高,相对误差较小。此外,该模型对步长较短的预测比较准确,误差较小,但对于步长较长的预测,误差较大。因此,该模型适合于短期预报。     

基于ARMA模型的四川省农村居民收入趋势预测

基于1978-2010年四川省农村居民人均纯收入的时间序列,先后通过平稳化检验及处理、自相关与偏相关分析,并结合SC准则定阶,建立模型为 ;然后确定模型参数并对模型的残差序列进行白噪声检验,检验通过,满足预测的要求;接着,运用该模型对四川省农村居民人均纯收入进行预测,结果显示平均绝对误差率为1.81%,其中2010年误差率为1.21%,精度较高;最后对2011-2013年进行短期预测,数据表明农村居民人均收入将继续增长。     

综合NDVI时序特征的冬小麦混合像元分解及面积估算

基于MODIS数据进行面积提取易受混合像元影响，为了降低因混合像元导致的错分和漏分误差，该文提出一种线性的混合像元分解模型，建立MODIS混合像元中冬小麦占比与MODIS/NDVI时间序列影像波峰波谷差值之间的定量关系。基于2017年保定市MODIS数据和GF数据进行了模型构建，基于2014年数据进行了模型验证，结果显示纯度指数（PPI）精确度均值为0.485，基于混合像元分解模型得到的2014年保定市冬小麦面积推算值为40.05万hm2，基于GF数据得的2014年保定市冬小麦面积“真值”为37.39万hm2，绝对误差为2.66万hm2，相对误差率为7.11%。利用河北省冬小麦广泛种植的8个地市对模型的适用性进行评价，结果表明不同地市的冬小麦面积推算值和冬小麦面积“真值”间平均误差率为3.69%。基于该模型的冬小麦面积推算值误差相对较低，数据可靠性较高，且受地域影响较小，具有较为普遍适用性。     

基于时间序列模型的洞头大型藻类碳汇强度预测分析

为掌握洞头大型藻类多年来固碳量的变化规律,对其固碳量做出准确预测,以1991—2011年洞头主要大型藻类紫菜、羊栖菜的产量为基础,对其固碳量进行了核算,2004年固碳量最高达3650.08 t/a,1991年固碳量最低为278.08 t/a,平均固碳量为1875.33 t/a。在核算的其固碳量的基础上建立了ARIMA (0,1,3)模型,并对固碳量进行了拟合及预测研究。结果显示：所建模型的残差序列为白噪声,用模型拟合1996—2011年洞头大藻固碳量,相对误差最大6.39%,最小0.75%,均在10%以内,拟合效果好,是一种短期预测精度较高的预测模型。经预测2012—2014年3个年度的洞头大藻固碳量为3697.21 t、3570.50 t、3773.32 t,表明未来3年洞头的大藻固碳量总体上呈上升趋势。大藻养殖的碳汇渔业具有巨大的经济效益、生态效益和社会效益,建议构建政府激励措施及机制,加强政策引导,创新渔业发展模式,促进大藻养殖碳汇渔业的健康发展。     

Statistical Model on the Time and Time Difference Error of Arrival in TDOA Location Estimation

Additional time delay of arrival resulted from non-line of sight(NLOS)propagation and muti-path effect in(cellular) environment is the dominant factor resulting in TDOA location error,study on the statistical property of TDOA error is useful for improving the accuracy of TDOA location.Based on the exponential distribution property of NLOS propagation delay,and considering the system error resulted from the signal detection,the authors establish a statistical model of arrival time and a model of TDOA error distribution by statistics when non-light of sight and multi-path propagation exist.The models reflect the statistical property of arrival time and TDOA error in cell network,the simulation shows that the proposed models have a good accuracy and a satisfactory behavior.     

Simulation of Wind Loading for Large - span Bridge Structures

Wind speed time series obtained with numerical simulation methods are adopted in structural analysis of wind - induced vibration. In this paper, based on the natural wind properties and the correlativity of nodal wind speed time series, AR model is employed to simulate wind speed time series, and the AIC rule is used to fix the rank of the AR model. Statistical analysis results of numerical wind speed time series show that AR model is efficient in simulating nodal wind speed time series which has time and space correlativity, and that the simulation precision, simulation efficiency and calculation stability answer the needs of practical engineering.     

Combining input uncertainty and residual error in crop model predictions: A case study on vineyards

As crop modelling has matured and been proposed as a tool for many practical applications, there is increased need to evaluate the uncertainty in model predictions. A particular case of interest that has not been treated before is that where one takes into account both uncertainty in the model explanatory variables and model residual error (the uncertainty in model predictions even when the explanatory variables are perfectly known). The specific case we consider is that of a model for predicting water stress of a vineyard. For many of the model explanatory variables, the vine grower (or the farmer advisor) has a choice between approximate values which are easily obtainable and more precise values that are more difficult (and more expensive) to obtain. We specifically discuss the explanatory variable “initial water stress” which is directly based on the initial soil water content and can be estimated or measured (precise but expensive). The vine grower is interested in the decrease in uncertainty that would result from measuring initial water stress, but it is the decrease in total uncertainty, including model residual error, that is of importance.We propose using accurate measurements of water stress over time in multiple vineyards, to estimate model residual error. The uncertainty in initial water stress can be estimated if one has approximate and precise values of initial water stress in several vineyards. We then combine the two sources of error by simulation thanks to an independence hypothesis; the model is run multiple times with a distribution of values for initial water stress, and on each day a distribution of model residual errors is added to the result.The results show that the resulting uncertainty is quite different in different fields. In some cases, uncertainty in initial water stress becomes negligible a short time after the start of simulations, in other cases that uncertainty remains important, compared to model residual error, throughout the growing season. In all cases, residual error is a substantial percentage of overall error and thus should be taken into account.     

一维稳态流非饱和土渗透系数垂直分布模型及其线性简化

渗透系数是控制地下水流动的重要参数,对渗透系数的空间分布规律进行研究具有重要的意义。基于Gardner模型获取了一维稳态流非饱和土渗透系数沿垂直分布模型,该模型用指数函数描述,受饱和渗透系数和无量纲的深度与流动率等因素的控制;该模型表明一维稳态流条件下均质典型土类的渗透系数其沿垂直方向变化趋势主要受比流量与饱和渗流系数的负数值二者之间的相对大小影响。分别采用泰勒级数方法和以地下水位处及地表处的渗透系数作为控制条件方法对一维稳态流非饱和土渗透系数沿垂直分布模型进行线性近似简化。采用泰勒级数方法获取的简化模型其计算误差随无量纲的深度增大而增大。简化后的模型具有形式简单、参数少等特点。通过算例对比简化模型与原模型的差异,计算结果表明:采用以地下水位处及地表处的渗透系数作为控制条件的方法进行线性近似简化的模型计算误差比采用泰勒级数方法获得的线性近似简化模型的计算误差小。     

Accounting for both parameter and model structure uncertainty in crop model predictions of phenology: A case study on rice

We consider predictions of the impact of climate warming on rice development times in Sri Lanka. The major emphasis is on the uncertainty of the predictions, and in particular on the estimation of mean squared error of prediction. Three contributions to mean squared error are considered. The first is parameter uncertainty that results from model calibration. To take proper account of the complex data structure, generalized least squares is used to estimate the parameters and the variance-covariance matrix of the parameter estimators. The second contribution is model structure uncertainty, which we estimate using two different models. An ANOVA analysis is used to separate the contributions of parameter and model uncertainty to mean squared error. The third contribution is model error, which is estimated using hindcasts. Mean squared error of prediction of time from emergence to maturity, for baseline +2 °C, is estimated as 108 days², with model error contributing 86 days², followed by model structure uncertainty which contributes 15 days² and parameter uncertainty which contributes 7 days². We also show how prediction uncertainty is reduced if prediction concerns development time averaged over years, or the difference in development time between baseline and warmer temperatures.     

ECMWF细网格对浙江沿海10 m风预报性能评估

为更好地开展浙江沿海海上大风的预报服务,本研究对2012年1月1日—2014年12月31日的ECMWF细网格10 m风场产品在浙江沿海的预报性能进行评估,并将其插值到沿海站点和观测数据进行对比分析。结果表明:冷空气影响下,浙江沿海的平均误差为负值,绝对误差分布和变化趋势与平均误差基本一致;台风影响下,24 h预报时效的平均误差为正值,随着预报时效的增加,平均误差逐渐转为负值,鱼山渔场、温台渔场及舟外渔场的绝对误差较其他区域大。另外,预报值和观测数据间的相关系数随预报时效的增加而减小,两者之间的相关系数平均值随海拔高度的增加而减小;ECMWF细网格对岱山和龙山村的预报偏大,对浪岗的预报偏小,预报偏差的离散度随预报时效增加而增大。     

Chaotic Time Series Analysis Based on Radial Basis Function Networks

A method based on radial basis function networks for forecasting chaotic time series is proposed.The nonlinear time series identification problem is formulated with a nonlinear autoregressive moving average(NARMAX)model then a new identification algorithm based on dynamic radial basis function networks is proposed.Then this method is applied to the estimation of embedding dimension for chaotic time series of Henon mapping and the confirmation of the chaotic phenomena in stock markets of China,from which one can get the desired results.Further research directions are also pointed out.     

基于气象条件的玉米发育期预报技术

为了给当地政府、农业部门、农户提供准确及时的玉米发育期预报,根据玉米日平均气温稳定通过10℃播种的温度指标,通过分析辽阳市1956—2010年日平均气温稳定通过10℃与5℃初日间隔日数和稳定通过5℃初日的相关性,利用日序与间隔日数建立回归方程,预测日平均气温稳定通过10℃初日,在此基础上做出玉米播种期预报;利用灯塔市农业气象观测站1981—2010年资料,分析玉米从播种到成熟各个发育期发育速度与气象条件的关系,经过对各时段的相关性检验,发现玉米从播种到成熟期间各发育期间隔日数与≥10℃活动积温相关性最好,从出苗到成熟玉米感温性、感光性都比较强,由此建立播种-开花及拔节-成熟的天数预报模型,预报玉米开花及成熟期。经检验,3款预报模型相关系数、复相关系数及F值均达0.01显著水平,回代与实况值有着比较一致的变化趋势,回代和预报绝对误差(ABSE)和标准误差(RMSE)在2.0~4.3天,可以满足农业气象服务需要,具有推广应用价值。     

Hyperspectral Estimation Model of Chlorophyll Content in Cotton Canopy Leaves under Drip Irrigation at Different Growth Stages

[Objective] This article is aimed to estimate the chlorophyll content of cotton canopy leaves in drip irrigation fields at different growth stages in northern Xinjiang and establish a model for estimating chlorophyll content in growth time series by using hyperspectral. [Method] Using Xinluzao 45 as the experimental material, the chlorophyll content and the corresponding spectral reflectance of cotton canopy leaves at different nitrogen levels and growing stages were measured, and the relationship between 12 indices and the chlorophyll content was analyzed. The estimation models of the chlorophyll content in cotton canopy leaves under drip irrigation were established. [Result] The correlation coefficient between the chlorophyll content of canopy leaves and Vogelmann red edge index was high in the four growing periods of cotton, and the correlation coefficient was 0.944, 0.907, 0.895, 0.930, respectively. And the spectral reflectivity was the highest at the flowering and boll period. The precision of the model established by the multivariate regression method is higher than that of the single exponential linear model with the determination coefficient more than 0.8 and the root mean square error smaller than that of the single exponential linear model. The model of the budding stage (y＝82.509_x1＋89.937_x2－94.438) has the best precision. [Conclusion] The chlorophyll content can be estimated by the models established at different growth stages, and the budding stage model has the best monitoring effect.     

基于HJ-1数据的冬小麦种植面积、长势与收割进度遥感监测

及时、准确地获取冬小麦种植面积、长势、收割进度等信息对冬小麦产量预报和农田精细管理具有重要意义。本研究采用HJ-1A/B CCD1/CCD2影像,结合S-G滤波算法、波段最大化的光谱角目标探测器、阈值分割等方法进行新乡市冬小麦种植面积、小麦长势和收割进度等遥感监测。结果表明：（1）利用HJ卫星NDVI时间序列数据结合波段最大化的光谱角度目标探测器方法,可以进行冬小麦种植面积提取,各县市冬小麦种植面积提取相对误差在10%以内;（2）单景影像NDVI值阈值分割可以反映冬小麦长势在时空差异;（3）冬小麦达到抽穗期的时间不一致,NDVI值在4月15日和21日附近达到最大值;（4）利用成熟期和收割期不同时间的NDVI影像,可监测各县市冬小麦收割进度的时空差异。各县市冬小麦在6月4日前开始收割与6月15日左右完成收割工作。HJ卫星时间序列遥感数据可以进行冬小麦种植面积、长势、收割进度等遥感监测。     

Neumann Series Solution of Newmark Method and Its Convergence for Solving Dynamic Equations

The convergence of Newmark Neumann series method for solving dynamic equations is discussed.Conditions of convergence and the estimating expressions of approximate solution and error for the series solution are also given.     

Application of Intelligent Neural Network on Prediction of Time Series Signal

The research of neural network has been maturated both in theory and practical application since 1980's, and also been employed into the prediction and analysis of nonlinear time series signal in the field of signal process system. Concerning with the problem of time series signal prediction based on traditional neural network, such as black box, poor accuracy, and facing the shortage of post knowledge, this paper presents a different neural network prediction model from the traditional ones, based on intelligent neural cell model and employing the iterative prediction method. Through the example on stock price prediction, the prediction accuracy and practical value are proved.     

The Genetic Algorithms Applied in the Combination Forecasting of Economic Chaos

Economic chaos is a random-like dynamic behavior resulting from definitive economic system . Although the time series from economic system isn't forecasted in a long term ,but can be predicted precisely in a short term , thus , by which a deterministic model can be constructed . To search chaos character in economic time series , we present a combined forecasting method of economic chaos basing on genetic algorithms and the reconstruction of phase space given by Wolf, this method takes advantage of genetic algorithms to determine the weighting coefficient of combined forecasting. It's greatly overcome the defaults that when traditional forecasting of economic chaos single model is mostly used so as to affect the forecasting precision. At last , an example is given to testify the validity and feasibility of this method.     

风电齿轮箱传动误差分析及振动噪声预估

针对一级行星两级平行轴风电齿轮传动系统,综合考虑齿轮时变啮合刚度、啮合阻尼、传递误差等因素,建立31个自由度的弯扭轴耦合集中参数动力学模型,采用变步长Runge-Kutta法对系统动力学微分方程进行求解,得出齿轮传动系统各级传动误差;借助软件建立风电齿轮箱刚柔耦合动力学模型,并导入传动误差,采用模态叠加法求得齿轮箱轴承支反力,并将其作为声振耦合模型的边界条件,采用声学有限元法对风电齿轮箱进行振动噪声预估,并与试验结果对比分析,两者吻合良好。     

The application of grey residual error GM(11)model based on Markvo to aging behavior prediction of plastic

To solve the problem of poor prediction accuracy of GM(11)model when the data are discretebased on the sign of residual error modification value of prediction value being determined by Markvo chaina residual error modification is presented.The grey residual error GM(11)model based on Markvo is constructed through the modification of simulation value and prediction value obtained from GM(11)model.The prediction of tensile strength of LDPE greenhouse film under natural aging condition is taken as an exampleand the applicability of constructed model in the prediction of plastic aging behavior is researched.The results show that the relative error between prediction value and the actual value of tensile strength of LDPE greenhouse film after aging 18 and 21 months obtained from constructed model are 1.49% and 4.96% respectively.Prediction accuracy are higher than that(relative error are 3.40% and 6.75% respectively)obtained from GM(11)model obviously.Grey residual error GM(11)model based on Markvo needs less original data and has high prediction accuracythus it is a simple and reliable method for plastic aging behavior prediction.