Modeling and Fault Monitoring of Bioprocess Using Generalized Additive Models （GAMs） and Bootstrap

Fault monitoring of bioprocess is important to ensure safety of a reactor and maintain high quality of products. It is difficult to build an accurate mechanistic model for a bioprocess, so fault monitoring based on rich historical or online database is an effective way. A group of data based on bootstrap method could be resampling stochastically, improving generalization capability of model. In this paper, online fault monitoring of generalized additive models （GAMs） combining with bootstrap is proposed for glutamate fermentation process. GAMs and bootstrap are first used to decide confidence interval based on the online and off-line normal sampled data from glutamate fermentation experiments. Then GAMs are used to online fault monitoring for time, dissolved oxygen, oxygen uptake rate, and carbon dioxide evolution rate. The method can provide accurate fault alarm online and is helpful to provide useful information for removing fault and abnormal phenomena in the fermentation.

Fitting Generalized Additive Logistic Regression Model with GAM Procedure

In dealing with nonparametric regression the GAM procedure is the most versatile of several new procedures. The terminology behind this procedure is more flexible than traditional parametric modeling tools. It relaxes the usual assumptions of parametric model and enables us to uncover structure to establish the relationship between independent variables and dependent variable in exponential family that may not be obvious otherwise. In this paper, we discussed two methods of fitting generalized additive logistic regression model, one based on Newton Raphson method and another based on iterative weighted least square method for first and second order Taylor series expansion. The use of the GAM procedure with the specified set of weights, using local scoring algorithm, was applied to real life data sets. The cubic spline smoother is applied to the independent variables. Based on nonparametric regression and smoothing techniques, this procedure provides powerful tools for data analysis.

Predictive Vegetation Mapping Approach Based on Spectral Data, DEM and Generalized Additive Models

This study aims to provide a predictive vegetation mapping approach based on the spectral data, DEM and Generalized Additive Models (GAMs). GAMs were used as a prediction tool to describe the relationship between vegetation and environmental variables, as well as spectral variables. Based on the fitted GAMs model, probability map of species occurrence was generated and then vegetation type of each grid was defined according to the probability of species occurrence. Deviance analysis was employed to test the goodness of curve fitting and drop contribution calculation was used to evaluate the contribution of each predictor in the fitted GAMs models. Area under curve (AUC) of Receiver Operating Characteristic (ROC) curve was employed to assess the results maps of probability. The results showed that: 1) AUC values of the fitted GAMs models are very high which proves that integrating spectral data and environmental variables based on the GAMs is a feasible way to map the vegetation. 2) Prediction accuracy varies with plant community, and community with dense cover is better predicted than sparse plant community. 3) Both spectral variables and environmental variables play an important role in mapping the vegetation. However, the contribution of the same predictor in the GAMs models for different plant communities is different. 4) Insufficient resolution of spectral data, environmental data and confounding effects of land use and other variables which are not closely related to the environmental conditions are the major causes of imprecision.

Modeling hot strip rolling process under framework of generalized additive model

This research develops a new mathematical modeling method by combining industrial big data and process mechanism analysis under the framework of generalized additive models(GAM)to generate a practical model with generalization and precision.Specifically,the proposed modeling method includes the following steps.Firstly,the influence factors are screened using mechanism knowledge and data-mining methods.Secondly,the unary GAM without interactions including cleaning the data,building the sub-models,and verifying the sub-models.Subsequently,the interactions between the various factors are explored,and the binary GAM with interactions is constructed.The relationships among the sub-models are analyzed,and the integrated model is built.Finally,based on the proposed modeling method,two prediction models of mechanical property and deformation resistance for hot-rolled strips are established.Industrial actual data verification demonstrates that the new models have good prediction precision,and the mean absolute percentage errors of tensile strength,yield strength and deformation resistance are 2.54%,3.34%and 6.53%,respectively.And experimental results suggest that the proposed method offers a new approach to industrial process modeling.

Simulating Potential Distribution of Tamarix chinensis in Yellow River Delta by Generalized Additive Models

There are typical ecosystems of littoral wetlands in the Yellow River Delta.In order to study the relationships between Tamarix chinensis and environmental variables and to predict T.chinensis potential distribution in the Yellow River Delta,641 vegetation samples and 964 soil samples were collected in the area in October of 2004,2005,2006 and 2007.The contents of soil organic matter,total phosphorus,salt,and soluble potassium were determined.Then,the analyzed data were interpolated into spatial raster data by Kriging interpolation method.Meanwhile,the digital elevation model,soil type map and landform unit map of the Yellow River Delta were also collected.Generalized Additive Models(GAMs) were employed to build species-environment model and then simulate the potential distribution of T.chinensis.The results indicated that the distribution of T.chinensis was mainly limited by soil salt content,total soil phosphorus content,soluble potassium content,soil type,landform unit,and elevation.The distribution probability of T.chinensis was produced with a lookup table generated by Grasp Module(based on GAMs) in software ArcView GIS 3.2.The AUC(Area Under Curve) value of validation and cross-validation of ROC(Receive Operating Characteristic) were both higher than 0.8,which suggested that the established model had a high precision for predicting species distribution.

Study of monthly variations in primary production and their relationships with environmental factors in the Daya Bay based on a general additive model

In this study, the horizontal and vertical distribution of primary production(PP) and its monthly variations were described based on field data collected from the Daya Bay in January–December of 2016. The relationships between PP and environmental factors were analyzed using a general additive model(GAM). Significant seasonal differences were observed in the horizontal distribution of PP, while vertical distribution showed a relatively consistent unimodal pattern. The monthly average PP(calculated by carbon) ranged from 48.03 to 390.56 mg/(m~2·h),with an annual average of 182.77 mg/(m~2·h). The highest PP was observed in May and the lowest in November.Additionally, the overall trend in PP was spring>summer>winter>autumn, and spring PP was approximately three times that of autumn PP. GAM analysis revealed that temperature, bottom salinity, phytoplankton, and photosynthetically active radiation(PAR) had no significant relationships with PP, while longitude, depth, surface salinity, chlorophyll a(Chl a) and transparency were significantly correlated with PP. Overall, the results presented herein indicate that monsoonal changes and terrestrial and offshore water systems have crucial effects on environmental factors that are associated with PP changes.

Inference Procedures on the Generalized Poisson Distribution from Multiple Samples: Comparisons with Nonparametric Models for Analysis of Covariance (ANCOVA) of Count Data

Count data that exhibit over dispersion (variance of counts is larger than its mean) are commonly analyzed using discrete distributions such as negative binomial, Poisson inverse Gaussian and other models. The Poisson is characterized by the equality of mean and variance whereas the Negative Binomial and the Poisson inverse Gaussian have variance larger than the mean and therefore are more appropriate to model over-dispersed count data. As an alternative to these two models, we shall use the generalized Poisson distribution for group comparisons in the presence of multiple covariates. This problem is known as the ANCOVA and is solved for continuous data. Our objectives were to develop ANCOVA using the generalized Poisson distribution, and compare its goodness of fit to that of the nonparametric Generalized Additive Models. We used real life data to show that the model performs quite satisfactorily when compared to the nonparametric Generalized Additive Models.

环境卫生与医院感染的时间序列研究:基于广义相加模型(GAM)

目的定量分析环境卫生对医院感染发生的影响。方法收集某三甲医院2018年1月—2022年12月医院感染与环境卫生学监测资料,采用时间序列的广义相加模型分析环境检出菌落形成单位(CFU)对医院感染发生的影响。结果单污染模型显示,医院感染与工作人员手细菌菌落数之间存在显著正相关性(β1=0.009,P=0.012),工作人员手月度平均菌落形成单位(MCFU/Dish)每升高1个四分位数间距(IQR),医院感染发生率增加13.28%(95%CI:2.82%~24.81%);亚组分析与滞后效应分析显示,工作人员手月度MCFU/Dish(卫生手消毒后)升高1个IQR,当月(lag0)医院感染超额风险(ER)为16.26%(95%CI:15.45%~17.09%)。多污染模型中,物体表面污染与医院感染的相关性同样具有统计学意义。结论医院环境卫生与医院感染之间存在显著相关性。

基于GAM模型的东海中南部大黄鱼资源密度分布及其与环境因子的关系

为探究大黄鱼(Larimichthys crocea)资源密度与海洋环境因子之间的关系,本文基于2018年秋季和2019年春季东海中南部两个航次渔业资源调查数据,采用广义加性模型(Generalized additive model,GAM)对大黄鱼资源密度及其与环境因子的关系进行了研究。研究表明,大黄鱼尾数密度和生物量密度秋季为106.2 ind/km^(2)和6.57 kg/km^(2),春季为7.9 ind/km^(2)和0.45 kg/km^(2)。春季大黄鱼分布重心位于鱼山渔场,秋季分布重心南移进入温台渔场。广义加性模型分析结果表明:春秋两季最优模型均由离岸距离、温度、盐度组成,其中秋季最优模型的温度和盐度因子均为底层水体,而春季环境因子均为表层水体;春秋两季大黄鱼生物量密度均随温度上升而下降;秋季随着底层盐度的增加先上升后下降,最适底层盐度为33.7;春季随表层盐度升高而呈单调增加趋势。春季和秋季大黄鱼分布受到了不同水层环境的影响,这是由其栖息和繁殖特性的季节变动所导致的。本研究结果阐明了春秋季东海中南部大黄鱼资源分布情况及其对海洋环境因子的响应。

基于广义加性混合模型(GAMM)的沙柳特征因子动态变化

【目的】对沙柳特征因子的动态变化进行研究,分析环境因子影响下的地径、枝高动态变化过程。【方法】采用样地调查、样本采集、气象数据收集等手段,基于广义加性混合模型(GAMM),以灌丛、枝条以及二者的嵌套作为随机效应,探究灌木地径、枝高与土壤水分(SM)、年平均降水量(MAP)、年平均气温(MAT)及年龄等影响因子的动态变化规律。【结果】1)对于不考虑随机效应的广义加性模型(GAM),灌木地径与影响因子呈较强的非线性关系(有效自由度E均大于8.20,且P<0.001),枝高仅与时间呈非线性关系,与其他影响因子均为线性关系。2)相较于GAM,GAMM在随机效应的影响下,地径与各影响因子之间非线性显著降低(E变小),但在以灌丛为随机效应的模型中,地径与年平均气温趋于线性关系(E为1),而枝高与时间的非线性关系更强,与其余影响因子仍呈线性关系。3)考虑随机效应的GAMM比GAM的拟合结果更优,且嵌套模式下的GAMM拟合效果最好。【结论】沙柳不同特征因子对环境因子的响应有差异,而相比枝高,地径的变化程度更大。研究结果有助于掌握沙柳特征因子的动态变化对环境因素的响应机制,为进一步探究沙地生境变化过程中植物种群变化、植被演替及植被管理提供科学依据。

Modeling the effect of stand and site characteristics on the probability of mistletoe infestation in Scots pine stands using remote sensing data

Over the past decade,the presence of mistletoe(Viscum album ssp.austriacum)in Scots pine stands has increased in many European countries.Understanding the factors that influence the occurrence of mistletoe in stands is key to making appropriate forest management decisions to limit damage and prevent the spread of mistletoe in the future.Therefore,the main objective of this study was to determine the probability of mistletoe occurrence in Scots pine stands in relation to stand-related endogenous factors such as age,top height,and stand density,as well as topographic and edaphic factors.We used unmanned aerial vehicle(UAV)imagery from 2,247 stands to detect mistletoe in Scots pine stands,while majority stand and site characteristics were calculated from airborne laser scanning(ALS)data.Information on stand age and site type from the State Forest database were also used.We found that mistletoe infestation in Scots pine stands is influenced by stand and site characteristics.We documented that the densest,tallest,and oldest stands were more susceptible to mistletoe infestation.Site type and specific microsite conditions associated with topography were also important factors driving mistletoe occurrence.In addition,climatic water balance was a significant factor in increasing the probability of mistletoe occurrence,which is important in the context of predicted temperature increases associated with climate change.Our results are important for better understanding patterns of mistletoe infestation and ecosystem functioning under climate change.In an era of climate change and technological development,the use of remote sensing methods to determine the risk of mistletoe infestation can be a very useful tool for managing forest ecosystems to maintain forest sustainability and prevent forest disturbance.

Probabilistic Precipitation Forecasting Based on Ensemble Output Using Generalized Additive Models and Bayesian Model Averaging

A probabilistic precipitation forecasting model using generalized additive models （GAMs） and Bayesian model averaging （BMA） was proposed in this paper. GAMs were used to fit the spatial-temporal precipitation models to individual ensemble member forecasts. The distributions of the precipitation occurrence and the cumulative precipitation amount were represented simultaneously by a single Tweedie distribution. BMA was then used as a post-processing method to combine the individual models to form a more skillful probabilistic forecasting model. The mixing weights were estimated using the expectation-maximization algorithm. The residual diagnostics was used to examine if the fitted BMA forecasting model had fully captured the spatial and temporal variations of precipitation. The proposed method was applied to daily observations at the Yishusi River basin for July 2007 using the National Centers for Environmental Prediction ensemble forecasts. By applying scoring rules, the BMA forecasts were verified and showed better performances compared with the empirical probabilistic ensemble forecasts, particularly for extreme precipitation. Finally, possible improvements and a^plication of this method to the downscaling of climate change scenarios were discussed.

Effect of climate factors on the incidence of hand, foot, and mouth disease in Malaysia: A generalized additive mixed model

Climate change is one of the critical determinants affecting life cycles and transmission of most infectious agents,including malaria,cholera,dengue fever,hand,foot,and mouth disease(HFMD),and the recent Corona-virus pandemic.HFMD has been associated with a growing number of outbreaks resulting in fatal complications since the late 1990s.The outbreaks may result from a combination of rapid population growth,climate change,socioeconomic changes,and other lifestyle changes.However,the modeling of climate variability and HFMD remains unclear,particularly in statistical theory development.The statistical relationship between HFMD and climate factors has been widely studied using generalized linear and additive modeling.When dealing with time-series data with clustered variables such as HFMD with clustered states,the independence principle of both modeling approaches may be violated.Thus,a Generalized Additive Mixed Model(GAMM)is used to investigate the relationship between HFMD and climate factors in Malaysia.The model is improved by using a first-order autoregressive term and treating all Malaysian states as a random effect.This method is preferred as it allows states to be modeled as random effects and accounts for time series data autocorrelation.The findings indicate that climate variables such as rainfall and wind speed affect HFMD cases in Malaysia.The risk of HFMD increased in the subsequent two weeks with rainfall below 60 mm and decreased with rainfall exceeding 60 mm.Besides,a two-week lag in wind speeds between 2 and 5 m/s reduced HFMD's chances.The results also show that HFMD cases rose in Malaysia during the inter-monsoon and southwest monsoon seasons but fell during the northeast monsoon.The study's outcomes can be used by public health officials and the general public to raise awareness,and thus,implement effective preventive measures.

Modeling deformation resistance for hot rolling based on generalized additive model

A model of deformation resistance during hot strip rolling was established based on generalized additive model.Firstly,a data modeling method based on generalized additive model was given.It included the selection of dependent variable and independent variables of the model,the link function of dependent variable and smoothing functional form of each independent variable,estimating process of the link function and smooth functions,and the last model modification.Then,the practical modeling test was carried out based on a large amount of hot rolling process data.An integrated variable was proposed to reflect the effects of different chemical compositions such as carbon,silicon,manganese,nickel,chromium,niobium,etc.The integrated chemical composition,strain,strain rate and rolling temperature were selected as independent variables and the cubic spline as the smooth function for them.The modeling process of deformation resistance was realized by SAS software,and the influence curves of the independent variables on deformation resistance were obtained by local scoring algorithm.Some interesting phenomena were found,for example,there is a critical value of strain rate,and the deformation resistance increases before this value and then decreases.The results confirm that the new model has higher prediction accuracy than traditional ones and is suitable for carbon steel,microalloyed steel,alloyed steel and other steel grades.

黑河上游天然草地蝗虫物种丰富度与地形关系的GAM分析

地形差异性导致的环境异质性为小尺度范围内生物空间格局的形成与维持提供了一种重要机制,是形成物种丰富度差异性的前提条件。借助GIS和S-PLUS软件,利用广义可加模型(GAM)于7-8月对影响蝗虫分布的地形因子进行了研究,在定量分析黑河上游祁连山区北坡地形的海拔分异特征的基础上研究了该区域蝗虫的丰富度与地形复杂度的关系。结果表明:在36个样方中共采集蝗虫3149头,隶属于3科10属13种;蝗虫丰富度受地形因子影响的顺序为海拔>坡向>坡度>剖面曲率>平面曲率>坡位;蝗虫的分布在平面曲率和剖面曲率各个梯度上的分布比较均衡,在海拔、坡向以及坡位的每个梯度上呈二次抛物线分布,坡度上呈递减趋势;从分布的区域上来看,蝗虫在整个区域都有较高的丰富度,但主要分布在海拔2600～2700m区域,坡向上则主要集中在西北坡和西坡,与实际观测情况相一致。蝗虫丰富度与地形因子之间的相互关系以及分布状态,反映了地形因子对水热条件的重分配使蝗虫分布格局出现多元化以及破碎化。

基于GAM模型研究金枪鱼围网沉降性能影响因素

金枪鱼围网渔业是现代金枪鱼渔业中捕捞效率最高的方法,研究围网沉降性能与影响因子之间的关系有利于提高围网捕捞效率。利用2011年9—12月金枪鱼围网渔船"金汇7号"在中西太平洋作业时所收集的数据,实验分析了围网沉降深度(H)与放网时间(T),放网速度(V0),括纲(L)及跑纲(L1)的投放长度,10、60和120 m 3个水层流速(V10,V60及V120)等因子之间的关系,并利用广义加性模型(GAM)评价了各因子对沉降深度的影响。结果表明:(1)网具中部沉降深度与时间的关系为H=-0.000 2 t2+0.408 6 t+1.809 9(R2=0.999 3);(2)沉降速度随着深度的增加而减小,网具中部沉降速度与时间的关系为V=2.5×107 t2-6×104 t+0.4412(R2=0.985 2);(3)GAM模型中的逐步回归分析表明,T、V60、V120和L 4个因子对H影响显著,且影响大小依次为V120、L、T和V60;(4)GAM模型分析表明,沉降深度随放网时间的增加而增大,放网时间集中在500～550 s;流速的大小与沉降深度呈负相关;括纲投放长度主要集中在1 800 m左右。

GLM和GAM模型研究东黄海鲐资源渔场与环境因子的关系

鲐是我国近海重要中上层鱼类,研究其资源变动、渔场分布与时空、海洋环境因子之间的关系有利于该资源的合理开发和利用。根据1998-2004年我国东黄海大型鲐围网渔业的生产统计和时间、空间、表温、表层盐度、表温梯度、表温的月差异等环境数据,利用广义可加模型(GAM)和广义线性模型(GLM)对鲐资源丰度和环境因子的关系进行研究。结果表明,在南部海域,作业渔场集中在122.5°E^124°E、26.5°N^28°N,适宜表温26.5~30℃,适宜表层盐度33.3~34.3,并明显集中在锋区周边海域;在北部海域,作业渔场集中在122.5°E^125.5°E、33°N^37.5°N,适宜表温15~20℃,适宜表层盐度31.3~32.3,集中在冷水区边缘海域。相对资源密度指数大于0.5的海域为122°30′E^124°30′E、26°30′N^28°N,122°30′E^125°30′E、33°N^34°30′N和124°E^125°E、34°30′N^37°N。研究认为,南北不同海域鲐分布的适宜表温和表层盐度差异明显。影响鲐资源丰度的环境因子重要性依次为时间、空间和海洋环境。

崇明北湖叶绿素a浓度与环境因子的GAM回归分析

以崇明北湖为例,采用广义加性模型(GAM)对该湖的叶绿素a浓度与相关环境因子进行分析.结果表明,叶绿素a浓度与总氮、总磷和水温之间存在较好的非线性关系(P<0.05),叶绿素a浓度与总磷之间的关系先为单调递增,当总磷浓度达到0.12mg/L时,变为单调递减;不同总氮浓度区间上,总氮对叶绿素a浓度的影响不同,氮浓度为0.6～1.8mg/L时,对叶绿素a浓度的影响不大;水温在24～26℃时,叶绿素a浓度最高.叶绿素a浓度与氮磷比之间也存在较好的非线性关系(P<0.1),氮限制时,叶绿素浓度与氮磷比呈反比;磷限制时,叶绿素a浓度随着氮磷比单调递减.

基于GAM模型分析夏秋季南奥克尼群岛南极磷虾渔场时空分布及与环境因子之间的关系

基于中国南极磷虾渔业科学观察员收集的数据,利用广义可加模型(GAM),对2009—2010年渔季和2010—2011年渔季夏秋季南奥克尼群岛周边水域南极磷虾渔场时空分布及其与表温和海况之间的关系进行了研究。结果表明,GAM模型对单位捕捞努力量渔获量(CPUE)总偏差解释率为41.78%,其中贡献最大的为月份,贡献率为15.53%;奥克尼群岛水域CPUE分布模式有着明显的月份差异。1月份平均CPUE值最低,而4月份平均CPUE值最高,各月份平均CPUE存在显著性差异。渔场主要集中在60°12'—60°30'S,45°30'—47°30'W。南极磷虾作业渔场的适宜海水表温范围为0.1—1.8℃,最适表温范围为0.5—1.5℃。各海况等级间及不同渔船的平均CPUE均存在显著性差异。逐步GAM模型分析结果表明,影响CPUE的因子按重要性从大到小依次为月份、渔船、经度、海表温度、纬度和海况。

基于GAM模型分析南京市循环和呼吸系统疾病死亡人数与体感温度的关系

研究了在混杂因子的共同作用下,体感温度的变化对循环和呼吸系统疾病死亡人数的影响,采用广义相加模型对两种系统疾病的死亡人数和体感温度的关系做了分析.结果显示,当以周为时间序列单位,体感温度在20℃以下时,每降低1℃,循环系统疾病的死亡人数增加1.31%,降低5℃,增加6.86%;当以日为时间序列单位,体感温度在14℃以下时,每降低1℃,循环系统疾病的死亡人数增加1.19%,降低5℃,增加5.79%.同样以日为时间序列,当体感温度在20℃以下时,每降低1℃,呼吸系统疾病的死亡人数增加1.41%,降低5℃,增加6.86%,说明体感温度的冷效应对死亡人数的影响较为明显.