A combined model based on principal components analysis (PCA) and generalized regression neural network (GRNN) was adopted to forecast electricity price in day-ahead electricity market. PCA was applied to mine the mai...A combined model based on principal components analysis (PCA) and generalized regression neural network (GRNN) was adopted to forecast electricity price in day-ahead electricity market. PCA was applied to mine the main influence on day-ahead price, avoiding the strong correlation between the input factors that might influence electricity price, such as the load of the forecasting hour, other history loads and prices, weather and temperature; then GRNN was employed to forecast electricity price according to the main information extracted by PCA. To prove the efficiency of the combined model, a case from PJM (Pennsylvania-New Jersey-Maryland) day-ahead electricity market was evaluated. Compared to back-propagation (BP) neural network and standard GRNN, the combined method reduces the mean absolute percentage error about 3%.展开更多
Cliff deformation behavior after conservation is of great significance for evaluating the conservation effect and discovering the dynamical law of soil. Modeling on deformation behavior is beneficial to the quantitati...Cliff deformation behavior after conservation is of great significance for evaluating the conservation effect and discovering the dynamical law of soil. Modeling on deformation behavior is beneficial to the quantitative evaluation of interactions between soil mass and structures as well as the forecast. Based on cliff conservation engineering of Jiaohe Ruins （the largest raw soil heritage site in the world）, data of horizontal deformation of the upper cliff were obtained by using Nanrui-made NDW-50 displacement device （precision： 0.01 mm, frequency： 15 min^-l）. Regression analysis indicates that deformation behavior models include exponential growth, linear growth and parabolic growth types, while daily deformation presents more intense periodicity （24 h）. The deformation is less than 1.5 mm during monitoring period, which has no impact on the stability of cliff. Deformation behavior provides the mutual duress and interaction between soil and engineering intervention. In addition, deformation mode attaches tensely to the damage pattern of the cliff. The conclusions are of importance to the stability evaluation of the carrier along Silk Road.展开更多
目的分析老年穿支动脉粥样硬化病患者血清微小RNA(micorRNA,miRNA)预测早期神经功能恶化的回归分析。方法选择2020年2月至2023年2月湖北医药学院附属随州市中心医院神经内科收治的老年穿支动脉粥样硬化病患者134例,依据早期神经功能恶...目的分析老年穿支动脉粥样硬化病患者血清微小RNA(micorRNA,miRNA)预测早期神经功能恶化的回归分析。方法选择2020年2月至2023年2月湖北医药学院附属随州市中心医院神经内科收治的老年穿支动脉粥样硬化病患者134例,依据早期神经功能恶化情况分为恶化组28例和未恶化组106例。入院时测定患者血清miR-130a、miR-210、miR-141-3p、miR-29a-3p水平,入院时及入院后7 d采用美国国立卫生研究院卒中量表(National Institutes of Health Stroke Scale,NIHSS)评分评估早期神经功能恶化情况。采用二元logistic回归分析法构建miR-130a、miR-210、miR-141-3p、miR-29a-3p预测老年穿支动脉粥样硬化病患者早期神经功能恶化模型,ROC曲线分析血清miR-130a、miR-210、miR-141-3p、miR-29a-3p水平对老年穿支动脉粥样硬化病患者早期神经功能恶化的预测价值。结果恶化组血清miR-130a、miR-210水平明显高于未恶化组,miR-141-3p、miR-29a-3p水平明显低于未恶化组,差异有统计学意义(P<0.01)。Logistic回归分析显示,血清miR-130a、miR-210、miR-141-3p、miR-29a-3p水平为老年穿支动脉粥样硬化病患者早期神经功能恶化的独立预测指标(P<0.05,P<0.01)。ROC曲线分析显示,血清miR-130a、miR-210、miR-141-3p、miR-29a-3p联合预测老年穿支动脉粥样硬化病患者早期神经功能恶化的曲线下面积为0.977(95%CI:0.936~0.995),敏感性为96.43%,特异性为90.57%,联合预测的效能明显优于各指标单独预测(P<0.01)。结论老年穿支动脉粥样硬化病患者血清miR-130a、miR-210、miR-141-3p、miR-29a-3p对预测早期神经功能恶化具有一定的价值,且四者联合检测可提高其预测效能。展开更多
为了解决高比例分布式电源(distributed generation,DG)大规模并网后实时量测数目缺失、传统预测辅助状态估计方法(forecasting-aided state estimation,FASE)估计精度有限等问题,提出了基于改进Crossformer伪量测构建的主动配电网FASE...为了解决高比例分布式电源(distributed generation,DG)大规模并网后实时量测数目缺失、传统预测辅助状态估计方法(forecasting-aided state estimation,FASE)估计精度有限等问题,提出了基于改进Crossformer伪量测构建的主动配电网FASE方法。首先,基于最大信息系数法(maximal information coefficient,MIC)筛选出高相关性的输入特征,提高预测模型的精度;然后,通过全变差正则化技术(total variation regularized,TV)优化鲁棒主成分分析法(robust principal component analysis,RPCA),构建TRPCA层,并将其嵌入到Crossformer中,以填补Crossformer无法有效处理非高斯噪声的空白;最后,利用改进的预测模型进行超短期负荷预测,经潮流计算得到节点伪量测,在量测不足情况下补全缺失数据,并结合扩展卡尔曼滤波器(extended Kalman filter,EKF)进行状态估计。在IEEE 33节点和IEEE 118节点标准配电网上进行仿真测试,结果表明所提方法在估计精度和鲁棒性等方面具有一定优势,可为主动配电网FASE提供参考。展开更多
基金Project(70671039) supported by the National Natural Science Foundation of China
文摘A combined model based on principal components analysis (PCA) and generalized regression neural network (GRNN) was adopted to forecast electricity price in day-ahead electricity market. PCA was applied to mine the main influence on day-ahead price, avoiding the strong correlation between the input factors that might influence electricity price, such as the load of the forecasting hour, other history loads and prices, weather and temperature; then GRNN was employed to forecast electricity price according to the main information extracted by PCA. To prove the efficiency of the combined model, a case from PJM (Pennsylvania-New Jersey-Maryland) day-ahead electricity market was evaluated. Compared to back-propagation (BP) neural network and standard GRNN, the combined method reduces the mean absolute percentage error about 3%.
基金Project(2010BAK67B16) supported by the National Science and Technology Pillar Program during the 11th Five-Year Plan Period of China
文摘Cliff deformation behavior after conservation is of great significance for evaluating the conservation effect and discovering the dynamical law of soil. Modeling on deformation behavior is beneficial to the quantitative evaluation of interactions between soil mass and structures as well as the forecast. Based on cliff conservation engineering of Jiaohe Ruins （the largest raw soil heritage site in the world）, data of horizontal deformation of the upper cliff were obtained by using Nanrui-made NDW-50 displacement device （precision： 0.01 mm, frequency： 15 min^-l）. Regression analysis indicates that deformation behavior models include exponential growth, linear growth and parabolic growth types, while daily deformation presents more intense periodicity （24 h）. The deformation is less than 1.5 mm during monitoring period, which has no impact on the stability of cliff. Deformation behavior provides the mutual duress and interaction between soil and engineering intervention. In addition, deformation mode attaches tensely to the damage pattern of the cliff. The conclusions are of importance to the stability evaluation of the carrier along Silk Road.
文摘目的分析老年穿支动脉粥样硬化病患者血清微小RNA(micorRNA,miRNA)预测早期神经功能恶化的回归分析。方法选择2020年2月至2023年2月湖北医药学院附属随州市中心医院神经内科收治的老年穿支动脉粥样硬化病患者134例,依据早期神经功能恶化情况分为恶化组28例和未恶化组106例。入院时测定患者血清miR-130a、miR-210、miR-141-3p、miR-29a-3p水平,入院时及入院后7 d采用美国国立卫生研究院卒中量表(National Institutes of Health Stroke Scale,NIHSS)评分评估早期神经功能恶化情况。采用二元logistic回归分析法构建miR-130a、miR-210、miR-141-3p、miR-29a-3p预测老年穿支动脉粥样硬化病患者早期神经功能恶化模型,ROC曲线分析血清miR-130a、miR-210、miR-141-3p、miR-29a-3p水平对老年穿支动脉粥样硬化病患者早期神经功能恶化的预测价值。结果恶化组血清miR-130a、miR-210水平明显高于未恶化组,miR-141-3p、miR-29a-3p水平明显低于未恶化组,差异有统计学意义(P<0.01)。Logistic回归分析显示,血清miR-130a、miR-210、miR-141-3p、miR-29a-3p水平为老年穿支动脉粥样硬化病患者早期神经功能恶化的独立预测指标(P<0.05,P<0.01)。ROC曲线分析显示,血清miR-130a、miR-210、miR-141-3p、miR-29a-3p联合预测老年穿支动脉粥样硬化病患者早期神经功能恶化的曲线下面积为0.977(95%CI:0.936~0.995),敏感性为96.43%,特异性为90.57%,联合预测的效能明显优于各指标单独预测(P<0.01)。结论老年穿支动脉粥样硬化病患者血清miR-130a、miR-210、miR-141-3p、miR-29a-3p对预测早期神经功能恶化具有一定的价值,且四者联合检测可提高其预测效能。
文摘为了解决高比例分布式电源(distributed generation,DG)大规模并网后实时量测数目缺失、传统预测辅助状态估计方法(forecasting-aided state estimation,FASE)估计精度有限等问题,提出了基于改进Crossformer伪量测构建的主动配电网FASE方法。首先,基于最大信息系数法(maximal information coefficient,MIC)筛选出高相关性的输入特征,提高预测模型的精度;然后,通过全变差正则化技术(total variation regularized,TV)优化鲁棒主成分分析法(robust principal component analysis,RPCA),构建TRPCA层,并将其嵌入到Crossformer中,以填补Crossformer无法有效处理非高斯噪声的空白;最后,利用改进的预测模型进行超短期负荷预测,经潮流计算得到节点伪量测,在量测不足情况下补全缺失数据,并结合扩展卡尔曼滤波器(extended Kalman filter,EKF)进行状态估计。在IEEE 33节点和IEEE 118节点标准配电网上进行仿真测试,结果表明所提方法在估计精度和鲁棒性等方面具有一定优势,可为主动配电网FASE提供参考。
