Research Article
A Deep Reinforcement Learning-Based Method for Economic Dispatch of Integrated Energy System
@INPROCEEDINGS{10.1007/978-3-032-00303-4_12, author={Ling Huang}, title={A Deep Reinforcement Learning-Based Method for Economic Dispatch of Integrated Energy System}, proceedings={Advanced Hybrid Information Processing. 8th International Conference, ADHIP 2024, Jiaxing, China, September 20--22, 2024, Proceedings, Part III}, proceedings_a={ADHIP}, year={2026}, month={11}, keywords={Deep Reinforcement Learning Integrated Energy System Economic Dispatch Dispatch Method Dispatch Model Multi-Objective Model Model Solving}, doi={10.1007/978-3-032-00303-4_12} }- Ling Huang
Year: 2026
A Deep Reinforcement Learning-Based Method for Economic Dispatch of Integrated Energy System
ADHIP
Springer
DOI: 10.1007/978-3-032-00303-4_12
Abstract
Enhancing the efficiency of dispatch within a combined energy framework is essential for fostering multi-energy synergies and improving economic outcomes. Nonetheless, the variable nature of renewable sources and the uncertainty surrounding consumer requirements lead to substantial variations in both the availability and consumption of energy, making conventional dispatch strategies insufficient for responding to the ever-changing environmental conditions. To address this, a novel economic dispatch approach leveraging deep reinforcement learning is introduced. Initially, a comprehensive economic dispatch model is formulated, tailored to the system’s structure and mathematical representation. This model aims to minimize expenses related to energy procurement and the operational costs of electric energy storage, including charging and discharging costs. Subsequently, deep reinforcement learning algorithms are employed to solve the model and derive an optimal economic dispatch strategy tailored to the system’s characteristics. Finally, simulations are conducted to validate the approach’s effectiveness and reliability, ensuring its feasibility in real-world scenarios.