Experimental and Numerical Collaborative Latching Control of Wave Energy Converter Arrays

Abstract

A challenge while applying latching control on a wave energy converter (WEC) is to find a reliable and robust control strategy working in irregular waves and handling the non-ideal behavior of real WECs. In this paper, a robust and model-free collaborative learning approach for latchable WECs in an array is presented. A machine learning algorithm with a shallow artificial neural network (ANN) is used to find optimal latching times. The applied strategy is compared to a latching time that is linearly correlated with the mean wave period: It is remarkable that the ANN-based WEC achieved a similar power absorption as the WEC applying a linear latching time, by applying only two different latching times. The strategy was tested in a numerical simulation, where for some sea states it absorbed more than twice the power compared to the uncontrolled WEC and over 30% more power than a WEC with constant latching. In wave tank tests with a 1:10 physical scale model the advantage decreased to +3% compared to the best tested constant latching WEC, which is explained by the lower advantage of the latching strategy caused by the non-ideal latching of the physical power take-off model.

Numerical and Experimental Investigation on a Moonpool-Buoy Wave Energy Converter

“…Thomas et al [31] applied a shallow artificial neural network (ANN) which is a kind of machine learning language to obtain optimal working times. Wu et al [32] put forward a new computational fluid dynamic method to predict the hydrodynamic characteristic of the Duck WEC, the results of which agree well with the experimental results. Kong et al [33] adopted a semi-analytical approach based on the potential flow to assess the wave energy efficiency of the moonpool platform WEC in the journal of Energies.…”

Advances and Challenges in Wave Energy Park Optimization—A Review

“…Real-time simulation was carried out by Balitsky et al (2014) to model the optimal configuration for a six-body array using measured wave data from the west coast of Ireland. Control models have been introduced both in experiments and numerical modeling of arrays Li and Belmont, 2014;Mercadé Ruiz et al, 2017;Nader et al, 2017;Thomas et al, 2018a), but it is beyond the scope of the current paper to go into the details of advanced generator simulations or wave-to-wire models. Instead, we refer to reviews such as Penalba and Ringwood (2016) and Wang et al (2018) and focus on the most commonly used PTO models used in wave energy park optimization.…”

“…Arrays of up to six point-absorber WECs, each moving in six degrees of freedom were carried out both at the Australian Maritime College by Nader et al (2017) and at the COAST laboratory at the University of Plymouth, UK, by Thomas et al (2018a) and Giassi et al (2019b). In the former experiment, the surface elevation was tracked by videogrammic measurements, and the experimentally measured interaction factor was presented for 1-2 floats moving in heave and surge (Nader et al, 2017).…”

“…In the latter, both a linear damping and an advanced control algorithm based on machine learning and artificial neural networks were used as PTO systems (Thomas et al, 2018b). The collaborative control algorithm required no previous knowledge of the incident waves but still increased the energy absorption through communication between the WECs (Thomas et al, 2018a), and the performance of the WECs with linear damping was compared for three array layouts by Giassi et al (2019b).…”

Advection-Based Coordinated Control for Wave-Energy Converter Array

“…Because the optimal harvesting condition of a WEC is a region defined by the amount of capture width information available, the MPPT control was modified to be the maximum capture width tracking (MCWT) control to ensure robust during the rapidly changing in ocean wave in [6]. In [7], a robust model-free collaborative learning approach was presented for a latchable WEC array. Furthermore, the machine-learning algorithm with a shallow artificial neural network (ANN) was applied to obtain the optimal latching times and absorb more power [8].…”