Coupling Smoothed Particle Hydrodynamics with Multi-Agent Deep Reinforcement Learning for Cooperative Control of Point Absorbers

TL;DR

This paper introduces a novel framework combining Smoothed Particle Hydrodynamics and multi-agent deep reinforcement learning to optimize the control of wave energy converters, significantly improving energy capture efficiency.

Contribution

It develops a GPU-accelerated multi-agent control platform integrating high-fidelity hydrodynamics with reinforcement learning for wave energy devices.

Findings

Energy capture increased by over 21% with learned control policies.

The platform enables real-time, large-scale 3D simulations for wave energy systems.

The approach is validated under various wave conditions, showing robustness.

Abstract

Wave Energy Converters, particularly point absorbers, have emerged as one of the most promising technologies for harvesting ocean wave energy. Nevertheless, achieving high conversion efficiency remains challenging due to the inherently complex and nonlinear interactions between incident waves and device motion dynamics. This study develops an optimal adaptive damping control model for the power take-off (PTO) system by coupling Smoothed Particle Hydrodynamics (SPH) with multi-agent deep reinforcement learning. The proposed framework enables real-time communication between high-fidelity SPH simulations and intelligent control agents that learn coordinated policies to maximise energy capture. In each training episode, the SPH-based environment provides instantaneous hydrodynamic states to the agents, which output continuous damping actions and receive rewards reflecting power absorption. The Multi-Agent Soft Actor Critic algorithm is employed within a centralised-training and decentralised-execution scheme to ensure stable learning in continuous, multi-body systems. The entire platform is implemented in a unified GPU-accelerated C++ environment, allowing long-horizon training and large-scale three-dimensional simulations. The approach is validated through a series of two-dimensional and three-dimensional benchmark cases under regular and irregular wave conditions. Compared with constant PTO damping, the learned control policy increases overall energy capture by 23.8% and 21.5%, respectively, demonstrating the strong potential of intelligent control for improving the performance of wave energy converter arrays. The developed three-dimensional GPU-accelerated multi-agent platform in computational hydrodynamics, is extendable to other fluid-structure interaction engineering problem that require real-time, multi-body coordinated control.