Climate is subject to fluctuations in the majority of the world, mainly caused by rainfall as well as temperature variations. Climate fluctuations in Kenya have resulted in the spread of desert-like conditions in the ...Climate is subject to fluctuations in the majority of the world, mainly caused by rainfall as well as temperature variations. Climate fluctuations in Kenya have resulted in the spread of desert-like conditions in the ASALs region, such as Marigat in Baringo County. As a county, Baringo experiences great variations in climate annually, as well as uncertainty in expected rains, thereby negatively impacting the production of crops such as sorghum. This study applied the rainfall anomaly index (RAI), standardised precipitation evapotranspiration index (SPEI), standard precipitation index (SPI), and Mann-Kendall (MK) statistical test for trends on historical climatic data in analysing both temperature and precipitation data over the period 1990 to 2022 to determine their trend, patterns and how they affect the production of sorghum crops. The machine learning method (R studio) with inputs was used to calculate the SPI, SPEI, RAI and MK trend test. The rainfall varied from below average to above average during the study period with no clear pattern in the RAI, SPEI and SPI values. The years 2020 and 2000 stood out as they had higher and lower rainfall than usual, respectively. The Marigat area generally experienced more rainfall during the high/long rainfall season (AMJJ). The MK trend test on average monthly rainfall, SOND, AMJJ, and annual precipitation confirmed a positive trend in precipitation. However, the short rainy season (SOND) was found to be the most variable period for rainfall, and there was a slight increase in daily average temperatures during this season.展开更多
大坝的位移趋势对其安全运行十分重要,而在非常规条件下快速得到大坝的预测模型的研究则相对较少。为使管理人员可以高效地掌握大坝位移的变化性态,基于多核支持向量机(Multiple-kernels support vector machine,MK-SVM)算法构建了一种...大坝的位移趋势对其安全运行十分重要,而在非常规条件下快速得到大坝的预测模型的研究则相对较少。为使管理人员可以高效地掌握大坝位移的变化性态,基于多核支持向量机(Multiple-kernels support vector machine,MK-SVM)算法构建了一种快速实现大坝位移预测的模型,并讨论该模型参数的范围,利用相关评价指标计算度量模型的精度。由工程案例可知,该模型预测大坝变形的变化趋势与实测变形基本一致,模型预测精度可以满足工程运行需要。结论可为大坝管理人员提供有益借鉴。展开更多
文摘玻璃生产线退火窑辊道系统轴承运行状态显著影响玻璃品质和生产效率,实时监测各轴承运行状态对确保退火窑系统的平稳运行具有重要意义,提出结合Inception模块和长短期神经网络(Long Short-term Memory,LSTM)的迁移诊断方法,对退火窑辊道系统中的辊道轴承和通轴轴承运行状态进行监测、诊断。首先,使用集合经验模态分解(Ensemble Empirical Mode Decomposition,EEMD)对轴承信号进行分解和重构降噪,并利用直方均衡化增强重构信号小波时频图的聚集性。然后,针对样本充足的辊道轴承,建立Inception-LSTM网络,提取多尺度特征并学习其中的时间依赖关系,实现状态诊断。再次,针对转速不同且样本量少的通轴轴承,以辊道轴承信号为源域,以通轴轴承信号为目标域,以Inception-LSTM网络为基础,使用多核最大均值差异(Multi-kernel Maximum Mean Discrepancies,MKMMD)减小分布差异,实现故障样本不平衡条件下的跨转速域不变特征提取和迁移诊断。最后,利用实验数据和实测数据验证本算法的有效性,结果表明,该方法能有效诊断出退火窑辊道系统轴承故障,且具有较高的准确率。
文摘Climate is subject to fluctuations in the majority of the world, mainly caused by rainfall as well as temperature variations. Climate fluctuations in Kenya have resulted in the spread of desert-like conditions in the ASALs region, such as Marigat in Baringo County. As a county, Baringo experiences great variations in climate annually, as well as uncertainty in expected rains, thereby negatively impacting the production of crops such as sorghum. This study applied the rainfall anomaly index (RAI), standardised precipitation evapotranspiration index (SPEI), standard precipitation index (SPI), and Mann-Kendall (MK) statistical test for trends on historical climatic data in analysing both temperature and precipitation data over the period 1990 to 2022 to determine their trend, patterns and how they affect the production of sorghum crops. The machine learning method (R studio) with inputs was used to calculate the SPI, SPEI, RAI and MK trend test. The rainfall varied from below average to above average during the study period with no clear pattern in the RAI, SPEI and SPI values. The years 2020 and 2000 stood out as they had higher and lower rainfall than usual, respectively. The Marigat area generally experienced more rainfall during the high/long rainfall season (AMJJ). The MK trend test on average monthly rainfall, SOND, AMJJ, and annual precipitation confirmed a positive trend in precipitation. However, the short rainy season (SOND) was found to be the most variable period for rainfall, and there was a slight increase in daily average temperatures during this season.
文摘大坝的位移趋势对其安全运行十分重要,而在非常规条件下快速得到大坝的预测模型的研究则相对较少。为使管理人员可以高效地掌握大坝位移的变化性态,基于多核支持向量机(Multiple-kernels support vector machine,MK-SVM)算法构建了一种快速实现大坝位移预测的模型,并讨论该模型参数的范围,利用相关评价指标计算度量模型的精度。由工程案例可知,该模型预测大坝变形的变化趋势与实测变形基本一致,模型预测精度可以满足工程运行需要。结论可为大坝管理人员提供有益借鉴。
