With the development of distributed photovoltaic energy in mining areas,the increasing proportion of new energy integration will gradually drive the structural transformation of coal mine parks into industrial active ...With the development of distributed photovoltaic energy in mining areas,the increasing proportion of new energy integration will gradually drive the structural transformation of coal mine parks into industrial active distribution networks.This paper takes a coal mine in Shaanxi Province as an empirical research subject,employing a multi-spatial hierarchical analysis method to systematically analyze the electricity usage characteristics across different spatial levels.It innovatively proposes a power line loss correction method for multi-level electrical equipment,effectively balancing the metering deviations of line losses between levels.Based on two types of carbon emission factors,it achieves precise evaluation of indirect carbon emissions from electricity usage in different levels of equipment or energy units.When calculating indirect carbon emissions from electricity using static carbon emission factors,the measured values show a positive correlation with peak electricity load.As the proportion of renewable photovoltaic power integration in industrial active distribution networks increases,the system’s indirect carbon emission factors exhibit a periodic decreasing trend.After adopting dynamic indirect carbon emission factors for accounting,the indirect carbon emissions from electricity in the first consumption level can be reduced by 31%-55%,while the overall plant-wide indirect carbon emissions decrease by 29068.94 kgCO_(2)/d,a reduction of approximately 40%.For the second consumption level in industrial production areas,indirect carbon emissions from electricity can be reduced by 5%-83%,with overall industrial park-wide indirect carbon emissions decreasing by 7376.48 kgCO_(2)/d,a reduction of about 31%.It is recommended to use advanced equipment and energy-saving technologies to suppress technical line losses in coal mine power supply systems and improve energy efficiency.Promoting the coordinated application of renewable energy and energy storage technologies can expand the scale of stable and reliable renewable energy supply,effectively reducing indirect carbon emissions from electricity.展开更多
针对腹部部分注释数据集的嗜铬细胞瘤图像分割缺乏不同器官间的特征学习,导致分割中难以准确区分肿瘤及周边器官边缘的问题,提出一种基于多尺度与空间频率特征的嗜铬细胞瘤图像分割网络(MF-Net)。首先,构建多尺度空间频率通道注意力模块...针对腹部部分注释数据集的嗜铬细胞瘤图像分割缺乏不同器官间的特征学习,导致分割中难以准确区分肿瘤及周边器官边缘的问题,提出一种基于多尺度与空间频率特征的嗜铬细胞瘤图像分割网络(MF-Net)。首先,构建多尺度空间频率通道注意力模块(MSFCA)对图像频域信息和相邻编码器的多尺度特征图进行加权融合,以强化器官间纹理和边界特征的捕捉,从而突出肿瘤区域的特征表示能力;其次,引入上采样多尺度特征融合模块(UMFF)通过结合上采样得到的不同尺度特征图,增强模型对图像中不同大小对象的识别能力;最后,利用自适应目标损失函数(AOb)对有注释腹部器官标签进行损失计算,并根据注释器官类别调整损失权重大小,从而优化分割网络的学习过程。实验结果表明,在腹部器官和嗜铬细胞瘤数据集上,MF-Net的分割准确率相较于单独训练的nnU-Net(no new U-Net)分别提升了3.33和3.18个百分点,而Dice系数(Dice)和归一化表面距离(NSD)分别为89.07%和92.85%;在域外数据集上,MF-Net的Dice和NSD分别为84.66%和90.55%。此外,可视化结果表明,MF-Net能更好地处理嗜铬细胞瘤图像中的复杂背景和模糊边界,为嗜铬细胞瘤的精确诊断和治疗提供了更好的技术支持。展开更多
基金supported by CHN Energy Investment Group(GJNY-23-138)。
文摘With the development of distributed photovoltaic energy in mining areas,the increasing proportion of new energy integration will gradually drive the structural transformation of coal mine parks into industrial active distribution networks.This paper takes a coal mine in Shaanxi Province as an empirical research subject,employing a multi-spatial hierarchical analysis method to systematically analyze the electricity usage characteristics across different spatial levels.It innovatively proposes a power line loss correction method for multi-level electrical equipment,effectively balancing the metering deviations of line losses between levels.Based on two types of carbon emission factors,it achieves precise evaluation of indirect carbon emissions from electricity usage in different levels of equipment or energy units.When calculating indirect carbon emissions from electricity using static carbon emission factors,the measured values show a positive correlation with peak electricity load.As the proportion of renewable photovoltaic power integration in industrial active distribution networks increases,the system’s indirect carbon emission factors exhibit a periodic decreasing trend.After adopting dynamic indirect carbon emission factors for accounting,the indirect carbon emissions from electricity in the first consumption level can be reduced by 31%-55%,while the overall plant-wide indirect carbon emissions decrease by 29068.94 kgCO_(2)/d,a reduction of approximately 40%.For the second consumption level in industrial production areas,indirect carbon emissions from electricity can be reduced by 5%-83%,with overall industrial park-wide indirect carbon emissions decreasing by 7376.48 kgCO_(2)/d,a reduction of about 31%.It is recommended to use advanced equipment and energy-saving technologies to suppress technical line losses in coal mine power supply systems and improve energy efficiency.Promoting the coordinated application of renewable energy and energy storage technologies can expand the scale of stable and reliable renewable energy supply,effectively reducing indirect carbon emissions from electricity.
文摘针对腹部部分注释数据集的嗜铬细胞瘤图像分割缺乏不同器官间的特征学习,导致分割中难以准确区分肿瘤及周边器官边缘的问题,提出一种基于多尺度与空间频率特征的嗜铬细胞瘤图像分割网络(MF-Net)。首先,构建多尺度空间频率通道注意力模块(MSFCA)对图像频域信息和相邻编码器的多尺度特征图进行加权融合,以强化器官间纹理和边界特征的捕捉,从而突出肿瘤区域的特征表示能力;其次,引入上采样多尺度特征融合模块(UMFF)通过结合上采样得到的不同尺度特征图,增强模型对图像中不同大小对象的识别能力;最后,利用自适应目标损失函数(AOb)对有注释腹部器官标签进行损失计算,并根据注释器官类别调整损失权重大小,从而优化分割网络的学习过程。实验结果表明,在腹部器官和嗜铬细胞瘤数据集上,MF-Net的分割准确率相较于单独训练的nnU-Net(no new U-Net)分别提升了3.33和3.18个百分点,而Dice系数(Dice)和归一化表面距离(NSD)分别为89.07%和92.85%;在域外数据集上,MF-Net的Dice和NSD分别为84.66%和90.55%。此外,可视化结果表明,MF-Net能更好地处理嗜铬细胞瘤图像中的复杂背景和模糊边界,为嗜铬细胞瘤的精确诊断和治疗提供了更好的技术支持。
