Distributed deep reinforcement learning-based gas supply system coordination management method for solid oxide fuel cell

In order to sustain solid oxide fuel cell (SOFC) net output power and prevent violation of oxygen excess ratio (OER) constraint and fuel utilization (FU) constraint, a data-driven gas supply system coordination management method is proposed. Accordingly, a population evolution-based multi-agent doub...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	Engineering applications of artificial intelligence Ročník 120; s. 105818
Hlavní autoři:	Li, Jiawen, Cui, Haoyang, Jiang, Wei
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	Elsevier Ltd 01.04.2023
Témata:	Distributed deep reinforcement learning Gas supply system coordination management method Population evolution multi-agent double delay deep deterministic policy gradient algorithm Solid oxide fuel cell Population evolution multi-agent double delay deep deterministic policy gradient algorithm Gas supply system coordination management method Distributed deep reinforcement learning Solid oxide fuel cell
ISSN:	0952-1976, 1873-6769
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	In order to sustain solid oxide fuel cell (SOFC) net output power and prevent violation of oxygen excess ratio (OER) constraint and fuel utilization (FU) constraint, a data-driven gas supply system coordination management method is proposed. Accordingly, a population evolution-based multi-agent double delay deep deterministic policy gradient (PE-MA4DPG) algorithm is introduced. The artificial intelligence design of the algorithm is guided by the concepts of imitation learning and curriculum learning, whereby different agents of different combinations are trained in different environments, thus improving the robustness of the coordination strategy. In this algorithm, the hydrogen controller and the air controller are treated as two agents. The centralized training enables agents with different objectives to coordinate with each other. The effectiveness of the proposed algorithm is demonstrated in three experiments, wherein the proposed algorithm is compared with a group of existing algorithms. •A 5kW SOFC gas supply system model considering various operating parameters is proposed.•A data-driven gas supply system coordination management method is proposed.•A novel large-scale deep reinforcement learning algorithm is proposed for this method.•The PE-MA4DPG algorithm proposed is characterized by superior robustness.•The proposed algorithm can guarantee better SOFC stability, performance, and efficiency, whilst satisfying the SOFC constraints.
ISSN:	0952-1976 1873-6769
DOI:	10.1016/j.engappai.2023.105818