A comprehensive review and simulation-based analysis of power market pricing mechanisms in energy systems with high penetration of distributed energy resources: an analysis of distribution locational marginal pricing, dynamic tariff, and machine learning-based approaches

• This study explores the integration of advanced machine learning methods, including Artificial Neural Networks, Extreme Gradient Boosting, Deep Reinforcement Learning, and clustering techniques, to improve price prediction and control strategy optimization for DERs. • The paper includes simulation-based validation of Machine Learning techniques for real-world pricing applications. • The paper offers a detailed comparison of two pricing mechanisms—Distribution Locational Marginal Pricing and Dynamic Tariffs. With the increasing penetration of Distributed Energy Resources (DERs) in Active Distribution Networks (AND), market-based approaches to techno-economic analysis and optimization of these flexible resources have become essential. Hence, determining appropriate price signals for controlling the active and reactive powers of these flexible resources is extremely important. This review article first examines two major and traditional pricing mechanisms in depth and compares some of their critical features: Distribution Locational Marginal Pricing (DLMP) and Dynamic Tariff (DT). This study undertakes an accurate investigation and literature review of different approaches established for optimally determining price signals in ADN. Accordingly, the advantages and disadvantages, possible applications, and proposed solutions for each approach in overwhelming the challenges above are addressed and discussed. Furthermore, this paper explores the key role of artificial neural networks (ANNs), extreme gradient boosting (XGBoost), deep reinforcement learning (DRL), and clustering algorithms in the electricity pricing process. It is also worth noting that this study goes beyond a review and performs simulations for each pricing method (conventional and ML-based methods) in MATLAB software. This professional work opens new horizons for developing innovative and efficient pricing models in future energy systems by covering critical research gaps in this field.