Two-Stage Optimization Model Based on Neo4j-Dueling Deep Q Network DOI Creative Commons
Tie Chen,

Pingping Yang,

Hongxin Li

et al.

Energies, Journal Year: 2024, Volume and Issue: 17(19), P. 4998 - 4998

Published: Oct. 8, 2024

To alleviate the power flow congestion in active distribution networks (ADNs), this paper proposes a two-stage load transfer optimization model based on Neo4j-Dueling DQN. First, Neo4j graph was established as training environment for Dueling Meanwhile, supply paths from point to source were obtained using Cypher language built into Neo4j, forming space that served action space. Secondly, various constraints process, reward and penalty function formulated establish DQN model. Finally, according ε−greedy selection strategy, actions selected interacted with environment, resulting optimal operation sequence. In paper, Python used programming language, TensorFlow open-source software library form deep reinforcement network, Py2neo toolkit complete linkage between python platform Neo4j. We conducted experiments real 79-node system, three scenarios validation. Under scenarios, time required obtain results 2.87 s, 4.37 s 3.45 respectively. For scenario 1 before after transfer, line loss, voltage deviation rate reduced by about 56.0%, 76.0% 55.7%, 2 41.7%, 72.9% 56.7%, 3 13.6%, 47.1% 37.7%, The experimental show trained can quickly accurately derive sequence under different conditions, thereby validating effectiveness of proposed

Language: Английский

Task Scheduling in Distributed Real-Time Systems Using Hybrid Model Based on ACO-GA DOI

Anchal Sharma,

Sangeeta Sharma,

Sanat Thakur

et al.

Communications in computer and information science, Journal Year: 2025, Volume and Issue: unknown, P. 448 - 463

Published: Jan. 1, 2025

Language: Английский

Citations

0
Deep reinforcement learning algorithm incorporating problem characteristics for dynamic multi-objective permutation flow-shop scheduling problem DOI
Yuanyuan Yang, Bin Qian,

Rong Hu

et al.

Swarm and Evolutionary Computation, Journal Year: 2025, Volume and Issue: 96, P. 101973 - 101973

Published: May 14, 2025

Language: Английский

Citations

0
Cloud continuum testbeds and next-generation ICTs: Trends, challenges, and perspectives DOI Creative Commons
Fran Casino, Peio López-Iturri,

Constantinos Patsakis

et al.

Computer Science Review, Journal Year: 2024, Volume and Issue: 56, P. 100696 - 100696

Published: Dec. 6, 2024

Language: Английский

Citations

2
Two-Stage Optimization Model Based on Neo4j-Dueling Deep Q Network DOI Creative Commons
Tie Chen,

Pingping Yang,

Hongxin Li

et al.

Energies, Journal Year: 2024, Volume and Issue: 17(19), P. 4998 - 4998

Published: Oct. 8, 2024

To alleviate the power flow congestion in active distribution networks (ADNs), this paper proposes a two-stage load transfer optimization model based on Neo4j-Dueling DQN. First, Neo4j graph was established as training environment for Dueling Meanwhile, supply paths from point to source were obtained using Cypher language built into Neo4j, forming space that served action space. Secondly, various constraints process, reward and penalty function formulated establish DQN model. Finally, according ε−greedy selection strategy, actions selected interacted with environment, resulting optimal operation sequence. In paper, Python used programming language, TensorFlow open-source software library form deep reinforcement network, Py2neo toolkit complete linkage between python platform Neo4j. We conducted experiments real 79-node system, three scenarios validation. Under scenarios, time required obtain results 2.87 s, 4.37 s 3.45 respectively. For scenario 1 before after transfer, line loss, voltage deviation rate reduced by about 56.0%, 76.0% 55.7%, 2 41.7%, 72.9% 56.7%, 3 13.6%, 47.1% 37.7%, The experimental show trained can quickly accurately derive sequence under different conditions, thereby validating effectiveness of proposed

Language: Английский

Citations

0