Research Worth Reading

ev-flow: A Reproducible, NHTS-Grounded Generator of Synthetic Plug-in Electric Vehicle Charging Behavior for Eight U.S. Regions — An open synthetic data generator for EV charging profiles grounded in the National Household Travel Survey, covering eight U.S. regions with regional, seasonal, and equipment heterogeneity. Useful for grid-integration studies where real telemetry is scarce or privacy-restricted, giving engineers realistic, privacy-free charging data for load forecasting, VPP simulations, and distribution system planning.

Rigorous uncertainty quantification of probabilistic AI weather forecasts with conformal prediction — Applies conformal prediction to provide rigorous, distribution-free uncertainty quantification for AI-based probabilistic weather forecasts. Addresses the gap between AI models’ ability to generate large ensembles and the statistical reliability of their uncertainty estimates — directly applicable to renewable generation forecasting and grid operations where calibrated probabilities matter.

Ramping Procurement and Bid-Cost Recovery in Real-Time Market — Analytical framework co-optimizing ramping procurement with economic dispatch under net-demand uncertainty, comparing single-interval and multi-interval flexible ramp product designs via rolling-window stochastic optimization. Relevant for real-time electricity market design and flexible resource procurement; shows how to model ramping products and bid-cost recovery in optimization.

Techno-Economic Analysis of Shared Mobile Storage for Demand Charge Reduction — High-fidelity fleet management framework evaluating shared EV fleets for demand charge reduction, explicitly modeling spatio-temporal coupling of energy and transit overheads. Moves beyond idealized models to assess practical logistical and operational constraints of mobile storage deployment — useful for fleet optimization, demand charge management, and vehicle-to-grid integration.

Open Source Projects

Eavor plots next step for novel geothermal project after rocky start — Eavor Technologies achieved a milestone with its closed-loop geothermal system (Eavor-Loop) delivering electricity to Germany’s grid, using a proprietary closed-loop design without hydraulic fracturing. Demonstrates a novel geothermal approach that avoids fracking and provides baseload power; relevant for subsurface modeling, drilling tech, and grid integration of firm renewable resources.

Here Comes Another New Addition To The US EV Battery Ecosystem — Canadian firm Graphite One is developing an onshore graphite supply chain for EV batteries with a mine in Alaska and a processing facility in Ohio. Addresses a critical bottleneck in the U.S. EV battery supply chain by reducing dependence on imported graphite; involves mining, processing, and logistics engineering for critical minerals.

Tesla offers discounted home batteries in New England VPP push — Tesla is offering discounted Powerwall batteries to homeowners in Massachusetts and Connecticut through a new virtual power plant (VPP) program, allowing the grid operator to tap residential storage capacity during peak demand. Real-world deployment of distributed battery aggregation for grid services; involves DER management, communication protocols, and market participation.

Today’s Synthesis

To integrate EV charging with renewable generation and demand‑charge management, an engineer could build a co‑optimization platform that first synthesizes regional EV load profiles using ev-flow . The synthetic data captures the heterogeneity of eight U.S. regions, seasons, and equipment types, providing privacy‑safe inputs for large‑scale simulations. Next, feed these loads into a rolling‑window stochastic optimizer that also consumes calibrated probabilistic weather forecasts generated by the conformal‑prediction‑enhanced AI model described in Rigorous uncertainty quantification of probabilistic AI weather forecasts with conformal prediction . The conformal layer guarantees distribution‑free confidence intervals, essential for reliable renewable output forecasts and risk‑aware dispatch. Finally, the optimizer schedules the shared mobile storage fleet outlined in Techno-Economic Analysis of Shared Mobile Storage for Demand Charge Reduction to shift charging loads and discharge storage during peak demand windows, minimizing demand charges while respecting transit constraints. This end‑to‑end pipeline turns synthetic charging data, robust weather forecasting, and mobile storage logistics into a deployable virtual power plant solution for fleet operators.