随着面向高比例可再生能源新型电力系统的转型,系统运行特性日趋复杂。暂态功角稳定(transientangle stability,TAS)与暂态电压稳定(transient voltage stability,TVS)问题相互耦合且频发,为系统安全稳定评估带来严峻挑战。研究首先采...随着面向高比例可再生能源新型电力系统的转型,系统运行特性日趋复杂。暂态功角稳定(transientangle stability,TAS)与暂态电压稳定(transient voltage stability,TVS)问题相互耦合且频发,为系统安全稳定评估带来严峻挑战。研究首先采用变步长二分法通过调用PSASP从时间维度上构建了暂态电压与暂态功角的稳定边界。研究了不同故障位置、感应电动机占比、负荷率对稳定边界的影响并依托边界确定主导失稳模式。其次提出一种基于注意力机制与一维卷积神经网络融合的电力系统功角稳定及电压稳定裕度评估的新方法。该方法直接面向测量数据,将节点稳态与暂态运行的电压幅值、有功功率、无功功率数据作为输入特征,节省了数据处理时间。通过一维卷积神经网络构建输入特征与极限切除时间的映射,利用注意力机制进一步提高了模型预测效果。通过新英格兰IEEE39节点系统进行分析验证,结果表明该方法可以实现暂态安全裕度的快速评估且具有较高的预测精度。展开更多
暂态电压稳定(transient voltage stability,TVS)与暂态功角稳定(transient angle stability,TAS)是电力系统安全运行的重要基础。随着新型电力系统的建设,电压与功角问题紧密耦合且频发,亟需高精度的一体化超前评估为紧急控制夯实基础...暂态电压稳定(transient voltage stability,TVS)与暂态功角稳定(transient angle stability,TAS)是电力系统安全运行的重要基础。随着新型电力系统的建设,电压与功角问题紧密耦合且频发,亟需高精度的一体化超前评估为紧急控制夯实基础。首先根据调研整合了表征功角稳定与电压稳定的综合特征,并根据极限梯度提升(extreme gradient boosting,XGBoost)衡量特征贡献度,根据贡献度生成含差分特征的特征集作为评估模型的输入。其次,提出了融合挤压激励模块的多尺度卷积门控循环单元模型(a multi-scale convolution gated recurrent unit integrated with squeeze excitation,SE-CGRU)。该模型通过挤压激励(squeeze&excitation,SE)模块实现特征通道权重的自适应调整,并利用多尺度卷积融合细节特征与宏观特征,实现了暂态功角与暂态电压的高精度一体化评估。在线评估时无需已知故障切除时间即可给出预测结果并输出系统当前状态下的安全裕度。通过引入带时间约束的损失函数与动态权重训练的方式,在保持现有精度的基础上大大缩减了响应时间,实现超前评估。采用多判据融合策略进一步减少了漏判与误判,提高了模型评估的可靠性。以新英格兰10机39节点系统和国内某区域省级互联系统为例验证分析,结果表明所提方法可实现高精度的功角和电压稳定一体化超前评估。展开更多
A physical modeling system of long slim tube was established. Several pressure measuring and sampling points were laid out at different positions along the tube. Through real-time measurements of pressures and chemica...A physical modeling system of long slim tube was established. Several pressure measuring and sampling points were laid out at different positions along the tube. Through real-time measurements of pressures and chemical concentrations at different points, the mass transfer and chemical concentration of ASP flooding in porous media are studied. The concentration of chemicals declines gradually during the fluid flow from the inlet to the outlet of the model. The concentration increases in the front edge of the slug faster than the concentration decreases in the rear edge of the slug. The concentration variation of the chemicals is an asymmetrical and offset process. The order of motion velocities of the chemicals from fast to slow is polymer, alkali and surfactant. The motion lag and comprehensive diffusion are strong in the vicinity of the inlet, the motion velocities of the chemicals are high, the difference of flow velocities among the three chemicals is significant and the chromatographic separation of the chemicals is obvious. In the area near the outlet, the comprehensive diffusion and motion lag become weak, the concentrations of the chemicals decrease, the motion velocities of the chemicals are slow, the difference among the motion velocities of the chemicals becomes small, the chromatographic separation is not obvious, the adsorption and retention of chemicals gradually increase as the chemical slug moves further along the tube, the adsorption and retention of polymer is the most serious.展开更多
文摘随着面向高比例可再生能源新型电力系统的转型,系统运行特性日趋复杂。暂态功角稳定(transientangle stability,TAS)与暂态电压稳定(transient voltage stability,TVS)问题相互耦合且频发,为系统安全稳定评估带来严峻挑战。研究首先采用变步长二分法通过调用PSASP从时间维度上构建了暂态电压与暂态功角的稳定边界。研究了不同故障位置、感应电动机占比、负荷率对稳定边界的影响并依托边界确定主导失稳模式。其次提出一种基于注意力机制与一维卷积神经网络融合的电力系统功角稳定及电压稳定裕度评估的新方法。该方法直接面向测量数据,将节点稳态与暂态运行的电压幅值、有功功率、无功功率数据作为输入特征,节省了数据处理时间。通过一维卷积神经网络构建输入特征与极限切除时间的映射,利用注意力机制进一步提高了模型预测效果。通过新英格兰IEEE39节点系统进行分析验证,结果表明该方法可以实现暂态安全裕度的快速评估且具有较高的预测精度。
基金Supported by the National Basic Research Program of China("973"Project)(Grant No.2005CB221300)
文摘A physical modeling system of long slim tube was established. Several pressure measuring and sampling points were laid out at different positions along the tube. Through real-time measurements of pressures and chemical concentrations at different points, the mass transfer and chemical concentration of ASP flooding in porous media are studied. The concentration of chemicals declines gradually during the fluid flow from the inlet to the outlet of the model. The concentration increases in the front edge of the slug faster than the concentration decreases in the rear edge of the slug. The concentration variation of the chemicals is an asymmetrical and offset process. The order of motion velocities of the chemicals from fast to slow is polymer, alkali and surfactant. The motion lag and comprehensive diffusion are strong in the vicinity of the inlet, the motion velocities of the chemicals are high, the difference of flow velocities among the three chemicals is significant and the chromatographic separation of the chemicals is obvious. In the area near the outlet, the comprehensive diffusion and motion lag become weak, the concentrations of the chemicals decrease, the motion velocities of the chemicals are slow, the difference among the motion velocities of the chemicals becomes small, the chromatographic separation is not obvious, the adsorption and retention of chemicals gradually increase as the chemical slug moves further along the tube, the adsorption and retention of polymer is the most serious.