Dynamic pricing and fair settlements for multi-microgrid energy trading

• Intermediary entity enables efficient pricing and fair market settlements. • SVM-based price forecasting achieves ∼10% RMSE accuracy. • Evolutionary stochastic optimization increases transaction profits by 33%. • Adaptive optimization estimates energy, sets fair prices, and mitigates risk. The scheduling of energy exchange between multi-microgrids (MGs) and energy markets is a complex task that requires balancing profitability, risk management, and operational efficiency. This paper introduces an intermediary entity (IE) to facilitate efficient price setting and ensure fair market settlements. Market prices are forecasted using a support vector machine model, achieving a root mean square error of approximately 10%. Transaction pricing is guided by IE-imposed tariffs in combination with an evolutionary stochastic decision optimization algorithm, which increases transaction profits by 33% compared with benchmark methods. Adaptive Optimization and Quantile Regression are further employed to estimate transaction energy, determine fair settlement prices, and mitigate financial risk. Simulation results demonstrate minimal fluctuations in transaction energy (0.4%–2.2%), a 15% improvement in overall market revenue stability, and MG net profits reaching 35.24% of total revenue. These results confirm that the proposed framework effectively integrates multi-MGs into energy markets, providing sustainable, risk-adjusted profitability through fair settlements while enhancing overall market efficiency, reliability, and decision-making under uncertainty.