Dynamic resource coordination can increase grid hosting capacity to support more renewables, storage, and electrified load growth

1. Problem Statement

The rapid integration of Distributed Energy Resources (DERs)—specifically solar PV, battery storage (BS), electric vehicles (EVs), and heat pumps (HPs)—along with the electrification of heating and transport, is straining low- and medium-voltage distribution grids.

• Hosting Capacity (HC) Limitations: Current grid planning relies heavily on static, deterministic methods that assume worst-case scenarios (e.g., maximum solar output coinciding with minimum load). These methods are overly conservative, often limiting DER penetration to avoid voltage violations (overvoltage/undervoltage) and thermal line overloads.
• The Gap: Existing literature often analyzes DERs in isolation (e.g., only solar or only EVs) or focuses on operational control rather than joint planning. There is a lack of frameworks that simultaneously optimize the siting, sizing, and coordination of multiple DER types to maximize grid capacity without requiring expensive physical infrastructure upgrades.

2. Methodology

The authors propose a comprehensive framework to evaluate and increase HC through dynamic coordination of DERs. They develop and compare three distinct approaches on a modified IEEE 123-node test feeder (representing a realistic US distribution system):

A. Deterministic Iterative (DI) Approach

• Static Case: Uses fixed, inflexible profiles for loads and generation. Checks AC power flow feasibility without optimization.
• Dynamic Case: Uses a multiperiod AC Optimal Power Flow (ACOPF) with global coordination. It optimizes the dispatch of flexible resources (batteries, heat pumps, EVs) in real-time to absorb excess solar generation or reduce peak loads, subject to device-level constraints (State of Charge, thermal comfort, charging limits).

B. Stochastic Programming (Two-Stage Stochastic Program - 2-SSP)

• Objective: Jointly optimizes DER siting and sizing (planning stage) and operational dispatch (operation stage) under uncertainty.
• Structure:
  • Stage 1 (Planning): Decides the penetration levels and locations of PV, BS, HP, and EVs.
  • Stage 2 (Operation): Solves ACOPF for multiple scenarios (realizations of weather, load, and prices) to ensure feasibility.
• Uncertainty Modeling: Incorporates uncertainties in solar irradiance, temperature (affecting HPs), load profiles, and electricity prices using Monte Carlo sampling.
• Acceleration: To handle computational complexity, the authors employ k-means clustering for scenario reduction and warm-starting the optimization solver.

C. Key Modeling Details

• Device Dynamics: The model includes detailed intertemporal constraints for batteries (SOC dynamics, charge/discharge efficiency), EVs (availability windows, V2G capability), and Heat Pumps (thermal dynamics, temperature setpoints).
• Objective Function: Minimizes PV curtailment, thermal line losses, and operational discomfort (temperature deviation, cycling costs) while maximizing DER usage.
• Grid Physics: Uses a branch flow model with Second-Order Cone Programming (SOCP) relaxation, which is exact for radial networks.

3. Key Contributions

1. Multi-DER Joint Optimization: Unlike prior works focusing on single DER types, this study simultaneously co-optimizes Solar, Batteries, Heat Pumps, and EVs.
2. Quantification of Complementarity: The study rigorously demonstrates how different DERs support one another (e.g., batteries charging during solar peaks to prevent overvoltage; heat pumps shifting load to evening peaks).
3. Three-Method Comparison: Provides a comparative analysis of static, deterministic dynamic, and stochastic dynamic approaches, highlighting the trade-offs between computational tractability and risk management.
4. Siting and Sizing Framework: Develops a novel framework that determines not just how much capacity the grid can hold, but where to place resources to maximize that capacity.

4. Key Results

A. Significant Capacity Increases

• Solar PV: Dynamic coordination increased the feasible solar penetration from 48% (static) to 83% (dynamic) without curtailment.
• Heat Pumps: HP penetration could be increased from 9% (static) to 55% (dynamic).
• Joint Penetration (Stochastic): The stochastic approach revealed that nodal colocation and complementary effects expanded the feasible region of simultaneous Solar, Battery, and Heat Pump penetrations by over 22× compared to static methods.
  • Achievable levels: 200% Solar, 100% Battery, and 90% Heat Pump penetration.

B. Mechanisms of Improvement

• Colocation: The optimization naturally favored co-locating batteries with solar and electrified loads. This reduces net power injections into the grid, alleviating voltage and thermal stress.
• Complementarity:
  • Batteries were identified as the most critical technology, providing the strongest complementary effect with both PV and HPs.
  • Heat Pumps provided significant flexibility by shifting cooling/heating loads, though they compete slightly with EVs for capacity.
  • EVs were less effective in midday solar absorption (due to unavailability) but crucial for evening load management.

C. Impact on Grid Metrics

• Voltage & Current: Dynamic coordination maintained voltages closer to the nominal 1.0 p.u. and significantly reduced mean/median line currents, even though worst-case values remained similar.
• Curtailment: The approach drastically reduced solar curtailment by utilizing excess generation to charge batteries or power heat pumps.
• Uncertainty: The study found that while dynamic coordination improves average outcomes, it introduces 46% higher volatility in grid metrics due to the propagation of input uncertainty through the optimization constraints.

5. Significance and Implications

• Deferred Infrastructure Investment: By leveraging software-defined coordination and flexibility, utilities can accommodate massive DER growth without immediate, costly upgrades to transformers or lines.
• Policy and Market Design: The results support the implementation of flexible interconnection agreements and retail market structures (e.g., dynamic pricing) that incentivize DER owners to participate in grid coordination.
• Decarbonization Pathway: The framework provides a technical roadmap to integrate the projected 262 GW of new DERs (2023–2027) and support full electrification of transport and heating, which would otherwise be blocked by grid constraints.
• Risk Management: The paper highlights that while dynamic approaches are superior, they require careful management of uncertainty to avoid increased volatility in grid operations.

Conclusion

The paper demonstrates that dynamic coordination transforms the distribution grid from a passive conduit into an active, flexible system. By jointly optimizing the siting, sizing, and real-time operation of diverse DERs, the grid's hosting capacity can be expanded by orders of magnitude, enabling a faster, more affordable, and reliable transition to a decarbonized energy system.