Safe Deep Reinforcement Learning-Based Real-Time Multi-Energy Management in Combined Heat and Power Microgrids
Bibliographic record
Abstract
Combined heat and power microgrids (CHPMGs) have become increasingly popular recently due to their ability to offer cost-effective and resilient solutions to support critical infrastructures. As the centerpiece of a CHPMG, an autonomous real-time multi-energy management system can leverage advanced metering infrastructure to monitor and dispatch various distributed energy resources (DERs), minimize operational costs, and improve the overall system reliability. In this paper, a novel real-time energy management system (EMS) for CHPMGs is proposed using a data-driven model-free safe deep reinforcement learning method. The energy management problem in CHPMGs is first formulated into a constrained Markov decision process (CMDP). A safe deep deterministic policy gradient (SDDPG) method is then applied to solve the developed CMDP. SDDPG features an actor-critic structure, in which the actor network learns the optimal control policy, and the critic network learns to evaluate the state-action value. To satisfy CHPMGs operational constraints during the training process, two sets of neural networks are proposed to approximate the constrained system parameters. A mathematical optimization-based safety layer is also constructed upon the actor network to analytically correct the agent’s actions into safe actions satisfying specific constraints. The proposed method is validated by case studies using real-world data; it is also compared with conventional deep reinforcement learning (DRL) and optimization-based approaches with and without accurate uncertainty data.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".