Risk-Aware Multi-Market Scheduling of Virtual Power Plants with Dynamic Network Tariffs

Researchers from ETH Zurich have developed a novel risk-aware scheduling framework for virtual power plants (VPPs) that integrates dynamic network tariffs and multi-market bidding. The two-stage stochastic optimization model, detailed in a paper accepted to PSCC 2026, accounts for detailed device-level constraints, network limitations, and operational uncertainties. By incorporating conditional value-at-risk (CVaR) and Benders decomposition, the framework remains tractable with large scenario sets while allowing operators to express risk preferences. The model jointly optimizes bids across energy and reserve markets, explicitly considering local flexibility procurement through dynamic network tariffs.

Results from a realistic case study reveal that both risk-neutral and risk-averse strategies exploit arbitrage opportunities, but risk aversion significantly reduces profit volatility by aligning more closely with physical dispatch. Dynamic tariffs effectively unlock local flexibility by shifting demand across the day, though strong tariff signals can reduce expected profitability by up to 65% while providing limited additional flexibility gains. This work highlights the trade-offs between profitability, risk, and flexibility in VPP operations, offering practical insights for grid operators and energy traders as distributed energy resource penetration increases.