基于ARIMA模型的湘江流域DO和NH4+–N含量贝叶斯预测     

刘潭秋  王巧玲 《湖南农业大学学报(自然科学版)》2017,43(5)

为实时把控湘江流域水质的变化趋势,采用污染比较严重的湘江流域长沙段和益阳段水质指标溶解氧(DO)和氨氮(NH_4~+-N)含量的监测数据,用贝叶斯方法推断经典的ARIMA时间序列模型,并用马尔可夫链蒙特卡罗(MCMC)模拟方法对DO和NH_4~+-N含量进行贝叶斯预测。结果表明,该模型的贝叶斯预测能实现对湘江流域长沙段和益阳段水质指标DO和NH_4~+-N含量的精确点预测、区间预测和概率预测。  相似文献   

基于高光谱成像技术识别水稻纹枯病     

李志伟  袁婧  丁为民  杨红兵  沈少庆  崔嘉林 《华南农业大学学报》2018,39(6):97-103

【目的】利用高光谱成像技术对水稻纹枯病进行早期的快速无损识别,结合判别分析方法建立相应的鉴别模型。【方法】以健康和感染纹枯病的水稻幼苗为研究对象,采集叶片和冠层各180个样本的380~1 030 nm波段的360条高光谱图像,剔除明显噪声部分后,以440~943 nm波段作为水稻样本的光谱范围,分别用不同的方法预处理获得水稻叶片的光谱曲线。采用偏最小二乘–判别分析(PLS-DA)对不同预处理的光谱建模。采用MNF算法对冠层的原始光谱数据进行特征信息提取,并基于特征信息建立线性判别分析(LDA)模型和误差反向传播神经网络(BPNN)判别模型。【结果】标准正态变量变换(SNV)预处理后建立的PLS-DA模型的预测集判别正确率最高,为92.1%。基于特征信息的LAD和BPNN模型的判别结果优于基于全波段的PLS-DA判别模型。基于最小噪声分离变换特征信息提取的BPNN模型取得了最优效果,建模集和预测集正确率分别达99.1%和98.4%。【结论】采用高光谱成像技术对水稻纹枯病生理特征进行无损鉴别是可行的,本研究为水稻纹枯病的识别提供了一种新方法。  相似文献   

基于贝叶斯网络的城市燃气管道安全失效概率   总被引：1，自引：0，他引：1  

郝永梅  邢志祥  沈明  邵辉  汪旭升 《油气储运》2012,31(4):270-273,327

贝叶斯网络对不确定性问题具有强大的处理能力和自我学习更新能力,而贝叶斯网络软件的应用提高了基于贝叶斯网络风险预测的有效性。建立了基于贝叶斯网络的城市燃气管道失效概率分析模型,运用HUGIN和MSBNX软件工具,结合某市天然气管道案例,计算多态故障顶事件安全失效概率和各失效因素的结构重要度。运用BN的推理能力,对造成管道安全失效的自然破坏因素和腐蚀因素分别进行单因素和双因素修正。修正后的贝叶斯网络模型更加符合实际,对提高城市燃气管道安全失效定量分析的系统性、预见性和准确性具有更好的现实意义,也充分显示了贝叶斯网络在处理复杂系统风险分析中独特的优越性和适用性。  相似文献   

基于超参数优化算法下的随机森林模型预测奶牛呼吸频率     

严格齐  赵婉莹  于镇伟  焦洪超  林海  李浩  施正香  王朝元 《农业工程学报》2024,40(11):195-203

奶牛呼吸频率是评估环境造成的奶牛热应激程度的重要指标之一。该研究基于随机森林（random forest,RF）算法提出了适用于生产条件下的奶牛个体呼吸频率准确预测模型,为了平衡模型精度与计算效率问题,利用遗传算法（genetic algorithm,GA）、差分进化（differential evolution,DE）算法、粒子群优化（particle swarm optimization,PSO）算法、贝叶斯优化（Bayesian optimization,BO）算法对模型超参数进行优化,并与网格搜索（grid search,GS）下的人工神经网络（artificial neural network,ANN）和极限梯度提升机（extreme gradient boosting,XGBoost）模型进行了对比分析。研究结果表明,使用融合环境参数的修正温湿指数（adjusted temperature-humidity index,ATHI）、时间区域、奶牛产奶量、泌乳天数、身体姿势以及胎次作为输入特征时,基准RF模型的预测性能最佳。在此基础上,4种智能优化算法下的RF模型性能优于GS-ANN和GS-XGBoost,其中BO-RF的综合性能最优,其决定系数、平均绝对误差、平均绝对百分比误差以及均方根误差分别为0.614、7.723、14.4%、9.737,超参数优化耗时约为DE-RF的1/220。特征重要性分析表明,输入因子对奶牛呼吸频率的影响程度不同,ATHI是影响力最高的因子,相对重要性（relative importance,RI）为0.71,其次是时间区域（RI=0.09）和奶牛产奶量（RI=0.07）。研究为奶牛生产、健康评价及牛舍环境精准调控提供了有效方法和基础。  相似文献   

Discrimination between Japanese and Chinese Products using trace elements concentration of kernel in pickled Japanese apricot     

《Soil Science and Plant Nutrition》2013,59(1)

pp. 875–880

The trace-element composition of kernel in pickled Japanese apricot (Prunus mume Sieb. et Zucc.) was determined using an inductively coupled plasma optical emission spectrometer in order to distinguish between Japanese products and Chinese products.

Strontium and barium concentrations in the kernels of Chinese products were 10 or more times those of the Japanese ones. When based on 8.0 mg kg^?1 of strontium concentration in kernel, 93.2% of sample was distinguished as Japanese products or Chinese ones.

Applying principal component analysis using 9 elements (Mn, Zn, Fe, Ni, Ba, Sr, Cu, Co, Cr), the pickled Japanese apricots tend to separate into two countries. Linear discriminant analysis (LDA) using 9 elements allowed a reasonable classification of pickled Japanese apricots according to the country of production.

The result of the analysis of K-nearest neighbors (KNN) was better than that of LDA.  相似文献   

Estimating spatial distribution of Atlantic salmon escapement using redd counts despite changes over time in counting procedure: application to the Allier River population     

Guillaume J.R. Dauphin  Catherine Brugel  Marion Hoffmann‐Legrand  Etienne Prévost 《Ecology of Freshwater Fish》2013,22(4):626-636

  相似文献   

Bayesian methods for estimating pathogen prevalence within groups of animals from faecal-pat sampling     

Clough HE  Clancy D  O'Neill PD  French NP 《Preventive veterinary medicine》2003,58(3-4):145-169

Pathogens such as Escherichia coli O157:H7 and Campylobacter spp. have been implicated in outbreaks of food poisoning in the UK and elsewhere. Domestic animals and wildlife are important reservoirs for both of these agents, and cross-contamination from faeces is believed to be responsible for many human outbreaks. Appropriate parameterisation of quantitative microbial-risk models requires representative data at all levels of the food chain. Our focus in this paper is on the early stages of the food chain-specifically, sampling issues which arise at the farm level. We estimated animal–pathogen prevalence from faecal-pat samples using a Bayesian method which reflected the uncertainties inherent in the animal-level prevalence estimates. (Note that prevalence here refers to the percentage of animals shedding the bacteria of interest). The method offers more flexibility than traditional, classical approaches: it allows the incorporation of prior belief, and permits the computation of a variety of distributional and numerical summaries, analogues of which often are not available through a classical framework. The Bayesian technique is illustrated with a number of examples reflecting the effects of a diversity of assumptions about the underlying processes. The technique appears to be both robust and flexible, and is useful when defecation rates in infected and uninfected groups are unequal, where population size is uncertain, and also where the microbiological-test sensitivity is imperfect. We also investigated the determination of the sample size necessary for determining animal-level prevalence from pat samples to within a pre-specified degree of accuracy.  相似文献   

A Bayesian approach to the accuracy of clinical observations     

Baadsgaard NP  Jørgensen E 《Preventive veterinary medicine》2003,59(4):189-206

The accuracy of clinical observations was estimated using Bayesian latent-class models with two or more independent tests. Four veterinarians carried out systematic independent clinical examinations on 155 pigs in three herds. Based on the results of binary recordings of clinical observations on dullness, poor body condition (PBC), skin lesions, lameness, respiratory disease, and diarrhea, a latent disease state for each clinical disease was estimated using Gibbs sampling.

The accuracy of the clinical observations differed for the four observers and for different clinical signs. Population parameters were estimated from a Bayesian hierarchical model, and the accuracy of a random observer was calculated. We concluded that the accuracy of the veterinarians in this study substantiated the need to pursue more-precise definitions of the clinical findings and that larger sample sizes would be needed to provide reasonable variance estimates. Finally, we concluded that the uncertainty in the clinical decision-making process (starting with the clinical examination) needs to be represented fully.  相似文献   

Selection for ovulation rate in rabbits     

Noelia Ibez  María Antonía Santacreu  Maria Martinez  Agueda Climent  Agustín Blasco 《Livestock Science》2006,101(1-3):126-133

An experiment of selection for ovulation rate was carried out. Animals were derived from a synthetic line first selected 12 generations for litter size, then 10 generations for uterine capacity. Selection was relaxed for 6 generations. Selection was based on the phenotypic value of ovulation rate with a selection pressure on does of 30%. Males were selected from litters of does with the highest ovulation rate. Males were selected within sire families in order to reduce inbreeding. Ovulation rate was measured in the second gestation by a laparoscopy, 12 days after mating. Each generation had about 80 females and 20 males. Results of three generations of selection were analyzed using Bayesian methods. Marginal posterior distributions of all unknowns were estimated by Gibbs sampling. Heritabilities of ovulation rate (OR), number of implanted embryos (IE), litter size (LS), embryo survival (ES), fetal survival (FS), and prenatal survival (PS) were 0.44, 0.32, 0.11, 0.26, 0.35, and 0.14, respectively. Genetic correlation between OR and LS was 0.56, indicating that selection for ovulation rate can augment litter size. Response to selection for OR was 1.80 ova. Correlated responses in IE and LS were 1.44 and 0.49, respectively. Selection for ovulation rate may be an alternative to improve litter size.  相似文献   

Fisher大间距线性分类器   总被引：1，自引：0，他引：1  

陈才扣  杨静宇 《植物保护》2007,(12):2143-2147

作为一种著名的特征抽取方法,Fisher线性鉴别分析的基本思想是选择使得Fisher准则函数达到最大值的向量(称为最优鉴别向量)作为最优投影方向,以便使得高维输入空间中的模式样本在该向量投影后,在类间散度达到最大的同时,类内散度最小。大间距线性分类器是寻找一个最优投影矢量(最优分隔超平面的法向量),它可使得投影后的两类样本之间的分类间距(Margin)最大。为了获得更佳的识别效果,结合Fisher线性鉴别分析和大间距分类器的优点,提出了一种新的线性投影分类算法——Fisher大间距线性分类器。该分类器的主要思想就是寻找最优投影矢量wbest(最优超平面的法向量),使得高维输入空间中的样本模式在wbest上投影后,在使类间间距达到最大的同时,使类内离散度尽可能地小。并从理论上讨论了与其他线性分类器的联系。在ORL人脸库和FERET人脸数据库上的实验结果表明,该线性投影分类算法的识别率优于其他分类器。  相似文献