基于BP神经网络的儿童卧室内灰尘PAEs浓度预测

在实测数据的基础上,以邻苯二甲酸酯(PAEs)的各类影响因素为自变量,PAEs浓度为因变量,采用Back-propagation(BP)神经网络建立儿童卧室内PAEs浓度预测模型.结果表明,该模型的预测效果较理想,其中,STD比值均＞0.5,NMB均接近0,EMR均＜19％.以室内环境与儿童健康(CCHH)课题组天津地...     

加热炉炉管穿孔事故处理及原因分析

以一起直接式原油加热炉炉管穿孔事故为例,介绍了事故处理程序,分析了炉管穿孔的原因是由于漏点腐蚀、坑点腐蚀以及炉管管卡损坏导致的,并提出了类似事故防范对策.     

A stochastic-advective transport model for NAPL dissolution and degradation in non-uniform flows in porous media

Remediation schemes for contaminated sites are often evaluated to assess their potential for source zone reduction of mass, or treatment of the contaminant between the source and a control plane (CP) to achieve regulatory limits. In this study, we utilize a stochastic stream tube model to explain the behavior of breakthrough curves (BTCs) across a CP. At the local scale, mass dissolution at the source is combined with an advection model with first-order decay for the dissolved plume. Field-scale averaging is then employed to account for spatial variation in mass within the source zone, and variation in the velocity field. Under the assumption of instantaneous mass transfer from the source to the moving liquid, semi-analytical expressions for the BTC and temporal moments are developed, followed by derivation of expressions for effective velocity, dispersion, and degradation coefficients using the method of moments. It is found that degradation strongly influences the behavior of moments and the effective parameters. While increased heterogeneity in the velocity field results in increased dispersion, degradation causes the center of mass of the plume to shift to earlier times, and reduces the dispersion of the BTC by lowering the concentrations in the tail. Modified definitions of effective parameters are presented for degrading solutes to account for the normalization constant (zeroth moment) that keeps changing with time or distance to the CP. It is shown that anomalous dispersion can result for high degradation rates combined with wide variation in velocity fluctuations. Implications of model results on estimating cleanup times and fulfillment of regulatory limits are discussed. Relating mass removal at the source to flux reductions past a control plane is confounded by many factors. Increased heterogeneity in velocity fields causes mass fluxes past a control plane to persist, however, aggressive remediation between the source and CP can reduce these fluxes.     

条带开采下沉系数计算与优化设计的神经网络模型

在综合分析条带开采地表下沉系数影响因素的基础上,采用神经网络方法建立了条带开采地表下沉系数的计算模型。模型以国内外成功的条带开采实例为学习训练样本和测试样本,对模型的计算结果与实测值进行了对比分析,分析结果表明,该模型的计算值更接近于实测值。在上述研究的基础上,在给定条带开采采出率的条件下,以条带开采的地表下沉系数最小为原则,运用该模型实现了对条带开采尺寸的优化设计。该研究的成果,为条带开采地表下沉系数的理论计算及条带开采尺寸的优化设计探索出了一种新的方法。     

Pipeline leakage detection and isolation: An integrated approach of statistical and wavelet feature extraction with multi-layer perceptron neural network (MLPNN)

Leakage diagnosis of hydrocarbon pipelines can prevent environmental and financial losses. This work proposes a novel method that not only detects the occurrence of a leakage fault, but also suggests its location and severity. The OLGA software is employed to provide the pipeline inlet pressure and outlet flow rates as the training data for the Fault Detection and Isolation (FDI) system. The FDI system is comprised of a Multi-Layer Perceptron Neural Network (MLPNN) classifier with various feature extraction methods including the statistical techniques, wavelet transform, and a fusion of both methods. Once different leakage scenarios are considered and the preprocessing methods are done, the proposed FDI system is applied to a 20-km pipeline in southern Iran (Goldkari-Binak pipeline) and a promising severity and location detectability (a correct classification rate of 92%) and a low False Alarm Rate (FAR) were achieved.     

CH_4/N_2/O_2预混气激波火焰结构数值预测

研究激波着火现象，推导激波加热点燃可燃气控制方程组。针对甲烷预混气激波火焰结构进行数值模拟，计算了激波燃烧时的压力、温度、及不同组分随时间的变化历程。其中甲烷燃烧采用美国BERKELEY大学GRI-MECH机理，该反应机理包含177个基元反应，涉及32种组分。程序采用美国SANDIA国家实验室发展的大型化学反应动力学软件包CHEMKIN III中相关的模型、子程序和热力学数据库。计算结果表明激波火焰有其自身的结构特征。     

九畹溪大桥大跨度钢管拱分段整体吊装施工安全措施探讨

在进行缆索吊装大跨度双肋钢管拱悬拼施工中,过去多采用空中焊接横联作业,该项施工难度大,尤其在桥高、风大、两岸地形陡峭、施工条件困难的情况下,施工安全和施工质量难以保证。是否能将空中焊接作业转化成地面作业,本文详细介绍缆索吊装大跨度双肋钢管拱进行悬拼施工新技术,并结合九畹溪大桥施工实例,探讨缆索吊装双肋悬拼钢管拱施工中工艺程序、质量控制、安全措施等工程施工新方法及施工效果。     

基于BP神经网络不同输入变量的爆破振速预测

以某隧道爆破开挖为实例,利用BP神经网络解决复杂非线性函数逼近问题的能力,以最大段药量、爆心距、爆破分段数、泊松比、岩石基本质量指标作为影响爆破振动速度的主要因素,选取不同维数的输入变量建立BP神经网络模型来预测爆破振动速度。对比分析各组预测值与实测值之间的相对误差,选取合理维数的输入变量建立了爆破振动危害预测的BP神经网络模型。     

基于证据理论的煤矿瓦斯涌出组合预测

为提高煤矿瓦斯涌出量预测的准确度,引入证据理论组合预测方法。根据瓦斯涌出量及其主要影响因素间的实验数据,采用3个不同的粒子群神经网络模型对涌出量进行初步预测。并由BP、RBF网络对预测误差及预测点的影响因素进行分析建模,以获取每个模型的可信度。再利用证据理论对其进行合成,确定组合模型的权值,最终实现对瓦斯涌出量的组合预测。实例结果表明,该组合预测方法的平均绝对误差、均方误差分别为18.5%、5.8%,均小于神经网络组合法及等权平均法的相应预测误差,适用于煤矿瓦斯涌出量预测。     

无起爆药雷管内管装药密度对燃烧转爆轰的影响研究

起爆药雷管生产中产生大量有毒废水,且其运输、贮存存在安全隐患。为克服这些困难,笔者自行制备无起爆药雷管。采用圆筒式金属内管中装填超细PETN(季戊四醇四硝酸酯),作为起爆元件,代替起爆药部分,研究装药密度对内管燃烧转爆轰(DDT)的影响。研究结果表明,在内管壁厚1～2 mm,内径Φ4.0 mm,长25 mm,装压密度为0.8～1.14 g/cm3的范围内,内管能可靠实现燃烧转爆轰。