IEEE Electrification Magazine - March 2020 - 64

elements from the wire harness that have only mechanical purposes (such as fixings, wire guides, and so on),
specifying the module and family for each one.

an entire procedure for successfully performing a reliable
power-flow simulation for onboard EDSs, taking into
account the available automotive-factory data formats.

Computational Tools for the Simulation
of Onboard EDSs

Tailored Power-Flow Simulation for Smart
Vehicular Electrical Networks

Simulation in automotive engineering was first used in
the domains of aerodynamics, vehicle collision, and
engine multibody dynamics. In recent years, these efforts
have expanded to areas such as artificial vision, virtual
reality, network communications, energy balance, electrical traction, and charging systems. However, as a consequence of the largely dispersed factory information and
sharp intricacy of vehicular EDSs, only a few commercial
platforms and even fewer academic efforts have proposed computational tools able to perform a detailed
simulation of these systems, considering the necessary
manufacturing data preprocessing, complex wiring, and
different electrical components forming the harnesses.
Commercial tools (such as EB Cable, Vesys, and EPLAN
Harness proD) permit only design duties but are not suited for simulation. Meanwhile, platforms (e.g., Power Net
Simulation by Bosch and Simulink from MATLAB) mainly
allow simulation of the overall energy balance, employing general models for the battery, alternator, and loads,
all operating under selected scenarios and driving conditions. Finally, software tools (for instance, Harness Studio,
Siemens Solid Edge, and Saber RD) have typically assisted in harness design but, recently, have also included
specific add-ons intended for simulation of the EDS.
However, not much information is available regarding
the features, models, and methodology employed by
those add-ons, which, in most cases, remain unknown
by the user and, thus, do not allow establishment of a
benchmarking between them. The next section details

QT
(User Tailored)

Data Preprocessing
The data containers with the most electrical details from
the harnesses are the WL and BOM files, which, in turn,
were formed from the KBL files, as previously explained.
However, as those files were primarily structured for
exchanging manufacturing information, they were not
conceived for electrical simulation. In addition, they do not
contain all of the necessary data, such as time-current
characteristics of fuses, the temperature class of each
cable, or pin-out information of the consumers, among
others. Therefore, to create a proper data container intended for power-flow studies, in addition to including the WL
and BOM archives as inputs, a customized multidimensional data structure has been created, referred to as QT.
This is an Excel file having condensed information from
the automatically generated BOM and WL but complemented with the aforementioned missing data, which
have been taken from other dispersed databases.
As the XML files were originally designed to contain all
of the possible modules that exist for a given family within
a car model, the newly generated data container provides,
by default, the full module information. In practice, a specific vehicle is described by the selection of only one module per family; thus, its simulation requires filtering of the
data container. To achieve this, the user introduces a list of
modules (one per family) in the form of a table, corresponding to the car configuration under study. Once this
table is introduced, together with the data container, a preprocessing algorithm removes the
unnecessary information. Finally,
the filtered data are used forward as
input for the power-flow solver. Figure 5 sketches the overall data preprocessing process.

Tailored Power-Flow Simulation
+
KBL File
List of Modules
(User Introduced)
BOM

Filtering
Data Container Suitable
for Electrical Simulation

Figure 5. The overall data preprocessing process.

I E E E E l e c t r i f i cati o n M agaz ine / MARCH 2020

Traditionally, the algorithm most
used for solving power-flow studies
in conventional electrical systems
relies on the Newton-Raphson criteria. However, the algorithm chosen for vehicular EDS features is
the method known as backward/
forward sweep (BFS). This method
uses the Kirchhoff's current and
voltage laws (Kirchhoff's current
law and Kirchhoff's voltage law)
iteratively. It was selected mainly
for the following reasons.

IEEE Electrification Magazine - March 2020

Table of Contents for the Digital Edition of IEEE Electrification Magazine - March 2020