A Review of Balancing Price Forecasting in the Context of Renewable-Rich Power Systems, Highlighting Profit-Aware and Spike-Resilient Approaches

Abstract

Balancing (real-time) market price forecasting is a vital enabler for renewable integration, storage arbitrage, and risk-aware trading, yet the literature remains fragmented and underdeveloped. This review addresses these shortcomings by systematically categorizing and evaluating studies across prediction horizons, modeling paradigms, and data-engineering practices. We show that enriching forecasts with auxiliary features, such as day-ahead prices, net imbalance volumes, renewable forecast errors, and meteorological inputs, substantially reduces error relative to price-only baselines. Probabilistic frameworks, while invaluable for providing risk envelopes in bidding strategies, are still underexploited. Typical reported accuracy spans mean absolute percentage errors of approximately 3–10% for very short-term (1–6 h ahead) horizons, 10–20% for mid-term horizons (12–24 h ahead), and around 25% for longer horizons (24–36 h ahead), with spikes and rapid ramps driving most residual error. From this synthesis, we identify the following four critical research gaps: (1) inadequate modeling of price spikes and ramps, (2) limited innovation in pre- and post-processing techniques, (3) sparse adoption of profit-driven (revenue-aware) evaluation, and (4) weak segmentation of distinct temporal regimes. By mapping prevailing methodologies, benchmarking performance, and highlighting emerging paradigms, such as feedback-driven, risk-aware, feature-enriched pipelines, this review delineates the state of the art and proposes a research agenda focused on maximizing economic value.