首页|New Computational Intelligence Study Findings Recently Were Reported by Research ers at University of Tulsa (Physics-informed Graph Capsule Generative Autoencode r for Probabilistic Ac Optimal Power Flow)
New Computational Intelligence Study Findings Recently Were Reported by Research ers at University of Tulsa (Physics-informed Graph Capsule Generative Autoencode r for Probabilistic Ac Optimal Power Flow)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – Fresh data on Machine Learning - Compu tational Intelligence are presented in a new report. According to news originati ng from Tulsa, Oklahoma, by NewsRx correspondents, research stated, “Due to the increasing demand for electricity and the inherent uncertainty in power generati on, finding efficient solutions to the stochastic alternating current optimal po wer flow (AC-OPF) problem has become crucial. However, the nonlinear and non-con vex nature of AC-OPF, coupled with the growing stochasticity resulting from the integration of renewable energy sources, presents significant challenges in achi eving fast and reliable solutions.” Our news journalists obtained a quote from the research from the University of T ulsa, “To address these challenges, this study proposes a novel graph-based gene rative methodology that effectively captures the uncertainties in power system m easurements, enabling the learning of probability distribution functions for gen eration dispatch and voltage setpoints. Our approach involves modeling the power system as a weighted graph and utilizing a deep spectral graph convolution netw ork to extract powerful spatial patterns from the input graph measurements. A un ique variational approach is introduced to identify the most relevant latent fea tures that accurately describe the setpoints of the AC-OPF problem. Additionally , a capsule network with a new greedy dynamic routing algorithm is proposed to p recisely decode the latent features and estimate the probabilistic AC-OPF proble m. Further, a set of carefully designed physics-informed loss functions is incor porated in the training procedure of the model to ensure adherence to the fundam ental physics rules governing power systems. Notably, the proposed physics-infor med loss functions not only enhance the accuracy of AC-OPF estimation by effecti vely regularizing the deep learning model but also significantly reduce the time complexity.”
TulsaOklahomaUnited StatesNorth an d Central AmericaComputational IntelligenceMachine LearningUniversity of T ulsa
NETL
NSTL
