Terra Daily — May 15, 2026

Research Worth Reading

• Optimal design of solar-battery hybrid resources considering multi-market participation under weather and price uncertainty — A stochastic optimization framework for sizing and operating co-located solar and battery storage across multiple electricity markets under realistic weather and price uncertainty. If you’ve done optimization under uncertainty in any domain, the modeling patterns here translate directly to resource planning for grid operators and renewable developers.
• Energy Management for Solar-Powered Electric-Bus Charging Station: A Data-Driven Method — Uses polynomial chaos expansion to build an energy management system for a solar-powered bus depot, jointly handling uncertainty in PV output, electricity prices, and bus schedule states. A clean example of how data-driven uncertainty quantification methods can be deployed in real transit electrification infrastructure.
• Frequency Nadir-Constrained Power System Restoration Planning with Energy Storage — Formulates black-start restoration sequencing as a frequency-constrained MILP, ensuring stability when bringing synchronous machines and storage online after a blackout. Directly useful for power systems engineers working on grid resilience — the constraint modeling approach is well-documented and reproducible.
• Distributionally Robust Model Predictive Control for Virtual Power Plants — Develops a distributionally robust MPC framework that coordinates buildings, batteries, and renewables within a virtual power plant under electricity price uncertainty. If you have control theory or optimization experience, this is a deployable architecture for DER aggregators operating in volatile wholesale markets.
• Battery-Assisted Operation of Hyperscale AI Data Centers under Connect-and-Manage Interconnection Practices — Models how multi-hundred-megawatt AI data centers can use on-site battery storage to operate within real-time power exchange limits imposed by emerging connect-and-manage grid interconnection rules. Sits at the intersection of data center infrastructure and grid integration — a problem space that’s only going to grow as AI compute demand scales.
• Optimizing Grid-Forming Controls using Relay-based Extremum Seeking to Enhance Transient Performance — Proposes a relay-based extremum seeking approach to auto-tune active power-frequency droop coefficients on grid-forming inverters, improving transient response without manual parameter sweeps. A practical controls contribution for anyone deploying inverter-based resources in weak or islanded grids where auto-tuning matters more than hand-crafted settings.

Technology & Innovation

• The world is installing grid batteries at a blistering pace — Global grid-scale battery installations hit 112 GW of new capacity in 2025, up 48% year-over-year per BloombergNEF. The deployment velocity is a concrete signal that storage is no longer a niche balancing tool — it’s becoming a standard grid infrastructure asset.
• Grid-Scale Virtual Power Plants are Here. Have Utilities Noticed? — Makes the case that aggregated distributed energy resources can substitute for new centralized generation, and that utilities need to incorporate VPPs into their planning models. Relevant reading if you’re considering work on DER orchestration or grid modernization software.
• Long-Haul Electric Trucking Is Almost Here — Covers the rollout of megawatt-scale charging infrastructure for heavy-duty electric trucks, a prerequisite for decarbonizing long-haul freight logistics. The power electronics and systems integration challenges here are substantial and largely unsolved — good territory for engineers who want to build physical infrastructure.

Today’s Synthesis

The distributionally robust MPC framework for virtual power plants and RMI’s argument that utilities must plan around aggregated DERs point to the same trajectory: coordinating distributed batteries, buildings, and renewables into dispatchable grid assets is moving from simulation to real-time operations. Meanwhile, the hyperscale data center battery paper introduces a new player — large, flexible loads that can shift consumption in response to grid conditions. Taken together, there’s a concrete engineering opportunity: take the robust MPC optimization approach, extend it to jointly coordinate behind-the-meter storage with interruptible data center loads, and build a dispatch engine for a DER aggregator participating in volatile wholesale markets. The uncertainty-handling machinery in the MPC translates directly to the price and weather volatility operators face daily, while the data center flexibility layer represents a resource most existing VPP platforms don’t model at all.