Peer-to-Peer Energy Exchange · Research Paper

Peer-to-Peer Energy Exchange Optimization

A non-cooperative game-theory method for building-to-building grid energy exchange, optimising welfare through strategic pricing and ESS degradation-aware trading.

Journal: Sustainable Energy, Grids and Networks (2024)

Authors: M.B. Gebremariam, P. García, Á. Navarro-Rodríguez, C. Blanco

Data: 4 buildings · Barcelona, Spain · Dec 2022 – Jun 2023

Building-to-Building Grid Energy Exchange

This research presents a non-cooperative game-theory method to optimise energy exchange within a building-to-building (B2B) grid. Each building with PV generation and energy storage strategically determines its energy trading decisions — including pricing — to maximise individual welfare while reducing reliance on the main grid.

Game Theory Stackelberg Game Non-cooperative ESS Degradation Barcelona Pilot

B2B Grid Model Schematic — **Fig. 1.** Schematics of the proposed decentralised B2B grid model with VSG-based interlinking inverters, PV generation, and ESS at each building.

System Architecture

The proposed system connects buildings through a Virtual Synchronous Generator (VSG) based controlled DC/AC interlinking inverter. Each building is equipped with in-building PV generation, ESS connected via DC/DC converter, and internal loads. The VSG control allows buildings to seamlessly participate in inertia support of the B2B grid network — a key differentiator for grid stability.

Building Classification

At each time step, buildings are classified as buyers or sellers based on their Generation-to-Load Ratio (GLR). Buildings with GLR > 1 become sellers, while buildings with GLR < 1 become buyers. The grid and ESS can also participate as sellers or buyers depending on preference and state of charge.

**Fig. 2.** Generation-to-Load Ratio (GLR) of the four participant buildings, showing how each building transitions between buyer and seller roles throughout the day.

Game Theory Optimization Model

The energy exchange is formulated as a non-cooperative Stackelberg game. Sellers announce prices and available energy; buyers select sellers and quantities to maximise their welfare function. Through iterative negotiation, the market reaches equilibrium where all participants benefit compared to grid-only operation.

Welfare Function & ESS Degradation

Each building's objective is to maximise a welfare function balancing utility from energy consumption against costs from B2B trading, grid imports, and ESS degradation. The utility function is quadratic and concave, ensuring satisfaction saturates when demand is fully supplied. ESS degradation cost — combining calendar and cyclic aging via rain-flow counting — is explicitly modelled, making this uniquely realistic among peer-to-peer trading approaches.

Energy Exchange Simulation

Daily Energy Exchange — **Fig. 5.** Daily energy exchange of the proposed B2B grid simulation for four buildings, showing PV generation, load demand, B2B trades, grid interaction, and ESS flows across a full day.

ESS SoC — **Fig. 6.** ESS State of Charge measurements across buildings, showing coordinated charge/discharge patterns.

Prices — **Fig. 7.** Energy prices set by sellers alongside grid retail, feed-in tariffs and ESS degradation costs.

Results & Cost Savings

The proposed B2B grid with ESS (considering degradation cost) is compared against building-to-grid (B2Grid) alternatives. Three cases are analysed: B2B without ESS, B2B with ESS (ideal), and B2B with ESS including degradation cost.

Key Performance Metrics

2–6%

Daily cost saving vs. B2Grid

165%

Peak-hour savings max

€567

Savings over 7 months (B4)

4 → 30

Scalable participant count

Exchange Detail — **Fig. 8.** Detailed energy exchange during peak availability hours (10:00–16:00) for all four buildings.

Cost Comparison — **Fig. 9.** Hourly cost comparison: B2Grid vs. B2B grid with and without ESS degradation cost for each building.

Real-World Validation: Barcelona Pilot

The system was validated using seven months of smart meter data from four interconnected buildings in Barcelona. The hourly energy exchange simulation confirms the model's ability to optimise real-world trading under actual market conditions, with savings ranging from €73 to €567 per building.

Real Case — **Fig. 14.** Hourly energy exchange simulation for four buildings in Barcelona from Dec 2022 to Jun 2023, showing seasonal variation in PV generation and B2B trading activity.

Full paper citation:
M.B. Gebremariam, P. García, Á. Navarro-Rodríguez, C. Blanco, "Energy exchange optimization for a building-to-building grid: A non-cooperative game-theory method," Sustainable Energy, Grids and Networks, vol. 39, 101420, 2024.

DOI: 10.1016/j.segan.2024.101420 ↗