IEEE Electrification Magazine - December 2019 - 107

impartial data comparison, thus optimizing the design
choice toward the fulfillment of the main design drivers
(such as pay load increase). Both the power system architecture (e.g., MVac power system, MVdc power system,
and MVac/MVdc hybrid system) and the main bus voltage
value (e.g., 6.6 or 11 kV) can be used as an integrated design
parameter, for example, for properly comparing the consequent variation in weights/volumes of distribution cabinets/electrical cables.
To prove the capability of the CSI software approach, the
refitting of a vessel's IPES section is presented as an example. In particular, the goal is to replace the actual LVac combat
system section with a new LVdc one to assess differences
in terms of power system weight and volume through the
3D parametric software. Thus, the given IPES LVac 440-V,
60-Hz section [illustrated in Figure 10(a)] is refitted with a
proposed LVdc 800-V section, as shown in Figure 10(b). The
main difference is in the transition between the MV and
the LV side. For the LVac section, a rotating converter is used
since the MV side has a 50-Hz system frequency, while for
the LVdc section, three static ac/dc devices of 750-kW each
replace the electromechanical converter.
Three power electronics components were used due
to both the market availability of the chosen device
and for purposes of redundancy. There is still the possibility to feed the emergency loads, not only with an
uninterruptible power supply but also with one of the
three converters, in case a fault occurs to the other
two, thus increasing the reliability of the system. Furthermore, the LV switchboard in the main vertical zone
1 (MVZ1) was duplicated for the newly designed dc
section to enhance the power system survivability, as
shown in the one-line diagram [Figure 10(b)]. Both
solutions are implemented in the CSI environment to
compare the different distribution systems. The MVZ1
LVac 440-V, 60-Hz switchboard is presented in Figure 11.
The cableways from other MVZs are depicted in yellow.
The LVac and LVdc section has been implemented to
the locals' switchboard only for the MVZ2 since the
main part of the combat system loads is located in that
MVZ. Finally, in Figure 12, the new proposed LVdc section

120

is shown. The cableways from other MVZs are portrayed
in green.
The CSI software is not only a CAD tool but also a realscale parametric model and thus is able to compare different power system layouts in terms of the weight and
volume of the electrical equipment and total weight and
length of the considered power system cables. Concerning
the former comparison, the output data of the parametric
CSI software are reported in Figure 13; the dc solution has
better performance both in weight and volume related to
the ac one. The weight of the proposed dc solution is only
69.9% of the ac one, and it assures a 16.7% volume reduction regarding the electrical equipment.
The consistent reductions of weight and volume are
mainly due to the removal of the electric machines providing frequency conversion. Concerning the latter comparison, the output data of the total cable weight and length
assessment are reported in Figure 14. The amounts of
weight and volume lowering are very large. The weight
reduction is more closely related to the chosen reference
voltage for the LVdc system, since the higher the voltage, the
lower the current for the same rated power to supply. Lowering the current is intended to decrease the copper/aluminum cross-section of the power cable and thus its weight.
Instead, the decreasing of the total cable's length is mainly
due to the reduction of the number of cables passing
through the MVZs. It must be noted that one of the main
cost items in the total ship cost production (about 9%) is
the staff-hours necessary to install the cables in the cableways around 9%.
In conclusion, the CSI software is a powerful means to
evaluate strengths and weaknesses of different power system design solutions. The onboard power system distribution can be selected based on objective and impartial
data comparison, thus optimizing the design choice
toward the fulfillment of the main design drivers. Finally,
exploiting the CSI software capabilities allows us to know
the cable lengths fairly accurately in the preliminary
design phase instead of in the functional design phase
(Figure 7) as well as the layout of cables and their
onboard paths (Figure 9). These data are fundamental to

Electrical Equipment

80
(%)

100

80
(%)

100
60

Weight (Kg)
ac Solution

Volume (m3)
dc Solution

Figure 13. The output data of the parametric CSI software: electrical
equipment weight and volume in the two compared solutions.

Total Cables

120

Weight (Kg)
ac Solution

Length (m)
dc Solution

Figure 14. The output data of the parametric CSI software: total
IPES cables weight and length in the two compared solutions.

IEEE Elec trific ation Magazine / D EC EM BE R 2 0 1 9

107

IEEE Electrification Magazine - December 2019

Table of Contents for the Digital Edition of IEEE Electrification Magazine - December 2019