Entries for Data-Driven Distributed (3D) Solar Visibility Prize

Neural Network & Kalman Filterbased hybrid algorithms offer true solar power predictions by combining deep learning and real-time estimation

This team (WVU-V&R-ComEd) will develop BTM PV estimation, pseudo-measurement, bad data detection and nonlinear state estimation.

Leveraging NLP techniques via an LLM, we're experimenting whether we can perform well in energy estimation tasks as inspired by MotionLM.

Climformatics Solution enables paradigm shift towards cheaper, greener energy, climate resilience and sustainability for distributed grids

Our hybrid algorithm combines traditional physics-based state estimation with data-driven machine learning to achieve superior performance.

Robust DSSE by using classification models, spatio-temporal correlations and time-dependent probabilities learnt from the historical data.

Our team is taking a classical and enhanced approach for our algorithm. We are using WLS and replacing state variables with branch currents.

Graph learning techniques will be integrated into the physical formulation for distribution system state estimation.

We introduce an algorithm that performs state estimation leveraging quantum AI. It detects anomalies and calculates an overall skill score.

SolarShield enhances distribution system visibility with DSSE algorithms, optimizing solar energy use for reliable, resilient electric grids

Transformer Based Multivariate Time Series forecasting model to achieve robust estimation of distributed energy system operating conditions.

Leveraging GCN-LSTM for spatiotemporal analysis to effectively capture and analyze complex patterns in the network data.

We have developed a machine learning (ML)-based DSSE framework that accounts for the unique attributes of modern distribution systems