The power output from wave energy converters (WECs) may be increased by controlling the oscillation in order to approach an optimum interaction between the WEC and the incident wave. Optimally controlled WECs, designed to operate at full capacity a rather large fraction of their lifetime, may improve the economic prospects for wave power significantly. An upper bound is given to the ratio between the converted power from a given wave, and the geometrical volume of the converter. One control strategy for maximising the converted power is based on measuring the incident wave, whereas another strategy utilises measurement of the WEC's own oscillation as input to the controller. In either case the measured quantity has to be predicted some seconds into the future because of non-causal control functions.
Most of the proposed devices for conversion of wave energy are oscillating systems with a frequency-dependent response showing the phenomenon of resonance. At resonance, that is when the wave period agrees with the natural (eigen) period, the fraction of converted energy attains maximum. With wave periods off resonance the conversion is less powerful, in particular so if the resonance bandwidth is narrow. Wave-energy converters (WECs) of large horizontal extension, socalled "terminators" and "attenuators", have rather broad bandwidths, whereas "point absorbers", for which the extension is very small compared to the predominant wavelength, have rather narrow bandwidth. On the other hand, an advantage with point absorbers is that the smaller the structural volume of the converter is, the larger is the ratio between the potentially converted power and the mentioned volume (Budal & Falnes, 1980). Hence, for a point absorber it is imperative that means are provided for optimum control of the oscillatory motion in order to achieve a maximum of power conversion.
