多因素耦合条件下硫化矿自燃神经网络动态预测模型研究

硫化矿石自燃是多种因素、多场耦合综合作用的结果,是一典型的非线性问题。笔者应用人工神经网络技术,以Matlab软件为平台,通过现场调查和理论分析,建立了矿石含硫量、通风强度、环境温度3因素与硫化矿石自燃之间的预测模型;通过数据样本学习与部分现场监测数据相结合进行模拟,研究表明预测数据与实测结果基本吻合,误差控制在10%以内,取得了较好的效果。该研究为预防硫化矿石自燃提供一个新的思路和方法,具有一定的理论意义和应用价值。     

FACTORS AFFECTING SALINITY REDUCTION IN LAKE TARPON,PINELLAS COUNTY,FLORIDA1

ABSTRACT: Following an enclosure of a sink-hole connecting Lake Tarpon to the Gulf of Mexico, the chloride concentration of lake waters decreased. Water and chloride budgets for the lake in 1975 were prepared, and predictions using the model of Lerman and Brunskill (1971) were made as to the time required for the lake to achieve fresh water status. Model verification indicated good agreement with predictions in 1976; however, data on [C1^-] for 1977 and 1978 are not as supportive of the model used. The information concerning the Lake Tarpon watershed provided by this latter fact is discussed.     

GEOMORPHIC PARAMETERS PREDICT HYDROGRAPH CHARACTERISTICS IN THE SOUTHWEST1

ABSTRACT: Critical design characteristics of ephermal runoff such as hydrograph rise time, duration, mean peak discharge, volume, peak-volume ratio, and maximum flood were related to physical basin parameters such as area, shape, slope, drainage density, basin relief, stream length, and combinations of these in intermontane watersheds representative of the Mexican Highland section of the Basin and Range Province. Parameters used were restricted to those easily obtainable from maps or aerial photographs. A parameter expressing basin shape and size was developed which proved to be as accurate a predictor as others used in existing prediction equations tested and was simpler and faster to derive. Simple prediction equations derived for hydrograph characteristics were all significant except for volume at the 5% level; three were significant at the 1% level. Relationships determined are applicable in semi-arid basins of the Southwest up to 60 square miles (155 km²) in area.     

贝纳德效应和地震预报

本文根据海城地震前后重力变化的实测资料提出重力加速度随时间变化的一种简化表示。在重力发生扰动时对贝纳德问题的动力学方程进行求解。计算结果表明,在强地震的孕育期,贝纳德花样的周期变化在一定时间后将自动停止。这个结论的验证可以作为一种新的地震预测方法的理论基础。     

多级模糊决策在地震预报中的应用

本文把多级模糊决策方法应用于地震预报当中。文中详细论证和推导了多级模糊决策方法与模型。该方法使决策部门可以根据输入前兆的变化随时作出合理的对策。文中还给出了地震预报多级模糊决策数字解实例,并用回顾法讨论了1976年松潘7.2级地震的模糊预报决策。     

基于灰色-马尔科夫模型的我国能源消费结构优化调整

本文以我国1997—2010年的消费结构数据为基础,运用灰色-马尔科夫链模型对数据进行了分析和预测,然后通过MATLAB预测了我国未来十年的能源消费结构,结合我国政府承诺的非化石能源目标对模型预测结果做了进一步的修正,最后根据能源发展趋势给出了能源消费结构的优化策略。     

基于ARIMA的安全生产事故预测与实证研究

为了减少企业安全管理者在生产作业中由于不确定性导致错误决策所产生的风险,在企业历年安全生产事故数据基础上进行预测具有一定的现实意义。以某企业2008年至2011年的安全生产事故次数时序数据,采用EViews 5统计分析软件,基于ARIMA时间序列预测模型更加关注对事故发生是否平稳而相对于其他预测模型更关注于趋势研究的良好特点,建立安全生产事故ARIMA时序预测模型,并对2012年的安全生产事故发生次数进行预测,通过效果检验发现该模型预测结果基本上能够反映该企业安全生产事故发生的实际情况。通过ARIMA方法在某企业安全生产事故预测具体案例的实现,是对现有安全生产事故预测方法的补充和完善,可为企业安全管理和决策提供一定的指导。     

基于MDE和分形优化SVM的周期来压预测

为了解决周期来压的预测问题,首先对已知支架周期来压荷载曲线使用多重差异进化算法(MDE)进行拟合,将每重拟合形成的单一正弦曲线与上次差余曲线(Ei)再作差余曲线(Ei+1)。将这些Ei图通过分形几何的盒子法计算维度和相关系数(r)。将每条Ei的维度、r和支架相对距离(L)作为输入值,对应的Ei的周期Ti、缩放系数Si和纵移系数Di作为目标值,使用支持向量机(SVM)进行训练。通过对维度和r规律的研究得到拟设置支架处荷载各Ei的维度和r,带入训练后的SVM模拟得到Ei的Ti、Si和Di,进而得到Ei的表达式。将上述Ei求和即为所求拟设置支架处的周期来压荷载。实例分析说明,该种方法预测结果可以大体反映支架周期来压的基本形式和变化规律。     

回采工作面瓦斯涌出量遗传投影寻踪回归预测

为快速、准确预测回采工作面瓦斯涌出量,基于投影降维思想,建立一种遗传算法(GA)投影寻踪回归预测方法。选取煤层瓦斯原始含量、埋藏深度、煤层厚度、煤层倾角、工作面长度、推进速度、采出率、临近层瓦斯含量、临近层厚度、临近层层间距、岩层岩性、开采深度作为评价因子,对某矿15个学习样本进行训练,建立GA投影寻踪回归预测模型。利用该矿3个实测样本对模型进行检验,并与主成分分析和BP神经网络方法结果进行对比。研究表明:利用GA投影寻踪回归预测回采工作面瓦斯涌出量,平均误差为3.43%,最大误差为5.7%,精度优于其他2种方法。     

Development and Operational Testing of a Super‐Ensemble Artificial Intelligence Flood‐Forecast Model for a Pacific Northwest River

Coastal catchments in British Columbia, Canada, experience a complex mixture of rainfall‐ and snowmelt‐driven contributions to flood events. Few operational flood‐forecast models are available in the region. Here, we integrated a number of proven technologies in a novel way to produce a super‐ensemble forecast system for the Englishman River, a flood‐prone stream on Vancouver Island. This three‐day‐ahead modeling system utilizes up to 42 numerical weather prediction model outputs from the North American Ensemble Forecast System, combined with six artificial neural network‐based streamflow models representing various slightly different system conceptualizations, all of which were trained exclusively on historical high‐flow data. As such, the system combines relatively low model development times and costs with the generation of fully probabilistic forecasts reflecting uncertainty in the simulation of both atmospheric and terrestrial hydrologic dynamics. Results from operational testing by British Columbia's flood forecasting agency during the 2013‐2014 storm season suggest that the prediction system is operationally useful and robust.