Abstract
Over the past decade, the controls design thinking within the Wave Energy Conversion (WEC) community has evolved from discrete techniques that have been shown to improve performance for specific device topologies, to generic controls approaches where losses and constraints in the system can be handled efficiently when maximizing power. This constrained optimization is an important feature for optimizing realistic and cost-effective WEC device topologies.
Over the past 5-years, Re Vision Consulting has developed optimal control systems for six different WEC topologies in collaboration with device developers. In the process we have developed a set of capabilities to enable optimal constrained control in any WEC device. This text summarizes some of the key lessons learned.
Most texts on controls are highly technical and require a deep background in controls theory and/or mathematics. The motivation for this white paper is to explain the key concepts in plain English, so that they can be understood by a wide audience. Optimal control needs to be at the center of any WEC device development process, and it is important for the entire design team to have a solid understanding of the challenges and opportunities. For the readers wanting to dive deeper into the controls approaches, references 1-7 listed provide some good background and application examples on the numerical algorithms referred to in this text.
The control system affects power capture, structural loads, and power take off (PTO) design. To achieve true economic optimality in a WEC system, optimal control needs to be considered as part of the design trade-off space. Simply adding controls to an existing WEC device topology will often not yield significant performance improvements, because the PTO may not be able to provide the capabilities needed to improve performance or the device envelope is not optimized to take advantage of advanced controls. Device and PTO attributes can only be optimized if their cost-drivers and their performance impacts are quantified. Constrained optimal control is a key tool in this optimization process.
There are two main controls approaches used within the WEC development community: (1) Causal controllers, which only leverage on-board measurements as feedback to the control law, and (2) Noncausal controllers that leverage a wave-excitation force forecast – typically implemented using Model Predictive Control (MPC). MPC with an accurate wave prediction remains the Tesla of controls approaches, because it has the most extensive set of capabilities that make it useful across the entire range of WEC and PTO topologies.
While causal control with acceptable performance has been demonstrated on a limited set of device topologies, it remains to be explored to what extent causal control laws can approximate the performance of MPC with a wave-forecast. It is important that controls performance does not only relate to energy capture, but also the capabilities of the algorithm to accommodate realistic devicespecific constraints such as PTO force, velocity, acceleration, and powerflow.
In general, our view is that during the device development process it is important to understand the fundamental upper limits of a particular configuration and use sensitivity studies to understand the trade-offs involved in arriving at an economically optimal configuration. MPC can serve as an important tool to explore this trade-off space, because it allows us to establish upper limits of constrained systems, which is not easily done using analytical methods. Once these trade-offs are fully understood, the designer can turn to the evaluation of simpler control strategies to further reduce complexity in the system. That could include the elimination of the wave-prediction required.
It should be pointed out that the cost of predicting ocean waves (a requirement for effective MPC implementation) is very small compared to the cost of the device itself at commercial scales. A simple 1% improvement in power capture would pay for the cost of the wave prediction system many times over in a wave farm. Causal controls approaches may be useful at smaller scales required for applications within the blue economy such as recharging unmanned vehicles at sea, where the economic calculus is driven by reliability and operational simplicity and not just performance.
This review aims to provide a synthesized overview of the critical challenges and opportunities in the design of control systems for WEC Devices and provide relevant examples. It is based on the experience of optimizing six different WEC topologies over the past five years at Re Vision Consulting. Theory is kept to a minimum to facilitate understanding by a broad audience.