Terra Daily — May 25, 2026

Research Worth Reading 🏊

• A Methodology for Impedance-based Stability Margin Analysis for Interconnected Offshore Wind Clusters — A methodology for assessing how integrating a new offshore wind plant shifts stability margins at the point of connection using impedance-based analysis. If you’ve ever debugged a coupled system, the parallels to grid cluster design are immediate. 🌌
• Holistic Grid-Forming Control to Enhance the Frequency Support from HVDC-Connected Offshore Wind Power Plants — Proposes a holistic grid-forming (GFM) control strategy so HVDC-connected offshore wind can provide inertial response and frequency containment reserve, tackling the stability gaps that come with high power-electronics penetration. A concrete control design, not just theory. 🐋
• Open-Source METANET Calibration for Reproducible Freeway Traffic Macroscopic Simulation — Fills the gap of missing open-source calibration tools for METANET, a second-order macroscopic traffic flow model. Gives engineers reproducible pipelines for traffic simulation, ramp metering, and variable speed limit control. 🌻
• OptiQU: Coordinated Multi-Level Voltage and Reactive Power Control for Enhanced Voltage Quality and Secure Grid Operation — Tackles voltage constraints in low-voltage grids loaded with PV, heat pumps, and EV chargers by coordinating voltage and reactive power control across multiple levels. Avoids expensive conventional grid reinforcement. 🏊
• The physics of AI weather models — Uses forecast-skill correlations and Centered Kernel Alignment to test whether AI weather models actually encode physical equations across architectures. Finds different models represent the atmosphere in similar ways — important for anyone building or benchmarking neural weather forecasters. 🌌

Technology & Innovation 🌻

• REPORT: U.S. Adds 10 GWh of New Energy Storage Capacity in 1st Quarter, Best Q1 on Record — The U.S. installed 9.7 GWh of new storage in Q1 2026, the strongest first quarter on record, with five-year deployment forecasts revised upward. A solid data point for capacity planning and tracking hardware trends. 🐋

Open Source Projects 🌌

• Denver has a plan to heat and cool buildings with — wait for it — sewage — Denver is piloting sewage-to-thermal-energy for building climate control, exploiting wastewater’s stable temperature to cut fossil fuel use. A pragmatic infrastructure hack that sounds like a punchline but is very real engineering. 🌻
• Boavizta/boaviztapi — RESTful API for BOAVIZTA reference data and LCA methodologies, letting you assess the environmental impact of digital infrastructure programmatically. Useful if you’re quantifying the carbon footprint of your own ML pipeline. 🏊
• lucasm/findto — A decentralized search assistant that aims to cut the carbon footprint of web search and AI queries by reducing redundant requests and improving efficiency. A small but direct attempt to make information retrieval less wasteful. 🌌

Implementation Guides 🐋

• Save Money on Electricity — Get to Know Your Time-of-Use Electricity Pricing Details — Breaks down how TOU pricing works and why understanding your rate structure can meaningfully reduce costs. Directly relevant when designing energy management systems, smart EV charging schedules, or home battery dispatch logic. 🌻

Today’s Synthesis

When you’re designing distributed energy management for buildings loaded with PV, heat pumps, and EVs, three pieces converge: you need grid-aware voltage control to avoid reinforcement costs (OptiQU: Coordinated Multi-Level Voltage and Reactive Power Control ), price signals to schedule loads intelligently (Save Money on Electricity — Get to Know Your Time-of-Use Electricity Pricing Details ), and a way to measure whether your optimization actually reduces carbon. The Boaviztapi API gives you programmatic LCA for digital infrastructure (Boavizta/boaviztapi ), so you can close the loop on lifecycle impact. Pair OptiQU-style reactive power coordination with TOU-aware dispatch logic in an open-source controller, then benchmark carbon outcomes. It’s the kind of system where control theory, economic signals, and reproducible measurement stack cleanly — and where your existing systems skills transfer immediately.