OCEAN - Open-access validated models from experimental data of wave energy system arrays

One of the key steps towards economic feasibility of wave energy conversion technology concerns scaling up to farms of multiple devices, in the attempt to reduce installation costs by sharing infrastructure, and a consequent drop in levelised cost of energy. Moreover, whenever wave energy systems are deployed in proximity (in so-called arrays), the exploitation of the hydrodynamic interactions between single devices is fully enabled, potentially increasing the final energy outcome. To achieve this in real (operational) time, the employed energy-maximising control strategies require control-oriented array models, able to efficiently describe the dynamics of these interconnected systems in a representative fashion. This can be, nonetheless, a difficult task when considering first principles alone, under small motion assumptions, for modelling purposes. Recognising the uncertainty associated to array numerical models obtained from the linearisation of simplified system equations around their equilibria, this dataset provides models of several array configurations identified following a frequency domain approach on the basis of experimental data. Tailored tests on laboratory-scale devices have been designed and conducted in the Aalborg University (Denmark) wave tank facility, with the purpose of performing representative system identification of the wave energy systems arrays. The obtained models are validated on different representative sea states configurations, in controlled and uncontrolled motion operational conditions. The validation results are fully discussed and analysed in terms of standard error measures and time lag, while the obtained models are made freely accessible via this repository (named OCEAN), in the attempt to openly provide validated models for different array configurations. The discussion, and validation analysis can be found in the paper by the same authors "Data-based modelling of arrays of wave energy systems: Experimental tests, models, and validation", 2024, Control Engineering Practice, doi: 10.1016/j.conengprac.2024.105949 .

波浪能转换技术（wave energy conversion technology）实现经济可行性的核心路径之一，在于将单台设备扩容为多设备阵列农场，通过共享基础设施降低安装成本，进而实现平准化度电成本（levelised cost of energy）的压降。此外，当波浪能系统以近距集群部署（即所谓的阵列）时，可充分利用单台设备间的水动力相互作用，有望提升最终能源产出。为实现运行工况下的实时优化控制，所采用的能量最大化控制策略需要面向控制的阵列模型，能够以具有代表性的方式高效刻画这类互联系统的动态特性。然而，若仅基于第一性原理并结合小幅度运动假设开展建模，该任务极具挑战性。鉴于基于简化系统方程在平衡点处线性化得到的阵列数值模型存在固有不确定性，本数据集提供了基于实验数据、通过频域方法辨识得到的多种阵列配置模型。研究团队在丹麦奥尔堡大学（Aalborg University）的波浪水池试验设施中，针对实验室规模的波浪能设备设计并开展了定制化试验，以实现波浪能系统阵列的代表性系统辨识。所得到的模型在受控与非受控运动的运行工况下，针对多种典型海况配置完成了验证。验证结果从标准误差指标与时滞两方面进行了充分讨论与分析；同时，本数据集（命名为OCEAN）公开了所有可获取的模型，旨在为不同阵列配置提供经过验证的公开模型资源。相关讨论与验证分析可参阅同一作者团队2024年发表于《Control Engineering Practice》的论文《基于数据的波浪能系统阵列建模：试验、模型与验证（Data-based modelling of arrays of wave energy systems: Experimental tests, models, and validation）》，DOI: 10.1016/j.conengprac.2024.105949。