Polyurethane polymer grouting materials were studied with conventional triaxial tests via the particle flow code in two dimensions(PFC^(2D)) method, and the simulation results agreed with the experimental data. Th...Polyurethane polymer grouting materials were studied with conventional triaxial tests via the particle flow code in two dimensions(PFC^(2D)) method, and the simulation results agreed with the experimental data. The particle flow code method can simulate the mechanical properties of the polymer. The triaxial cyclic loading tests of the polymer material under different confining pressures were carried out via PFC^(2D) to analyze its mechanical performance. The PFC^(2D) simulation results show that the value of the elastic modulus of the polymer decreases slowly at first and fluctuated within a narrow range near the value of the peak strength; the cumulative plastic strain increases slowly at first and then increases rapidly; the peak strength and elastic modulus of polymer increase with the confining pressure; the PFC^(2D) method can be used to quantitatively evaluate the damage behavior of the polymer material and estimate the fatigue life of the materials under fatigue load based on the number and the location of micro-cracks. Thus, the PFC^(2D) method is an effective tool to study polymers.展开更多
建设智能教育平台是推动教育智能化的一个重要过程,但智能教育平台依赖的人工智能模型在训练过程中会消耗大量电力,因此,开展短期电力负荷预测对建设智能教育平台具有重要意义.针对在考虑多个属性开展短期电力负荷预测时,由于部分属性...建设智能教育平台是推动教育智能化的一个重要过程,但智能教育平台依赖的人工智能模型在训练过程中会消耗大量电力,因此,开展短期电力负荷预测对建设智能教育平台具有重要意义.针对在考虑多个属性开展短期电力负荷预测时,由于部分属性与电力负荷数据的相关性不强并且Transformer无法捕捉电力负荷数据的时间相关性,而导致电力负荷预测不够准确的问题,基于SR(Székely and Rizzo)距离相关系数、融合时间定位编码和Transformer,提出了一种短期电力负荷预测模型SF-Transformer.SF-Transformer通过SR距离相关系数对影响电力负荷数据的属性进行筛选,选择与电力负荷数据之间SR距离相关系数较大的属性.SF-Transformer采用一种全局时间编码与局部位置编码相结合的融合时间定位编码,有助于模型全面获取电力负荷数据的时间定位信息.在数据集上开展了实验,实验结果表明SF-Transformer与其他模型相比,在两种时长上进行电力负荷预测具有更低的均方根误差和平均绝对误差.展开更多
基金the National Key R&D Program of China(No.2017YFC0405002)
文摘Polyurethane polymer grouting materials were studied with conventional triaxial tests via the particle flow code in two dimensions(PFC^(2D)) method, and the simulation results agreed with the experimental data. The particle flow code method can simulate the mechanical properties of the polymer. The triaxial cyclic loading tests of the polymer material under different confining pressures were carried out via PFC^(2D) to analyze its mechanical performance. The PFC^(2D) simulation results show that the value of the elastic modulus of the polymer decreases slowly at first and fluctuated within a narrow range near the value of the peak strength; the cumulative plastic strain increases slowly at first and then increases rapidly; the peak strength and elastic modulus of polymer increase with the confining pressure; the PFC^(2D) method can be used to quantitatively evaluate the damage behavior of the polymer material and estimate the fatigue life of the materials under fatigue load based on the number and the location of micro-cracks. Thus, the PFC^(2D) method is an effective tool to study polymers.
文摘建设智能教育平台是推动教育智能化的一个重要过程,但智能教育平台依赖的人工智能模型在训练过程中会消耗大量电力,因此,开展短期电力负荷预测对建设智能教育平台具有重要意义.针对在考虑多个属性开展短期电力负荷预测时,由于部分属性与电力负荷数据的相关性不强并且Transformer无法捕捉电力负荷数据的时间相关性,而导致电力负荷预测不够准确的问题,基于SR(Székely and Rizzo)距离相关系数、融合时间定位编码和Transformer,提出了一种短期电力负荷预测模型SF-Transformer.SF-Transformer通过SR距离相关系数对影响电力负荷数据的属性进行筛选,选择与电力负荷数据之间SR距离相关系数较大的属性.SF-Transformer采用一种全局时间编码与局部位置编码相结合的融合时间定位编码,有助于模型全面获取电力负荷数据的时间定位信息.在数据集上开展了实验,实验结果表明SF-Transformer与其他模型相比,在两种时长上进行电力负荷预测具有更低的均方根误差和平均绝对误差.