Intensity modulated radiation therapy (IMRT) is a highly accurate technique that is usually implemented in either dynamic or step-and-shoot fashion with many segments each having low monitor units (MUs). The present s...Intensity modulated radiation therapy (IMRT) is a highly accurate technique that is usually implemented in either dynamic or step-and-shoot fashion with many segments each having low monitor units (MUs). The present study evaluated the effects of beam startup characteristics on the dose delivery accuracy for each segment at low MUs for step-and-shoot IMRT with an Elekta Precise accelerator at the highest dose rates. We used a two-dimensional semi-conductor detector for the dose measurements. The field size of each segment was assumed to be 20 ×20 cm2 and each segment was set to deliver 1 - 10 MUs. Our results show a variation in dose delivery accuracy between segments for the same IMRT beam, which can be attributed to the beam startup characteristics. This variability is attributed to the changes in the transient changes in the temperatures of the electron gun filament and the magnetron. That is, the transient increase in the temperature of the filament leads to increasing doses with time and that of the magnetron leads to decreasing doses with time during the first few MUs.展开更多
A big step forward to improve power system monitoring and performance, continued load growth without a corresponding increase in transmission resources has resulted in reduced operational margins for many power system...A big step forward to improve power system monitoring and performance, continued load growth without a corresponding increase in transmission resources has resulted in reduced operational margins for many power systems worldwide and has led to operation of power systems closer to their stability limits and to power exchange in new patterns. These issues, as well as the on-going worldwide trend towards deregulation of the entire industry on the one hand and the increased need for accurate and better network monitoring on the other hand, force power utilities exposed to this pressure to demand new solutions for wide area monitoring, protection and control. Wide-area monitoring, protection, and control require communicating the specific-node information to a remote station but all information should be time synchronized so that to neutralize the time difference between information. It gives a complete simultaneous snap shot of the power system. The conventional system is not able to satisfy the time-synchronized requirement of power system. Phasor Measurement Unit (PMU) is enabler of time-synchronized measurement, it communicate the synchronized local information to remote station.展开更多
Deep reinforcement learning(DRL) has demonstrated significant potential in industrial manufacturing domains such as workshop scheduling and energy system management.However, due to the model's inherent uncertainty...Deep reinforcement learning(DRL) has demonstrated significant potential in industrial manufacturing domains such as workshop scheduling and energy system management.However, due to the model's inherent uncertainty, rigorous validation is requisite for its application in real-world tasks. Specific tests may reveal inadequacies in the performance of pre-trained DRL models, while the “black-box” nature of DRL poses a challenge for testing model behavior. We propose a novel performance improvement framework based on probabilistic automata,which aims to proactively identify and correct critical vulnerabilities of DRL systems, so that the performance of DRL models in real tasks can be improved with minimal model modifications.First, a probabilistic automaton is constructed from the historical trajectory of the DRL system by abstracting the state to generate probabilistic decision-making units(PDMUs), and a reverse breadth-first search(BFS) method is used to identify the key PDMU-action pairs that have the greatest impact on adverse outcomes. This process relies only on the state-action sequence and final result of each trajectory. Then, under the key PDMU, we search for the new action that has the greatest impact on favorable results. Finally, the key PDMU, undesirable action and new action are encapsulated as monitors to guide the DRL system to obtain more favorable results through real-time monitoring and correction mechanisms. Evaluations in two standard reinforcement learning environments and three actual job scheduling scenarios confirmed the effectiveness of the method, providing certain guarantees for the deployment of DRL models in real-world applications.展开更多
广域测量系统(wide area monitoring system,WAMS)的发展为电力系统低频振荡在线辨识奠定了基础。WAMS采集的信号含有高斯白噪声,经低通滤波处理后会产生高斯色噪声,因此会对模式识别的准确性产生不利影响。针对这一问题,提出以实测信...广域测量系统(wide area monitoring system,WAMS)的发展为电力系统低频振荡在线辨识奠定了基础。WAMS采集的信号含有高斯白噪声,经低通滤波处理后会产生高斯色噪声,因此会对模式识别的准确性产生不利影响。针对这一问题,提出以实测信号的四阶混合平均累计量(fourth-order mixed mean cumulant,FOMMC)的对角切片来代替实测信号,并结合矩阵束(matrix pencil,MP)算法对振荡模式进行识别的方法。仿真结果表明,FOMMC-MP算法能够有效从色噪声环境中辨识出系统主导模态。展开更多
同步相量估计算法对同步相量测量单元(phasor measurement unit,PMU)的性能有重要影响,如何获得准确、实时性好、具有良好动态性能的相量估计算法值得研究。为此,提出四分之三基波周期最小二乘(three-quarter fundamental period least ...同步相量估计算法对同步相量测量单元(phasor measurement unit,PMU)的性能有重要影响,如何获得准确、实时性好、具有良好动态性能的相量估计算法值得研究。为此,提出四分之三基波周期最小二乘(three-quarter fundamental period least squares,TQLS)相量估计算法。TQLS法的估计周期为四分之三基波周期,即对该时间窗内的信号进行采样,根据公式推导的结论,可以得到采样结果与基波相量之间的关系。利用最小二乘法,可以由采样结果估计出相量。在整个估计过程中没有泰勒级数展开。理论分析表明,所提TQLS算法可以准确估计基波相量,同时,该算法在基频附近频率响应的幅值平坦,可以准确估计动态相量。仿真结果分别从频域和时域角度验证了TQLS算法的有效性,并且仿真结果显示TQLS算法的计算量与傅里叶变换法相当。展开更多
文摘Intensity modulated radiation therapy (IMRT) is a highly accurate technique that is usually implemented in either dynamic or step-and-shoot fashion with many segments each having low monitor units (MUs). The present study evaluated the effects of beam startup characteristics on the dose delivery accuracy for each segment at low MUs for step-and-shoot IMRT with an Elekta Precise accelerator at the highest dose rates. We used a two-dimensional semi-conductor detector for the dose measurements. The field size of each segment was assumed to be 20 ×20 cm2 and each segment was set to deliver 1 - 10 MUs. Our results show a variation in dose delivery accuracy between segments for the same IMRT beam, which can be attributed to the beam startup characteristics. This variability is attributed to the changes in the transient changes in the temperatures of the electron gun filament and the magnetron. That is, the transient increase in the temperature of the filament leads to increasing doses with time and that of the magnetron leads to decreasing doses with time during the first few MUs.
文摘A big step forward to improve power system monitoring and performance, continued load growth without a corresponding increase in transmission resources has resulted in reduced operational margins for many power systems worldwide and has led to operation of power systems closer to their stability limits and to power exchange in new patterns. These issues, as well as the on-going worldwide trend towards deregulation of the entire industry on the one hand and the increased need for accurate and better network monitoring on the other hand, force power utilities exposed to this pressure to demand new solutions for wide area monitoring, protection and control. Wide-area monitoring, protection, and control require communicating the specific-node information to a remote station but all information should be time synchronized so that to neutralize the time difference between information. It gives a complete simultaneous snap shot of the power system. The conventional system is not able to satisfy the time-synchronized requirement of power system. Phasor Measurement Unit (PMU) is enabler of time-synchronized measurement, it communicate the synchronized local information to remote station.
基金supported by the Shanghai Science and Technology Committee (22511105500)the National Nature Science Foundation of China (62172299, 62032019)+2 种基金the Space Optoelectronic Measurement and Perception LaboratoryBeijing Institute of Control Engineering(LabSOMP-2023-03)the Central Universities of China (2023-4-YB-05)。
文摘Deep reinforcement learning(DRL) has demonstrated significant potential in industrial manufacturing domains such as workshop scheduling and energy system management.However, due to the model's inherent uncertainty, rigorous validation is requisite for its application in real-world tasks. Specific tests may reveal inadequacies in the performance of pre-trained DRL models, while the “black-box” nature of DRL poses a challenge for testing model behavior. We propose a novel performance improvement framework based on probabilistic automata,which aims to proactively identify and correct critical vulnerabilities of DRL systems, so that the performance of DRL models in real tasks can be improved with minimal model modifications.First, a probabilistic automaton is constructed from the historical trajectory of the DRL system by abstracting the state to generate probabilistic decision-making units(PDMUs), and a reverse breadth-first search(BFS) method is used to identify the key PDMU-action pairs that have the greatest impact on adverse outcomes. This process relies only on the state-action sequence and final result of each trajectory. Then, under the key PDMU, we search for the new action that has the greatest impact on favorable results. Finally, the key PDMU, undesirable action and new action are encapsulated as monitors to guide the DRL system to obtain more favorable results through real-time monitoring and correction mechanisms. Evaluations in two standard reinforcement learning environments and three actual job scheduling scenarios confirmed the effectiveness of the method, providing certain guarantees for the deployment of DRL models in real-world applications.
文摘广域测量系统(wide area monitoring system,WAMS)的发展为电力系统低频振荡在线辨识奠定了基础。WAMS采集的信号含有高斯白噪声,经低通滤波处理后会产生高斯色噪声,因此会对模式识别的准确性产生不利影响。针对这一问题,提出以实测信号的四阶混合平均累计量(fourth-order mixed mean cumulant,FOMMC)的对角切片来代替实测信号,并结合矩阵束(matrix pencil,MP)算法对振荡模式进行识别的方法。仿真结果表明,FOMMC-MP算法能够有效从色噪声环境中辨识出系统主导模态。
文摘同步相量估计算法对同步相量测量单元(phasor measurement unit,PMU)的性能有重要影响,如何获得准确、实时性好、具有良好动态性能的相量估计算法值得研究。为此,提出四分之三基波周期最小二乘(three-quarter fundamental period least squares,TQLS)相量估计算法。TQLS法的估计周期为四分之三基波周期,即对该时间窗内的信号进行采样,根据公式推导的结论,可以得到采样结果与基波相量之间的关系。利用最小二乘法,可以由采样结果估计出相量。在整个估计过程中没有泰勒级数展开。理论分析表明,所提TQLS算法可以准确估计基波相量,同时,该算法在基频附近频率响应的幅值平坦,可以准确估计动态相量。仿真结果分别从频域和时域角度验证了TQLS算法的有效性,并且仿真结果显示TQLS算法的计算量与傅里叶变换法相当。
