Deep reinforcement learning approach for hybrid renewable energy systems optimization
Bibliographic record
Abstract
The sizing of hybrid renewable energy systems (HRES) is a major challenge faced in contemporary energy research. The optimal configuration based on the specific consumption requirements is essential for strategic energy planning. Effective sizing must balance the investment costs, reliability, environmental impacts, and greenhouse gas emissions while satisfying the expected energy requirements. This study proposes a novel multi-criteria sizing approach based on deep reinforcement learning (DRL). The DRL agent is guided by a reward function that integrates three essential performance metrics: energy cost (LCOE), renewable energy fraction (REF), and the loss of power supply probability (LPSP). A penalty function is also included to consider the reliance on external sources, such as diesel generators and the public grid, promoting greater autonomy and renewable usage. The DRL-based approach was implemented and tested on three distinct demand profiles, using hourly data for one year. A comparative analysis was conducted against three established methods: particle swarm optimization (PSO), multi-objective PSO (MOPSO), and non-dominated sorted genetic algorithm (NSGA-II). The results indicate that DRL significantly outperforms all the benchmark methods in terms of economic efficiency. DRL achieves a significant reduction in the energy costs, ranging from 21.33 % to 30.09 % when compared with PSO, 27.89 %–30.27 % when compared with MOPSO, and 27.63 %–28.47 % when compared with NSGA-II. These findings demonstrate that DRL presents a robust and adaptive framework for the sizing and operational control of HRES. DRL presents more autonomous, cost-effective, and scalable renewable energy solutions by minimizing the energy costs while maintaining the system reliability.
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 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".