CIS & MINDS Seminar - Sijia Geng

<p><strong>Recorded Seminar Link:</strong></p><p><a href=" link:</strong></p><p><a href="">... and Analysis Methods in Energy and Electrified Transportation Systems</strong></p><p><strong>SijiaGeng, Assistant Professor, Electrical and Computer Engineering, Johns HopkinsUniversity</strong></p><p><strong> </strong></p><p><strong>Abstract:</strong></p><p>Electricenergy systems are currently experiencing a profound transformation, includinga considerable degree of uncertainty driven by renewable energy sources, newforms of dynamics due to inverter-based resources (IBRs), and large-scale integrationof electric vehicles (EVs) that integrates various energy sectors andtransportation. </p><p> </p><p>Inthe first part of the talk, I will propose a novel “integer-clustering”approach to model a large number of EVs that manages vehicle charging andenergy at the fleet level yet maintain individual trip dispatch. The model isthen used to develop a spatially and temporally-resolved decision-making tool foroptimally planning and/or operating EV fleets and energy infrastructure (e.g., fast-charging,hydrogen-fueling, and distributed energy resources). The tool comprises atwo-stage framework where a tractable disaggregation step follows theinteger-clustering problem to recover an individually feasible solution. Weestablish theoretical lower and upper bounds on the true individual formulationwhich underpins a guaranteed performance of the proposed method. The optimalityaccuracy and computational efficiency of the integer-clustering formulation arenumerically validated on a real-world case study of Boston’s public transitnetwork. Substantial speedups with minimal loss in solution quality aredemonstrated. By using a real geospatial timetable dataset for bus schedule andrenewable generation, we provide insights into different pathways fordecarbonizing heavy-duty EV fleets and their impacts on energy systems.</p><p> </p><p>Inthe second part of the talk, I will focus on an important problem of voltagestability in power systems. The problem is related to finding the singularsolution space boundary (SSB) of power flow equations. We propose a novelmethod rooted in differential geometry to approximate the SSB of power systemsunder high variability of renewable generation. Conventional methods mostly relyon either expensive numerical continuation at specified directions or numericaloptimization. Instead, the proposed approach constructs the Christoffel symbolsof the second kind from the Riemannian metric tensors to characterize thecomplete local geometry which is then extended to the proximity of the SSB withefficient computations. As a result, this approach is suitable to handlehigh-dimensional variability in operating points. We demonstrate advantages ofthe proposed method using various case studies and provide additional insightson voltage stability in renewable-rich power systems.</p><p><br><strong>Bio: </strong>Sijia Geng is an Assistant Professor in the Department ofElectrical and Computer Engineering at Johns Hopkins University. Before joiningJHU in January 2023, she was a Postdoctoral Associate at the Laboratory forInformation &amp; Decision Systems (LIDS) at MIT in 2022. She received her <a href=""></a> Electrical and Computer Engineering from the University of Michigan, AnnArbor, where she also received the M.S. in Mathematics and M.S. in ECE. Herresearch integrates methodologies from system and control theory, analysis, andoptimization to address pressing and fundamental challenges in complex andnetworked energy systems. She aims at driving the widespread utilization ofrenewable energy resources while enhancing the resiliency and efficiency of energysystems through developing rigorous theory and scalable computational tools. Sheis the recipient of a Best Paper Award at the MIT/Harvard Applied EnergySymposium in 2022 and was named a Barbour Scholar in 2021.</p><p> </p>

Tuesday, October 31, 2023 - 12:00 to 13:00

Clark, 110