Terra Daily — June 29, 2026

Research Worth Reading

• Real-Time State Estimation in Smart Grids over 5G Networks: Experimental Validation Using Raspberry Pis and Typhoon HIL — Demonstrates real-time state estimation for smart grids using Raspberry Pi hardware validated on Typhoon HIL over 5G, providing latency and reliability benchmarks for 5G‑enabled grid monitoring.
• Decentralized Stability of IBR-dominated Power Grids Using Block Diagonal Dominance — Introduces a block diagonal dominance criterion for decentralized small‑signal stability assessment in grids with high inverter‑based resource penetration, offering a scalable, model‑agnostic method for renewable integration planning.
• Repair-before-veto control for safe lithium-ion fast charging under unknown ambient and cooling-fault conditions — Implements a repair‑before‑veto control strategy that handles uncertain ambient temperature and cooling degradation, preventing thermal runaway in EV fast‑charging systems.
• Cement’s Future Is Less Portland, Not One Magic Cement — Surveys cement decarbonization pathways—including supplementary cementitious materials, electrochemical processes, hydrogen kilns, low‑carbon limestone, recycled cement, and carbon‑capture retrofits—and argues for a portfolio approach rather than a single breakthrough.

Open Source Projects

• pvlib-python: Photovoltaic Energy System Simulation Library — A Python library providing functions for simulating photovoltaic system performance, irradiance modeling, and loss estimation used in solar energy research and engineering.
• OpenFAST: NREL Wind Farm Simulation Code — NREL‑supported aeroelastic simulation codes for individual turbines and wind farms, essential for turbine design, load analysis, and farm layout optimization.
• Energinet Model Testbench (MTB): Grid Compliance Automation — Automates grid compliance studies in PSCAD and PowerFactory for renewable integration, aiding power‑systems engineers in connection studies for solar, wind, and HVDC projects.
• CodeCarbon: Compute Emissions Tracking — Tracks carbon emissions from compute workloads and suggests reduction strategies, helping ML/AI engineers monitor and lower the carbon footprint of training and inference.
• Green Metrics Tool: Software Energy & Carbon Measurement — Measures software energy consumption and carbon emissions with timeline data, git integration, dashboards, and optimization recommendations for green‑software teams.
• CESM: Community Earth System Model — NCAR’s comprehensive climate model for simulating Earth’s climate system, used for long‑term climate projections and Earth‑system research.

Today’s Synthesis

A practical project for engineers entering climate tech is to create an end‑to‑end smart‑grid monitoring platform that fuses renewable generation forecasts with real‑time grid state estimation while tracking the carbon cost of its own computation. Start by integrating pvlib-python to generate high‑resolution PV output predictions for distributed solar assets. Feed these predictions into the Real-Time State Estimation in Smart Grids over 5G Networks pipeline, using Raspberry Pi nodes and Typhoon HIL validation to compute grid voltages and power flows with sub‑second latency over a 5G link. Parallelize the estimation and forecasting workloads on a cluster instrumented with CodeCarbon to capture GPU/CPU energy draw and emit reduction suggestions. The system can alert operators to overloads, optimize inverter set‑points, and provide a transparent carbon‑footprint report for each monitoring cycle. This combination leverages existing open‑source tools, proven 5G‑enabled hardware validation, and compute‑emission tracking to deliver a deployable, data‑driven solution for grid operators managing high renewable penetration.