基于时间序列神经网络预测的研究

阐述了BP神经网络基本原理,并对G．P算法进行描述,提出了一种基于BP神经网络的时间序列的预测和方法,通过应用实例的分析,该方法可以得到BP网络应用于非线性时间序列预测是可行的,结果表明：神经网络方法可以成功地用于分析预测时间序列变量．参4．     

应用于水文预报的优化BP神经网络研究

利用广东省滨江流域的水文观测资料，建立了以前期降水量为预报因子、以水位为输出的BP人工神经网络水文预报模型。首先采用了合理的方法进行样本组织，进而利用最优子集回归技术进行输入因子的确定，然后进行了不同隐层节点数、不同转移函数、不同训练算法的组合试验，确定了应用于水文预报中的优化BP神经网络：网络结构为8-9-1；转移函数的组合方式为tansig-线性函数；训练算法为采用evenberg-Marquardt(Lm)算法。为便于精度分析，还采用了最优子集回归模型作了研究。结果表明，优化BP网络模型无论在拟合精度还是在预测精度上都高于最优子集模型。总的来说BP网络是一种精度较高的水文预测模型。     

BP模型的改进及其在大气污染预报中的应用

针对传统BP模型存在着训练速度较慢、局部极值以及最佳网络结构无法准确确定的不足，进行了改进，应用于城市空气污染预报，建立大气污染浓度的神经网络预测模型。计算结果表明，应用改进的BP模型进行大气污染预报能够得到更好的预测结果，具有很强的实用性。     

基于多模型优选的区域土壤重金属含量空间预测方法研究

土壤重金属含量空间预测研究对实现区域土壤资源的优化利用以及土壤环境的保护和污染防治具有重要意义。以江苏省常州市金坛区为例,基于源、汇和空间分异因子,利用BP神经网络(back-propagation network)建模方法,分别构建了源汇模型(BP-S)、空间分异模型(BP-K)和改进的多因素综合模型(BP-SK),模拟预测了区域土壤重金属Cd、Pb、Cr、Cu和Zn含量的空间分布,并对各模型预测精度进行对比分析。针对不同模型在不同区域和元素间预测精度的差异,优选出预测精度最高的模型组合,以此探求区域重金属含量空间分布最优预测结果。结果表明:(1)BP-SK模型对Cd、Cr、Cu和Zn含量预测精度均高于BP-S和BP-K模型,仅在对Pb含量的预测中BP-S、BP-K模型精度高于BP-SK模型,BP-SK模型比其他模型更能突出局部特征,包含的信息更丰富。(2)优选后最优模型预测精度较原单一模型均有不同程度的提高,对Cd、Pb、Cr、Cu和Zn含量的预测精度分别提高15.15%、20.71%、19.19%、1.75%和9.24%。(3)各模型对Cd、Pb、Cr、Cu和Zn含量的空间预测高值区均位于区域中部和东北部,低值区位于西部丘陵山区,BP-SK模型在人为影响剧烈的地区预测效果更好,而BP-K模型在自然因素影响较大的丘陵山地区的适用性更好。     

基于BP神经网络算法的密云水库水质参数反演研究

密云水库是北京市重要的地表饮用水源地,在保障首都水源安全方面起着重要作用,而密云水库水质参数的区域监测已成为当前亟待解决的问题。为了实现密云水库水质参数大范围、实时获取,该文基于遥感和GIS技术,采用BP神经网络算法,结合地面监测数据和Landsat 8遥感影像,分别建立了反演总磷、总氮、氨氮和COD(化学需氧量)4个水质参数的BP神经网络模型,并反演了密云水库2013-2018年非结冰期主要水质参数,分析了密云水库主要水质参数的年际变化特征、季节变化特征和空间分异特征。结果表明,(1)水质参数的Landsat 8敏感波段分别为:总氮为1、4波段,氨氮为1-7波段,总磷为1、3-7波段,COD为2-5波段。(2)密云水库主要水质参数在2013-2018年总体呈下降趋势,氨氮和COD为Ⅰ类水质,总磷为Ⅱ类水质,总氮为Ⅲ类水质。(3)4个水质参数指标春季最高、秋季次之、夏季最低,总氮、总磷、氨氮和COD的春季值分别是夏季值的1.08、1.36、1.6、1.45倍。(4)密云水库不同水质参数的空间差异性较大,总体来看,水库北部和东部的4个水质参数含量相对较高,这种分布与北部和东部村庄密集以及密云水库两大入库河流有关。综上所述,基于BP神经网络算法的密云水库水质反演研究是可行的,且得到了较为可信的研究结果,该研究可为密云水库水质管理与政策制定提供重要的科学依据。     

Candida utilis生物合成谷胱甘肽的营养及环境条件

研究了摇瓶条件下Candida utilis WSH02—08生物合成谷胱甘肽(GSH)的营养条件，确定以葡萄糖作为碳源，硫酸铵和尿素作为混合氮源．在L6(4^5)正交试验的基础上，选用BP神经网络对营养条件进行优化和预测，得出较优的营养组合为：葡萄糖30gL^-1(NH4)2SO4gL^-1，尿素4gL^-1，KH2PO4 3gL^-1，MgSO4 0．25gL^-1．研究了GSH发酵的环境条件，得出较优的组合为：初始pH6．0，装液量30mL／250mL，接种量10％．利用优化后的营养及环境条件进行摇瓶发酵实验，结果表明，GSH产量和细胞干重均有较大幅度的提高．图7表2参13     

酚类化合物臭氧氧化速率的神经网络研究

酚类化合物(BP)是重要的工业原料或中间体,但工业废水含有的酚类化合物会对环境造成污染。为建立酚类化合物臭氧氧化速率的QSPR(quantitative structure-property relationship)预测模型,分析了23种酚的分子结构与臭氧氧化速率之间的相关关系,计算了这些酚的分子连接性指数和分子形状指数,优化筛选了连接性指数的1χ和2χ、分子形状指数的K1和K2共4种参数,将其作为BP神经网络的输入层变量,臭氧氧化速率作为输出层变量,采用4:2:1的网络结构,获得了令人满意的QSPR神经网络预测模型,模型总相关系数r为0.976,计算得到的臭氧氧化速率的预测值与实验值较为吻合,平均残差仅为0.05;为检验结构参数建立模型的普适性,同样方法建立对酚类化合物的辛醇-水分配系数的预测模型,模型总相关系数r达到0.993,辛醇-水分配系数的预测值与实验值吻合度较为理想,结果表明,本法建构的神经网络模型具有良好的稳健性和预测能力。     

基于植被指数和神经网络的热带人工林地上蓄积量遥感估测

热带森林作为陆地生态系统的组成成分之一,研究其蓄积量估测对我们了解其在全球碳循环中的地位和作用有很重要的意义.但遥感估测森林生态参数的精度如何,还是个不确定的问题.利用LANDSAT-TM数据,基于森林清查数据和遥感技术,以尾叶桉和加勒比松为例,对中国南方地区人工林蓄积量估测进行了尝试研究.首先,通过测量样方胸径、树高,建立森林蓄积量估算模型.其次,通过对比分析不同植被指数与森林蓄积量之间的关系,选择合适植被指数组合,建立多元回归和神经网络模型.结果表明:单波段TM数据和大多数植被指数与蓄积量相关性并不好.神经网络比回归分析模拟效果好.而多元回归和神经网络模型大大提高预测精度.本研究方法对大面积的森林蓄积量估测具有一定的参考价值.     

邯郸采暖期环境污染特征及空气质量预报方法研究

利用河北邯郸气象和环境监测资料,分析了邯郸采暖期空气质量和环境气象条件特征;同时利用线性回归和BP神经网络统计方法对采暖期空气质量进行了预报研究。结果表明,PM_(10)、PM_(2.5)、SO_2、NO_2、CO(95)(CO日均值的第95百分位数)的空气质量指数(AQI)在冬季最高,夏季最低,O_3-8(90)(O_3日最大8 h值的第90百分位数)的AQI则相反。邯郸采暖期首要污染物以PM_(2.5)和PM_(10)为主,除O_3-8(90)外,其他5种污染物采暖期AQI均高于其年均值;同时采暖期降水少,温度低,小风出现频率明显高于非采暖期,而且局地逆温强,静稳天气指数高,是全年环境气象条件最差的时期。邯郸采暖期的环境气象条件1月最差,且夜晚差于白天,尤其是局地5—7时。邯郸采暖期首要污染物浓度与前一日污染物浓度、静稳指数、逆温、相对湿度和露点温度等呈正相关,与气温、风速、能见度和混合层高度等呈负相关。BP神经网络模型对污染物浓度的预报效果优于线性回归模型,可尝试应用于邯郸空气质量预报工作。     

公路交通噪声预测的神经网络模型

利用多条国道、省道公路交通噪声的实测数据，建立了一个公路交通噪声预测的神经网络模型，能够同时预测公路交通噪声的等效连续声级、统计声级和标准偏差，预测值和实测值基本相符，该模型为公路交通噪声的预测、治理和控制提供了良好的条件。     

Determination of the principal factors of river water quality through cluster analysis method and its prediction

In this paper, an artificial neural network model was built to predict the Chemical Oxygen Demand (COD_Mn) measured by permanganate index in Songhua River. To enhance the prediction accuracy, principal factors were determined through the analysis of the weight relation between influencing factors and forecasting object using cluster analysis method, which optimized the topological structure of the prediction model input items of the artificial neural network. It was shown that application of the principal factors in water quality prediction model can improve its forecasting skill significantly through the comparison between results of prediction by artificial neural network and the measurements of the COD_Mn. This methodology is also applicable to various water quality prediction targets of other water bodies and it is valuable for theoretical study and practical application.     

Pesticide prediction in ground water in North Carolina domestic wells using artificial neural networks

This paper illustrates the application of artificial neural networks (ANN) for prediction of pesticide occurrence in rural domestic wells from the available limited information. Among the three ANN models (a feed-forward back propagation [BP], a radial basis function [RBF] and an adaptive neural network-based fuzzy inference system [ANFIS]) employed for this investigation, the BP neural network was found to be superior to RBF and ANFIS type networks for the detection of pesticide occurrences in wells. For improved model prediction efficiency, optimization of network structure (e.g., number of hidden layers and number of nodes in each hidden layer) and spread (the width of the radial basis function) are important for BP and RBF type of network, respectively. A four layer BP network with a 3:2 neurons ratio of the first hidden layer to the second hidden layer produced better prediction performance efficiencies in terms of the pesticide detection efficiency (E_f), the root mean square error (RMSE), and the correlation coefficient (R) and the overall E_f of the BP neural network was found greater than 85%. Sensitivity analysis was performed to measure the relative importance of one input parameter over the other in pesticide occurrence in wells. It was shown in terms of the prediction efficiencies (E_f, RMSE, and R) of a four-layer BP neural network that the time of sample collection (TSC; month of the year), the depths of wells, and pesticide travel times (PTT) were more important parameters in the prediction of the pesticide occurrences in rural domestic wells. This means that the wells having shallow ground water table are more vulnerable to pesticide occurrence.     

基于人工神经网络方法的河北省近海沉积物重金属污染综合评价

利用误差反相传播神经（BP）网络对河北省近海沉积物中的铅、镉、锌、汞、砷5种重金属元素的污染水平进行分析,利用自组织特征映射（SOFM）网络对上述重金属元素分布特征进行分类,通过分类与污染水平量化值的结合,进行综合评价。SOFM把52个沉积物样品分别划分为3、4、6类和9类。对比各种分类,分为3类的物理意义较明确。每个类别分别对应高中低不同的污染物浓度水平,差异显著、分类方式比较合理。通过此种分类可以判断河北省近海的沉积物重金属污染在不同海域存在一定的差别,整体上是离海岸越远,沉积物的重金属污染水平越高,距海岸较近的海域内,沉积物的重金属污染水平较低,但渤海湾内的重金属污染水平高于其他海域。     

Water quality prediction of copper-molybdenum mining-beneficiation wastewater based on the PSO-SVR model

● Data acquisition and pre-processing for wastewater treatment were summarized. ● A PSO-SVR model for predicting COD_eff in wastewater was proposed. ● The COD_eff prediction performances of the three models in the paper were compared. ● The COD_eff prediction effects of different models in other studies were discussed. The mining-beneficiation wastewater treatment is highly complex and nonlinear. Various factors like influent quality, flow rate, pH and chemical dose, tend to restrict the effluent effectiveness of mining-beneficiation wastewater treatment. Chemical oxygen demand (COD) is a crucial indicator to measure the quality of mining-beneficiation wastewater. Predicting COD concentration accurately of mining-beneficiation wastewater after treatment is essential for achieving stable and compliant discharge. This reduces environmental risk and significantly improves the discharge quality of wastewater. This paper presents a novel AI algorithm PSO-SVR, to predict water quality. Hyperparameter optimization of our proposed model PSO-SVR, uses particle swarm optimization to improve support vector regression for COD prediction. The generalization capacity tested on out-of-distribution (OOD) data for our PSO-SVR model is strong, with the following performance metrics of root means square error (RMSE) is 1.51, mean absolute error (MAE) is 1.26, and the coefficient of determination (R²) is 0.85. We compare the performance of PSO-SVR model with back propagation neural network (BPNN) and radial basis function neural network (RBFNN) and shows it edges over in terms of the performance metrics of RMSE, MAE and R², and is the best model for COD prediction of mining-beneficiation wastewater. This is because of the less overfitting tendency of PSO-SVR compared with neural network architectures. Our proposed PSO-SVR model is optimum for the prediction of COD in copper-molybdenum mining-beneficiation wastewater treatment. In addition, PSO-SVR can be used to predict COD on a wide variety of wastewater through the process of transfer learning.