人工神经网络模型及其在农业和生态学研究中的应用

对于一些复杂的农业生态系统,人们对其生态过程了解较少,且这些系统的不确定性和模糊性较大,用传统的方法难以模拟这些系统的行为,神经网络模型因为能较精确地模拟这些系统的行为,而引起生态学者们的广泛兴趣。该文着重介绍了误差逆传神经网络模型的结构、算法及其在农业和生态学中的应用研究。误差逆传神经网络模型一般采用三层神经网络模型结构,三层的神经网络模型能模拟任意复杂程度的连续函数,而且因为它的结构小而不容易产生与训练数据的过度吻合。误差逆传神经网络模型算法的主要特征是:利用当前的输入误差对权值进行调整。在生态学和农业研究中,误差逆传神经网络模型通常作为非线性函数模拟器用于预测作物产量、生物生产量、生物与环境之间的关系等。已有的研究表明:误差逆传神经网络模型的模拟精度要远远高于多元线性方程,类似于非线性方程,而在样本量足够的情况下,有一定的外推能力。但是误差逆传神经网络模型需要大量的样本量来保证所求取参数的可靠性,但这在实际研究中很难做到,因而限制了误差逆传神经网络模型的应用。近年来人们提出了强制训练停止、复合模型等多种技术来提高误差逆传神经网络模型的外推能力,也提出了Garson算法、敏感性分析以及随机化检验等技术对误差逆传神经网络模型的机理进行解释。误差逆传神经网络模型的真正优势在于模拟人们了解较少或不确定性和模糊性较大系统的行为,这些是传统模型所无法实现的,因而是对传统机理模型的重要补充。     

城市边缘区景观生态规划的人工神经网络模型

景观生态规划是景观生态学的一个重要应用领域,本文在地理信息系统的辅助下引入了人工神经网络这一新兴应用技术,建立了城市边缘区景观生态规划的BP神经网络模型,模型以区域的高程、高程离差、坡度、坡度离差、地貌分区、离黄河距离、居民点数七个要素作为输入变量,选取斑块密度、分维数、Shannon多样性指数和聚集度指数作为输出变量,精心采集了20个样本对网络进行训练,结果表明网络收敛效果理想,泛化能力强,为景观生态规划提供了一个新的模拟分析手段。     

基于BP人工神经网络的小城镇生态环境质量评价模型

针对中国小城镇生态环境质量综合评价存在的问题,以生态环境质量指标体系作为神经网络的输入、以生态环境等级评分作为输出,基于BP人工神经网络,建立了具有20个隐含层节点、3层网络的小城镇生态环境质量评价模型;以生态环境指标的各级评价标准作为模型的训练样本,以训练样本数量的10%以及各指标各等级的临界值、中间值作为检验样本,以研究区生态环境质量的实际监测值作为预测样本,利用MATLAB软件对BP人工神经网络进行训练,并对鄂州市杜山镇生态环境质量等级进行了模式识别.结果表明：利用BP人工神经网络方法对小城镇生态环境质量进行预测是可行的、可靠的,它不仅能很好地评价区域生态环境质量,而且能够与区域生态环境的实际特征相结合.     

基于BP人工神经网络的兴安落叶松天然林全林分生长模型的研究

以大兴安岭地区兴安落叶松天然林为研究对象,基于688块固定标准地数据,采用MATLAB中log-sigmoid型函数(logsig)和线性函数(purelin)为神经元的作用函数,依据全林分生长模型的概念,以年龄（A）、地位级指数（SCI）和林分密度指数（SDI）作为输入变量,以林分每公顷蓄积量（M）作为输出变量,构建和训练了全林分生长的BP人工神经网络模型,并与常规建模方法进行了对比研究。结果表明,BP人工神经网络模型的拟合精度高达99.6%,检验精度为98.9%,说明与其它建模方法相比人工神经网络建模具有较高的拟合精度和适应性,对林分生长具有更好的预测能力。     

用于昆虫分类鉴定的人工神经网络方法研究:主成分分析与数学建模

为探讨人工神经网络(ANN)在昆虫分类上的可行性,本文提出利用主成分分析和数学建模等方法相结合改进ANN,并以鳞翅目夜蛾科6种蛾类昆虫为样本进行了验证.首先利用Bugshape1.0特征提取软件获取6种蛾180个右前翅样本的13项数学形态特征数据,再运用主成分分析对蛾翅数学形态特征变量重新组合生成新的综合变量,然后结合主成分分析建立BP神经网络分类器.主成分分析结果表明,前5个主成分的累积贡献率为85.52%,已基本包含了全部特征变量具有的信息.在主成分分析的基础上,建立具有5个输入层节点,12个隐含层节点和1个输出层节点的三层BP神经网络分类器.每种蛾20个样本共120组特征数据对分类器进行训练和仿真,其余60组特征数据对分类器进行验证,仿真输出值与目标值的相关系数R=0.997,分类正确率达到了93.33%.较之未经过主成分分析而单独使用BP神经网络建立的分类器,基于主成分分析的BP神经网络分类器具有更优的性能和更准确的分类能力.研究结果表明本文提出的方法具有很好的分类和鉴别作用,为蛾种类的鉴别提供了一种可行的方法.     

BP人工神经网络模拟杨树林冠蒸腾

利用2008和2010年的气温、饱和差、总辐射和叶面积指数作为模型输入,液流法观测的蒸腾速率作为模型输出,建立了用于杨树林冠蒸腾模拟的BP人工神经网络模型,利用2009年的观测数据对模型的模拟能力进行了检验,并应用连接权值计算得到的输入变量对输出变量的相对贡献进行了敏感性分析。结果表明:建立的BP人工神经网络蒸腾模型可以很好的模拟林冠蒸腾大小和季节变化,模拟的绝对误差和绝对相对误差的平均值分别为0.11 mm/d和9.5%,纳什效率系数为0.83;输入变量对蒸腾的相对贡献以及蒸腾与输入变量之间的相关性大小顺序相同,均为总辐射叶面积指数饱和差气温。     

水稻生长动态模拟研究进展

在查阅了国内外水稻生长动态模拟研究领域大量献的基础上,主要就气候变化对水稻生长的影响的模拟,水稻生产潜力的估算,生育期预测,氮肥的优化管理,水稻群体质量指标的模拟与优化以及水稻干物质生产模拟等6个方面的研究动态进行了综述。提出了水稻生长动态模拟模型研究和应用中存在的建模方法,参数确定和生产应用等3个方面的问题,最后对该领域今后的攻关内容进行了探讨,认为进一步研制和完善包括营养元素,病虫害在内的,以作物生理生态为基础的水稻生产系统综合性模拟模型,充分利用以信息技术为主体的现代科学技术,组织全国范围的协作试验以建立水稻品种参数数据库和研制估算水稻品种参数的数学方法,将水稻生长动态模拟模型和专家系统结合,组建水稻生产优化管理决策支持系统,是提高水稻生长动态模拟模型实用性的关键。     

基于熵准则的鲁棒的RBF谷胱甘肽发酵建模

在谷胱甘肽的发酵过程建模中, 当试验数据含有噪音时, 往往会导致模型预测精度和泛化能力的下降。针对该问题, 提出了一种新的基于熵准则的RBF神经网络建模方法。与传统的基于MSE准则函数的建模方法相比, 新方法能从训练样本的整体分布结构来进行模型参数学习, 有效地避免了传统的基于MSE准则的RBF网络的过学习和泛化能力差的缺陷。将该模型应用到实际的谷胱甘肽发酵过程建模中, 实验结果表明: 该方法具有较高的预测精度、泛化能力和良好的鲁棒性, 从而对谷胱甘肽的发酵建模有潜在的应用价值。     

基于熵准则的鲁棒的RBF谷胱甘肽发酵建模

在谷胱甘肽的发酵过程建模中, 当试验数据含有噪音时, 往往会导致模型预测精度和泛化能力的下降。针对该问题, 提出了一种新的基于熵准则的RBF神经网络建模方法。与传统的基于MSE准则函数的建模方法相比, 新方法能从训练样本的整体分布结构来进行模型参数学习, 有效地避免了传统的基于MSE准则的RBF网络的过学习和泛化能力差的缺陷。将该模型应用到实际的谷胱甘肽发酵过程建模中, 实验结果表明: 该方法具有较高的预测精度、泛化能力和良好的鲁棒性, 从而对谷胱甘肽的发酵建模有潜在的应用价值。     

遗传算法优化真菌深层培养过程神经网络模型物的研究

提出一种简单的真菌深层培养过程网络模型。输入变量为可在线测量的排气中的二氧化碳浓度,网络权数采用遗传算法进行优化训练。所获神经网络模型能准确预测培养过程的状态变量。研究表明遗传算法训练此类神经网络系统是可行的。     

Application of neural computing methods for interpreting phospholipid fatty acid profiles of natural microbial communities

The microbial community compositions of surface and subsurface marine sediments and sediments lining burrows of marine polychaetes and hemichordates from the North Inlet estuary (near Georgetown, S.C. ) were analyzed by comparing ester-linked phospholipid fatty acid (PLFA) profiles with a back-propagating neural network (NN). The NNs were trained to relate PLFA inputs to sediment type outputs (e.g., surface, subsurface, and burrow lining) and worm species (e.g., Notomastus lobatus, Balanoglossus aurantiacus, and Branchyoasychus americana). Sensitivity analysis was used to determine which of the 60 PLFAs significantly contributed to training the NN. The NN architecture was optimized by changing the number of hidden neurons and calculating the cross-validation error between predicted and actual outputs of training and test data. The optimal NN architecture was found to be four hidden neurons with 60-input neurons representing the 60 PLFAs, and four output neurons coding for both sediment types and worm species. Comparison of cross-validation results using NNs and linear discriminant analysis (LDA) revealed that NNs had significantly fewer incorrect classifications (2.7%) than LDA (8.4%). For the NN cross-validation, both sediment type and worm species had 3 incorrect classifications out of 112. For the LDA cross-validation, sediment type and worm species had 7 and 12 incorrect classifications out of 112, respectively. Sensitivity analysis of the trained NNs revealed that 17 fatty acids explained 50% of variability in the data set. These PLFAs were highly different among sediments and burrow types, indicating significant differences in the microbiota.     

Predictive modeling of competitive biosorption equilibrium data

This paper compares regression and neural network modeling approaches to predict competitive biosorption equilibrium data. The regression approach is based on the fitting of modified Langmuir-type isotherm models to experimental data. Neural networks, on the other hand, are non-parametric statistical estimators capable of identifying patterns in data and correlations between input and output. Our results show that the neural network approach outperforms traditional regression-based modeling in correlating and predicting the simultaneous uptake of copper and cadmium by a microbial biosorbent. The neural network is capable of accurately predicting unseen data when provided with limited amounts of data for training. Because neural networks are purely data-driven models, they are more suitable for obtaining accurate predictions than for probing the physical nature of the biosorption process.     

Application of Nonlinear Analysis Methods for Identifying Relationships Between Microbial Community Structure and Groundwater Geochemistry

The relationship between groundwater geochemistry and microbial community structure can be complex and difficult to assess. We applied nonlinear and generalized linear data analysis methods to relate microbial biomarkers (phospholipids fatty acids, PLFA) to groundwater geochemical characteristics at the Shiprock uranium mill tailings disposal site that is primarily contaminated by uranium, sulfate, and nitrate. First, predictive models were constructed using feedforward artificial neural networks (NN) to predict PLFA classes from geochemistry. To reduce the danger of overfitting, parsimonious NN architectures were selected based on pruning of hidden nodes and elimination of redundant predictor (geochemical) variables. The resulting NN models greatly outperformed the generalized linear models. Sensitivity analysis indicated that tritium, which was indicative of riverine influences, and uranium were important in predicting the distributions of the PLFA classes. In contrast, nitrate concentration and inorganic carbon were least important, and total ionic strength was of intermediate importance. Second, nonlinear principal components (NPC) were extracted from the PLFA data using a variant of the feedforward NN. The NPC grouped the samples according to similar geochemistry. PLFA indicators of Gram-negative bacteria and eukaryotes were associated with the groups of wells with lower levels of contamination. The more contaminated samples contained microbial communities that were predominated by terminally branched saturates and branched monounsaturates that are indicative of metal reducers, actinomycetes, and Gram-positive bacteria. These results indicate that the microbial community at the site is coupled to the geochemistry and knowledge of the geochemistry allows prediction of the community composition.     

Protein secondary structure prediction using three neural networks and a segmental semi Markov model

Prediction of protein secondary structure is an important step towards elucidating its three dimensional structure and its function. This is a challenging problem in bioinformatics. Segmental semi Markov models (SSMMs) are one of the best studied methods in this field. However, incorporating evolutionary information to these methods is somewhat difficult. On the other hand, the systems of multiple neural networks (NNs) are powerful tools for multi-class pattern classification which can easily be applied to take these sorts of information into account.To overcome the weakness of SSMMs in prediction, in this work we consider a SSMM as a decision function on outputs of three NNs that uses multiple sequence alignment profiles. We consider four types of observations for outputs of a neural network. Then profile table related to each sequence is reduced to a sequence of four observations. In order to predict secondary structure of each amino acid we need to consider a decision function. We use an SSMM on outputs of three neural networks. The proposed SSMM has discriminative power and weights over different dependency models for outputs of neural networks. The results show that the accuracy of our model in predictions, particularly for strands, is considerably increased.     

Application of artificial neural network model for the development of optimized complex medium for phenol degradation using <Emphasis Type="Italic">Pseudomonas pictorum</Emphasis> (NICM 2074)

Biodegradation of phenol using Pseudomonas pictorum (NICM 2074) a potential biodegradant of phenol was investigated for its degrading potential under different operating conditions. The neural network input parameter set consisted of the same set of four levels of maltose (0.025, 0.05, 0.075 g/l), phosphate (3, 12.5, 22 g/l), pH (7, 8, 9) and temperature (30°C, 32°C, 34°C) on phenol degradation was investigated and a Artificial Neural Network (ANN) model was developed to predict the extent of degradation. The learning, recall and generalization characteristic of neural networks was studied using phenol degradation system data. The efficiency of the model generated by the ANN, was tested and compared with the results obtained from an established second order polynomial multiple regression analysis (MRA). Further, the two models (ANN and MRA) were used to predict the percentage of degradation of phenol for blind test data. Performance of both the models were validated in the cases of training and test data, ANN was recommended based on the following higher coefficient of determination R ²; lower standard error of residuals and lower mean absolute percentage deviation.     

Prediction of the disulfide bonding state of cysteines in proteins with hidden neural networks

A hybrid system (hidden neural network) based on a hidden Markov model (HMM) and neural networks (NN) was trained to predict the bonding states of cysteines in proteins starting from the residue chains. Training was performed using 4136 cysteine-containing segments extracted from 969 non-homologous proteins of well-resolved 3D structure and without chain-breaks. After a 20-fold cross-validation procedure, the efficiency of the prediction scores as high as 80% using neural networks based on evolutionary information. When the whole protein is taken into account by means of an HMM, a hybrid system is generated, whose emission probabilities are computed using the NN output (hidden neural networks). In this case, the predictor accuracy increases up to 88%. Further, when tested on a protein basis, the hybrid system can correctly predict 84% of the chains in the data set, with a gain of at least 27% over the NN predictor.     

程海富营养化机理的神经网络模拟及响应情景分析

揭示湖泊的富营养化发生机制、定量了解关键生源要素与藻类爆发的因果关联对有效改善湖泊水质和富营养化状况具有重要的科学与决策意义。本研究以云南省程海为例,建立了基于神经网络的响应模型,对富营养化机理进行了研究,并从富营养化核心驱动因子识别、神经网络模型构建与架构分析以及叶绿素a(Chl a)与TN、TP浓度降低的响应模拟几个方面对面临的科学问题进行探索。模拟结果表明,神经网络模型必须在适当的架构下才能产生科学合理的结果;程海的富营养化机制由一个氮(N)、磷(P)共限制的营养盐-藻类动力结构主导,但在此主导结构下拥有氮型限制的次级结构。基于神经网络模型模拟,推导出一系列基于湖体水质控制的Chl a响应的非线性函数,为程海的富营养化控制提供了快速决策支持。     

OutbreakFlow: Model-based Bayesian inference of disease outbreak dynamics with invertible neural networks and its application to the COVID-19 pandemics in Germany

Mathematical models in epidemiology are an indispensable tool to determine the dynamics and important characteristics of infectious diseases. Apart from their scientific merit, these models are often used to inform political decisions and interventional measures during an ongoing outbreak. However, reliably inferring the epidemical dynamics by connecting complex models to real data is still hard and requires either laborious manual parameter fitting or expensive optimization methods which have to be repeated from scratch for every application of a given model. In this work, we address this problem with a novel combination of epidemiological modeling with specialized neural networks. Our approach entails two computational phases: In an initial training phase, a mathematical model describing the epidemic is used as a coach for a neural network, which acquires global knowledge about the full range of possible disease dynamics. In the subsequent inference phase, the trained neural network processes the observed data of an actual outbreak and infers the parameters of the model in order to realistically reproduce the observed dynamics and reliably predict future progression. With its flexible framework, our simulation-based approach is applicable to a variety of epidemiological models. Moreover, since our method is fully Bayesian, it is designed to incorporate all available prior knowledge about plausible parameter values and returns complete joint posterior distributions over these parameters. Application of our method to the early Covid-19 outbreak phase in Germany demonstrates that we are able to obtain reliable probabilistic estimates for important disease characteristics, such as generation time, fraction of undetected infections, likelihood of transmission before symptom onset, and reporting delays using a very moderate amount of real-world observations.     

Modeling of phenol degradation system using artificial neural networks

Pseudomonas pictorum (NICM-2077) an effective strain used in the biodegradation of phenol was grown on various nutrient compounds which protect the microbes while confronting shock loads of concentrated toxic pollutants during waste water treatment. In the present study the effect of glucose, yeast extract, (NH₄)₂SO₄ and NaCl on phenol degradation has been investigated and a Artificial Neural Network (ANN) Model has been developed to predict degradation. Also the learning, recall and generalization characteristics of neural networks has been studied using phenol degradation system data. The network model was then compared with a Multiple Regression Analysis model (MRA) arrived from the same training data. Further, these two models were used to predict the percentage degradation of phenol for a blind test data. Though both the models perform equally well ANN is found to be better than MRA due to its slightly higher coefficient of correlation, lower RMS error value and lower average absolute error value during prediction.