Experimental assessment of an unknown-input estimator for a nonlinear wave energy converter

IFAC PAPERSONLINE(2023)

引用 0|浏览0
暂无评分
摘要
In the wave energy field, one of the main challenges towards commercialisation of wave energy devices is the development of suitable control laws, able to maximise the absorbed energy while guaranteeing effective satisfaction of any required physical constraint. However, one of the main characteristics of this optimal control problem is that the system behaviour is strongly influenced by the external (uncontrollable) input arising from the wave source, i.e. the wave excitation, which is often unmeasurable. As such, computation of optimal control solutions for WEC systems requires availability of instantaneous knowledge of the wave excitation, and hence input-unknown estimators are developed within the control loop. State-of-the-art estimation strategies are based on the knowledge of control-oriented linearized models of the system, often neglecting the influence of nonlinear phenomena within the system description. We propose, in this paper, an approach inspired by disturbance observer-based control, able to accommodate well-known hydrodynamic nonlinear effects in the process of estimating the unknown excitation force acting on the device. This strategy, which in contrast to the usually applied estimators does not require an implicit/explicit model of the wave excitation force, is tested on a hardware-in-the-loop facility in different sea state conditions, to realistically assess its performance in terms of estimation error and delay. The experimental appraisal show satisfactory results in terms of normalized root mean square error and average delay, which, together with the simplicity of the method, positions the proposed strategy as a promising candidate for hardware implementations in real environments. Copyright (c) 2023 The Authors.
更多
查看译文
关键词
Wave energy,unknown-input estimation,nonlinear systems,experimental assessment,energy-maximising control,disturbance observer-based control
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要