基于GA-ELM的冲击地压危险性预测研究

为提高冲击地压预测的效率和准确率,在分析冲击地压影响因素的基础上,提出了一种将遗传算法(GA)与极限学习机(ELM)相结合的冲击地压预测的新方法。为了避免ELM受输入权值矩阵和隐含层偏差随机性的影响,算法采用GA对ELM的输入权值矩阵和隐含层偏差进行优化,建立GA-ELM冲击地压预测模型。利用某矿冲击地压统计数据对该模型进行了实例分析,将ELM、SVM和BP算法预测结果与该模型进行了对比分析。结果表明:GA-ELM模型具有较高的预测精度,可以相对准确、有效地对冲击地压发生的可能性进行预测。     

SDA脱硫工艺在烧结烟气脱硫中的应用

介绍了沙钢集团4号360 m2烧结机全烟气脱硫项目的工艺选择、脱硫原理、工艺流程及投产后生产情况,通过对生产后运行分析,认为旋转喷雾干燥法(SDA)作为一种新型的脱硫工艺,适合作为烧结烟气脱硫的主工艺进行推广。     

The role of social learning for soil conservation: the case of Amba Zuria land management,Ethiopia

Social learning plays key roles in sustainable natural resource management; however, studies on its role show mixed results. Even though most current studies highlight positive outcomes, there are also negative effects of social learning with respect to natural resource management. This paper explores the influence of social learning outcomes on the adoption of soil conservation practices in Amba Zuria, Ethiopia. Data were collected through semi-structured interviews, group discussions and in workshops. The interviews were transcribed and analysed with the qualitative data analysis software ATLAS.ti. In addition, a review of pertinent documents and literature complements the analysis. We found that social learning results in positive outcomes in interactions among the actors, broader understanding on soil conservation, trust and mutual understanding, and the creation of opportunities to apply both indigenous and scientific knowledge. These outcomes, in turn, foster the adoption of stone terraces to conserve soils. No such positive results were found for farmers who were not involved in the learning process and who did not adopt stone terraces. The non-adoption can partly be explained by their exclusion from sharing the positive outcomes of social learning. The study also shows limitations related to the short duration of the learning platforms, insufficient representation of women and irregular patterns of learning schedules. Hence, Ethiopian soil conservation strategies should put emphasis on social learning in order to strengthen the collective processes of soil conservation measures.     

The vulnerable do‐gooders: security strategies of German aid agencies

Humanitarian and development agencies have confronted growing insecurity in some of the regions in which they work over the course of the past decade. Consequently, aid agencies are changing their approach to security issues. Compared to aid agencies from other countries, especially the United Kingdom and the United States, German aid agencies have been relatively slow to adopt security measures. In addition, the security measures they have selected differ from one agency to another, even in comparable security contexts. The literature on organisational learning helps to explain these differences. The findings show that external and internal organisational factors influence an aid agency's choice of a particular security measure. The different responses of German aid agencies to the changing security environment demonstrate that security measures are not only influenced by an organisation's identity but also can be a way for them to demonstrate their identity and thereby distinguish themselves from other actors that deliver aid.     

烧结机头电除尘灰的处理与利用

针对烧结机头除尘灰的组成和特点进行分析,探讨了目前国内钢铁企业对烧结机头除尘灰处理的现状和存在的问题,对烧结机头除尘灰固体废弃物的处理与利用技术的发展提出了一些建议和展望。     

基于GA-ELM浆体管道输送临界流速预测模型研究

针对浆体管道输送临界流速预测难度大、精确度低等技术难题,提出了基于极限学习机(ELM)的临界流速预测模型,用训练集对模型进行训练,以验证集预测值的均方误差作为适应度函数,利用遗传算法(GA)对ELM模型参数进行优化,应用优化得到的ELM模型对预测集进行预测。以某矿山为例,模型参数优化结果如下:隐含层节点数L为400,输入权值ai、偏置向量bi最优组合下预测结果适应度为0.0201。采用优化的ELM模型对预测集进行预测,预测结果的最大相对误差x=3.96%,平均相对误差y=1.58%,对比BP神经网络(x=12.95%)和SVM模型(x=3.19%),表明ELM模型更加精确、高效。     

基于PCA-ELM模型的露采爆破振动对民房破坏的预测分析

针对露天采矿爆破振动对民房破坏的预测问题,采用主成分分析(PCA)和极限学习机(ELM)方法,选取爆破振幅、主频率、主频率持续时间、灰缝强度、砖墙面积率、房屋高度、屋盖形式、圈梁立柱、施工质量、场地条件10个主要影响因素。引入相关性分析在主成分分析过程中,对相关性高的指标进行降维,把得到的3个综合因子和爆破振幅、主频率、主频率持续时间、砖墙面积率作为输入变量,构建露天煤矿PCA-ELM预测模型。选取露天矿实际爆破过程中测量的100组数据作为模型训练样本,用另外20组数据作为测试样本进行预测。结果表明:对民房破坏影响因素中灰缝强度、房屋高度、屋盖形式、圈梁立柱、施工质量、场地条件之间具有较高的关联度。该模型处理高维数据时较传统的ELM算法具有预测精度高、稳定性好等特点,可准确预测爆破振动对民房的破坏程度,误判率为1/20。     

基于EMD-MFOA-ELM的瓦斯涌出量时变序列预测研究

为准确分析工作面绝对瓦斯涌出量的非平稳特征,实现瓦斯涌出量的准确预测,基于经验模态分解(EMD)、修正的果蝇优化算法(MFOA)和极限学习机(ELM)基本原理,构建瓦斯涌出量的EMD-MFOA-ELM多尺度时变预测模型。通过EMD将瓦斯涌出量时变序列进行深层次分解,获得多尺度本征模态函数(IMF);采用MFOA-ELM对各IMF时变序列建立动态预测模型,等权叠加各预测值,得到模型最终预测结果。以晋煤某矿瓦斯涌出量监测时序样本为例进行研究分析,结果表明:EMD能充分挖掘出监测数据隐含信息,有效降低数据复杂度;该模型预测相对误差为0.024 3%~0.651 0%,平均值仅为0.252 6%,预测精度和泛化能力高于未经EMD分解模型,能很好地适用于非平稳时变序列预测。     

气喷旋冲烟气脱硫装置运行关键控制

气喷旋冲湿式石灰石-石膏法烧结烟气脱硫装置(XPB)是宝钢针对烧结烟气特性自主研发的新型脱硫技术,较详细论述该系统稳定、经济运行的关键控制:增压风机风门或叶轮的控制、吸收塔进气箱温度的控制、吸收塔反应池pH值的控制、曝气管浸没深度的控制等。     

毒死蜱低剂量重复染毒致学习记忆功能障碍及CREB信号途径的抑制(Repeated Exposures to Low-Level Chlorpyrifos Result in Learning Memory Impairment and Inhibition of the CREB Signal System)

毒死蜱(CPF)是我国推荐使用的低毒有机磷农药之一,目前在全世界广泛应用.选择不引起动物出现全身系统毒性的CPF剂量(1、5、10mg·kg-1)对大鼠连续染毒4周,评价CPF对动物的学习记忆功能和CREB信号通路的影响.结果显示:CPF连续染毒后动物出现记忆能力的损伤,随着染毒剂量的增加,海马、皮质和纹状体CREB及ERKI/Ⅱ磷酸化蛋白表达逐渐减少,这表明ERK/CREB信号通路可能是参与毒死蜱低剂量重复暴露致学习记忆功能改变的重要途径.