In an offshore environment, the design of the wind energy conversion systems (WECS) requires taking into account not only efficiency and reliability but also size and weight, as expensive platforms must be placed to support each component. A contribution of around 15% in active volume and around 10% in active weight is normally reported for power electronics converters in wind turbine applications (
Blaabjerg et al., 2006). Therefore power density is also a performance index of paramount importance, especially when most of the electrical power conversion components are going to be located in the nacelle or tower of the wind turbine (WT).
The efficiency (
η) of an electrical system is the ratio of power output and power input (
Pin) (
Eq. [9.1]), and the power output is the difference between power input and the total losses from the input to output stages of the WECS, including power semiconductors and passive elements like inductors and capacitors.
On the other hand, the power density (
ρ) defined by
Eq. [9.2] characterizes the degree of compactness of a WECS.
ρ depends on the total converter volume and power losses of the system. The converter volume (Vol
Total) is the summation of the individual volumes Vol
(i) of the components, and the utilization of the Vol
Total by active parts is characterized by the volume utilization factor
CPV, which has typical values between 0.5 and 0.7 (
Kolar et al., 2010).
The WECS should be located in the nacelle, tower or pillar of the WT, so the weight of the converter is also relevant in order to compare different designs. The power-to-mass ratio (
γ), defined by
Eq. [9.3], indicates the level of heaviness of the WECS. The total converter active mass (Mass
Total) is calculated via summation of the individual masses (Mass
(i)) of the components.
A base topology known as the two-level voltage source converter (2L-VSC) has been considered in order to illustrate the evaluation procedure. The nominal η, ρ and γ are obtained for a set of design parameters and constraints. Even more, the η–ρ Pareto-front and the ρ–γ Pareto-front are considered in order to compare different parameters of design.