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.     

基于改进水平集算法的锅炉承压视觉预警技术

传统锅炉承压预警技术主要依靠人工巡查发现问题,在检测过程中受工作人员自身操作不稳定因素影响,存在故障预警实时性差、盲点多等问题,达不到相应的预警要求.针对此问题,提出基于改进采用水平集算法的锅炉承压预警技术.利用图像处理技术从图像中分割出锅炉承压关键部件,统计关键部件灰度值,并利用灰度值设定特征图像及初始轮廓,设计采用...     

应急状态下矿井瓦斯爆炸致灾因子传播快速分类器研究*

为了在矿井瓦斯爆炸灾变发生后,快速确定瓦斯爆炸冲击波的压力、温度、有毒有害气体等致灾因子在井巷网络中的传播情况。利用CFD数值模拟或爆炸实验获得瓦斯爆炸冲击波的压力、温度、有毒有害气体等致灾因子传播大数据,将影响瓦斯爆炸传播的因素以及观测点等参数作为人工神经网络的输入节点,压力、温度等致灾因子作为输出节点,建立瓦斯爆炸致灾因子传播快速预测机器学习模型,解决CFD数值模拟的建模、计算及数据分析处理等过程耗时大、不适应灾变应急的快速响应等问题。研究结果表明:在给定爆炸位置和爆炸当量的均直巷道,获得任一点的爆炸冲击波压力、温度以及有毒有害气体所需时间是瞬时的,人工神经网络平均训练误差为6.92 %,有训练样本的验证误差为5.24 %,无训练样本的验证误差为6.88 %。     

A new method for broad-scale modeling and projection of plant assemblages under climatic,biotic, and environmental cofiltering

There is increasing interestin broad-scale analysis, modeling, and prediction of the distribution and composition of plant species assemblages under climatic, environmental, and biotic change, particularly for conservation purposes. We devised a method to reliably predict the impact of climate change on large assemblages of plant communities, while also considering competing biotic and environmental factors. To this purpose, we first used multilabel algorithms in order to convert the task of explaining a large assemblage of plant communities into a classification framework able to capture with high cross-validated accuracy the pattern of species distributions under a composite set of biotic and abiotic factors. We applied our model to a large set of plant communities in the Swiss Alps. Our model explained presences and absences of 175 plant species in 608 plots with >87% cross-validated accuracy, predicted decreases in α, β, and γ diversity by 2040 under both moderate and extreme climate scenarios, and identified likely advantaged and disadvantaged plant species under climate change. Multilabel variable selection revealed the overriding importance of topography, soils, and temperature extremes (rather than averages) in determining the distribution of plant species in the study area and their response to climate change. Our method addressed a number of challenging research problems, such as scaling to large numbers of species, considering species relationships and rarity, and addressing an overwhelming proportion of absences in presence–absence matrices. By handling hundreds to thousands of plants and plots simultaneously over large areas, our method can inform broad-scale conservation of plant species under climate change because it allows species that require urgent conservation action (assisted migration, seed conservation, and ex situ conservation) to be detected and prioritized. Our method also increases the practicality of assisted colonization of plant species by helping to prevent ill-advised introduction of plant species with limited future survival probability.     

基于RF-ELM模型的边坡稳定性预测研究

为实现对边坡稳定性的有效预测,将极限学习机算法与旋转森林算法相结合,并依据影响边坡稳定性的六项重要因素,建立了边坡稳定性预测的RF-ELM预测模型。该模型是以极限学习机算法为基分类器,以旋转森林算法为框架的集成学习模型,利用UCI数据库中三组数据集验证了该集成模型确实提高了ELM的预测性能。将RF-ELM模型应用于边坡稳定性的预测问题中,结合39组工程实例数据进行预测实验,结果表明该模型具有较高的预测精度,可有效的对边坡稳定性进行预测。     

The expansion of tree-based boom crops in mainland Southeast Asia: 2001 to 2014

Over the past half century, countries of Mainland Southeast Asia (MSEA) – Cambodia, Laos, Myanmar, Thailand, and Vietnam – have witnessed increases in commercialized agriculture with rapid expansions of boom-crop plantations. We used MODIS EVI and SWIR time-series from 2001–2014 to classify tree-cover changes across MSEA and performed a supervised change detection using an upscaling approach by deriving samples from existing Landsat classifications. We used the random forest classifier and distinguished 24 classes (16 representing boom-crops) with an accuracy of 82.2%. Boom-crops occupy about 18% of the landscape (8% of which is rubber). Since 2003 74,960 km² of rubber have been planted; 70% of rubber is planted on former forest land, and 30% on low vegetation area (mainly former croplands). Timing, patterns of change, and deforestation rates, however, differ among the MSEA countries and the high spatial and temporal detail of our classification allowed us to quantify dynamics and discuss political and socio-economic drivers of change.     

基于深度置信网络的管网泄漏故障诊断方法研究

针对当前管网系统数据量大不利于传统模型方法诊断故障的问题,设计了1种基于深度置信网络的管网故障诊断算法。首先,对管网数据结构以及管网系统运行状态进行分析,选取管网主要数据作为故障诊断网络的输入,确定相应运行状态作为诊断网络输出;其次,设计了基于多个受限制玻尔兹曼机与Softmax分类器级联的深度置信网络,并且利用对比散度算法和BP算法对模型进行预训练与调优,使模型参数达到全局最优;最后,通过实验测试确定所设计的深度置信网络的训练迭代次数与网络层数,使算法诊断准确率达到最优。研究结果表明:提出的基于深度置信网络的管网故障诊断算法对管网故障诊断可以达到良好的诊断结果,泄漏预测准确率在验证集样本上可达96.87%,在管网泄漏检测方面,相较于传统基于模型的方法优势明显。