An active-controlled heaving plate breakwater trained by an intelligent framework based on deep reinforcement learning

Xie, Yulin; Zhao, Xizeng; Luo, Min

doi:10.1016/j.oceaneng.2021.110357

Cited by 13 publications

(1 citation statement)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…More significantly, Xie and Zhao [54] applied a DRL framework to find an optimal control strategy to suppress tank sloshing with two active-controlled horizontal plates, which showed that the sloshing loads with active controlled baffles could be greatly reduced compared to the ones with fixed baffles. Subsequently, Xie et al [55] built a two-dimensional wave flume with an actively controlled plate to dissipate regular waves with different periods, where the motion strategy of the plate was generated and analyzed by DRL. This was the first time that DRL was applied in the action control of breakwaters for wave dissipation.…”

Section: Introductionmentioning

confidence: 99%

A Model Coupling CFD and DRL: Investigation on Wave Dissipation by Actively Controlled Flat Plate

et al. 2022

View full text Add to dashboard Cite

In order to protect structures along the offshore and off the coast, breakwaters are commonly applied to reduce the influence of waves. Flat plate breakwater is studied frequently due to its great performance near the water surface, however, traditional passive methods such as fixed and floating flat plate breakwaters usually fail to give full play of effective wave dissipation when encountering variable unknown incoming waves. Therefore, this paper develops an interdisciplinary model coupling computational fluid dynamics (CFD) and deep reinforcement learning (DRL) to study the wave dissipation of a submerged movable flat plate breakwater against regular waves. An in-house numerical wave tank (NWT) is built to simulate the fluid-structure interaction between the plate and regular waves. The fluid domain in NWT is regarded as environment while the flat plate breakwater is agent. In addition, the wave dissipation strategy is learned by the artificial neural network (ANN) through the continuous wave-plate interaction. It is excited to find that the coupling model is able to learn the control strategies automatically and exhibits good adaptability to the changes of environment. 13 14 INDEX TERMS Wave dissipation, deep reinforcement learning, flat plate breakwater, active controlling, CFD and DRL. culated, of which the results showed that a suitable geo-49 metrical porosity of the upper plate leads to a low level of 50 uplift wave forces on both plates. Reference [17] investi-51 gated the flow field passing over a submerged horizontal 52 flat plate by means of a free surface potential flow model, 53 a form discretized by the Boundary Element Method (BEM), 54 from which the wave dissipation mechanism of a plate-55 type breakwater and the influence of plate thickness, length 56 and submergence was concluded. Reference [18] presented 57 an integrated CFD+CSM approach to fully simulate the 58 hydroelastic interaction between nonlinear ocean waves and 59 a deformable SHPB device, making two major contributions: 60 (a) the research work demonstrated a valid computer-aided 61 approach that can perform hydroelastic offshore and coastal 62 designs; (b) the researching work demonstrated a hydroelastic 63 design methodology for breakwaters for the first time and 64 showed that the deformation of the structure has a significant 65 effect on the wave-damping performance. References [19] 66 and [20] conducted an experiment to research the wave dissi-67 pation performance of a twin-plate considering various wave 68 conditions. References [21] and [22] investigated the interac-69 tion between freak waves based on numerically investigating 70 the interaction characteristics of horizontal deck structures 71 and freak waves, which revealed the corresponding phenom-72 ena, wave loads, and structural response. Reference [23] 73 further studied the interaction between Peregrine breather-74 based freak waves and twin-plate breakwaters, including the 75 characteristics of interaction phenomena, transmitted wave 76 surface, wave loads, and st...

show abstract

Section: Introductionmentioning

confidence: 99%

A Model Coupling CFD and DRL: Investigation on Wave Dissipation by Actively Controlled Flat Plate

et al. 2022

View full text Add to dashboard Cite

show abstract

Analysis of Key Disciplinary Parameters in Floating Offshore Wind Turbines with An AI-Based SADA Method

Chen

2022

China Ocean Eng

View full text Add to dashboard Cite

Floating offshore wind turbines (FOWTs) are a promising offshore renewable energy harvesting facility but requesting multiple-disciplinary analysis for their dynamic performance predictions. However, engineering-fidelity level tools and the empirical parameters pose challenges due to the strong nonlinear coupling effects of FOWTs. A novel method, named SADA, was proposed by Chen and Hu (2021) for optimizing the design and dynamic performance prediction of FOWTs in combination with AI technology. In the SADA method, the concept of Key Disciplinary Parameters (KDPs) is also proposed, and it is of crucial importance in the SADA method. The purpose of this paper is to make an in-depth investigation of the characters of KDPs and the internal correlations between different KDPs in the dynamic performance prediction of FOWTs. Firstly, a brief description of SADA is given, and the basin experimental data are used to conduct the training process of SADA. Secondly, categories and boundary conditions of KDPs are introduced. Three types of KDPs are given, and different boundary conditions are used to analyze KDPs. The results show that the wind and current in Environmental KDPs are strongly correlated with the percentage difference of dynamic response rather than that by wave parameters. In general, the optimization results of SADA consider the specific basin environment and the coupling results between different KDPs help the designers further understand the factors that have a more significant impact on the FOWTs system in a specific domain.

show abstract

A simple approach for wave absorbing control of plunger wavemakers using machine learning: Numerical study

Xie

Zhao²,

Liu³

2023

Coastal Engineering

View full text Add to dashboard Cite

An active-controlled heaving plate breakwater trained by an intelligent framework based on deep reinforcement learning

Cited by 13 publications

References 51 publications

A Model Coupling CFD and DRL: Investigation on Wave Dissipation by Actively Controlled Flat Plate

A Model Coupling CFD and DRL: Investigation on Wave Dissipation by Actively Controlled Flat Plate

Analysis of Key Disciplinary Parameters in Floating Offshore Wind Turbines with An AI-Based SADA Method

A simple approach for wave absorbing control of plunger wavemakers using machine learning: Numerical study

Contact Info

Product

Resources

About