Postal departments are actively taking part in e commerce, of which logistics is a key joint. Computerized routing and scheduling of postal transportation operations offers significant potential for cost decreases an...Postal departments are actively taking part in e commerce, of which logistics is a key joint. Computerized routing and scheduling of postal transportation operations offers significant potential for cost decreases and productivity gains. Routing and scheduling hierarchy model is initially built and demonstrated in detail on the basis of statements of specific requirements of postal logistics in this paper, and the realized software is proved to be practical and reliable.展开更多
Deep coal seam mining floor strata water bursting is a complicate nonlinear system, whose factors are coupling and influencing themselves. It built the analytic structure model for deep coal seam mining floor strata w...Deep coal seam mining floor strata water bursting is a complicate nonlinear system, whose factors are coupling and influencing themselves. It built the analytic structure model for deep coal seam mining floor strata water bursting, the judgment matrix was found by the expert scoring method, the contribution weights of the influenced factors were given out by the equation analytic process. The thirteen controlling factors and five main controlling factors were put award by analyzing weights, so the result was basically conform to the field practice. The expert scoring method and analytic process can convert the objective fact to the subjective cognition, so it is a method that can turn the qualitative into the quantitative. This can be relative objectively and precisely to study the question of many factors and grey box.展开更多
叙述了一个实验性多任务操作系统 XINUOS 的基本结构、设计及其在 PC 微机上的实现过程,讨论了它的分层观点和几个设计特点。XINUOS 用 C 语言和部分8088汇编语言写成,由于系统分层,模块量少,设备依赖性小,移植容易,可以作为计算机操作...叙述了一个实验性多任务操作系统 XINUOS 的基本结构、设计及其在 PC 微机上的实现过程,讨论了它的分层观点和几个设计特点。XINUOS 用 C 语言和部分8088汇编语言写成,由于系统分层,模块量少,设备依赖性小,移植容易,可以作为计算机操作系统的教学,实验和开发的极好实体和环境。展开更多
为准确预测电力市场中的短期电价,将最大信息系数(maximal information coefficient,MIC)相关性分析与改进多层级门控长短期记忆网络(multi-hierachy gated long shortterm memory,MHG-LSTM)相结合,提出一种新型短期电价预测方法。该方...为准确预测电力市场中的短期电价,将最大信息系数(maximal information coefficient,MIC)相关性分析与改进多层级门控长短期记忆网络(multi-hierachy gated long shortterm memory,MHG-LSTM)相结合,提出一种新型短期电价预测方法。该方法首先对备选序列与预测电价序列做MIC相关性分析,在此基础上筛选备选序列并经小波变换合成神经网络输入序列,有效增加了输入中与预测电价相关的信息密度;其次,对传统LSTM进行创新性改进,提出用两级遗忘门和输入门替换传统的一级门控机构的MHG-LSTM模型,提高了神经网络选择和提取高频电价序列特征的能力。在PJM市场日前电价数据集上对所提方法进行仿真实验,实验结果表明,该方法的预测误差仅为4.506%,相比已有预测方法有效提升了短期电价的预测精度,且具有很强的普适性,可应用于电力市场短期电价预测,为市场参与者和监管机构提供有力决策依据。展开更多
文摘Postal departments are actively taking part in e commerce, of which logistics is a key joint. Computerized routing and scheduling of postal transportation operations offers significant potential for cost decreases and productivity gains. Routing and scheduling hierarchy model is initially built and demonstrated in detail on the basis of statements of specific requirements of postal logistics in this paper, and the realized software is proved to be practical and reliable.
文摘Deep coal seam mining floor strata water bursting is a complicate nonlinear system, whose factors are coupling and influencing themselves. It built the analytic structure model for deep coal seam mining floor strata water bursting, the judgment matrix was found by the expert scoring method, the contribution weights of the influenced factors were given out by the equation analytic process. The thirteen controlling factors and five main controlling factors were put award by analyzing weights, so the result was basically conform to the field practice. The expert scoring method and analytic process can convert the objective fact to the subjective cognition, so it is a method that can turn the qualitative into the quantitative. This can be relative objectively and precisely to study the question of many factors and grey box.
文摘为准确预测电力市场中的短期电价,将最大信息系数(maximal information coefficient,MIC)相关性分析与改进多层级门控长短期记忆网络(multi-hierachy gated long shortterm memory,MHG-LSTM)相结合,提出一种新型短期电价预测方法。该方法首先对备选序列与预测电价序列做MIC相关性分析,在此基础上筛选备选序列并经小波变换合成神经网络输入序列,有效增加了输入中与预测电价相关的信息密度;其次,对传统LSTM进行创新性改进,提出用两级遗忘门和输入门替换传统的一级门控机构的MHG-LSTM模型,提高了神经网络选择和提取高频电价序列特征的能力。在PJM市场日前电价数据集上对所提方法进行仿真实验,实验结果表明,该方法的预测误差仅为4.506%,相比已有预测方法有效提升了短期电价的预测精度,且具有很强的普适性,可应用于电力市场短期电价预测,为市场参与者和监管机构提供有力决策依据。