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.

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.

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.

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.

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.

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.

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.

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。

High-resolution peak demand estimation using generalized additive models and deep neural networks

This paper covers predicting high-resolution electricity peak demand features given lower-resolution data.This is a relevant setup as it answers whether limited higher-resolution monitoring helps to estimate future high-resolution peak loads when the high-resolution data is no longer available.That question is particularly interesting for network operators considering replacing high-resolution monitoring by predictive models due to economic considerations.We propose models to predict half-hourly minima and maxima of high-resolution(every minute)electricity load data while model inputs are of a lower resolution(30 min).We combine predictions of generalized additive models(GAM)and deep artificial neural networks(DNN),which are popular in load forecasting.We extensively analyze the prediction models,including the input parameters’importance,focusing on load,weather,and seasonal effects.The proposed method won a data competition organized by Western Power Distribution,a British distribution network operator.In addition,we provide a rigorous evaluation study that goes beyond the competition frame to analyze the models’robustness.The results show that the proposed methods are superior to the competition benchmark concerning the out-of-sample root mean squared error(RMSE).This holds regarding the competition month and the supplementary evaluation study,which covers an additional eleven months.Overall,our proposed model combination reduces the out-of-sample RMSE by 57.4%compared to the benchmark.

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.

基于GAM的喀斯特植被覆盖与驱动因素非线性关系分析

利用Theil-Sen斜率估计和Mann-Kendall显著性检验对2000~2018年西南五省的归一化植被指数(NDVI)进行研究,探究植被覆盖的时空变化特征.用MK(Mann-Kendall)突变检验寻找NDVI与夜间灯光数据的突变时间点.借助广义可加模型(GAM)对社会经济驱动因素,自然驱动因素分别拟合植被覆盖的响应曲线,探究喀斯特地区与非喀斯特地区的非线性响应关系的差异.结果表明:研究区中部、东部的植被恢复效果整体比西部地区好;利用夜间灯光数据作为表征城镇化的替代变量,城镇化与植被覆盖都在2009~2010年之间发生了突变,城镇化的突变时间早于植被覆盖的突变;喀斯特地区有88.54%的植被覆盖增加,其中有48.15%显著增加;非喀斯特地区有80.08%的植被覆盖增加,其中显著增加的占32.34%.喀斯特地区的植被恢复整体要比非喀斯特地区好.植被指数与国内生产总值(GDP)、路网密度、建成区面积等人类影响因素呈现不同的非线性响应关系.整体而言,非喀斯特地区受到的气温影响要比喀斯特地区的大,降雨、土层厚度对喀斯特地区植被恢复的影响更大一些.对比评价喀斯特与非喀斯特地区的植被恢复效果,分区探索植被覆盖变化背后驱动机制的非线性关系,对生态恢复规划等具有启示意义.

基于GAM模型的昆明地面臭氧与气象要素拟合分析

基于昆明2018-2021年O_(3)日最大8 h滑动平均值[ρ(O_(3)-8 h)]和气象要素数据,采用广义相加模型(GAM)中的平滑样条函数拟合单要素、交互项的平滑回归函数拟合多要素与ρ(O_(3)-8 h)的影响关系。引入相对危险度概念,用分布滞后非线性模型(DLNM)分析气象要素和ρ(O_(3)-8 h)的滞后效应。构造滞后项和交互项的GAM模型,进行ρ(O_(3)-8 h)拟合预测。结果表明:当地面气压>818 hPa或平均风速<2.0 m/s时,ρ(O_(3)-8 h)出现1~3 d的滞后效应;GAM模型的交互项平滑回归函数优于单要素平滑样条函数的效果;干冷、湿热、低压大风、高压小风天气以及适当的气温和适中的水汽压有利于ρ(O_(3)-8 h)的增加;纳入交互项(包含滞后项)的GAM模型的拟合效果好于其他模型。该模型的判定系数达到0.672,广义交叉验证得分为352,拟合误差为13.7μg/m^(3),准确率达71.1%,特别在拟合因变量峰值和谷值时优势明显。

基于曲线拟合误差估计的阻抗优化识别方法

在雷达系统、舰船系统、数据中心等大型复杂系统中,供配电系统变流器端口阻抗信息的完整、快速、准确获取,是研判系统可靠性的重要基础。现有研究大都使用分段线性拟合值方法进行阻抗识别,不仅需要大量阻抗测量才能获得完整的阻抗特性,并且可能因过度测量而对系统的正常运行造成干扰。针对这些问题,从阻抗特性曲线的非线性拟合入手,提出一种“非线性曲线拟合、全局误差估计、优选测量点频率”的阻抗识别方法,实现基于曲线拟合误差估计的测量点优化选择,以较少阻抗测量代价获取宽频带阻抗特性。与现有的方法相比,所提方法更有助于测量点的合理选择、阻抗谐振峰的准确识别,从而更高效地实现宽频带阻抗的识别,为复杂供配电系统运行状况的研判提供可靠依据。算例仿真证实了所提方法的有效性。

中国新质生产力多因素交互效应研究

加快形成和发展新质生产力既是我国当下关注的关键议题,也是亟须探究的重要命题。基于中国30个省(区、市)面板数据,本文构建广义相加模型(GAM)探讨创新生产力、技术生产力、资源节约型生产力、环境友好型生产力、数字产业生产力和产业数字生产力对新质生产力水平的因素交互影响效应。研究表明:各因素对新质生产力均具有非线性正向效应;多因素共同影响新质生产力且存在非线性交互作用,其中资源节约型生产力是主导影响因素;因素间的交互效应对新质生产力的影响存在显著差异,创新生产力、数字产业生产力与其他因素结合时产生的影响差异最为明显。基于此,提出促进新质生产力发展的相关启示。本文研究有利于拓展新质生产力提升路径的理论研究,为政府制定发展政策提供参考。

中西太平洋海山特征对延绳钓渔业和围网渔业黄鳍金枪鱼CPUE的影响

海山是海底重要的生物栖息地类型之一,是研究海洋生物多样性的热点区域。黄鳍金枪鱼(Thunnus albacares)广泛分布于中西太平洋,具有极高的生态和经济价值,然而,鲜有关于海山及其相关特征对黄鳍金枪鱼资源丰度和分布影响的研究。基于2010—2021年中西太平洋渔业委员会(Western and Central Pacific Fisheries Commission,WCPFC)汇总的延绳钓和围网渔业数据结合海山特征数据,采用广义加性模型(Generalized additive model,GAM)分析两种不同捕捞方式的黄鳍金枪鱼单位捕捞努力量渔获量(Catch per unit effort,CPUE)与海山相关特征之间的关系。结果表明,中西太平洋两种渔业方式的黄鳍金枪鱼渔获量主要来源于海山区域,海山特征对两种渔业黄鳍金枪鱼的CPUE均产生了极显著性影响(P<0.001)。在延绳钓渔业中,较高的CPUE出现在山顶深度、粗糙度、底面积和海山密度较小、坡度较缓的区域;而在围网渔业中,较高的CPUE则出现在粗糙度较小、山顶深度较大、底面积较大、较陡峭且密集的海山区域。研究探讨了中西太平洋海山特征对黄鳍金枪鱼不同群体的影响机制,为今后进一步探索黄鳍金枪鱼种群分布和资源丰度变化与海洋环境的关系提供了参考与新思路。

佳芦河流域设计年径流和设计洪水的广义可加模型

受气象要素变化和人类活动的影响,径流序列的一致性常遭到破坏,需要直接针对非一致性水文序列进行研究。本文以黄土高原区佳芦河流域的年径流量和年最大洪峰流量为研究对象,基于广义可加模型(GAMLSS),优选气温、降水和库坝控制下的流域有效产流面积作为协变量进行模拟分析,并采用平均设计寿命水平法(ADLL)与等效可靠性法(ER)估算非平稳序列的设计年径流及洪峰流量。结果表明:两个水文序列均具有非一致性,所构建的多协变量年径流模型的低分位数曲线和高分位数曲线的拟合效果较好,洪峰流量模型整体模拟效果均较好;GAMLSS模型及相应的ADLL和ER方法在特定重现期或概率区间能提供更高精度的洪峰流量及年径流量设计值,且较好地体现了水文序列变化趋势对设计值的影响。

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.